The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
[SUBCOMMITTEE PRINT] COMPENDIUM ON MONETARY POLICY GUIDELINES AND FEDERAL RESERVE STRUCTURE PURSUANT TO H.R. 11 SUBCOMMITTEE ON DOMESTIC FINANCE OF THE COMMITTEE ON BANKING AND CURRENCY HOUSE OF REPRESENTATIVES 90th Congress, Second Session D E C E M B E R 1968 Printed for the use of the Committee on Banking and Currency U.S. GOVERNMENT PRINTING OFFICE 21-570 WASHINGTON : 1968 C O M M I T T E E ON B A N K I N G A N D CURRENCY WRIGHT PATMAN, Texas, Chairman WILLIAM A. BARRETT, Pennsylvania LEONOR K. SULLIVAN, Missouri HENRY S. REUSS, Wisconsin THOMAS L. ASHLEY, Ohio WILLIAM S. MOOBHEAD, Pennsylvania ROBERT G. STEPHENS, JR., Georgia FERNAND J. ST GERMAIN, Rhode Island HENRY B. GONZALEZ, Texas JOSEPH G. MINISH, New Jersey RICHARD T. HANNA, California TOM S. GETTYS, South Carolina FRANK ANNUNZIO, Illinois THOMAS M. REES, California JONATHAN B. BINGHAM, New York NICK GALIFIANAKIS, North Carolina TOM BEVIL, Alabama LESTER L. WOLFF, New York CHARLES H. GRIFFIN, Mississippi WILLIAM B. WIDNALL, New Jersey PAUL A. FINO, New York FLORENCE P. DWYER, New Jersey SEYMOUR HALPERN, New York W. E. (BILL) BROCK, Tennessee DEL CLAWSON, California ALBERT W. JOHNSON, Pennsylvania J. WILLIAM STANTON, Ohio CHESTER L. MIZE, Kansas SHERMAN P. LLOYD, Utah BENJAMIN B. BLACKBURN, Georgia GARRY BROWN, Michigan LAWRENCE G. WILLIAMS, Pennsylvania CHALMERS P. WYLIE, Ohio $»ADL NELSON, Clerk and Staff Director CURTIS A . PRINS, Chief BE NET D . GELLMAN, JAMES F . DOHERTY, ROBERT E . WEINTRAUB, Investigator Counsel Counsel Economist ORMAN S. FINK, Minority Staff Member SUBCOMMITTEE ON DOMESTIC FINANCE WRIGHT PATMAN, Texas, Chairman JOSEPH G. MINISH, New Jersey WILLIAM B. WIDNALL, New Jersey RICHARD T. HANNA, California W. E. (BILL) BROCK, Tennessee TOM S. GETTYS, South Carolina J. WILLIAM STANTON, Ohio FRANK ANNUNZIO, Illinois SHERMAN P. LLOYD, Utah THOMAS M. REES, California BENJAMIN B. BLACKBURN, Georgia NICK GALIFIANAKIS, North Carolina LESTER L. WOLFF, New York (II) LETTER OF TRANSMITTAL Transmitted herewith for the use of the Subcommittee on Domestic Finance and other members of the Banking and Currency Committee and the Congress are the replies received from the Federal Reserve, the Treasury, the Council of Economic Advisers and 71 academic, bank and research monetary economists in response to a questionnaire sent out on July 9, 1968. Respondents were asked to express their opinions on questions pertaining to H.R. 11, a bill "To make the Federal Reserve System responsive to the best interests of the people of the United States and to improve the coordination of monetary, fiscal, and economic policy." (A reproduction of H.R. 11 introduced on January 10, 1967, in the first session of the 90th Congress is found on pages 1-5.) Questionnaires were sent to the seven members of the Board of Governors of the Federal Reserve System and the 12 Reserve bank presidents, the Secretary of the Treasury, the members of the President's Council of Economic Advisers and 125 prominent academic, bank and research monetary economists; representing all schools of thought on the fundamental question of how to manage the Nation's money and credit. Replies are printed verbatim, as they were received, with only minor editorial changes. In addition, the staff letter of transmittal which follows contains a question-by-question summary of the replies and analysis of the response of the Federal Reserve whose 19 highest officials replied as one man to the questions asked. Neither the staff analysis nor any of the responses to the questionnaire which follow herein necessarily represent the views of any member of the subcommittee. I am sure that all members of the subcommittee join me in expressing gratitude and appreciation to those who took the time to think about our questions and submit replies thereby giving us the benefit of their valuable experience and training. Reform of our monetary policy system is needed now. Our economy has been in the expansion phase of the business cycle now for nearly 8 jrears. The expansion, however, has been marred by a minirecession in late 1966 and early 1967 and by inflations of prices and interest rates first in late 1965 and early 1966 and more recently in 1968. Monetary policy has played an important role in all these movements. Favorable monetary trends contributed substantially, to the powerful upsurge which has dominated our economic performance since February 1961, and perverse monetary developments contributed to the recent short-lived minirecession and low-level inflationary episodes that have flawed this performance. We must realize that continuation of the upswing and minimization of future destabilizing developments, whether in the direction of recession or inflation, depend strategically on our achievement in future years of favorable monetary trends and avoidance of perverse departures from these trends. (in) IV Monetary trends emerge primarily from policy actions of our monetary authorities, that is, the policy making officers of the Federal Reserve System. Nothing can be plainer, therefore, than the need to structure the Federal Reserve and define its role in the context of the totality of the Government's economic policies so that we are assured of monetary trends favorable to stable- economic growth and avoid destabilizing monetary developments. H.R. 11 was conceived for this purpose. I believe that this compendium furnishes an indisputable basis for reform of the Federal Reserve System and amendment of the Employment Act, essentially as provided for by H.R. 11. H.R. 11 provides for: (1) Reducing the number of members of the Federal Reserve Board to 5 and their term of office to 5 years and making the term of the Chairman of the Board coterminous with that of the President of the United States; (2) Vesting all power to direct open market operations in the Federal Reserve Board and coordinating such operations with the economic programs and policies of the President pursuant to the Employment Act; (3) Requiring that the President, in submitting his annual economic report pursuant to the Employment Act, shall include, along with his recommendations on fiscal and debt-management policy, guidelines concerning monetary policy including the growth of the money supply as defined by him; (4^ Retiring Federal Reserve bank stock; (5) Annual audit of the Federal Reserve Board and banks and their branches; and (6) Funds to operate the Federal Reserve System to be appropriated by the Congress of the United States. Still another reform which the compendium demonstrates we must adopt is the transfer of all but a small fraction of the Federal Reserve's portfolio of U.S. Government securities—nowr totaling more than $53 billion—to the Treasury. As long as the Federal Reserve holds these securities many persons, as the compendium shows, wTill fail to see that open market purchases of U.S. Government securities increase the public's financial wealth or net worth, and thereby cause increased spending and activity in the economy at large. The keystone of the matter is that open market purchases reduce the public's holdings of the Federal debt and liability for it by equal amounts, and thus the net effect is to increase net worth by the amount of currency and reserves the Federal Reserve uses to pay for its purchases of U.S. Government securities. What many do not seem to understand is that taxpayers are liable for U.S. Government securities and the interest thereon only so long as these securities are held outside the Federal Reserve. When the Federal Reserve uses the Nation's money and credit to buy U.S. Government securities it retires them just as surely as if the Treasury had bought them. Taxpayer liability ceases. The law must be changed so that no one fails to recognize this. Transfer of all but a small fraction of the Federal Reserve's portfolio to the Treasury and automatic transfer of all new purchases is urgently needed therefore. With these transfers there will be no failure to see that open market operations directly change the public's financial wealth or net worth, and thereby change spending and economic activity. And seeing this will make it V possible at long last- to develop a viable monetary policy—a policy tuned in to the realities of monetary and economic processes. In transmitting this compendium to the subcommittee it is my hope that it will be read and discussed not only by members of the Banking and Currency Committee but also by the entire Congress and the general public as well. The questionnaire and analysis which follows was done under the immediate supervision of Dr. Robert E. Weintraub, professor of economics at the University of California at Santa Barbara. Sincerely yours, WRIGHT PATMAN, Chairman, House Committee on Banking crnd Currency. CONTENTS Page Statement of Respondents: Letter of transmittal Text of H.R. 11 Summary and analysis Gov. William McC. Martin, Chairman, for the members of the Board of Governors and the Reserve bank presidents of the Federal Reserve System Honorable Henry H. Fowler, Secretary of the Treasury Members of the Council of Economic Advisers—Dr. Arthur Okun (Chairman), Dr. Merton J. Peck, Dr. Warren Smith Dr. E. Sherman Adams, First National City Bank Prof. Carl T. Arlt, University of Illinois Prof. Joseph Aschheim, George Washington University Prof. George L. Bach, Stanford University Prof. Martin Bronfenbrenner, Carnegie-Mellon University Prof. Karl Brunner, Ohio State University Prof. Meyer L. Burstein, Warwick University and Aspen, Colo Prof. Phillip Cagan, Columbia University Dr. Gregory C. Chow, Thomas J. Watson Research Center, I B M Prof. Carl F. Christ, Johns Hopkins University Prof. Jacob Cohen, University of Pittsburgh Prof. Robert L. Crouch, University of California, Santa Barbara Prof. John M. Culbertson, University of Wisconsin Prof. Paul Davidson, Rutgers University Prof. William G. Dewald, Ohio State University Prof. James S. Earley, University of California, Riverside Prof. Otto Eckstein, Harvard University Prof. David I. Fand, Wayne State University Prof. William Fellner, Yale University Prof. Leo Fishman, West Virginia University Prof. William J. Frazer, Jr., University of Florida Prof. Milton Friedman, University of Chicago Prof. Gary Fromm, The Brookings Institution Dr. Tilford C. Gaines, Manufacturers Hanover Trust Co Prof. William I. Greenwald, CCNY Prof. Herschel I. Grossman, Brown University Prof. Seymour E. Harris, Emeritus, Harvard University and University of California, San Diego Prof. Lowell Harriss, Columbia University Dr. E. C. Harwood, American Institute for Economic Research Dr. Gabriel Hauge, Manufacturers Hanover Trust Co Prof. Thomas M. Havrilesky, University of Maryland Prof. Donald D. Hester, University of Wisconsin Dr. Walter E. Hoadley, Bank of America Prof. George Horwich, Purdue University Prof. William R . Hosek, University of New Hampshire Prof. Dudley W. Johnson, University of Washington, Seattle Prof. Edward J. Kane, Boston College Prof. Norman F. Keiser, San Jose State College Prof. Raymond P. Kent, Notre Dame University Dr. Leon H. Keyserling, Washington, D.C Prof. Henry A. Latane, University of North Carolina Prof. Axel Leijonhufvud, UCLA Dr. Michael E. Levy, The National Industrial Conference Board Prof. Dudley G. Luckett, Iowa State University Prof. George Macesich, Florida State University (VII) III 1 7 29 55 70 85 87 91 94 98 100 104 106 106 109 1^5 119 127 130 142 147 151 154 158 160 166 203 227 227 235 236 242 250 264 272 273 283 291 293 304 313 338 344 356 364 408 409 421 431 437 VIII Dr. Carl H. Madden, Chamber of Commerce of the United States Prof. Thomas Mayer, University of California, Davis Prof. Paul W. McCracken, University of Michigan Prof. Stephen L. McDonald, University of Texas Prof. Jacques Melitz, Tulane University Prof. Allan H. Meltzer, Carnegie-Mellon University Prof. Hyman P. Minsky, Washington University, St. Louis Prof. George R . Morrison, University of California, San Diego Prof. Walter A. Morton, University of Wisconsin Dr. Guy E. Noyes, Morgan Guaranty Trust Co Prof. Boris P. Pesek, University of Wisconsin Prof. Howard N. Ross, CUNY, Bernard M. Baruch College Prof. Robert H. Scott, University of Washington, Seattle Dr. Beryl Sprenkel, Harris Trust and Savings Bank Prof. Roland Stucki, University of Utah Prof. Ronald L. Teigen, University of Michigan Prof. Earl A. Thompson, UCLA Prof. Richard S. Thorn, University of Pittsburgh Jerry Voorhis, Cooperative Foundation & National Association of Housing Cooperatives, Chicago, 111 Dr. Charls E. Walker, American Bankers Association Staff Report on, "The Federal Reserve System After 50 Years," parts II and IV Prof. Henry C. Wallich, Yale University Dr. Clark Warburton, McLean, Va Prof. Sydney Weintraub, University of Pennsylvania Prof. C. R. Whittlesey, Emeritus, University of Pennsylvania Dr. Frazar B. Wilde, Chairman Emeritus, Connecticut General Life Insurance Co Dr. Leland Yeager, University of Virginia Page 457 460 472 474 479 488 491 493 501 502 508 520 526 534 546 548 553 559 573 574 593 630 634 644 648 650 651 TEXT OF H.R. 11 [H.R. 11, 90th Cong., first sess.], A BILL To make the Federal Reserve System responsive to the best interests of the people of the United States and to improve the coordination of monetary, fiscal, and economic policy It provides f o r : ( a ) Retiring Federal Reserve bank stock; (b) Coordinating Federal Reserve bank policies and programs with those of the President of the United States in keeping with the provisions of the Employment Act of 1946; ( c ) Reducing the number and term of office of members of the Federal Reserve Board; (d) Making the term of Chairman of the Board coterminous with that of the President of the United States; (e) An audit for each fiscal year of the Federal Reserve Board and the Federal Reserve banks and their branches by the Comptroller General of the United States; ( f ) Funds to operate the Federal Reserve System to be appropriated by the Congress of the United States. Be it enacted "by the Senate and House of Representatives of the United States of America in Congress assembled, RETIREMENT OF FEDERAL RESERVE BANK STOCK 1. ( a ) The last sentence of the first paragraph of section 2 of the Federal Reserve Act (12 U . S . C . 222) is amended by changing "subscribing and paying for stock" to read "obtaining a certificate of membership". (b) The last sentence of the third paragraph of such section 2 (12 U.S.C. 282) is amended by changing "subscribe to the capital stock of such Federal reserve bank in a sum equal to 6 per centum of the paid-up capital and surplus of such bank, one-sixth of the subscription to be payable on call of the organization committee or of the Board of Governors of the Federal Reserve System, one-sixth within three months and one-sixth within six months thereafter, and the remainder of the subscription, or any part thereof, shall be subject to call when deemed necessary by the Board of Governors of the Federal Reserve System, said payments to be in gold or gold certificates." to read "obtaining a certificate of membership pursuant to the provisions of this Act." (c) The fourth paragraph of such section 2 (12 U.S.C. 502) is repealed. (d) The paragraphs which, prior to the repeal made by subsection (c) of this section, were the eighth, ninth, tenth, eleventh, and twelfth paragraphs of such section 2 (12 U.S.C. 283-286) are repealed, (e) The first sentence of the last paragraph of such section 2 (12 U.S.C. 281) is repealed. SEC. 2. (a) The last sentence of the first paragraph of section 4 of the Federal Reserve Act is amended by changing "a subscription to the capital stock o f " to read "an application for a certificate of membership in". (b) The second paragraph of such section is amended (1) by changing "when the minimum amount of capital stock prescribed by this Act for the organization of any Federal reserve bank shall have been subscribed and allotted," to read "when the organization committee shall deem that a sufficient proportion of eligible banks have applied for membership in a Federal Reserve bank in process of organization,", (2) by striking "the amount of capital stock and the number of shares into which the same is divided,", (3) by changing "subscribed to the capital stock o f " to read "applied for membership in", (4) by striking "and the number of shares subscribed by each", and (5) by changing "subscribed or may thereafter subscribe to the capital stock o f " to read "applied or may thereafter apply for membership in". SECTION (1) 10 (c) The subparagraph numbered "Eighth" of the fourth paragraph of such section 4 (12 U.S.C. 341) is amended by striking "stock". (d) The tenth paragraph of such section 4 is amended by changing "stockholding" to read "member". (e) The second sentence of the twelfth paragraph of such section 4 is amended by changing "subscriptions to the capital stock" to read "applications for membership". SEC. 3. Section 5 of the Federal Reserve Act (12 U.S.C. 287) is amended to read: "CERTIFICATES OF MEMBERSHIP "SEC. 5. (a) The Federal Reserve banks shall have no capital stock. " ( b ) A bank applying for membership in the Federal Reserve System at any time after the date of enactment of this subsection shall submit such application, in accordance with the regulations of the Federal Reserve Board, to the Federal Reserve bank of its district. Such application shall be accompanied by a membership fee of $10, which shall not be refundable unless such application is disapproved or withdrawn before approval. " ( c ) Upon the approval of an application submitted pursuant to subsection (b) of this section, the Federal Reserve bank shall issue to the applicant a certificate attesting the membership of the applicant in such Federal Reserve bank and in the Federal Reserve System. " ( d ) Wiheni a member bank voluntarily liquidates, it shall surrender its certificate of membership and cease to be a member of the Federal Reserve bank of its district and of the Federal Reserve System." SEC. 4. (a) The first paragraph (12 U.S.C. 288, first paragraph) of section 6 of the Federal Reserve Act is repealed. (b) The second sentence of the paragraph which, prior to the repeal made by subsection (a) of this section, was the second paragraph (12 U.S.C. 288, second paragraph) of such section 6, is amended to read: "The certificate of membership held by said national bank shall be surrendered to the Federal Reserve bank of its district, and said national bank shall cease to be a member of such Federal Reserve bank and of the Federal Reserve System." SEC. 5. (a) The first paragraph (12 U.S.C. 289) of section 7 of the Federal Reserve Act is amended by striking "the stockholders shall be entitled to receive an annual dividend of 6 per centum on the paid-in capital stock, which dividend shall be cumulative. After the aforesaid dividend claims have been fully met,". (b) The second sentence of the second paragraph (12 U.S.C. 290) of such section 7 is amended by striking "dividend requirements as hereinbefore provided, and the par value of the stock.". (c)' The third paragraph (12 U S.C. 531) of such section 7 is amended by striking "capital stock and". SEC. 6. (a) The first paragraph (12 U.S.C. 321, first paragraph) of section 9 of the Federal Reserve Act is amended (1) by changing, in the first sentence, "the right to subscribe to the stock o f " to read "membership in", (2) by striking the second and third sentences, and (3) by changing, in the last sentence, "stockholder" to read "member". (b) The first sentence of the second paragraph (12 U.S.C. 321, second paragraph) of such section 9 is amended by changing "Federal reserve bank stock owned by the national bank shall be canceled and paid for as provided in section 5 of this Act." to read "membership of such national bank shall be extinguished and the certificate of membership canceled as provided in section 5 of this Act." (c) The first sentence of the third paragraph (12 U.S.C. 321, third paragraph) of such section 9 is amended (1) by changing "stockholder" to read "member", and (2) by changing "stock" to read "membership". (d) The fifth paragraph (12 U.S.C. 323) of such section 9 is repealed. (e) The first sentence of the paragraph which, prior to the repeal made by subsection (d) of this section, was the ninth paragraph (12 U.S.C. 327) of such section 9, is amended by striking out "stock" and inserting in lieu thereof "certificate of membership". (f) The paragraph which, prior to the repeal made by subsection (d) of this section, was the tenth paragraph (12 U.S.C. 328) of such section 9, is amended (1) by changing, in the first sentence thereof, "all of its holdings of capital stock" to read "its certificate of membership", (2) by striking the second proviso of the first sentence thereof, (3) by changing, in the last sentence thereof, "stock 3 holdings" to read "certificate of membership", and (4) by striking, in the last sentence thereof, "a refund of its cash paid subscription with interest at the rate* of one-half of 1 per centum per month from date of last dividend, if earned,, the amount refunded in no event to exceed the book value of the stock at that time, and shall likewise be entitled to". (g) The paragraph which, prior to the repeal made by subsection (d) of this; section, was the sixteenth paragraph (12 U.S.C. 333) of such section 9, isamended (1) by striking, in the first sentence thereof, except that any such; savings bank shall subscribe for capital stock of the Federal reserve bank in an amount equal to six-tenths of 1 per centum of its total deposit liabilities as shown by the most recent report of examination of such savings bank preceding its admission to membership", (2) by striking all of the remaining sentences of such paragraph except the last sentence thereof, and (3) by striking, in the last sentence of such paragraph, ", except as otherwise hereinbefore provided with respect to capital stock". (h) The paragraph which, prior to the repeal made by subsection (d) of this section, was the twenty-second paragraph (12 U.S.C. 337) of such section 9, is amended (1) by changing, in the third sentence thereof, "stock" to read "certificate of membership", and (2) by changing, in the last sentence thereof, "stock" to read "certificates of membership". (i) The last paragraph (12 U.S.C. 338) of such section 9 is amended by changing, in the last sentence thereof, "stock" to read "certificates of membership". SEC. 7. The amendments made by the first six sections of this Act shall take effect on the thirty-first day after the date of enactment of this Act SEC. 8. (a) Not later than thirty-one days after the date of enactment of this Act, each holder of stock in any Federal Reserve bank shall surrender such stock to such bank, which shall, as of the thirty-first day after the date of enactment of this Act, cancel and retire the same and pay or credit to such former holder the par value thereof, plus interest at the rate of one-half of one per centum per month from the date of the last dividend, less a membership fee of $10, which shall not be refundable. (b) Upon the cancellation and retirement of Federal Reserve bank stock as provided in subsection (a) of this section, each Federal Reserve bank shall issue to each such former holder thereof a certificate attesting its membership in such Federal Reserve bank and in the Federal Reserve System. SEC. 9. The eleventh paragraph of section 9 of the Federal Reserve Act is amended to read: "Any applying bank shall be eligible for membership if it is an insured bank as defined in subsection (h) of section 3 of the Federal Deposit Insurance Act. The capital stock of a State member bank shall not be reduced except with the prior consent of the Federal Reserve Board." COORDINATION OF MONETARY POLICIES AND PROGRAMS SEC. 10. (a) Section 12A of the Federal Reserve Act (12 U.S.C. 263) is amended to read: "SECTION 12A. OPEN MARKET OPERATIONS " ( a ) No Federal Reserve bank shall engage or decline to engage in openmarket operations under section 14 of this Act except in accordance with the direction of and regulations adopted by the Board. The Board shall consider, adopt, and transmit to the several Federal Reserve banks regulations relating to the open-market transactions of such banks. " ( b ) All purchases and sales by Federal Reserve banks of paper described in section 14 of this Act as eligible for open-market operations, as well as all other actions and policies of the Federal Reserve banks and the Board in the field of monetary affairs, shall be conducted in accordance with the programs and policies of the President pursuant to the Employment Act of 1946 and other provisions of law. " ( c ) The Board shall submit a quarterly report to the Congress stating, in comprehensive detail, its past and prospective actions and policies under this section and otherwise with respect to monetary affairs, and indicating specifically how such actions and policies facilitate the economic program of the President." 4 ABOLITION OF FEDERAL OPEN MARKET COMMITTEE (b) The Federal Open Market Committee is abolished. FEDERAL RESERVE BOARD MEMBERSHIP AND TENURE SEC. 11. (a) The first and second paragraphs ( 1 2 U . S . C . 2 4 1 and 2 4 2 ) of section 10 of the Federal Reserve Act are amended to read as follows: "The Federal Reserve Board (hereinafter referred to as the 'Board') shall be composed of five members appointed by the President by and with the advice and consent of the Senate. Each member shall be appointed for a term expiring on June 30 of one of the first five calendar years succeeding the year in which he is appointed, as designated by the President at the time of nomination, subject to the limitation that not more than one member of the Board may have a term scheduled to expire within the same calendar year. The members of the Board shall devote their entire time to the business of the Board. "The members of the Board shall be ineligible during the time they are in office and for two years thereafter to hold any office, position, or employment in any member bank, except that this restriction shall not apply to a member who has served the full term for which he was appointed. The President shall designate one member as Chairman, to serve as such until the expiration of his term of office as a member, or until the President shall designate another member to serve as Chairman, whichever is earlier. The Chairman of the Board, subject to its supervision, shall be its active executive officer. The Chairman may designate one member as Vice Chairman, who shall have power to act in the temporary absence or disability of the Chairman, or in the event of the death, resignation, or permanent incapacity of the Chairman, to act as Chairman pending appointment of his successor. Each member of the Board shall within fifteen days after notice of appointment make and subscribe the oath of office. Upon the expiration of their terms of office, members of the Board shall continue to serve until their successors are appointed and have qualified." (b) The Board of Governors of the Federal Reserve System established under authority of the Federal Reserve Act as in effect prior to the effective date of the amendment made by subsection (a) of this section is abolished. Each member of the Board of Governors of the Federal Reserve System in office immediately prior to the taking effect of such amendment shall be paid one year's salary at his then current rate. ( c ) On and after the effective date of the amendment made by subsection ( a ) of this section, any reference (other than the reference in subsection (b) of this section) to the Board of Governors of the Federal Reserve System in any law, rule, or regulation of the United States or any department or agency thereof shall be deemed a reference to the Federal Reserve Board. AUDIT OF FEDERAL RESERVE SYSTEM BY COMPTROLLER GENERAL SEC. 12. (a) The Comptroller General shall make, under such rules and regulations as he shall prescribe, an audit for each fiscal year of the Federal Reserve Board and the Federal Reserve banks and their branches. (b) In making the audit required by subsection ( a ) , representatives of the General Accounting Office shall have access to all books, accounts, financial records, rep>rts, files, and all other papers, things, or property belonging to or in use by the entities being audited, including reports of examinations of member banks, and they shall be afforded full facilities for verifying transactions with balances or securities held by depositaries, fiscal agents, and custodians of such entities. 5 ( c ) The Comptroller General shall, at the end of six months after the end of the year, or as soon thereafter as may be practicable, make a report to the Congress on the results of the audit required by subsection ( a ) , and he shall make any special or preliminary reports he deems desirable for the information of the Congress. A copy of each report made under this subsection shall be sent to the President of the United States, the Federal Reserve Board and the Federal Reserve banks. In addition to other matters, the report shall include such comments and recommendations as the Comptroller General may deem advisable, including recommendations for attaining a more economical and efficient administration of the entities audited, and the report shall specifically show any program, financial transaction, or undertaking observed in the course of the audit which in the opinion of the Comptroller General has been carried on without authority of law. (d) The Comptroller General is authorized to employ such personnel and to obtain such temporary and intermittent services as may be necessary to carry out the audit required by subsection ( a ) , at such rates as he may determine, without regard to the civil service and classification laws, and without regard to section 15 of the Act of August 2,1946, as amended (5 U.S.C. 55a). RECEIPTS AND EXPENDITURES OF FEDERAL RESERVE SYSTEM SEC. 13. Section 7 of the Federal Reserve Act is amended by inserting immediately after the section heading the following new paragraph: "The full amount of all interest, discounts, assessments, and fees received by Federal Reserve banks shall be paid or credited by such banks to the Secretary of the Treasury and covered into the Treasury as miscellaneous receipts. The expenses of such banks may be paid only from such funds as may be specifically authorized or appropriated for that purposes." SEC. 14. (a) The third paragraph (12 U . S . C . 243) of section 10 of the Federal Reserve Act is amended to read : "There are hereby authorized to be appropriated such sums as may be necessary to pay the expenses of the Federal Reserve Board and the salaries of its members and employees. Subject to the availability of appropriations, the Board may maintain, enlarge, or remodel its office building in the District of Columbia and shall have sole control of such building and space therein." (b) The fourth paragraph (12 U.S.C. 244) of section 10 of the Federal Reserve Act is amended by striking the third sentence. EFFECTIVE DATE I ACCOUNTING PERIOD SEC. 15. Sections 13 and 14 of this Act shall take effect on the first day of the first fiscal year which begins after the date of enactment of this Act. During the period between the date of enactment of this Act and the effective date of such sections, the several Federal Reserve banks and the Federal Reserve Board shall take such steps as may be necessary to change their accounting period from the calendar year to the fiscal year and otherwise to bring their accounting practices and procedures into conformity with those employed by other agencies of the United States operated with appropriated funds. AMENDMENT OF EMPLOYMENT ACT OF 1946 SEC. 16. Subsection (a) of section 3 of the Employment Act of 1946 (15 U.S.C. 1022(a)) is amended by adding the following new sentence at the end thereof: "Such program shall include the President's recommendations on fiscal and debt management policy and guidelines concerning monetary policy, domestic and foreign, including the growth of the money supply as defined by him." COMPENDIUM ON MONETARY POLICY GUIDELINES AND FEDERAL RESERVE STRUCTURE Staff Report for the Subcommittee on Domestic Finance of the Committee on Banking and Currency S U M M A R Y AND ANALYSIS D E A R M R . C H A I R M A N : Replies to the questionnaire have been received from the Federal Reserve, the Secretary of the Treasury, the Council of Economic Advisers, and 71 academic, bank, and research monetary economists in response to your letter of inquiry on H.R. 11 of July 9, 1968. These replies are transmitted herewith along with a summary of the responses and an analysis of the reply of the Federal Reserve. I . VIEWS ON COORDINATING MONETARY AND FISCAL POLICIES The first two questions of the questionnaire concerned the matter of coordinating fiscal, debt management, and monetary policies. Specifically, respondents were asked: 1. Do you believe that a program coordinating fiscal, debt management, and monetary policies should be set forth at the beginning of each year for the purpose of achieving the goals of the Employment Act, or alternatively, should we treat monetary crndfiscalpolicies as independent, mutually exclusive stabilization policies f If you believe a program should be specified, do you believe that the President should be responsible for drawing up this program, or alternatively, should such responsibility be dispersed between the Federal Reserve System and agencies responsible to the President? 1. Summary of respondents' views By more than a 3-to-2 majority respondents favored the principle of requiring the President to present annually an economic program coordinating fiscal, debt management, and monetary policies. Moreover, only half of the dissenting respondents—comprising only about onefifth of all respondents—favored the system now in force. Under this regime, monetary policy in no way is constrained or even guided by an economic program or monetary rule but rather is used flexibly for purposes of cushioning unexpected shocks and reversing emerging undesired economic trends, and fiscal policy is used only by way of trying to correct major disequilibriums. The other half of the dissenting group— also comprising one-fifth of all respondents—opposed discretionary management of our money and credit whether orchestrated, as now, by the Federal Reserve authorities with reference to the fiscal policy extant, or as provided for in H.R. 11, by the President together with his fiscal and debt management policies. Respondents in this group (7) 8 regarded fiscal policy as too slow and cumbersome for use as a stabilization tool, and were not sanguine about discretionary monetary management. They therefore favored the development of a clearly defined monetary strategy or rule. Thus, if respondents are divided by their views on the present system of uncoordinated discretionary monetary management, we find they are opposed by a nearly 4-to-l majority. For readers' convenience, table I lists respondents by their broad views on questions 1 and 2. Respondents are classified by whether they (1) oppose the present schema wherein the monetary authorities have full discretion and act independently of the fiscal authorities and, if so, favor (a) requiring the President annually to present a program coordinating monetary and fiscal policies, though on a provisional basis, or (6) constraining the use of discretion in monetary management by adopting some clearly defined strategy or rule, or (2) favor the present system. Of course the finer points of respondents' views on these complex questions are not captured by our category titles, and therefore some respondents' views may be misinterpreted in table I. We hope not. In any case respondents' views should be read in full. TABLE I.—TABULATION OF VIEWS ON COORDINATING MONETARY AND FISCAL POLICIES Respondents' views Opposed to the present regime wherein the Federal Resdrve is neither guided by a program coordinating monetary; and fiscal! policies on a provisional basis, nor constrained by a monetary rule Favor a coordinated program Favor a rule * Chairman Okun Arlt Bach 2 Burstein * Chow 2 Cohen Davidson Dewald Earley Fand Fishman Frazera Gaines Greenwald* Harris, S. Harriss, L. Hauge Havrilesky 2 Hoadley Horwich Hosek2 Johnson 2 Keiser Kent Keyserling2 Leijonhufvud 2 Luckett 2 Mayer2 McCracken McDonald Morrison 2 Morton Noyes Scott Sprenkel 2 Stucki2 Thompson? Thorn Voorhis2 Warburton2 Weintraub Yeager2 Aschheim Bronfenbrenner Brunner Cagan Christ Crouch Culbertson Friedman Grossman Harwood Melitz Meltzer Pesek In favor of the present regime Governor Martin Secretary Fowler Adams Eckstein Hester Kane Madden Minsky Ross Teigen Walker Wallich Whittlesey Wilde 1 Dr. Harwood proposed adopting a full-bodied gold money. The others in this group favored a percent per annum monetary growth rule, or at least constraining the Fed to focus on money supply. 2 However, also favor constraining Federal Reserve actions by imposing a clearly defined money supply strategy or alternatively, a monetary growth rule valid for the year. 9 A large number of the 3-to-2 majority favoring coordination of stabilization policies as provided by H.R. 11 based their support on the assumption that fiscal and monetary policies are substitutable one for the other, and therefore, unless they are coordinated, sometimes will work at cross purposes and other times to compound disequilibriums in the economy at large.1 The validity of this assumption is undeniable if fiscal and monetary actions are distributed through the future in similar time patterns, with repercussions from both policies occurring in the current quarter and the bulk of all effects occurring within 9 months or a year. Under this regime, it would be irresponsible not to coordinate monetary andfiscalpolicies. But the case for coordinating also is strong if the lags between actions and effects—the so-called impact or outside lags—differ for monetary and fiscal actions. If the outside lag of monetary policy is shorter than that of fiscal policy, the success of current fiscal actions will depend significantly on future monetary policy. Under this regimefiscalpolicy cannot be programed rationally to achieve the goals of the Employment Act without some idea of future monetary actions. Clearly, in this case, if fiscal policy is used for stabilization purposes, those who present the program for current fiscal policy must also present at least a general near-future monetary policy program. The alternative to doing this is our present system in which, as was noted above, fiscal policy is used only to correct major disequilibriums and discretionary authority characterizes monetary management. Discussion of this alternative is resumed later in analyzing the Federal Reserve's views on coordinating monetary and fiscal policies. Conversely, if the outside lag of monetary policy is longer than that of fiscal policy, monetary policy cannot be programed rationally even from day to day without knowledge of futurefiscalpolicy. The alternative to coordination in this case is to establish a neutral monetary strategy to endure regardless of the economic winds. Discussion of this alternative also is resumed later. Respondents who favored coordinating monetary and fiscal policies recognized that any annual economic program presented in January had to be both general and provisional to permit adapting to undesired changes in economic trends. To this there can be no disagreement. To remove anty doubt that may exist about the intent of H.R. 11 in this respect it is recommended that section 10(b) be amended, as was suggested by Mr. Keyserling, to read that open-market operations "shall be conductedfoisofaras feasible in accordance with the programs and policies of the President pursuant to the Employment Act of 1946 and other provisions of law." [Emphasis supplied.] It also is noteworthy that several of the respondents who favored the principle of requiring the President to coordinate macroeconomic policies urged that our action options for coordinating monetary and fiscal policies be widened by delegating limited power to change tax rates to the President. This idea, however meritorious, takes us afield from the committee's jurisdiction and the immediate subject at hand. ll The danger of monetary and fiscal policies working at cross-purposes often has been recognized. For example, in 1964, many feared that the Federal Reserve would cancel the stimulus of the taix cut by tightening money. 'The danger, under the present system that monetary policy will compound an undesired thrust from fiscal policy has not been so widely recognized. But it exists. To illustrate, in the first half of this year, 1968. monetary policy was extremely expansionary in respect to the growth of the money supply (conventionally defined) and thereby compounded the inflationary thrust of the fiscal policy then extant. IThe Federal Reserve authorities apparently decided the 1968 inflation had to be tackled by fiscal policy, and failed to reverse their inflationary policy. 21-750—68 2 10 Also, many respondents who favored requiring that the President coordinate monetary and fiscal policies as provided by H.R. 11 stipulated that this requirement should be coupled with a statutory directive instructing the Federal Reserve to regulate the money supply to achieve maximum employment and price level stability. It is important to recognize that this suggestion is similar to the recommendation to develop a clearly defined monetary strategy or rule which was made by half of the respondents who opposed requiring the President to make recommendations concerning monetary policy along with his recommendations on fiscal and debt management policies. The similarity of these views indicates that coordination can be carried out in the context of a clearly defined monetary strategy. To further pursue this matter, some respondents argued that there is little advantage to coordinating monetary and fiscal policies inasmuch as neither the President and his advisers nor the Federal Reserve authorities have yet bothered to acquire adequate knowledge of how monetary policy affects economic activity. Instead of discretionary coordination we now need, in the view of these respondents, a clearly defined strategy or rule for the conduct of monetary policy. The staff shares this group's concern for developing an appropriate strategy for monetary policy, and also joins with them in deploring the fact that the Federal Reserve authorities have neglected to develop a validated theory of how monetary policy works. However, we believe that the development of a clearly defined monetary strategy is not inconsistent with coordination. In support of this belief we note again that many of the respondents who favored coordination also wanted a statutory instruction to regulate the money supply to achieve maximum employment and price-level stability. In this regard, H.R. 11 directs the President to specify guidelines for the growth of the money supply along with his other stabilization recommendations. In other words, the operational assumption for monetary policy of H.R. 11 is that the quantity of money is the crucial variable by which Federal Reserve actions are transmitted to the economy in the large. Thus, it is the clear intent of H.R. 11 that the President's program for achieving the goals of the Employment Act be centered on a money supply growth strategy. In the later review of respondents' views on monetary policy guidelines we will see that the overwhelming majority of respondents, including many who favor that the President coordinate monetary and fiscal policies, favor the development and specification of a money supply growth strategy. The second group of respondents who were opposed to coordination, comprising once again about one-fifth of all respondents, held the view that the monetary authorities must retain virtually unlimited freedom to take whatever actions they deem wise. The Federal Reserve was among those respondents favoring the fullest use of discretionary authority in monetary management. The argument of this group is analyzed below in considering the Federal Reserve's views on coordinating monetary and fiscal policies. 2. The Federal Reserve's views on coordinating monetary and fiscal policies In replying to the two questions on coordination the Federal Reserve concluded that for purposes of achieving full employment, price-level stability and balance-of-payments equilibrium, there is a natural 11 division of labor and responsibility between monetary policy and fiscal- and debt-management policies (hereafter simply fiscal policy). In reaching this conclusion the Federal Reserve observed that "major changes in the intensity of fiscal stimulation or restraint are not everyday occurrences," and therefore fiscal policy is not well suited for adjusting the economy to minor swings in business activity and reacting to unexpected events in the short run. Rather, the Federal Reserve's view is that fiscal policy is the appropriate tool for countering gross maladjustments in the macroeconomy, for example, mass unemployment and rapid inflation. But, concerning monetary policy on the other hand, the Federal Reserve's view is that it "is well suited to rapid and marginal response to the emerging requirements of the economy. It is continually under review and subject to gradual, flexible and even reversible adjustments. It is the very essence of monetary policy that it can respond to the unexpected developments and that it can adjust for divergencies between unfolding economic events and projections." Assuming the validity of this argument "it would seem," as the Federal Reserve asserted, "most unwise to commit monetary policy in advance. * * * To do so would rob it of the very flexibility and adaptability that constitute the unique contribution of monetary policy to the economic stabilization instruments at the Government's disposal." Rather, given this argument, optimal stabilization policy requires that fiscal policy be set at the beginning of each year and that monetary policy be used flexibly within the year to adjust to changing business and international conditions. Responsibility should be divided accordingly—fiscal policy with the President and monetary policy with the Federal Reserve. It is asserted that, "This division of responsibilities in the field of economic policy is one of the desirable checks and balances in our system of government." The Federal Reserve's argument, however, is not persuasive. To begin with it calls for operational procedures which are the antithesis of democratic procedures. For, if we accept the premise that monetary policy is "unique"—the only flexible instrument at the Government's disposal for achieving economic stabilization, then it is just plain wrong that control of monetary policy should be vested in authorities (Federal Reserve officers) wTho are only remotely responsible to the people. The details of the structure of the Federal Reserve are discussed later. Here our only concern is that if the premise is accepted that the economic state of the union rests so strategically on the satisfactory use of monetary policy, then surely, under our form of government, the President must control or at least guide the monetary authorities in their use of the only flexible instrument we have for achieving economic stabilization. Furthermore, the operational procedures called for by the Federal Reserve's argument contravene the requirements of existing law. For it is impossible for the President to discharge the responsibilities assigned him by the Employment Act of 1946 if he cannot guide the use of the only effective tool at the Government's disposal for achieving "Maximum employment, output and purchasing power." Second, as a matter of economics and logic the Federal Reserve's argument is not persuasive. It rests on the fact tha/t, under present institutional arrangements, monetary policy can be changed more rapidly than fiscal policy. But there is nothing sacred about these arrange- 12 ments. If the Congress so desires it can give the President clearly defined limited powers to change tax rates, a course of action many persons have recommended. This would make fiscal policy just as well suited as monetary policy "to rapid and marginal response * * * and subject to gradual, flexible, and even reversible adjustments." Moreover, if the so-called "impact" or "outside lag" between actions of the Federal Reserve and changes in employment, production, and purchasing power is longer than the outside lag for fiscal policy, then effective economic stabilization strategy would in fact require using fiscal policy counter-cyclically, not monetary policy. That is, under this structure of outside lags the Federal Reserve's argument should be turned around. Fiscal policy should be used flexibly and monetary policy changed only infrequently and within clearly prescribed limits, if at all. The Federal Reserve, of course, must live in the world as it is, not in some theoretically ideal world. And, in the world as it is, there are constraints on changing fiscal policy promptly but none on changing monetary policy promptly. However, this does not mean that monetary policy should be used flexibly—only that it can be. In fact, for the world as it is, many economists argue that the degree of monetary stimulation should be kept relatively constant over time because we lack both foresight about future economic trends and knowledge about the outside lag for monetary policy, which are required if we are to benefit from changes in the degree of monetary stimulation. The Federal Reserve does not claim ability to forecast. Indeed the Federal Reserve's reply asserts that "the possibility of error in forecasting * * * remains disturbingly high." Nor does the Federal Reserve claim knowledge of the outside lag for monetary policy. The System's reply does not cover this important subject in any substantive detail. Thus it is curious that the Federal Reserve argues against limiting "the flexibility and adaptability that constitute the unique contribution of monetary policy to the economic stabilization instruments at the Government's disposal." For, clearly, given both the primitive state of the art of economic forecasting and our lack of knowledge on the outside lag for monetary policy, using monetary policy flexibly involves awesome risks as well as a high potential for serving well the public interest. For example, today a trend to recession may be forseen and monetary policy eased to prevent it. But by the time today's action takes effect the problem may be inflation and we will wish that the monetary authorities had tightened when they eased. But, if the impact lag is short or no other change occurs, today's shift to monetary ease will work effectively, preventing the predicted recession without contributing to inflation. Because the flexible use of monetary policy involves risks as well as potential benefits it is imperative to safeguard against monetary policy being used unwisely while at the same time not eliminating its potential for good. This is the purpose of the provision in H.R. 11 requiring the President to set guidelines for monetary policy at the beginning of the year along with his recommendations on fiscal and debt management policy. The guidelines would serve as a warning against unduly frequent or large changes in the degree of monetary stimulation or restraint without interfering with the "rapid and marginal response" the Federal Reserve argues monetary policy is well suited to. The case for such guidelines appears indisputable. 13 I I . VIEWS ON MONETARY POLICY GUIDELINES Question 3 concerned the nature of details of Presidential guidelines on monetary policy. Specifically, respondents were asked: A. Should monetary policy be used to try to achieve the goals of the Employimnt Act via intervention of money supply (defined as desired) as provided in H.R. 11, or alternatively should H.R. 11 be amended to make some other variable or variables the immediate target of monetary policy; for example, interest rates, bank credit, liquidity, high-powered or base-money, total bank reserves, excess reserves, and free reserves? * * * It ivould be most helpful if, in providing the reasons for yowr choice, you list the actions the Federal Reserve should take to control the target variable (or variables) and also explain the link between your recommended target of monetary policy and the goals of the economy as defined by the Employment Act. B. Should the guidelines of monetary policy be specified in terms of some index of past, present, or future economic activity, or alternatively in terms of the target variable's value or growth? C. For only those persons who recommend that some index of economic activity be used to guide the monetary authorities in controlling the target variable: Should we use a leading (forward looking), lagging (backward looking) or coincident indicator of economic activity? D. For only those persons who recommend that the guidelines be put in terms of the target variable's value or growth: Should the same guidelines be used each year into the foreseeable future, or alternatively\ should new guidelines be issued at the beginning of each year conditioned on expected private investment, Government spending, taxes, et cetera? E. For only those persons icho recommend that the guidelines be put in terms of the target variable's value or growth and who also recommend that the same guidelines be used year after year into the foreseeable future: What band of values or range of growth do you recommend? F. For all those persons recommending that the guidelines be put in terms of the target variable's value or growth, * * * Under what circumstances, if any, should the monetary authorities be permitted during the year to adjust the target variable so that it exceeds or falls short of the band of values or range of growth defined by the guidelines issued at the beginning of the year? 1. Summary of respondents'' vietvs By a more than 2 to 1 majority, respondents favored making the growth of the money supply or its cognate, base money, the target of monetary policy. The larger part of the minority was eclectic in its approach to the kinds and means of monetary management. Respondents in the majority group differed in respect to the details of managing the growth of the money supply. To begin with a few of these respondents wanted to use base money, defined as bank reserves plus publicly held currency and coin, as the target variable. But the overwhelming number in the majority urged that policy focus on some money supply measure. On this question the staff concurs with the larger number of these respondents who believe it would 14 be more productive to use a money supply measure than to use base money as the target of monetary policy. Base money affects economic activity largely by changing the money supply and the correspondence between changes in base money and money supply is not a constant. At different times a given change in, say, the conventional money supply requires different input of base money. Thus, though the Federal Reserve controls money supply largely by changing base money, money supply is the appropriate vehicle for transmitting monetary policy actions to the economy in the large. Second, there were differences about the most useful definition of money. These differences centered on whether to include time deposits in commercial banks or to count as money only publicly held demand deposits and currency and coin. But no one argued that this question is crucial. In fact, many of the respondents in the money supply group did not specify which measure to use and many others indicated that either one might be used. The staff agrees that this matter is not crucial. Last, there were differences about how to specify the guidelines for money supply growth. Roughly half of the group favored specifying a target percentage change in money supply for 6 months to a year ahead in terms of the economy's expected or actual performance. A fairly popular plan of this type, advanced by several respondents, requires the President to specify every January the estimated change in money supply that is needed to enable us to achieve our full employment real gross national product in the year ahead without generating inflation. Under this plan the Federal Reserve would be allowed to vary the growth of the money supply around the target growth rate. The President would set the limits, say plus or minus 2 percentage points, around his target percent per annum growth rate. Other plans of this type which were advanced by respondents would require the monetary authorities to generate whatever money supply growth it takes (1) to keep the rate of unemployment under some desired maximum, say 4 percent, or (2) to prevent the price level from rising faster than some minimum rate, say 3 percent per year for the The other half of the many respondents urging adoption of a money supply target recommended that the Congress or the President set guidelines for money supply growth in terms of a band or range of percent per annum values. The major argument for this strategy is that it would mute the development of economic disequilibriums because of mistakes in monetary management. The most popular bands of values recommended were 3 to 5 and 2 to 6 percent per annum. Several respondents among those urging the specification of a percent per year range for money supply growth suggested that the President also set a target growth rate within the guideline range every 6 months or year. This could be done using econometric techniques, if desired. The Federal Reserve would be allowed to use discretion to regulate the growth of the money supply around the target rate but not enough to violate the guideline range. A few respondents here recommended setting a quasi-permanent relatively-narrow band of values for monetary growth and instructing the Federal Reserve to stay within this range. The range would be 15 adjusted outward only if it was proved to be clearly inappropriate by a persistent inflationary trend or persistent unemployment. But others wanted the range reviewed each year. Still another strategy that was suggested called for specifying a fairly broad range of allowable money supply growth and using triggers to collapse the range. Thus, the maximum allowable range of money supply growth might be set as zero to 10 percent per year. And the Federal Reserve would be directed to reduce the upper limit to, say, 8 percent when the CPI advances more rapidly than 2 percent per year and by 1 additional percentage point for every additional point of inflation. In the same way the lower limit of allowable money supply growth would be set at, say, 2 percent per year when the rate of unemployment reached 3 percent and raised one point for every point rise in unemployment. Last, some suggested trying to hit an interest rate target subject to the constraint that monetary growth stay within a specified range. The staff sees no need at this time to choose among the various strategies recommended by respondents for setting monetary policy guidelines. Rather, the conclusion that should be drawn from this listing of possible montary policy strategies is that issuing guidelines for money supply growth, as provided for by H.R. 11, keeps the door open for fruitful innovations in monetary policy while at the same time protecting against major errors in monetary management. As was earlier noted, the major alternative to adopting a money supply target for monetary policy which was advanced by respondents calls for eclectism in monetary management. The eclectic approach to monetary policy is discussed next in analyzing the Federal Reserve's views on monetary guidelines, for the System was a strong advocate of this approach. 2. The Federal Reserve's views on monetary policy The Federal Reserve's reply to this series of questions is in a sense a nonreply. The Federal Reserve's view is that it is necessary to be eclectic in managing the Nation's money and credit. Neither the kind nor even the means of management can be specified. For, as asserted by the Federal Reserve, "monetary policy cannot be formulated solely in terms of any single financial variable or any single class of variables." Rather the kind of monetary management, and the means of manage-, ment, must be adapted to the particular requirements of each new crisis, new situation, new day. For each particular crisis, situation, day, in the Federal Reserve's view, "incoming information on both financial quantities and financial prices must be assimilaited and interpreted. Movements in financial quantities—such as total bank reserves, the money stock, commercial bank time deposits—and claims against nonbank intermediaries—on the one hand, together with indications of cost and availability of credit—including interest rates and nonprice terms of credit—on the other, must be evaluated jointly to assess what effects monetary policy currently is having * * *." To justify its eclectism the Federal Reserve argued that, "The effects that stem from any given monetary policy depend fundamentally on private reactions to the policy, and these are not static. They change over time * * *." Thus, beginning in the 1950's, "the monetary authorities have had increasingly to take into consideration the effects of changes in policy on a broad range of financial assets * * *. In par- 16 ticular, monetary policy decisions have had to take into account the potential effect of variations in time deposit growth * * *. [Also] we cannot afford to exclude the major nonbank thrift institutions from consideration in formulating monetary policy * * *. Still another complexity arising in the late 1950's and continuing throughout the 1960's has been the serious imbalance in the U.S. balance of payments." Distilled to its essence, the Federal Reserve's reply here argues that because there nearly always is something undesirable happening (e.g., an outflow of funds from nonbank thrift institutions, imbalance in our external payments, etc.), and also because there are many possible target variables or vehicles for transmitting monetary actions to the macroeconomy (e.g., money supply, interest rates, etc.), the monetary authorities must be allowed to "play it by ear"—to use a familiar analogy. The plea should be denied. To say that something undesirable nearly always happens and that there are many possible monetary policy targets is no substitute for the difficult theoretical analysis and hard empirical research that would have led the Federal Reserve to provide a validated or at least verifiable theory of how their actions affect employment, production, and prices. Manifestly, the Federal Reserve's eclectic views on the nature of monetary policy guidelines in no way whatever casts doubt on the usefulness of requiring the President to specify monetary guidelines for the Federal Reserve "including the growth of the money supply as defined by him," as provided by H.R. 11. The staff's view is that the purposes of the Employment Act, which we conceive as the minimization of both unemployment and inflation, will be served by the President's setting money supply guidelines, as provided by H.R. 11. In principle, changes in money supply that originate in open market operations change spending and economic activity by changing the size and composition of the public's nominal or financial wealth. When the Federal Reserve buys U.S. Government securities on the open market the public's assets are unchanged since increases in holdings of base-money are offset by decreases in holdings of securities but taxpayers' liabilities fall by the amount of Federal debt retired and hence there is a rise in net worth. In turn, the rise in net worth acts directly to increase consumption and investment. Added stimulus is provided because increases in money supply necessarily change the composition of financial wealth. As a result the return to holding money falls relative to other returns and spending on the whole spectrum of assets (real and financial) and on goods increases as the public attempts to realign returns. Moreover, there is at least a prima facie empirical case tht perverse changes in money supply have contributed substantially to past episodes of inflation and recession. Guidelines will impel, but not compel, the Federal Reserve to dampen and perhaps even prevent sharp destabilizing changes in money supply in future years. As was observed earlier, the overwhelming number of persons responding to the committee's questions share this view. The staff sees no technical problem in using money supply as the target variable of monetary policy. In this connection the staff recognizes that money supply tends to fall in recessions and rise in periods 17 of economic expansion. But this does not disqualify money supply from being used as the target variable of monetary policy. The Federal Reserve has ample powers to overwhelm cyclical movements of money supply and make monetary growth whatever it desires from quarter to quarter though not day to day. Thus, the fact that money supply has a procyclical component in no way disqualifies it from being used as the target variable. Indeed, this property makes money supply especially well suited to serve as the target variable of monetary policy. For, because money supply has a procyclical component, the Federal Reserve cannot be deceived into thinking it has tightened (or eased) when it has not if money supply is used as the target variable. In expansion periods when the natural tendency is for monetary growth to accelerate the goal of policy is to decelerate the growth of the money stock, and only such restraint can be regarded as proof that monetary policy has been tightened successfully. Conversely, in recessions when the money stock tends to fall, the goal of policy is to increase monetary growth, and only this acceleration can be regarded as proof that monetary policy has been eased sufficiently. Thus, money supply is a technically usable as well as a potentially useful vehicle for transmitting monetary actions to the macroeconomy.2 One final remark is in order here. The Federal Reserve, having asserted that "monetary policy cannot be formulated in terms of any single financial variable or any single class of variables," did not, of course, reply to the questions (3.B), (3.C), (3.D), (3.E), and (3.F) concerning guideline details. But clearly the setting of money supply guidelines, as provided by H.R. 11, will involve consideration of (3.B) whether the growth of the money stock should be tied to some index of economic activity, or alternatively, whether percentage per annum growth guidelines should be specified without regard to the behavior of economic indexes, and (3.C), if the former, what index, or (3.D)? (3.E), and (3.F) if the latter, whether the growth rate should be reviewed annually, what band or range of percentage growth rates should be specified and what circumstances, if any, should trigger violations of the guidelines. H.R. 11, wisely in the opinion of the staff, leaves these details to the President. Hopefully, the replies of many of the respondents to questions (3.B), (3.C), (3.D), (3.E), and (3.F), which were summarized in the preceding section and are printed as received in the text of this report, will throw light on how they should be worked out. For readers' convenience table 2 lists respondents bv their broad views on question 3. Respondents are classified by whether thev favored (1) a money supply target, (2) an interest rate or bank credit target, or (3) the eclectic approach to monetary management. The staff recommends reading respondents' replies to question 3 to capture the full flavor of their views. 2 Our objection to using interest rates as the target variable may now be noted. It is that, though interest rates undeniably help to transmit monetary actions to the macroeconomy, movements of interest rates may provide misleading information about the thrust of monetary policy. In expansions when the aim of monetary policy is to tighten money and credit we can be deceived into believing policy was tight when it wasn t because in such periods interest rates tend to rise because of increases in credit demand. In the same way, in recessions we might believe that policy was easy when it wasn t because interest rates tend to fall in such periods as a result of decreases in credit demand. 18 TABLE I I . — T A B U L A T I O N OF VIEWS ON MONETARY POLICY TARGETS Respondents' targets Money supply, more specifically, percent per annum growth of the money stock, or a money supply cognate The rate of interest or credit flows or both Chairman Okun i Adams Cohen Gaines Hauge Horwich Scott Arlt Aschheim 3 Bach Bronfenbrenner Brunner Cagan Chow Christ Crouch Culbertson Davidson Dewald Fand Fellner Fishman Frazer Friedman Greenwald Grossman 3 Harris, S. 3 Harriss, L. Havrilesky Hosek Johnson Keiser 3 Kent Keyserling 3 Latang 3 Leijonhufvud Levy 3 Luckett Mayer McCracken 3 McDonald Melitz Meltzer Morrison Morton 3 Pesek Sprenkel Stucki Thompson Thorn Voorhis Wallich 3 Warburton Weintraub Yeager Eclectic Governor Martin Secretary Fowler Burstein Earley Eckstein Hester Hoadley Kane Madden Minsky Noyes Ross Teigen 1 Walker 2 Wilde Whittlesey » Subject, however, to not using free reserves as the target and requiring that the rate of growth of the money stock be greater than zero. 2 Favors, however, that the Federal Reserve explain monetary growth outside the 2 to 6 percent per year range. 3 Supplemented by or in association with interest rates or bank credit or both. I I I . VIEWS ON DEBT M A N A G E M E N T Question 4 concerned debt management policy. Specifically, respondents were asked: Given the goals of the Employment Act, what can debt management do to help their implementation? 1. Summary of respondents' views Roughly 25 percent of all respondents did not comment on this question. In principle, debt management can influence aggregate demand by shortening maturities in recessions, which would increase the public's liquidity and thereby propensities to consume and invest, and conversely, increasing the maturity of the debt in inflations to decrease liquidity and hence spending. But only about 15 percent of all respondents recommended pursuing this strategy aggressively. At the other extreme about 40 percent of all respondents opposed changing 19 the maturity of the debt—shortening it in recessions and lengthening it in inflations—by way of attempting to offset cyclical movements in economic activity. Some persons in this group favored rather managing the debt to minimize the carrying cost even though this entails procyclical changes in the public's liquidity. The argument underlying this view is that the Federal debt is too small a part of total financial wealth and too narrowly held to be able to affect the public's liquidity by altering its age-mix, and hence debt management is not a useful stabilization tool. Under this assumption it is eminently sensible to adopt a policy that minimizes carrying costs. But the majority in the group of respondents who opposed using debt management for stabilization purposes did not recommend using it to minimize carrying costs. Rather they favored adopting a passive strategy, one of keeping the maturity-composition of the debt relatively constant and thus not interfering with the stabilization effects of monetary and fiscal policies. It w^ould appear that this group, though not believing that the public's liquidity and/or propensities to spend and invest could be changed by altering the age-mix of the Federal debt, did not want to risk affecting aggregate demand procyclically by altering the age-mix to minimize carrying costs—i.e., by lengthening the maturity of the publicly held debt in recessions and shortening it in inflations. Roughly 20 percent of all respondents, including the Federal Reserve, viewed debt management as having "some" or "limited" potential for influencing economic trends via intervention of liquidity and the propensities to consume and invest. They recognized, however, that the usefulness of debt management as a stabilization tool is constrained both by the purely housekeeping requirement of holding down carrying costs and the fact that holdings of Federal debt comprise only a small part of a small part of the public's total financial wealth. Some of the respondents in this group observed that because of the housekeeping goal there was some danger that debt management would be destabilizing. They recommended, therefore, that at minimum the debt not be managed to minimize carrying costs over the cycle. This viewpoint is discussed further immediately below in presenting the Federal Reserve's views on debt management. Here it is noted only that, with respect to policy, 60 percent of all respondents and 75 percent of those who commented on debt management would appear to agree that the important contribution debt managers can make to economic stability is simply not to interfere with other stabilization policies. This majority consists of the respondents who stated that debt management has no potential as a stabilization policy yet recommended keeping the age-mix of the debt constant, and those respondents who concluded that debt management would be destabilizing if used to minimize carrying costs and recommended that it definitely not be used for this purpose. 2. The Federal Reserved views on debt management On this matter the Federal Reserve replied that, "Shifts in the maturity composition of the Federal debt * * * alter the liquidity of the debt and/or term structure of interest rates. [Thereby] they will have some impact on financial flows and private spending * * *." [Emphasis supplied.] However, the Federal Reserve, correctly we think, concluded that, regardless of the potential for influencing "financial flows and private spending" by changing the age-mix of the Federal debt, the role that debt management can play as a stabiliza- 20 tion tool is clearly circumscribed. As the Federal Reserve's reply noted, it is limited because technical and housekeeping considerations make debt lengthening "most feasible in periods of declining interest rates * * * [but] when interest rate declines are associated with an undesirable slackening of economic activity, the economic goals of the country may indicate the desirability of keeping debt lengthening to moderate dimensions so as to encourage investors to lend more to finance capital outlays of private sectors of the economy." In the same way from the standpoint of housekeeping considerations it is most feasible to shorten the debt in periods of rising interest rates. But just as interest rate declines are symptomatic of recessions so rising interest rates are a symptom of inflation, and manifestly, it would be unwise to add to the public's liquidity by debt-shortening operations during inflationary periods. H.R. 11 neither specifies nor suggests debt management targets. Respondents were questioned about the potential benefits from debt management policies to determine whether H.R. 11 should be amended to require the President to make recommendations on the term structure of interest rates or the age-mix of the publicly held Federal debt. No substantial reason has been developed for such an amendment. IV. VIEWS ON POLICY INSTRUMENTS Question 5 explored several aspects of the use of policy instruments. Specifically, respondents were asked: 5.A Do you see any merit in using open-market operations for defensive purposes or should they be used only to facilitate achievement of the Presidents economic program and the goals of the Employment Act? 5.B Do you believe that monetary policy can be effectively and efficiently implemented solely by open-market operations? 5.0 For what purposes, if any, should (a) rediscounting, (b) changes in reserve requirements, and (c) regulation Q be used? 5.D Do you see any merit m requiring the Federal Reserve Board to make detailed quarterly reports to the Congress on past and prospective actions and policies? 5.E What costs and benefits would accrue if representatives of the Congress, the Treasury, and the CEA were observers at Open Market Committee meetings? 1. Summary of respondents' views One-fourth of all respondents did not comment on the merits of using open-market operations for so-called defensive purposes. Those who did approved defensive open-market transactions by a nearly 4to-1 margin. Many respondents pointed out in support of their view that defensive transactions to absorb certain transient influences are essential in order to closely control the rate of growth of the money supply. Monetary growth can be influenced perversely at any point in time by sudden, unexpected, and ephemeral changes in such elements as U.S. gold holdings, the public's preferences for currency and time deposits and banks' desires to hold excess reserves. But open-market operations can be used to prevent these influences from modifying significantly desired money supply growth. Clearly, in the limited sense of maintaining desired money supply growth against perverse influences defensive open-market operations have merit. Respondents who opposed defensive operations, however, would not appear to have had this meaning in mind. Rather, their opposition is to using open-mar- 21 ket operations to assist Treasury financing operations and otherwise maintain order in the money market. The staff agrees both that openmarket operations should not be used for such purposes and must be used to maintain desired money supply growth in the face of perverse transient influences. A fourth of all respondents also failed to comment on questions 5.B and 5.C. Of the remainder, half stated that monetary policy can be effectively and efficiently implemented solely by open-market operations. But many in this group recommended nevertheless retaining some of the other currently used tools of monetary policy, especially rediscounting, for such special purposes as providing a sure source of short-term funds to banks. The other half of respondents commenting on questions 5.B and 5.C concluded that monetary policy would be less effectively and efficiently implemented by using only open-market operations than by using, in addition, some of the other policy instruments now being used. However, even among this group there was strong sentiment for rescinding regulation Q. The staff believes that questions concerning monetary policy instruments or tools are a secondary matter compared to the questions of monetary and fiscal policy coordination, the target and guideline specification for monetary policy, and the structure of the Federal Reserve. These latter are the subjects of H.R. 11. Respondents were questioned about instruments to determine whether there was reason for amending H.R. 11 to modify the currently used kit of monetary policy instruments. But substantial argument was not developed for amending the bill to modify the Federal Reserve's existing powers to set rediscount rates and rediscount eligible paper, change reserve requirements, and regulate interest paid on time deposits. On the other hand the bill might be amended to assure that these powers, as open market powers, are used insofar as feasible to implement the President's economic program pursuant to the Employment Act. However, in view of the heavy sentiment against regulation Q, the committee might want to take up the question of reevaluating regulation Q separately. Roughly one-sixth of all respondents did not comment on the merits of requiring the Federal Reserve to make detailed reports on its actions to the Congress. Respondents who commented on this question favored the reporting requirement by a nearly 4-to-l majority. Some, however, wanted any report confined to past actions. Others recommended that the report be limited to explaining money supply growth. Another recommendation called for full discussion of proposed changes in regulations covering rediscounting, reserve requirements, and commercial banking activities. One-third of all respondents did not comment on the costs and benefits of having administration observers at OMC meetings. Those who commented were opposed to the idea by more than a 5-to-l majority. 2. The Federal Reserve*s views Concerning the usefulness of defensive open-market operations, the Federal Reserve replied that "if the financial markets are to respond as intended to national policy action, the [money and credit] mechanism must be protected from short-run swings in such factors as the public's demand for currency, the speed of check collections, international currency flows, or the size of Treasury balances held at Federal Reserve banks." Because all the factors listed affect money supply growth the staff has no quarrel with this view. As was earlier noted, in 22 the limited sense of offsetting undesired autonomous influences on money supply growth defensive open-market operations are an essential part of monetary policy. It is gratifying that the Federal Reserve did not attempt to also justify using open-market operations to maintain order in the money market. It makes little or no sense to use open-market operations to insulate money-market variables—and thereby players—against the economic winds of the da j . On the questions pertaining to the instruments of monetary policy, the Federal Reserve argued that monetary policy can be more effectively implemented by using rediscounting, reserve requirement changes, and changes in regulation Q along with open-market operations than by open-market operations alone. The Federal Reserve's argument that changes in reserve requirements have advantages over market operations in special circumstances that "require a massive and immediate tightening or easing of bank reserve positions" is especially compelling. The Korean war inflation was quickly and substantially damped by increasing reserve requirements in January 1951. More important, now might be another time when an upward revision of reserve requirements would be an effective way of decelerating inflation. But the System's argument on regulation Q is not persuasive. The Federal Reserve would retain regulation Q to protect "thrift institutions" whose "earning power is limited by their necessarily heavy commitment in long-term assets * * * [which commitment] has limited the ability of these institutions to meet the competition of rising open market rates." The staff shares the Federal Reserve's concern for thrift institutions. But, if desired, these institutions and the mortgage and other markets they serve can be protected in periods of disintermediation by Federal Reserve purchases of the obligations of Federal home loan banks and its members, the Federal National Mortgage Association and other agencies. Concerning the merits of monetary policy reports, the Federal Reserve argued that "it could be seriously misleading to the public for the Federal Reserve to present, at the beginning of a quarter, a detailed prospectus of future actions and policies when in fact the actual policies adopted would depend so heavily on the extent to which domestic and foreign developments within the quarter alter the System's assessment of future monetary and credit needs." The Federal Reserve is not, of course, opposed to reporting to the Congress about its past actions. It does so now. The staff believes, however, that there is merit in providing a projection of the money stock and of the broad actions that will be taken to achieve this target along with the explanation of past money supply growth in the Fed's quarterly reports to the Congress. Such requirement will impel development of validated theory on money supply and of the relationships of employment and prices to money supply. Few persons care to make wrong forecasts. To pursue this matter, it is vital that the reports be substantive. On this the Federal Reserve's response indicates that its future reports will be as meaningless as its past and current ones have been. The Federal Reserve stated that "such reports, to be useful, should include an analysis of all major monetary and financial developments of the preceding calendar quarter." [Emphasis supplied.] We see no advantage in covering the spectrum of major monetary and financial developments. Rather, to obtain maximum benefits from 23 monetary policy reports they must focus on the behavior of the money supply. Specifically, they must explain both the proximate causes of money supply growth in the preceding quarter or 6 months and how the observed growth has or will implement the President's economic program and the goals of the Employment Act. The educational value—to the Federal Reserve authorities—of having to prepare and discuss such reports will be large. The System, therefore, should not be permitted in reporting to the Congress to substitute extensive descriptions of monetary and financial developments for meaningful, empirically verifiable, statements about the policy of the immediate past. The Federal Reserve was opposed to having administration observers at meetings concerned with open market policy. The System argued that this would restrict "full and frank" discussion. H.R. 11 does not now call for administration observers at these meetings. And, in view of the fact that H.R. 11 requires the President to make monetary policy recommendations, including guidelines on money supply growth, the staff sees no need for amending H.R. 11 to provide for such observers, especially inasmuch as their presence might inhibit discussion. For reader's convenience, table 3 lists respondents' votes, "yes" or "no," on questions 5.D. and 5.E. Respondents' views, especially on 5.D., should be read in full. V. VIEWS ON T H E FEDERAL RESERVE^ STRUCTURE H.R. 11 provides for the following structural changes in the Federal Reserve System : 1. Retiring Federal Reserve bank stock; 2. Reducing the number of members of the Federal Reserve Board to five and their terms of office to no longer than 5 years; 3. Making the term of the Chairman of the Board coterminous with that of the President of the United States; 4. An audit for each fiscal year of the Federal Reserve Board and the Federal Reserve banks and their branches by the Comptroller General of the United States; and 5. Funds to operate the Federal Reserve System to be appropriated by the Congress of the United States. Respondents were asked: Please comment freely on these several provisions. In particular, it would be most helpful if you would indicate any risks involved in adopting these provisions and discuss whether their adoption would facilitate the grand aim of H.R. 11 j which is to provide for coordination by the President of monetary and fiscal policies. By heavy majorities respondents favored all provisions except No. 5. Respondents' votes on these matters, including the Federal Reserve's votes, are recorded in table 4. In addition, H.R. 11 provides for the vesting of all open market powers in the five-man Federal Reserve Board. Respondents were not asked to comment on this provision directly, though many recognized that any surviving open market committee would have to be drastically reduced in size if the Board were reduced to five members. Among those who commented, some respondents favored doing away entirely with Reserve bank representation in formulating open market policy but others favored retaining some representation. 24 TABLE I I I . — T A B U L A T I O N OF VOTES ON QUESTIONS 5 . 0 AND 5.E Respondents' votes On the desirability of reporting past and prospective policies and actions For Governor M a r t i n 1 Secretary Fowler 1 Chairman O k u n 1 Arlt1 Aschheim Bach 1 Brunner Cagan 2 Cohen Crouch 2 Culbertson 2 Dewald Ear ley Fand Fishman 3 Frazer Friedman Fro mm Gaines 1 Harris, S. Harriss, L. Hauge1 Havrilesky Horwich Hosek 1 Johnson Kane Keiser Kent 2 Keyserling Luckett 1 Madden * McDonald Meltzer Minsky 1 Morrison 3 Noyes 1 Ross Sprenkel Stucki Thorn Voorhis Walker » Warburton Weintraub 1 Whittlesey Yeager 2 Against Burstein Chow Davidson Hester Hoadley Leijonhufvud Melitz Morton Pesek Scott Teigen WaOich On the desirability of having observers at OMC meetings For Davidson Ear ley Havrilesky Keiser Keyserling Melitz Warburton Against Governor Martin Secretary Fowler Chairman Okun Aschheim Bach Burstein Cagan Chow Crouch Culbertson Dewald Fand Frazer Gaines Greenwlad Harris, S. Hauge Hester Hoadley Horwich Hosek Johnson Kent Leijonhufvud Madden Mayer McDonald Morrison Morton Noyes Pesek Ross Scott Sprenkel Stucki Teigen Walker Wallich Weintraub Wilde 1 Confined, however, to past actions. 2 Reporting, however, not essential if guidelines are issued; serve to check conformity with guidelines. Explain only (past) variations in monetary growth. 3 Putting aside consideration of the requirement that funds to operate the Fed be appropriated by Congress, we consider below the other structural provisions of H.R. 11. The case for the bill's other provisions would appear indisputable. To begin with rigorous analysis and hard empirical work can play no role in monetary policy as long as open market policy is set by general agreement of 19 men. Federal Reserve policy is eclectic in the kinds, means, and timing of monetary atcions because it is a consensus policy that is reached by softening and blending the opinions of all participants at open market committee meetings. In this connection, we note how few dissents there are to OMC decisions, a fact which supports that the aim of the OMC decision process is not to produce a socially optimal policy but rather to conceal differences of opinion. If monetary policy ever is to be based on validated theory the Federal Reserve's decisionmaking machinery will have to be overhauled. H.R. 11 provides necessary and sufficient streamlining by vesting all open market powers in a five-man Board of Governors. 25 Second, the Federal Reserve's many ties to the commercial banking business and to the New York money market tend inevitably to produce in our monetary authorities a limited and ofttimes dangerously deceptive view of what is in the national interest and how best to achieve these goals. Commercial bank members of the Fed elect two-thirds of their Reserve bank's directors and not surprisingly there is a strong banking orientation among those chosen to direct the affairs of the Reserve banks and select their presidents—men who now serve on the Federal Open Market Committee. In addition, the Federal Reserve serves and supervises commercial banks in a variety of ways from clearing checks to rediscounting eligible paper. Also, the Federal Reserve is the largest transactor in Government securities doing business on a day-to-day basis. Inescapably Federal Reserve authorities obtain a substantial proportion of their information and feel about the economy's trends and problems from members of the banking community and transactors in the money market. More important, they get an exaggerated notion of the remedial effects of using monetary policy to solve the problems encountered by loan and other bank officers, bank examiners, and money market technicians and transactors. This is an intellectual form of myopia: viz, that the problems of the banking business and money market are problems the monetary authority must solve and to which they must furthermore give the highest priority. It is not a sound working hypothesis for the exercise of monetary policy. The principal operating mistake deriving from the Federal Reserve's ties to commercial banking and the money market is that too much attention is given to interest rates, free reserves, and other money market and credit variables, and too little is paid to the money supply. The money supply has behaved erratically because the growth of M has emerged as a byproduct of the Federal Reserve's emphasis on credit variables and especially interest rates. In the 1953-60 period low monetary growth often was consistent with the Federal Reserve's interest rate targets in this period. At times in the years after 1953 the achieved low monetary growth doubtless was desired; for example, in the second half of 1956. But at other times; for example, in the fall and winter of 1957-58, it was not. More recently rapid monetary growth has been consistent with the Federal Reserve's interest rate and other credit targets. Because interest rates have been high and free reserves low by historical standards the Federal Reserve has been deceived lately (mid-1967 to mid1968) into thinking it has been following a tight money policy. But in fact the thrust of policy judged by changes in money supply has been aggressively expansionary, and inflation of both prices and interest rates has resulted. A change in the priority target of monetary policy definitely is in order. But it is naive and romantic to believe that under present structural arrangements the Federal Reserve authorities will make money supply their target the variable for transmitting its actions to the economy at large and achieving the goals of the Employment Act. We cannot expect money to be the target of monetary policy as long as the Federal Reserve's ties to the banking business remain in force. Further, we cannot expect appropriate coordination of monetary and fiscal policies as long as the members of the Board of Governors, by reason of their 14-year terms and the lack of effective appoint21-570—68 3 26 ment control by the incumbent President^ have no financial or political incentive to correct their mistakes and misconceptions. The restructuring provisions of H.R. 11 do not guarantee a sensible and sound monetary policy but unless they are adopted, sensible and sound monetary developments will emerge only as happy accidents. The national interests can be more rationally implemented. TABLE IV.—TABULATION OF VIEWS ON THE STRUCTURE OF THE FEDERAL RESERVE SYSTEM Respondents' votes Retiring the capital stock For Chairman Okun Arlt Aschheim Cohen Crouch Davidson Dewald Earley Fand Fish man Frazer Friedman Gaines Hauge Havrilesky Hester Horwich Hosek Johnson Kane Keiser Kent Keyserling Leijonhufvud Macesich Madden Mayer Melitz Meltzer Morrison Morton Pesek Ross Scott Teigen Thorn Voorhis Wallich Warburton Weintraub Whittlesey Yeager Decreasing the number of Governors and their t e r m s 1 Against Governor Martin Secretary Fowler Adams Bach Fellner Hoadley McCracken McDonald Noyes Sprenkel Walker See footnotes at end of table. For Secretary Fowler Chairman Okun Aschheim Bach Brunner Burstein Chow Cohen Crouch Dewald Earley Eckstein Fand Fishman Frazer Friedman Harris, S. Havrilesky Hosek Johnson Kane Keiser Kent Keyserling Leijonhufvud Mayer McCracken Melitz Meltzer Morrison Ross Stucki Teigen Thorn Voorhis Wallich Warburton Weintraub Whittlesey Wilde Yeager Against Governor Martin Adams Arlt Fellner Gaines Hauge Hester Hoadley Horwich Madden McDonald Minsky Morton Noyes Pesek Scott Sprenkel Walker Making the term of the FRB Chairman coterminus with that of the President For Governor Martin Secretary Fowler Chairman Okun Adams Arlt Aschheim Bach Brunner Chow Cohen Crouch Davidson Dewald Earley Eckstein Fand Fishman Friedman Gaines Harris, S. Hauge Havrilesky Hester Johnson Kane Keiser Kent Keyserling Leijonhufvud Mayer McCracken McDonald Melitz Meltzer Minsky Morrison Ross Scott Stucki Teigen Voorhis Walker Wallich Warburton Weintraub Whittelsey Yeager Against Burstein Fellner Hoadley Horwich Madden Morton Noyes Pesek Sprenkel Thorn 27 TABLE IV.—TABULATION OF VIEWS ON THE STRUCTURE OF THE FEDERAL RESERVE SYSTEM—Continued Respondents' votes Auditing Federal Reserve spending For Aschheim Chow Cohen Crouch Dewald Fand Fishman Frazer Friedman Gaines Greenwald Harris, S. Hauge Havnlesky Hosek Johnson Keiser Keyserling Leijonhufvud McCracken McDonald Melitz Meltzer Morrison Stuck! Teigen Voorhis Warburton Weintraub Wilde Yeager Against Governor Martin Secretary Fowler Chairmau Okun Adams Arlt Bach Brunner Burstein Earley Hester Hoadley Kent Madden Mayer Minsky Morton Noyes Pesek Ross Scott Sprenkel Thorn Walker Wallich Whittlesey Providing that funds to operate the Federal Reserve be appropriated by Congress For Chow Cohen Crouch Davidson Dewald Fand Fishman Friedman Harris, S. Hosek Keiser Keyserling Leijonhufvud Melitz Meltzer Morrison Morton Voorhis Yeager Against Governor Martin Secretary Fowler Chairman Okun Adams Arlt Aschheim Bach Brunner Burstein Earley Frazer Gaines Greenwald Hauge Havrilesky Hester Hoadley Horwich Johnson Kent Madden Mayer McCracken McDonald Minsky Noyes Pesek Ross Scott Sprenkel Stucki Teigen Thorn Walker Walhch Warburton Weintraub Whittlesey Wilde 'Not necessarily exactly as provided by H.R. 11. We note here also that some respondents expressed their own ideas on restructuring the Federal Reserve System. Brofenbrenner stated that the restructuring proposals of H.R. 11 were "matters of subsidiary importance. I should, instead be interested in procedures for identifying and disciplining members of the Board of Governors, or subsidiary staff members, responsible for egregious and continued breaches of the proposed monetary rule." (Brofenbrenner proposed that monetary growth be calculated each year, based on expected labor force, output and velocity changes, and that exchange rates be competitively determined.) Grossman was against Federal Reserve independence but did not specify structural changes or comment on the provisions of H.R. 11. Luckett favored making the Secretary of the Treasury ana Chairman of the CEA voting members of the Federal Reserve Board. Last, several respondents observed that the restructuring details in H.R. 11 would be unnecessary if there was a clearly defined monetary growth strategy which the Federal Reserve was instructed to follow, including Culbertson, Friedman, Leijonhufvud, and Yeager. REPLY OF HON. WILLIAM McC. MARTIN, CHAIRMAN, BOARD OF GOVERNORS OF THE FEDERAL RESERVE SYSTEM, FOR THE MEMBERS OF THE BOARD OF GOVERNORS AND THE RESERVE BANK PRESIDNTS OF THE FEDERAL RESERVE SYSTEM BOARD OF GOVERNORS OP THE FEDERAL RESERVE SYSTEM, Washington, September 9,1968. H o n . WRIGHT PATMAN, Chairman, Committee on BamJc and Currency, House of Representatives, Washington, D.C. DEAR MR. CHAIRMAN : Each of the members of the Board of Governors and each of the presidents of the Reserve banks has received your letter of July 1968 regarding plans of the Subcommittee on Domestic Finance to hold hearings on H.R. 11 later this year, and requesting views by September 1 on certain questions pertaining to monetary policy guidelines and open-market operations, some aspects of the structure of the Federal Reserve System, and recent monetary developments. The members of the Board and the Reserve bank presidents have each considered your questions. As you know, most of these issues have been raised on previous occasions and have been carefully reviewed within the System. The members of the Board and Reserve bank presidents concluded that for the present review and in view of the time limit prescribed it would be suitable to join in submitting to you a single document, a copy of which is enclosed. The enclosed answers reflect the views generally held by the members of the Board and the presidents, although understandably some of us may have different shadings of view and emphasis on some points. Sincerely yours, W M . MCC. MARTIN, Jr. STATEMENT OF W I L L I A M McC. MARTIN, JR., CHAIRMAN OF T H E BOARD OF GOVERNORS OF T H E FEDERAL RESERVE SYSTEM FOR THE MEMBERS OF T H E BOARD AND T H E RESERVE B A N K PRESIDENTS Question I J. Do you believe that a program coordinating fiscal, debt management, and monetary policies should be set forth at the beginning of each year for the purpose of achieving the goals of the Employment Act, or alternatively should vie treat monetary and fiscal policies as independent mutually exclusive stabilization policies? Answer. It is important that monetary policy and fiscal policy be coordinated in the promotion of our national economic goals. Pursuant to the Employment Act of 1946, early each year the President transmits to the Congress, among other things, an economic report, a review of the economic program of the Federal Government and a program for carrying out the policy declared in the act, together with such recommendations for legislation as he may deem necessary or desirable. In addition, the Council of Economic Advisers presents its appraisal of the various programs and activities of the Federal Government and its recommendations regarding national economic policy. (29) 30 The President's program customarily specifies the fiscal actions needed, in his judgment, to achieve the goals of the Employment Act, and often notes the monetary policies that he believes would appropriately accompany the proposed fiscal and debt management actions. Any economic program submitted must of necessity be provisional. While the art of economic projection has progressed significantly in postwar years, the possibility of error in forecasting the timing and detail of actual economic performance—and, on occasion, in the whole profile of developments to come—remains disturbingly high. Also there is a high incidence with which unexpected events having major economic implications take place—international political and economic disturbances, civil disorders, strikes, and the like. It must also be emphasized that any proposals with respect to future monetary policies must also be provisional since the choice of appropriate monetary policy will be contingent on the extent to which actual economic developments conform to those projected, and on the extent to which actual fiscal and debt management actions conform, in both substance and timing, to those proposed in the President's program. Finally, any overall stabilization program must recognize the inherent advantages and disadvantages of alternative economic tools of public policy. Some policy instruments are capable of gradual and continuous shadings in degree of impact, while others require specific actions involving time-consuming procedures. The major influence of some on the economy appears only with a considerable lag, others achieve their influence more promptly. The nature of the linkages is both variable and imprecise. Generally, massive present or prospective economic imbalances are best dealt with through adjustments in fiscal policy. The distortions introduced when monetary policy is pushed to extremes—in terms of effects on financial values, investment incentives, potential cyclical instability, and international relationships— are large. But major changes in the intensity of fiscal stimulation or restraint are not everyday occurrences; they take time to plan, enact, and implement, as experience with both the tax cuts of 1964 and the surcharge of 1968 attest, and, once made, they are not quickly reversible. Similarly, expenditure programs—based on a history of political determination of social and national needs—are not usually susceptible to abrupt and reversible changes of pace. Monetary policy, on the other hand, is well suited to rapid and marginal response to the emerging requirements of the economy. It is continuously under review and subject to gradual, flexible, and even reversible adjustments. It is the very essence of monetary policy that it can respond to the unexpected development and that it can adjust for divergencies between unfolding economic events and projections. Given the lags involved in some of the effects of monetary policy, it is important that it be free to respond to the earliest indicators of a need for action. While it is possible to describe, in general terms, the profile of monetary policy that would be consistent with a given economic projection, and that, in combination with an appropriate fiscal program, should help to achieve the Nation's overall economic goals, it does not seem desirable to specify in advance the precise combination of stabilization tools. In particular, it would seem most unwise to com- 31 mit monetary policy in advance to a particular course of development without regard to the varying and frequently conflicting economic and financial tendencies—domestic and foreign—that do in fact emerge with the passage of time. To do so would rob it of the very flexibility and adaptability that constitute the unique contribution of monetary policy to the economic stabilization instruments at the Government's disposal. Question L2. If you believe a program should be specified, do you believe that the President should be responsible for drawing up this program, or alternatively should such responsibility be dispersed between the Federal Reserve System and agencies responsible to the President? {Please note that informal consulting arrangements can be made as desired whether responsibility is assigned to the President or divided between the President and the Federal Reserve. The concern here is with the assignment of formal responsibility for drawing up the economic program.) Answer. The responsibility for recommending to the Congress changes in Federal expenditure and revenue programs clearly rests with the President. Suggestions and advice may be sought from interested Federal agencies as to specific content, of course, and frequently the Federal Reserve has contributed to this process. In the President's report there often is reference to monetary as well as fiscal policy, and the Council's report customarily discusses monetary policy developments at some length. We believe that such references are wholly appropriate, in view of the importance of financial developments to economic conditions generally, and in recognition of the role of monetary policy in the Government's economic stabilization effort. Views as to what constitutes appropriate monetary policies must of necessity be provisional for the reasons stated in answer to question 1.1, but such policies must be taken into account as an important factor conditioning, and conditioned by, the economy's prospects. We believe, however, that any specifications as to monetary policy should continue to be regarded in the nature of suggestions of what constitutes appropriate policy under clearly stated assumptions— which may or may not prove correct—rather than as instructions to the Federal Reserve System. The System was created by Congress, and is answerable for its actions to the Congress; its role is that of advising and cooperating with the executive branch of Government in the public management of economic affairs, without being formally a part of it. This division of responsibilities in the field of economic policy is one of the desirable checks and balances of our system of government, and we do not believe that the Congress should cede its ultimate authority in the monetary sphere to the executive branch. Question 1.3. Concerning monetary policy guidelines: A. Should monetary policy be used to try to achieve the goals of the Employment Act via intervention of money supply {defined o,s desired) as provided in H.R. 1U or alternatively should H.R. 11 be amended to make some other variable or variables the immediate target of monetary policy; for example, interest rates, bank credit, liquidity, high-powered or base-monkey, total bank reserves, excess reserves and free reserves? Please define the target variable or combination of variables recommended and state the reasons for your choice. {If desired, recommend a target variable 32 or variables not listed here.) It would be most helpful if, in providing the reasons for your choice., you list the actions the Federal Reserve should take to control the target variable (or variables) and also explain the link between your recommended target of monetary policy and the goals of the economy as defined by the Employment Act. B. Should the guidelines of monetary policy be specified in terms of some index of past, present, or future economic activity, or alternatively in terms of the target variable's value or growth? For example, should the President's 1969 program for achieving the goals of the Employment Act be formulated to require consistency with some set of overall indicators of economic activity, or alternatively so that your target variable attains a certain value or growth regardless of the economic winds? Please indicate the reasons for your preference. G. For only those persons who recommend that some index of economic activity be used to guide the monetary authorities in controlling the target variable: Should we use a leading (forivard looking), lagging (backward looking), or coincident indicator of economic activity? It would be most helpful also if you would identify the index you would like to see used and specify how the target variable should be related to this index. D. For only those persons who recommend that the guidelines be put in terms of the target variable's vafoie of growth: Should the same guidelines be used each year into the foreseeable future, or alternatively, should new guideline be issued at the beginning of each year conditioned on expected private investment, Government spending, taxes, etc.? Please indicate the reasons for your preference. E. For only those persons who recommend that the guidelines be put in terms of the target variable's value or growth and who also recommend that the same guidelines be used year after year into the foreseeable future: What band of values or range of growth do you recommend? (By way of clarification, a band of values appears appropriate if your target variable is, say, free reserves, whereas a range of growth is appropriate if it is, say, money supply.) F. For all those persons recommending that the guidelines be put in terms of the target variable's value of growth (regardless of whether you recommend using the same guidelines year after year or revising them each year in light of expected private investment and fiscal policy): Under what circumstances, if any, should the monetary authorities be permitted during the year to adjust the target variable so that it exceeds or falls short of the band, of values or range of growth defined by the guidelines issued at the beginning of the year? Answer: Sumvmary.—In a dynamic and complex economy, with a particularly dynamic and complex financial system, monetary policy cannot be formulated solely in terms of the behavior of any single financial variable or any single class of variables. Over the postwar decades, there have been major shifts in the financial structure and financial environment: shifts in savers' preference among the rapidly proliferating 33 variety of financial assets available through institutions and financial markets; changes in borrowers' dependence on particular sources and types of credit; changes in spheres and intensity of competition among financial institutions; and shifts in emphasis in the monetary/fiscal mix of stabilization policy. In light of these major structural and behavioral changes, it would have been most unwise to have determined policy targets and objectives solely in terms of desired levels or changes in a particular financial quantity or a particular financial price. Policy decisions have always been based, and must continue to rest, on assessments of the impact of policy changes on a wide range of financial markets and institutions, and on interpretations of the significance of these changes for the ultimate goals of policy relating to employment, prices, growth, and international equilibrium. Background.—The philosophy of economic stabilization policy that has developed in the United States over the past several decades increasingly has come to recognize that governmental policies can moderate the business fluctuations normally experienced in a dynamic economy. The Employment Act of 1946 envisaged that the Federal Government had clear responsibilities for adopting stabilization policies that would temper economic fluctuations, and thereby foster conditions conducive to the attainment of high-level employment and output, and the maximum rate of economic growth that can be sustained without inflation. The task of realizing the goals set forth in the Employment Act of 1946 is not, of course, an easy one. The sources of disturbance to sustainable, noninflationary economic expansion are numerous. The sources of instability often are difficult to identify, in particular when they are associated with shifts in spending propensities in the private sector. Disturbances originating in the monetary and financial sectors of the economy are also potential initiating sources of economic instability. For example, shifts in the public's financial asset preferences— between currency and bank deposits, between classes of bank deposits, or between bank deposits and other types of financial assets—may lead to disruptive changes in financial market conditions. At the same time, the possibility always exists that central bank policies could themselves be an initiating source of economic instability. The tools of monetary policy are powerful and must be administered with care if our economic objectives are to be achieved. The Employment Act of 1946 did not in fact attempt to prescribe any specific policies or techniques for achieving the goals it seeks. Wisely, it recognized that stabilization policies would have to be adapted to the needs of an ever-changing economy and that—in any case—our understanding of economic fluctuations, and how to moderate them, had not reached the stage at which the precise amount and combination of monetary-fiscal policies needed to assure stable economic growth could be really determined. We have learned much during the years since its passage about what can be accomplished with timely application of the tools of economic stabilization. One important lesson has been that there is no simple touchstone by which to guide the conduct of monetary policy. In an economy as dynamic as ours, no single measure of monetary stimulus or restraint has been sufficient to serve adequately as an 34 exclusive indicator of what monetary policy has been, or as an exclusive guide to what it should be. The effects that stem from any given monetary policy depend fundamentally on private reactions to the policy, and these reactions are not static. They change over time, partly because of the adaptive behavior of the private economy to the policy measures themselves. Similarly, the international impact of domestic policy measures, including monetary polic}^ actions, cannot always be gaged precisely. There have been considerable changes in the structure of financial markets and in financial market responses to monetary policy during the postwar years. They have affected significantly the variables that the Federal Reserve must be concerned with in its assessment of monetary policy and its effects on economic activity and the international position of the dollar. During the early years of the postwar period, Federal Reserve policies were directed principally at preventing variations in the prices and yields of Treasury securities, rather than at the more fundamental objectives of economic stabilization. Part of the excess liquidity inherited from World War I I was worked off during this period; nontheless, when monetary policy turned from pegging bond prices to accomplish the objectives of stabilization policy in 1951, its operations for a number of years took place in the context of a financial climate that heavily reflected the influence of the enormous wartime buildup of liquid assets in the hands of the public and of financial institutions. During the first decade of the postwar period, therefore, the effects of Federal Reserve policies on financial institutions were confined more heavily to the commercial banks—and, indeed, to the impact on commercial bank demand deposits—than has subsequently been the case. Time deposits of commercial banks, during those years wrere held primarily by small savers whose financial asset holdings were relatively insensitive to changes in monetary policy. Major nonbank financial institutions, with ample amounts of Treasury securities to sell in order to meet private credit demands, felt only moderate effects of changes in monetary policy on the growth of their resources. The complexity of financial behavior that began to develop in the latter part of the 1950's further complicated the task of central banking. On the one hand, financial institutions have become much more aggressive in their competition for funds, largely as a consequence of the progressive decline in their liquidity since the end of World War II interacting with mounting demands for credit to finance expenditures. This increased competition has resulted in markedly higher interest rates paid by the various competing institutions on their deposits or shares, and it has also produced a diversification in types of instruments offered by the institutions for the financial investor to hold. Another development of major importance in financial markets relates to the increased sensitivity of financial investors to considerations of yield in the placement of their financial savings, and their growing willingness to substitute among a wider range of financial assets. As a consequence, the monetary authorities have had increasingly to take into consideration the effects of changes in policy on a broad range of financial assets, including savings and loan shares, mutual savings bank deposits, time deposits of commercial banks, policy loans of 35 insurance- companies, and intermediate- and long-term securities issued by the Federal Government, by States and municipalities, and by corporations. In particular, monetary policy decisions have had to take into account the potential effect of variations in time deposit growth on creditfinanced spending, and hence on the pace of economic expansion. The accelerated growth in time deposits that has normally accompanied an increase in rates paid by banks, for example, to some degree represents funds that otherwise might have been invested in market securities, or in the deposits and shares of nonbank thrift institutions. To that extent, the effects of larger supplies of funds made available to borrowers by commercial banks have been offset by the lesser rise in funds supplied to credit markets by nonbank intermediaries, or directly by nonfinancial businesses and consumers. But to some degree, the accelerated growth of time deposits reflects direct substitutions by businesses and consumers of time deposits for demand balances in their liquid asset portfolios. Substitutions of that kind, since reserve requirements are lower on commercial bank time deposits than on demand balances, permits commercial bank credit to grow without a corresponding reduction in credit supplies through other channels. To that extent, the effect of increased time deposit growth rates is expansionary. The significance of any given change in the growth rate of time deposits, therefore, depends on whether it does or does not imply changes in the total supply of credit, and in credit cost and availability that are detrimental to the maintenance of economic stability. And these effects in credit markets depend, in turn, on the factors motivating the change in the public's demand for time deposits during any particular period of time. The spreading of the effects of monetary policy beyond the commercial banking system to include the major nonbank financial institutions has complicated the problems of monetary decisionmaking still more, although the substantive issues involved are similar to those to be dealt with in connection with commercial bank time deposits. It has become amply evident since 1966 that we cannot ^SFord to exclude the major nonbank thrift institutions from consideration in formulating monetary policy. Flows of deposits and shares to these institutions, and hence the amount of credit supplied by them to finance spending, tend to be reduced markedly by policies of monetary restraint that lead to increasing yields on substitutes for the liabilities of these institutions, since the ability of these institutions to increase the rates they pay is limited by the fact that their current incomes respond very slowly to changing market rates of interest. Similarly, easing conditions in the money and capital markets stimulate inflows into these institutions. Because these institutions are heavily specialized in supplying funds for homebuilding, such variations in flows of funds through them can have major effects on residential construction. The home-building industry is heavily dependent upon a steady flow of mortgage money from nonbank intermediaries. The growth rate of credit supplied to borrowers through these nonbank intermediaries need not be associated closely with growth rates of the money stock, or of commercial bank time deposits, or of total bank reserves, or of other variables that are at times suggested as sufficient guides for the conduct of monetary policy. Indeed, at critical times 36 in the recent past it has not. During the early months of 1966, for example, the effects of monetary policy working through the inflows to nonbank thrift institutions began to exert a major restrictive impact on the supply of mortgage money and hence on housing starts. The money stock, on the other hand, continued to grow quite rapidly through April. To establish monetary policy by fixing the growth rate of any single variable, ignoring such evidence as may be available on the effects of monetary policy through other channels, would be to court disaster. Still another complexity arising in the late 1950's and continuing throughout the 1960's has been the serious imbalance in the U.S. balance of payments. One effect has been the development of new financial instruments and markets, such as the Eurodollar market, in which the effect of domestic policy actions are registered. It is sometimes suggested that the way around the problems posed by the increasing complexity of financial market behavior is to adopt even broader definitions of "money," in the hope that a definition might be found that would somehow cope with the broader influence of monetary policy in financial markets. Undoubtedly, monetary processes are better understood by expanding analytic horizons to include variables other than currency and demand balances in our efforts to understand the effects of monetary policy on the economy. But since the interpretation of changes in nonmonetary financial asset holdings depends on an understanding of the sources from which funds flow into these assets, and the reasons for these flows, we cannot expect to develop an adequate guide for the conduct of monetary policy simply by the construction of a new definition of money. Our monetary history does not indicate that there is any single financial asset, or combination of financial assets, that uniquely satisfies the public's liquidity preference. As noted earlier, the significance to be attached to the growth of the public's holdings of particular kinds of liquid assets, and combinations of them, depends importantly on whether the associated changes in credit market conditions are in the interests of economic stability. Changes in interest rates and in other dimensions of loan contracts, can therefore provide useful information on the course of monetary policy. Credit market conditions must always be given close attention in establishing guidelines for monetary policy, since it is through the credit markets that monetary policy has its most direct effect on spending. But like changes in the money stock, changes in credit market conditions are partly the result of Federal Reserve policy, and also partly the result of decisions of commercial banks, of nonbank financial institutions, and of nonfinancial businesses and consumers. For that reason credit market conditions cannot be an exclusive guide in the formulation of policy decisions. In seeking guidance for the conduct of monetary policy, therefore, incoming information on both financial quantities and financial prices must be assimilated and interpreted. Movements in financial quantities—such as total bank reserves, the money, stock, commercial bank time deposits, and claims against nonbank intermediaries—on the one hand, together with indications of cost and availability of credit— including interest rates and nonprice terms of credit—on the other, must be evaluated jointly to assess what effects monetary policy cur- 37 rently is having on the total supply of funds, its distribution among the various sectors of the economy, and hence on the availability of funds to finance spending. This interpretation must, of necessity, seek to evaluate the behavior of financial variables in light of underlying real developments in markets for goods and services. It is particularly important to distinguish between the variations in demands for and supplies of credit that are produced by changes in decisions to spend on goods and services, and those associated with the public's desires to rearrange financial asset portfolios, corporate mergers, and similar transactions. Decisions giving rise to the first kind of variation in credit conditions can lead directly to economic instability. The latter class of decision does not directly alter the pace of economic expansion, but the resulting side effects in financial markets may do so. The appropriate monetary policies to be followed, in response to an observed variation in credit demands or supplies, depend on which of these two classes of decisions is responsible. In the final analysis, evalution of whether monetary policy has contributed positively to economic stabilization cannot be judged simply on the behavior of financial variables, no matter how carefully they are intepreted. The ultimate test of monetary policy is the extent to which it has succeeded in promoting our national economic goals of maximum sustainable economic growth, maximum practicable employment, reasonable price stability, and a strong dollar at home and abroad. Question H. Concerning debt management policy: Given the goals of the Employment Act what can debt management do to help their implementation? (If you believe that debt management has no role to play in this matter, please explain why.) Answer. As a stabilization tool, debt management can play a useful although circumscribed, role in implementing the goals of the Employment Act, as a complement to fiscal and monetary policies. In the early 1960's, for example, debt management contributed to maintenance of upward pressures on short-term interest rates for balance-of-payments purposes, thus giving monetary policy somew^hat greater flexibility for adapting its policies to domestic credit needs. But a number of considerations, including budgetary and legal constraints and the need for a balanced debt structure, tend at most times to limit the contribution that debt management can make to economic stabilization. In any event, any contribution of debt management may be overweighted by ongoing fiscal and monetary policies, which between them tend to have larger, more pervasive, and more sustained effects on interest rates and credit availability. Debt management policies are those related to the structure—primarily the maturity composition—of the outstanding publicly held Federal debt. Shifts in the maturity composition of the Federal debt, which is the main aspect of debt management, alter the liquidity of the debt and/or the term structure of interest rates. Market expectations as to the future course of interest rates also play an important role in affecting the term structure. In the degree that changes in the relative supply of short- and longterm securities affect the interest rate structure and the availability of funds, they will have some impact on financial flows and private 38 spending in various sectors of the economy. In particular, the distribution of liquid assets in the hands of commercial banks and elsewhere in the economy may have a considerable effect on the nature and timing of the responses of the financial markets to monetary policy. Changes in market interest rates, particularly short- and intermediateterm rates, may affect the ability of savings and loan associations and mutual savings banks to compete for the public's savings, and thereby influence the cost and availability of credit for housing. Changes in short-term interest rates also particularly affect the ability of major commercial banks to attract funds from corporations through large negotiable time certificates of deposit, and influence bank ability to finance other businesses. On the other hand, a debt management policy that stresses debt lengthening operations would affect the availability of funds for investment in State and local government and corporate bonds, since insurance companies, trust funds, and others who purchase long-term Government debt may be using funds that would otherwise be placed in private markets. Because various segments of the domestic credit market, as well as international flows of funds, may be affected by debt management decisions, the debt managers are always faced with the need to evaluate how the cash raising and refunding problems connected with the Government debt interact with economic and credit trends in the country and how they phase in with fiscal and monetary policies. There are, however, important technical problems of orderly debt management procedure which sometimes tend to conflict with economic goals. It is generally desirable to maintain a balanced debt structure— with maturities reasonably spaced and not excessively large at any one time—so as to avoid the continuous or awkwardly large refunding operations that might tax the market's capacity to absorb Treasury issues, given the steady flow of private issues into the market. Since the public debt continuously shortens with passage of time, a balanced debt structure requires the Treasury to be alert to opportunities for debt lengthening operations. These operations are most feasible in periods of declining interest rates (when rising bond prices make them an attractive investment to hold). However, when interest rate declines are associated with an undesirable slackening of economic activity, the economic goals of the country may indicate the desirability of keeping debt lengthening to moderate dimensions so as to encourage investors to lend more to finance capital outlays of private sectors of the economy. Thus, considerations aiming at achieving an appropriate debt structure must be reconciled with the objectives of the Employment Act. It must also be kept in mind that the flexibility of debt management in maintaining a balanced debt structure is limited by the 414-percent interest rate ceiling on Treasury bonds, which eliminates sales of long-term debt at times when the market may be receptive, and when the absorption of savings into long-term Treasury issues would be consistent with economic stabilization goals. Question I.S.A. H.R. 11 makes no provision whatever for conducting open market operations for so-called defensive or road-clearing p\irposes, that is to counteract seasonal and other transient factors affecting money market and credit conditions. Do you see any merit in using open market operations for defensive purposes or should they be used only to facilitate achievement of the Presidents economic pro- 39 gram and the goals of the Employment Act? What risks and costs, if any, must be faced and paid if open market transactions are used to counteract transient influences ? Answer. The Federal Reserve System must be concerned both with providing an efficient monetary system, which handles routinely the multiplicity of daily payments of economic life, and with fostering economic growth at a high level of employment while seeking to maintain the purchasing power of the currency at home and abroad. Section 12A of the Federal Reserve Act points up this dual responsibility and focus of open market operations in the following language : 3. Governing principles.— (c) The time, character, and volume of all purchases and sales of paper described in Section 14 of this Act as eligible for open-market operations shall be governed with a view to accommodating commerce and business and with regard to their bearing upon the general credit situation of the country. Indeed, the two aims of a smoothly f u n c t i o n i n g monetary mechanism on the one hand, and a monetary policy geared to the achievement of national economic goals on the other, arc not readily separable. An efficient monetary system is needed if monetary policy is to be used effectively, for if the financial markets are to respond as intended to national policy actions the mechanism must be protected from shortrun swings in such factors as the public's demand for currency, the speed of check collections, international gold or currency flows, or the size of Treasury balances held with the Federal Reserve banks. A striking recent example is the statement week encompassing the July 4 holiday in 1968 when open market operations routinely compensated for the $740 million outflow of currency into the hands of the public. Another important example occurred recently when the United Kingdom repaid its short-term swap drawings by funds obtained through the International Monetary Fund, causing an absorption of £700 million of member bank reserves which had to be offset to maintain the overall policy posture. So-called defensive operations, then, simply prevent operating transactions from interfering with the effective implementation of monetary policy. In our flexible financial system the Federal Open Market Committee directs open market operations to aim at maintaining a degree of ease or pressure on the banking system that is intended at the same time (1) to foster monetary and credit conditions appropriate to national economic objectives, and (2) to insulate the monetary system from the effects of various factors, including temporary influences, that are unrelated to policy. In this way, changes in monetary and credit conditions over a period of time flow from conscious decisionmaking on the part of the Federal Open Market Committee, rather than being subject to sharp up-and-down swings as a result of short-term or other special influences that have no policy relevance. To exert its desired influence over the growth path of the banking system's reserve base, the Federal Reserve must take account of all the forces affecting reserves. If the central bank permitted wide week-to-week fluctuations in reserve availability to take place, and did not attempt to offset those fluctuations as is done now, the cost would be considerable. Fluctuations in reserves would generate changes in bank credit that might well be perverse from the standpoint of monetary policy objectives. Another highly probable effect would be a sharp increase in short-term varia- 40 tions in interest rates, as markets generally sought to adjust to quick ebbs and flows in reserve availability. In order to compensate investors for the great uncertainty of sharply fluctuating interest rates, it is likely that average levels of interest rates, particularly short-term rates, would tend to be higher than otherwise. The U.S. Treasury, as the largest borrower in the short-term market (with over $60 billion of bills outstanding), might bear a particularly heavy share of the higher interest costs. While the question cited above refers to the "risks and costs" of using open market operations to counteract transient influences, it would rather seem to be the case that the significant risks and costs lie on the side of not using oj>en market operations for defensive purposes. Nor is the task of mapping out defensive operations, and executing them, a significant drain on resources that could otherwise be better employed. For as developed above, the planning and execution of these operations is in practice inseparable from, and essential to, the carrying out of operations designed to achieve national economic objectives. Question I.5.B. Do you believe that monetary policy can be effectively and efficiently implemented solely by open market operations? Answer. Sole reliance by the Federal Reserve on open market operations in the conduct of monetary policy would greatly reduce the effectiveness and flexibility of such policy. Even if reserve requirements were not subject to change and discounting were abolished, the System would of course still be able to influence the volume of bank reserves through open market policy. Nevertheless, there are many situations in which the conduct of policy is greatly improved by the availability of the other policy instruments and some situations that can properly be treated only through the use of these other instruments. Open market operations are the preferred technique for day-to-day operations and, in many situations, as a vehicle of policy change. The special advantages of the other instruments and the situations in which they assist the development of effective monetary policies are developed below in answer to question I.5.C. Question I.5.C. For what purposes, if any, should (a) rediscounting, (b) changes in reserve requirements, and (c) regulation Q be used? How might H.R. 11 be amended to implement your recommendations? Answer, (a) Rediscounting.—Discounting and changes in the rate charge on discounts constitute the oldest tool of monetary policy and the one whose use is most widespread among the world's central banks. The discounting mechanism permits the performance of the central bank's role of lender of last resort and allows a broader variety of debt to be monetized than if only open market operations were permitted. From the point of view of the individual bank it provides a means of meeting temporary reserve needs which frequently, in the nature of the banking business, are unforeseen. Second, the existence of rediscounting provides a means through which the Federal Reserve can supply reserves immediately and directly to the banks under pressure. Open market operations do not provide such assurance. Third, the existence of a discount mechanism cushions the impact of open market operations not only on individual banks but also on the money market generally, and thus permits such operations to be undertaken more aggressively without fear that they will have seriously disruptive effects. Fourth, the existence of a discounting mechanism is an important di- 41 rect channel of communication between the Federal Reserve and its member banks which increases the System's knowledge of trends and developments in the market and in the banking system. Changes in the discount rate are an important instrument of monetary policy. The precise role of discounting as a part of policy in the future depends upon what actions may be taken in light of the reappraisal of the discount mechanism currently underway in the System. But as long as a discount facility exists there must be a discount rate and policies with regard to changing its level. There have been occasions m the past, and may well be m the future, when the decisive influence of a change in the discount rate on market psychology, interest rates, and expectations generally, can be quite useful. This might perhaps be especially the case when dealing with problems in foreign exchange markets where clear-cut and massive action is sometimes required to stem the tide of adverse developments. Most of the major central banks of the world have indeed used discount rate changes as a principal means of dealing with foreign exchange market problems. It might be noted, incidentally, that the usefulness of discounting would be increased if proposed legislation removing the technical requirements for the eligibility of paper for discounting were enacted. (i) Changes in reserve requirements.—In principle, any change in the overall credit-creating capacity of the banking system that can be accomplished through changes in reserve requirements could also be accomplished through open market operations. There are at least four situations in which reserve-requirement changes may have particular advantages. First, special situations might require a massive and immediate tightening or easing of bank reserve positions. Such situations are hardly likely to be frequent and, indeed, are difficult to spell out; but, as long as the possibility exists, there are obvious advantages in holding changes in reserve requirements available for use. A second advantage of reserve-equipment changes, in some circumstances, is the fact that they impinge immediately on every member bank, whereas open-market operations impinge first on banks in the money market centers, with the influence spreading only gradually to the rest of the banking system. Third, changes in reserve requirements may be designed to have specific effects on the composition of bank assets and on the structure of interest rates. This sort of consideration has been an important reason for changes in reserve requirements during this decade, especially when it was desired to supply reserves without encouraging an outflow of volatile short-term funds or when it was desired to maintain flows of bank funds into mortgage finance in a context of overall credit restraint. Finally, a change m reserve requirements can be used in appropriate circumstances to give a clear indication to the public that the System intends to change the direction of policy or to pursue further an existing path. (c) Regulation Q.—We believe that the rates paid by financial institutions to attract funds ideally should be completely free to reflect market forces, and that healthy competition among financial institutions in this respect, as well as others, should be encouraged. The financial conditions of the past few years, however, have made interestrate ceilings unavoidable. High and rapidly rising interest rates have at times in recent years put great pressure on financial institutions such as the mutual savings banks and the savings and loan associations. The 21-570—68- 4 42 earning power of these institutions is limited by their necessarily heavy commitment in long-term assets bearing the lower interest rates characteristic at the time they were issued. This has limited the ability of these institutions to meet the competition of rising open-market rates by raising the rates they offer to savers on their own liabilities. Thus, for the proper functioning of these institutions and of the markets they serve, notably the mortgage market, it has been necessary to reduce the pressure on them by limiting the rates they are allowed to offer and, at the same time, by limiting the rates that could be offered by the commercial banks as their competitors. As long as our financial institutions are so varied in size, scope, and powers, it is not feasible to eliminate the power to establish ceilings on interest rates paid on time deposits when this appears necessary. Once the need for such ceilings is acknowledged, the need to change them from time to time must also be admitted in view of the substantial fluctuations that are often experienced in open-market rates. Under the circumstances, then, flexibility in setting these rates has been necessary to increase the efficiency of monetary policy and to protect the health of financial institutions. With regard to ways in which H.R. 11 might be amended the System has, on a number of occasions, recommended that the Congress modify the laws relating to "nonpar clearance" of checks, those limiting the System's flexibility in fixing reserve requirements, and those relating to the eligibility of member bank assets for discounting. These matters continue to deserve congressional consideration. Question I.5.D. Do you see any merit in requiring the Federal Reserve Board to make detailed quarterly reports to the Congress on past and prospective actions and policies? Are there any risks amd costs in this procedure? In what ways, if any, would you modify the reporting provision? What information do you believe should be included in such reports as you recommend the Federal Reserve submit to the Congress? Answer. The Federal Reserve welcomes opportunities for full and frequent interchanges of view with appropriate committees of the Congress regarding the discharge of its responsibilities for monetary policy. It does not see merit, however, in a legislative requirement for "detailed quarterly reports to the Congress on past and prospective actions and policies." The Board now makes public the records of recent policy actions of the Federal Open Market Committee, prepared for inclusion in the Board's annual report to the Congress, on a current basis throughout the year, with a lag of approximately one-quarter. Information on changes in discount rates and on changes in Board regulations, including those relating to reserve requirements and ceiling rates on time and savings deposits is, of course, released at the time of the actions. Discussions are currently underway with the Joint Economic Committee of the Congress regarding possible arrangements for quarterly reports by the Board to that committee. It is the tentative view of the Board that such reports, to be most useful, should include an analysis of all major monetary and financial developments of the preceding calendar quarter. In any case, the Federal Reserve believes that the purposes of such reports are most likely to be best served 43 if their content, frequency, and timing remain subject to modification over time m light of accumulated experience. A specific legislative requirement in this area would sacrifice the flexibility that may be important in insuring that the reports are of maximum usefulness. With respect to prospective policy actions, we believe that a legislative requirement for detailed quarterly reports would involve major risks and costs. Advance commitments as to policy would seriously damage the ability of the Federal Reserve to formulate and implement appropriate monetary policies. Furthermore, such commitments could possibly generate unwarranted expectations in financial markets, in which expectations play such an important role. Monetary policies are formulated by the Board and the FOMC in light of a wide spectrum of current information available at the time on current economic conditions, including data that are often preliminary, and on the economic outlook as best as it can be assessed at the time. But policies are modified when conditions are found to depart from expectations, and/or when the expectations themselves change. In the nature of the case, then, monetary authorities should not commit themselves on policy actions beyond the immediate future. The deliberations of the Board and the FOMC concerning the policy responses that might be appropriate, at later times, if events follow specified alternative courses, hinge on specific assumptions, and the range of alternative policy responses is often modified as economic developments unfold. Thus, it could be seriously misleading to the public for the Federal Reserve to present, at the beginning of a quarter, a detailed prospectus of future actions and policies when in fact the actual policies adopted would depend so heavily on the extent to which domestic and foreign developments within the quarter alter the System's assessment of future monetary and credit needs. Moreover, regular prognostication by the Federal Reserve regarding its future policy actions—such as would be involved in the proposed quarterly reports—would be likely to stimulate large anticipatory swings in financial market conditions. Market participants are themselves always speculating—in their actions as well as assessments—about the possible course of monetary policy and the prospects for particular policy actions. Such activity frequently has significant effects on short-run changes in financial market conditions, including interest rates. Market conditions might well come to be more strongly influenced by the System's quarterly statements than primarily by the basic underlying forces of supply and demand. This, in turn, would not only damage the ability of financial markets to perform their essential function of resource allocation, it would also interfere with the ability of the Federal Reserve to assess the underlying strength of market demands and supplies and to formulate policies appropriate to the basic domestic and international economic situation. Question I.5.E. What costs and benefits would accrue if representatives of the Congress, the Treasury and the CEA were observers at Open Market Committee meetings? Answer. Congress and the public are, of course, entitled to know what actions are taken as a result of the discussions at meetings of the FOMC, and the reasons for these actions. This information is made public in a variety of ways as promptly as feasible, as noted in the 44 answer to the preceding question (I.5.D). We are inclined to believe, therefore, that it would be unproductive for representatives of the Congress, the Treasury, and the Council of Economic Advisers to sit in at meetings at which, as observers, they would have no right to vote or otherwise participate actively, and the results of which are, in any event, made public m extensive detail. Neither does it appear likely that the proposal would improve on present procedures for coordinating monetary policy with fiscal and debt management policies. One obvious drawback to the proposal would be that the presence of such observers might inhibit the full and frank exchange of views that are essential to enable the Federal Open Market Committee to operate effectively. We believe that all similar bodies that are assigned comparable policy responsibilities are also given the opportunity to meet in private to discuss how best to carry out their responsibilities. This principle clearly extends to various regulatory agenices and commissions of Government, as well as to committees of Congress, and we think it should apply to the FOMC as well. Question II. H.R. 11 provides for the following structural changes in the Federal Reserve System,: 1. Retiring Federal Reserve bank stock; Reducing the number of members of the Federal Reserve Board to five and their terms of office to no longer than 5 years; 3. Making the term of the Chairman of the Board coterminous with that of the President of the United States; An audit for each fiscal year of the Federal Reserve Board and the Federal Reserve banks and their branches by the Comptroller General of the United States; and 5. Funds to operate the Federal Reserve System to be appropriated by the Congress of the United States. Please comment freely on these several provisions. In particular it would be most helpful if you would indicate any risks involved in adopting these provisions and discuss whether their adoption woidd facilitate the grand aim of H.R. 11, which is to provide for coordination by the President of monetary and fiscal policies. Answer. II.l. Retiring Federal Reserve bank stock. While ownership of Federal Reserve bank stock by member banks of the Federal Reserve System is not essential, there appears to be no compelling reason for eliminating such ownership. Such a change would involve the loss of some intangible but important advantages that result from such ownership of Reserve bank stock and could involve a risk of diminishing the effectiveness of the System's operations. In addition, retirement 01 Federal Reserve bank stock could be construed, both at home and abroad, as indicating a change in the structure and character of the Federal Reserve System. There is clearly no foundation for any assumption or inference that ownership of Reserve bank stock by member banks enables them to "control" the operations of the Reserve banks or to determine System policies. The true effect and the advantages of such ownership of Reserve bank stock were described in one of Chairman Martin's replies to the 1952 Patman Questionnaire: As a consequence of the public nature of the Federal Reserve banks, ownership of their stock does not carry with it the same attributes of control and financial interest usually attached to stock ownership in private corporations. The amount of Reserve bank stock which a member bank must own is fixed by law 45 in relation to the member bank's own capital and surplus. Such stock may not be transferred or hypothecated. Ownership of stock entitles the member banks to no voice in the management of the affairs of the Reserve bank other than the right to participate in the election of six of the nine directors of the Reserve bank. As the result of the election procedure prescribed by the Federal Reserve Act, each member bank votes for only two of the nine directors. Under the law, dividends on Federal Reserve bank stock are limited to 6 percent per annum; and in the event of the liquidation of a Federal Reserve bank, any remaining surplus would be paid to the United States. Ownership of Federal Reserve bank stock by member banks is an obligation incident to membership in the System—in effect, a compulsory contribution to the capital of the Reserve banks. It was not intended to, nor does it, vest in member banks the control of the Reserve banks or the determination of System policies. Such control would obviously be inappropriate in view of the functions exercised by the Reserve banks. Stock ownership by the member banks has certain definite advantages. It provides a wide decentralized base for the organization of a Federal Reserve bank. The element of member bank interest, though without control, has contributed to a breadth of judgment and experience on the part of the Reserve bank directors in evaluating business-like methods in the operations of the Reserve banks as public institutions. It gives to each member bank a tangible interest in, and direct connection with, the Federal Reserve bank of its district, and this has real psychological value. It helps to create in member banks a greater interest in the affairs of the System and understanding of its purposes and operations than would be the case in the absence of such ownership. (Joint Committee Print of the Joint Committee on the Economic Report regarding "Monetary Policy and the Management of the Public Debt," 1952, pt. 1, pp. 261, 262.) II.2. Reducing the number of members of the Federal Reserve Board to five and their terms of office to no longer than 5 years. The original Federal Reserve Act provided for a Board of two ex officio members, the Secretary of the Treasury and the Comptroller of the Currency, and five members appointed by the President for 10year terms. In 1922, provision was made for an additional appointive member. The Banking Act of 1933 increased the terms of the six appointive members to 12 years. The Board was reconstituted by the Banking Act of 1935, effective February 1, 1936, to eliminate the ex officio members and to provide for a Board of seven appointive members with staggered terms of 14 years, with a prohibition against reappointment of a member after serving a full term. With respect to the size of the Board, the possible advantages and disadvantages of reducing the number of members were stated by Chairman Martin in one of his replies to the 1952 Patman questionnaire : Over a considerable period of time there have been proposals that the membership of the Board be reduced from seven to some lesser number, such as five or three. The reason most commonly advanced for such proposals is that greater importance would be attached to individual membership and that the position would be more attractive to men of higher caliber. Another reason is that Board decisions probably would be made more promptly. The timeliness of policy decisions is often extremely important and the need for expediting such decisions is strongly stressed by those students of monetary policy who have come to feel that the chief shortcoming of reserve banking policy over the years has been that important decisions have frequently come too late. Against proposals to reduce the size of the Board, it has been maintained that the advantages of collective deliberation and judgment would be correspondingly lessened, that there is at least safety and perhaps greater wisdom in numbers, and that a reduction in the size of the Board would necessarily require reconsideration of the composition and possibly even the status of the Federal Open Market Committee. Moreover, it is believed that a smaller board would find it more difficult to operate effectively and promptly on some occasions because of necessary absences, from illnesses or other causes, and the resulting lack of 4(3 a quorum. (Joint Committee Print of the Joint Committee on the Economic Report regarding "Monetary Policy and the Management of the Public Debt," 1952, pt. 1, pp. 302, 303.) On balance, the disadvantages of a reduction in the membership of the Board from seven to five would outweigh any possible advantages. With respect to the length of terms of Board members, it is important to bear in mind that the relatively long term of office provided for Board members since the establishment of the Federal Reserve System has always been regarded as a means of protecting members from political pressures. It is possible that the accomplishment of this objective does not require a term as long as 14 years; but it is questionable whether a term as short as 5 years would achieve this purpose. Moreover, if reappointment should be precluded after service of a full term, as under present law, a qualified candidate for membership might be reluctant to interrupt his career for that period of time. If the prohibition against reappointment should be eliminated, on the other hand, considerations relating to possible reappointment could conceivably inhibit objective public interest considerations. Again balancing the pros and cons, we are inclined to believe that a term of 5 years, with or without provision for reappointment would appear to be undesirable. It is noted that the provisions of H.R. 11 reducing the number of members of the Board would be accompanied by provisions that would abolish the Federal Open Market Committee and transfer regulation of open market operations to the reconstituted Board. Such a transfer of authority over open market operations to the Board would not be desirable. As Chairman Martin pointed out in replying to the 1952 Patman questionnaire: The present arrangement under which open market operations are placed under the jurisdiction of a committee representing the Reserve banks as well as the Board is consistent with the basic concept of a regional Federal Reserve System. It provides a means whereby the viewpoints of the presidents of the Federal Reserve banks located in various parts of the country, with their technical experience in banking and with their broad contacts with current credit and business developments, both indirectly and through their boards of directorsr may be brought to bear upon the complex credit problems of the System. It promotes System-wide understanding of these problems and closer relations between the presidents and the Board in the determination of System policies. In practice the open market policies of the Open Market Committee and the credit policies of the Board have been coordinated and the existing arrangement has worked satisfactorily. (Joint Committee Print of the Joint Committee on the Economic Report regarding "Monetary Policy and the Management of the Public Debt," 1952, pt. 1, p. 294.) II.3. Making the term of the Chairman of the Board coterminous with that of the President of the United States. It would be desirable to amend the law to make the terms of the Chairman and Vice Chairman of the Board more nearly coterminous with the term of the President. In 1952, in his replies to the Patman questionnaire, Chairman Martin noted that, when the Board was reconstituted by the Banking Act of 1935, it was specifically provided that the Chairman and Vice Chairman should be designated for terms of 4 years and that the possible purpose of this provision was to afford a new President an 47 opportunity to designate the Chairman and Vice Chairman. He stated, however, that: * * * Assuming such a purpose, the provision has not worked out satisfactorily in practice because it has not been feasible to make appointments so that they would coincide with the term for which the President is elected. It might be preferable if the law were changed to provide that the President shall designate the Chairman and Vice Chairman for terms expiring on a selected date, say March 31, 1953, and on March 31 of every fourth year thereafter. (Joint Committee Print of Joint Committee on the Economic Report regarding "Monetary Policy and the Management of the Public Debt," pt. 1,1952, p. 301.) On April 17, 1962, President Kennedy submitted to Congress a message recommending such an amendment to the law that would make the terms of the Chairman and Vice Chairman of the Board generally coterminous with the term of the President. Stating that Chairman Martin concurred in this proposal, the President's message contained the following paragraph: Federal Reserve monetary policies affect, and are affected by, the economic and financial measures of other Federal agencies. Federal Reserve actions are an important part, but not the whole, of Government policies for economic stabilization and growth at home and for the defense of the dollar abroad. Therefore, as has been recognized throughout the history of the Federal Reserve, the principal officer of the System must have the confidence of the President. This is essential for the effective coordination of the monetary, fiscal, and financial policies of the Government. It is essential for the effective representation of the Federal Reserve System itself in the formulation of Executive policies affecting the System's responsibilities. In a letter dated October 6, 1966, to Representative Abraham J. Multer, chairman of the Subcommittee on Bank Supervision and Insurance of the House Banking and Currency Committee, Chairman Martin stated that the Board believed that the terms of the Chairman and Vice Chairman of the Board should be related to the President's term of office and that a new President should be able to appoint a Chairman of his own choice and should not be limited in his selection to incumbent Board members. A change in the law enabling the President to appoint a Chairman of his own choice shortly after his inauguration would provide a practical basis for effective coordination of Federal Reserve monetary policies with the fiscal and financial policies of the executive branch of the Government without affecting the exercise of independent judgment by the Board in the discharge of the responsibilites imposed upon it by Congress. Such an arrangement would in fact afford a means by which the Federal Reserve, through the Chairman of the Board, would be better able to participate, at the highest level of the executive branch, in continuing efforts to promote the sound conduct of the Government's financial affairs. In order to accomplish the objectives of such a change in the lawT any amendment for this purpose should provide for an adjustment in the terms of members of the Board so that the term of one member would expire in each odd year instead of an even year, thereby causing a vacancy to occur in the membership of the Board in the year of a President's inauguration. Any such amendment should also provide for a reasonable time lag, perhaps as much as 6 months, between the time a newly elected President takes office and the expiration of the terms of the incumbent Chairman and Vice Chairman. 48 II.4. An audit for each fiscal year of the Federal Reserve Board and the Federal Reserve banks and their branches by the Comptroller General of the United States. It would be unnecessary and unwise to provide for audit of the Federal Reserve Board and Federal Reserve banks by the Comptroller General of the United States. The most recent public statement by a Federal Reserve official of the reasons for this judgment was made Tby Gov. J. L. Robertson when he testified on September 14, 1967, before the Committee on Banking and Currency of the House of Representatives with respect to H.R. 12754. The pertinent portion of Governor Robertson's testimony is set forth below : Let me try briefly now to set forth the present procedures for audit and examination of the Board and the Reserve banks, and add a few comments as to why section 2 of H.R. 12754 is unnecessary and unwise. Manifestly, Federal Reserve operations should be conducted with maximum efficiency and economy. To that end Congress has placed upon the Board of Governors, an arm of the Congress, direct responsibility for general supervision and periodic examination of the Reserve banks. The Federal Reserve Act also provides that each Reserve bank shall have a board of nine directors chosen from its district. They are outstanding in their communities; many have had broad experience in business and professional life, and are therefore able to apply to the Reserve banks the high standards of efficiency prevalent in private enterprise. Thus the Federal Reserve combines advantages of governmental control with advantages of private business management. Since 1952, the Board has been audited annually by independent public accounting firms, and their audit reports have been submitted to the Banking and Currency Committees of both House of Congress. We have endeavored to select topflight auditing firms for this work. The firms selected have been Arthur Anderson & Co., Price Waterhouse & Co., Haskins & Sells, and, most recently, Lybrand, Ross Broa and Montgomery. The Federal Reserve Act provides that the Board "shall at least once a year, order an examination of each Federal Reserve bank." The Board maintains a staff of examiners who devote themselves exclusively to this work. The Board's instructions to its examiners require, briefly, that the examination shall look to (a) each bank's financial condition through appraisal of its assets and vertification of its assets and liabilities: (b) its proper discharge of all its responsibilities; and (c) its compliance with all applicable provisions of law and regulations. Each year, an outside commercial auditing firm (Haskins & Sells for 1967) is engaged to accompany the Board's examiners on their examination of one of the Reserve banks, to review, observe, and submit recommendations for improving, the examination procedures. Also, each Reserve bank has a resident auditor, responsible directly to the bank's board of directors and not dependent on any of the bank's officers for security of position. Throughout the year, he and his staff make comprehensive audits of all phases of the bank's operations, reporting directly to the board of directors of the bank. Copies of these reports are reviewed by the Board of Governors of the Federal Reserve System. In sum, then, we have in each Reserve bank an internal audit program conducted the year round by the bank's resident auditor and his staff, who, by a deliberately established plan of organization, are directly responsible to the board of directors and independent of the bank's operating management. In addition, a staff of examiners directly employed by the Board of Governors in Washington examines each bank every year and reports directly to the Board of Governors. We have the statements of certified public accountants of national repute that the examination procedures employed by the Board's staff conform to generally accepted auditing standards. This combination of internal and external scrutiny provides an objective audit coverage of the Reserve banks that is unexcelled in any other organization. In addition, the System is subject to congressional scrutiny, a responsibility which this committee and its distinguished chairman take very seriously indeed, as you know. But some of the newer members of the committee may not fully appreciate how thoroughly the committee and its staff, including the capable and conscientious investigators who have been on loan to the committee from GAO in recent years, have examined into expenditures by the Reserve banks. Not only 49 have the reports of examination of the Reserve banks been furnished to these investigators, but when they asked to see the working papers used in the course of the examinations these, too, were furnished, including the contents of our examiners' locked work trunks- Detailed breakdowns of expenditures in each of four categories were requested and furnished, together with descriptive material and justifications for thousands of items selected from these categories by the committee's investigators. They have visited several of the Reserve banks, where, they reported, they "were courteously received and given all reasonable cooperation by bank personnel in accomplishing [their] work." (The quotation is from page 317 of this committee's 1964 hearings on the Federal Reserve System after 50 years.) The Board of Governors, then, stands ready to provide any information yott seek concerning expenditures by the System. We take our responsibilities seriously, too, as the Government agency designated by the Congress to make sure that the Reserve banks are carrying out efficiently the duties assigned to them by law. Direct expenditures for audit and examination of the Reserve banks in 1966 totaled approximately $4 million. What GAO does for the Post Office, we d o for the Reserve banks, reporting directly to you. This seems to us a sensible arrangement, since we have the particular expertise related to Reserve bank operations. If another arm of Congress were directed to do the same job, the end result would be duplication and overlapping of responsibilities, with attendant increases in costs and deterioration in operating efficiency and no apparent offsetting benefits. Let me add a few comments about the wording of section 2. It provides that GAO "shall have access to all books, accounts, financial records, reports, files and all other papers, things, of property belonging to or in use by the entities being audited, including reports of examinations of member banks." This provision raises serious questions about whether the System would be able to maintain relationships such as those presently in effect with foreign central banks, which depend on our ability to assure others that we can maintain confidentiality when they request it. As to one particular aspect of this problem, section 2 is crystal clear: it specifically requires that we make available to GAO the reports of examination of member banks. As I have indicated before, the System stands ready to answer any question about its own expenditures. But we believe that the long-established tradition that reports of examination of commercial banks should be kept confidential is not only essential to maintain effective supervision, but also to protect the privancy of customers of the member banks in their personal and business staffs." II.5. Funds to operate the Federal Reserve System to be appropriated by the Congress of the United States. This proposal contemplates that all earnings of the Federal Reserve banks would be transferred to the Treasury of the United States and that the expenses of such banks, as well as the expenses of the Board of Governors, would be paid only from appropriated funds. Adoption of this proposal would represent a radical alteration of the basic concept of the Federal Reserve System and prevent the System from discharging its statutory functions in the most effective manner, which requires the exercise of independent judgment and freedom from political and partisan pressures or the possibility of such pressures. Since the inception of the Federal Reserve System, the law has provided that the expenses of the Board of Governors, including the salaries of its members and employees, shall be paid out of semiannual assessments levied upon the 12 regional Federal Reserve banks. The expenses of the Federal Reserve banks are paid from the earnings of the Reserve banks which are derived principally from Government securities acquired pursuant to open market operations designed by law to aid in the maintenance of a sound basic economy and sound credit conditions. Since 1947, under direction of the Board of Governors pursuant to provisions of the Federal Reserve Act, the Reserve banks 50 have paid the greater part of their earnings to the Treasury of the United States. At present, all of such earnings, after payment of dividends to member banks and current expenses, and the maintenance of the Reserve banks' surplus in an amount equal to their paid-in capital, are transferred to the U.S. Treasury. Under this practice, the Reserve banks since 1947 have paid over to the Treasury more than $14 billion. In 1967 alone, the amount of such payments was nearly $2 billion. A requirement that the expenses of the Board and the Reserve banks be paid only from funds appropriated by Congress would create unnecessary and hampering rigidities in the performance of the public service functions of the System. More importantly, however, it would inject political pressures and considerations into the formulation of monetary and credit policies. One of the major purposes of the Banking Act of 1933 was to strengthen the Federal Reserve System by increasing the independence of the Federal Reserve Board. (See Report of Senate Banking and Currency Committee, April 22,1932, 72d Cong., p. 12.) Among other amendments to the Federal Reserve Act made by the 1933 act in order to accomplish this purpose was a change in section 10 of the Federal Reserve Act to provide specifically that the Board should determine the manner in which its obligations are incurred and its disbursements and expenses paid and to provide specifically that funds of the Board derived from assessments on the Federal Reserve banks "shall not be construed to be Government funds or appropriated moneys." It would be unfortunate if, after so many years, Congress should abandon the basic principle that the expenses of the Board, as well as those of the Reserve banks, should not be subjected to the limitations inherent in the appropriations process. Any change in the law that would make the Federal Reserve subject to the appropriations process would be logically inconsistent with the following conclusions reached by the Subcommittee on General Credit Control and Debt Management in 1952: The independence of the Federal Reserve System is based, not on legal right, but on expediency. Congress, desiring that the claims of restrictive monetary policy should be strongly stated on appropriate occasions, has chosen to endow the System with a considerable degree of independence, both from itself and from the Chief Executive. This independence is in no way related to the unsettled question of whether the Board of Governors is or is not a part of the executive branch of the Government. It is naturally limited by the overriding requirement that all of the economic policies of the Government—monetary policy and fiscal policy among them—be coordinated with each other in such a way as to make a meaningful whole. The independence of the Federal Reserve System is desirable, not as an end in itself, but as a means of contributing to the formulation of the best overall economic policy. In our judgment, the present degree of independence of the System is about the best suited for this purpose under present conditions. (Joint Committee Print of Report of the Subcommittee on General Credit Control and Debt Management of the Joint Committee on the Economic Report, 82d Cong., June 26,1952, p. 4.) Question III. Your analysis of monetary developments, since 196U, including policy induced changes and their effects on economic activity, is invited. Answer. For most of the period since 1964, the United States has been experiencing both the benefits and the problems of a full employment economy. The benefits have included an exceptionally low level of unemployment, maintained even during the short slowdown in 51 early 1967. The problems have included an unacceptable and unsustainable rate of price inflation, which has offset the bulk of the gains in money incomes, and has resulted in a sharp deterioration in the U.S. international trading position and overall balance of payments. These problems have been exacerbated by the delay in achieving adequate fiscal restraint, putting too large a share of the stabilization burden on monetary policy. The economy approaches its potential: 1964 mid-1965 During 1964 the Federal Eeserve sought to influence the cost and availability of funds and the growth in the Nation's liquidity so as to contribute both to continued orderly expansion in the domestic economy and to further improvement m the U.S. balance of payments. Throughout most of the year, the posture of monetary policy remained one of ease with respect to domestic credit conditions, supporting the stimulative effect of the March cut in Federal income taxes. During this period, the unemployment rate declined to about 5 percent, as economic activity expanded further. Beginning in the fall of 1964, with the economy strong enough to permit it, the Federal Reserve reduced the degree of ease slightly in order to minimize the outflow of funds attracted by higher shortterm interest rates abroad. In mid-August it became clear that the balance-of-payments deficit in the second and third quarters was running much larger than the quite low total achieved in the first quarter. In this situation, the Federal Open Market Committee moved in the direction of slightly firmer money market conditions in order to raise short-term interest rates somewhat, keeping them more in line with short-term rates abroad. In November, the Bank of England raised its discount rate 2 percentage points, to 7 percent, in response to the speculative attack on sterling. Federal Reserve bank discount rates were raised from Sy2 to 4 percent, to counter possible capital outflows that might be prompted by any wideing spread between money rates in this country and the higher rates abroad. In response, short-term domestic market rates moved up somewhat, after having remained relatively stable earlier in the year. So that this action would not unduly restrict the availability of bank credit, for domestic purposes, the Federal Reserve continued to supply reserves to banks through open-market operations and the maximum rates payable on time and savings deposits by banks were raised. As a result of the sustained growth over the preceding 4 years, the economy began approaching full utilization of labor resources as 1965 developed. While plant utilization was not pushed up to the optimum level, excess capacity did not prevent spending on new plant and equipment from remaining high. With continued expansion in demand, output, and employment, upward pressures on prices began to emerge. Moreover, in the early months of 1965, the deficit in the U.S. balance of payments continued to be large. The Federal Reserve participated in the administration's credit restraint program, announced in February 1965, to alleviate the balance-of-payments situation. To help reduce outflows of capital to foreigners, the voluntary foreign credit restraint program was set up, under which the Board of Governors issued guidelines to banks and 52 other financial institutions which were designed to restrain their lending and investing abroad. In an effort to reduce inflationary pressures that might develop as economic growth was extended, and to reinforce the voluntary foreign credit restraint program, the Federal Reserve began to supply less reserves through open-market operations, relative to demands, so as to encourage more moderate growth in the reserve base, bank credit, and money supply. Member bank borrowings from the Federal Reserve rose in the first half of 1965, and the banking system moved from a position of free reserves (excess reserves greater than borrowings) to one of net borrowed reserves (borrowings greater than excess reserves). Emergence of inflationary pressures: mid-1965 late 1966 In the latter half of the year, although the balance of payments was showing an improvement, demand pressures were increasing in the domestic economy largely in response to stimulative fiscal developments—including the military buildup in Vietnam, the reduction in Federal excise taxes, and the increase in social security benefit payments. Domestic price increases became more widespread, the unemployment rate moved down toward 4 percent, and plant capacity utilization was high. With confidence that further rapid economic expansion was in prospect, business speanding for inventories and fixed capital rose rapidly, resulting in heavy demands for credit. As a result interest rates, which had shown little change during the first half of the year from the levels of late 1964, also began to increase. With unused resources moving nearer to critically low levels, and with indications of continuing pressures from the business investment boom and an acceleration in defense spending, the Federal Reserve raised the discount rate by one-half of a percentage point, to 4 ^ percent, effective December 6. At the same time, to avoid a developing constriction in the flow of funds in credit markets, the Board of Governors raised interest rate ceilings on time deposits under regulation Q by a full percentage point to 5y2 percent. The expansionary forces in the economy, which had gathered momentum after mid-1965, accelerated in early 1966. In a further effort to blunt the inflationary impact of credit-financed spending, the Federal Reserve, in February, further increased the pressure on bank reserves through more restrictive open-market operations. With demand for credit still strong, interest rates rose sharply further through the summer. These higher market interest rates, together with intensified bank competition for funds, led to a sizable reduction in net inflows of savings to nonbank savings institutions and thence to the mortgage market. As a result, a heavier share of the impact of monetary restraint fell on the home building industry than on other sectors of the economy; industrial and other business concerns were still obtaining a considerable amount of credit, though at rising interest rates, to finance their increasing outlays for fixed capital and inventories. During the summer the Federal Reserve took a variety of steps to attempt to redress the uneven effects of financial restraint. These measures were designed to help prevent rate competition for savings among financial institutions from adding to the upward thrust of interest rates, to reduce the rapid growth of business loans at banks,, and to moderate pressures on the mortgage market. 53 In July, the regulation Q ceilings on new multiple-maturity time deposits were reduced. In addition, between July and September, reserve requirements on time deposits in excess of $5 million at each member bank were raised twice. An attempt by some banks to avoid ceilings by issuance of promissory notes maturing in less than 2 years was forestalled by bringing them under reserve requirement and interest ceiling regulations. Furthermore, on September 1 a letter was sent to each member bank which, while stating that reserves would be provided to meet seasonal or emergency needs, requested their cooperation in curtailing the business loan expansion. Indeed, the continued rapid expansion in business loans in a period of overall restraint on bank credit expansion was seriously limiting the availability of bank funds to meet other needs and threatened to cause excessive strains in the market for obligations of State and local governments. Finally, in late September, new temporary authority was enacted by Congress which broadened the basis for interest rate ceilings on time and savings deposits. Promptly thereafter the Federal Reserve and other regulatory authorities acted to limit further, or to reduce, interest rates payable on certain types of deposits and shareholdings at commercial banks, mutual savings banks, and federally insured savings and loan associations. Economic pause and resumption of expansion: Late 1966-summer 1967 By early fall it became evident that monetary policy; aided by certain fiscal restraints—including the suspension of the investment tax credit and accelerated depreciation privileges—had considerable success in achieving the objectives of curbing excessive aggregate demand and of damping inflationary pressures. Defense spending did continue to rise sharply in the fall, but residential construction activity had already fallen sharply, the rate of increase in consumer expenditures had slowed, there were signs that business plant and equipment outlays would moderate, and business inventory accumulation appeared to be reaching a peak. Expansion in bank credit and money supply slowed considerably. Federal Reserve open market operations in the early fall were modified so as to reduce some of the pressure on banks, and as fall progressed, the Federal Open Market Committee shifted its policy so as to stimulate moderate renewed expansion of bank credit and easier conditions in financial markets. By the end of 1966, pressures on financial markets had eased significantly and most market rates of interest had declined sharply from their late summer peaks. In the early months of 1967, with economic activity slackening, the Federal Reserve extended the shift toward greater monetary ease initiated in the fall of 1966. Open market operations wrere increasingly directed toward easing domestic credit conditions. Furthermore, in March, the Board of Governors authorized a reduction in reserve requirements on savings deposits and on the first $5 million of other time deposits at each member bank, and in the following month the Federal Reserve discount rate was reduced from 4% percent to 4 percent. Pressure on most financial markets continued to ease and corporations, banks, and nonbank savings institutions were able to improve liquidity positions that had been eroded during the previous year's monetary restraint. The shift toward ease beginning in the fall of 1966 was set in motion early enough so that a,n upswing in construction began in early 1967 and was a factor tending to offset the weakening in overall economic 54 activity. Moreover, credit was readily available to finance consumeroutlays on durable goods and to provide a cushion against businesses' tendency to reduce the rate of inventory accumulation. By early summer the prospects of a more rapid increase in economic activity suggested the desirability of greater emphasis on restraint, in the mix of fiscal and monetary policies, preferably through fiscal measures such as the administration's proposed tax increase. The inventory adjustment was relatively short lived, and in the summer the rate of business inventory accumulation began to rise again. Moreover, heavy Government expenditures continued to be a major stimulative force in the economy. And finally, inflationary price pressures were becoming more widespread, as rapid economic expansion resumed, sizable wage settlements were reached in key industries, and industrial prices rose. Inflationary pressures mount: summer 1967-^mid-1968 The highly stimulative Federal budgetary policy continued in the latter half of 1967 and the first half of 1968, leading to upward pressures on the economy. Since by the fall of 1967, no action had yet been taken on the administration's recommendation for a tax increase, the Federal Reserve System saw no choice but to move toward restraint. Indeed, the U.S. balance-of-payments position appeared to be worsening, international confidence in the dollar was ebbing, and domestic price increases were accelerating. At the time of the devaluation of the pound, in mid-November, the Federal Reserve raised the discount rate back to 4y2 percent, and open market operations were adjusted in the direction of restraint. Late in December, the Board of Governors also announced a one-half percentage point increase in reserve requirements against demand deposits in excess of $5 million at each member bank effective around mid-January 1968. During the first half of 1968, additional measures were taken to moderate the sustained domestic and international pressures on the dollar. The Federal Open Market Committee limited further the flow of reserves to banks. In March, in response to large gold outflows stemming from heightened speculation as to the possible devaluation of the dollar, the Federal Reserve again raised the discount rate, from 41/2 to 5 percent. In April, the discount rate wras again raised—to 51/2 percent—and regulation Q ceilings were increased on all but the shorter term CD's as holders of these instruments began to shift their funds into higher yielding market securities. The series of monetary actions beginning in late 1967, coupled with continued large demands for funds by governments and the private sectors of the economy, contributed to a sustained rise in market interest rates. Along with the rise in rates was a slowing in the expansion of bank credit, and reduced net inflows of funds to nonbank savings institutions. Less funds became available to finance construction, and activity in this sector of the economy began to slow down. There w^as evidence that some State and local government bond offerings were postponed as a result of interest costs, with spending possibly affected to a marginal degree. In the latter part of June, Congress enacted a program of fiscal restraint—including a tax increase and governmental expenditure reductions. This long-awaited move changed market expectations here and abroad and interest rates declined from their earlier peaks. Overall pressures in capital and money markets have been reduced since mid-1968, and the ability of banks and thrift institutions to obtain funds for lending appears to have improved in some degree. REPLY OF HON. HENRY H. FOWLER, SECRETARY OF THE TREASURY T H E SECRETARY OF THE TREASURY, Washington, September 5,1968. Hon. WRIGHT PATMAN, Chairman, House Banking and Currency Committee, House of Representatives, Washington, D.C. DEAR MR. CHAIRMAN : In reply to your letter of July 9 with regard to the hearings to be held on H.R. 11 by the Domestic Finance Subcommittee of the Committee on Banking and Currency, I am enclosing answers to the questions submitted. Sincerely yours, HENRY H . FOWLER. STATEMENT OF HON. HENRY H. FOWLER, SECRETARY OF THE TREASURY Question 1. Do you believe that a program coordinating fiscal debt management and monetary policies should be set forth at the beginning of each year for the purpose of achieving the goals of the Employment Act, or alternatively should we treat monetary and fiscal policies as independent mutually exclusive stabilization policies? Answer—Treasury response. The question implies that there are only two alternatives: (a) set forth a detailed program of fiscal and monetary measures at the beginning of each year geared to a forecast of the economy, financial markets and balance of payments, or (b) to treat monetary and fiscal policies as independent mutually exclusive stabilization policies. In our opinion, choice of either of the extreme alternatives would be unlikely to further achievement of the goals of the Employment Act of 1946 and other important goals, and under some circumstances, might actually impede achievement of the desired goals. There should be little need to argue at any length the case against treating monetary and fiscal policies as "independent mutually exclusive stablization policies." Economic theory and actual experience with stabilization policy demonstrat-e conclusively that the attainment of multiple, and frequently conflicting goals requires the coordinated use of policy instruments. We are not so rich in policy instruments that we can afford to let monetary and fiscal policy go their separate ways. Instead, these and other policy instruments must be combined in such overall proportion as to promote a proper matching between policies and objectives. Even then, the attainment of the goals of the Employment Act and other important goals will be a continuing task, requiring the best efforts of the executive, the Congress, and the private community. But, without the coordinated use of policy tools, the chances for success would be drastically reduced. (55) 56 It is one thing to recognize that major policy tools must be used in a coordinated way and quite another to argue that it would be useful to set forth a very detailed monetary and financial program at the beginning of each year. The budget message of the President, the Economic Report of the President and the Annual Report of the Council of Economic Advisers already go a long way toward specifying an economic and financial program. It may be possible to go somewhat further in spelling out financial assumptions underlying the economic projections in these documents. But we do not believe that it would be either possible or desirable to spell out prospective monetary and debt management steps in a great deal more detail than is now the case. One reason is the inherent difficulty of the forecasting process. It is hard1 enough to project the probable movements of major economic series. Our ability to project financial variable tends at present to be even more circumscribed. A requirement to be highly precise might under these conditions simply result in frequent large errors, and would nto necessarily be a real aid in forward planning. A second reason for questioning the usefulness of attempting to specify the details of monetary policy for a year or so ahead relates to the basic character of the monetary policy tool. One of the chief advantages of monetary policy as a stabilization tool is an ability to make prompt changes of direction in response to a changing pattern of events. Any monetary projection should typically be much provisional than projections in the fiscal area, where discretionary changes in policy are less frequent and less closely attuned to minor swings in economic activity. If there is an attempt to pin down future monetary actions too precisely, policymakers may lose the flexibility they need. It might Tbe argued that little harm would result from presenting very detailed projections of the economy, financial flows, and the balance of payments along with a proposed package of fiscal, monetary, and other policies, even if the projections turned out to be very wide of the mark. This, however, is not a convincing line of argument. The publication of such official projections would inevitably tend to suggest greater certainty as to the future course of events than could actually be the case and might even tend to reduce the needed "freedom of maneuver" of monetary policy. For these reasons as well as because of the traditional "independence" of the Federal Reserve within the Government, we believe that the monetary projections that underlie the Economic Report of the President must necessarily remain conditional, cannot be overly precise, and must typically recognize the need for monetary policy to be used flexibly in the light of changing circumstances at home and abroad. Question 13. If you believe a program should be specified, do you believe that the President should be responsible for drawing up this program, or alternatively should such responsibility be dispersed between the Federal Reserve System and agencies responsible to the President? (Please note that informal consulting arrangements can be made as desired whether responsibiltv is assigned to the President or divided between the President and the Federal Reserve. The concern here is with the assignment of formal responsibility for drawing up the economic program,.) 57 Answer—Treasury response. The President already has the responsibility for drawing up, at the beginning of each year, a detailed economic program that is incorporated in his budget and Economic Report messages. In this context, he usually does spell out, in a general way, his assumptions regarding the monetary policies that would be consistent with the proposed fiscal and economic program and that he would regard as appropriate. In working out these assumptions, the President usually takes account of the views of various agenices as well as those of the Federal Reserve. Responsibility for the presentation of such a set of economic recommendations, based on specified assumptions with respect to financial developments and policies, should in our view continue to rest with the President. For the reasons spelled out in our response to the previous question, however, statements regarding assumed or desired monetary policies must necessarily be provisional and leave ample room for the flexible use of such policies. Moreover, given the traditional arrangements under which the Federal Reserve is directly answerable to the Congress, formal responsibility for the determination and execution of monetary policy must remain with the Federal Reserve and, ultimately, the Congress. Question 1.3. Concerning monetary policy guidelines: A. Should monetary policy be used to try to achieve the goals of the Employment Act via intervention of money supply (defined as desired) as provided in H.R. 11, or alternatively should H.R. 11 be amended to make some other variable or variables the immediate target of monetary policy; for example, interest rates, bank credit, liquidity, high-powered or base-money, total bank reserves, excess reserves and free reserves? Please define the target variable or combination of variables recommended and state the reasons for your choice. (If desired, recommend a target variable or variables not listed here.) It would be most helpful if, in providing the reasons for your choice, you list the actions the Federal Reserve should take to control the target variable (or variables) and also explain the link between your recommended target of monetary policy and the goals of the economy as defined by the Employment Act. B. Should the guidelines of monetary policy be specified in terms of some index of past, present, or future economic activity, or alternatively in terms of the target variable's value or growth? For example, should the President's 1969 program for achieving the goals of the Employment Act be formulated to require consistency with some set of overall indicators of economic activity, or alternatively so that your target variable attains a certain value or growth regardless of the economic winds? Please indicate the reasons for your preference. C. For only those persons who recommend that some index of economic activity be used to guide the monetary authorities in controlling the target variable: Should we use a leading (forward looking), lagging (backward looking) or coincident indicator of economic activity? It would be most helpful also if you would identify the index you, would like to see used and specify how the target variable should be related to this index. 21-570—68 5 58 D. For only those persons ivho recommend that the guidelines be put in terms of the target variable's value or growth: Shoidd the same guidelines be used each year into the foreseeable future, or alternatively, should new guidelines be issued at the beginning of each year conditioned on expected private investment, Government spending, taxes, et cetera? Please indicate the reasons for your preference. E. For only those persons who recommend that the guidelines be put in terms of the target variable's value or growth and who also recommend that the same guidelines be used year after year into the foreseeable future: What band of values or range of growth do you recommend? (By way of clarification, a band of values appears appropriate if your target variable is, say, free reserves, whereas a range of growth is appropriate if it is, say, money supply.) F. For all those persons recommending that the guidelines be put in terms of the target variable's value or growth (regardless of whether you recommend using the same guidelines year after year or revising them each year in light of expected private investment and fiscal policy): Under what circumstances, if any, should the monetary authorities be permitted during the year to adjust the target variable so that it exceeds or falls short of the band of values or range of growth defined by the guidelines issued at the beginning of the year? Answer—Treasury response. It is clear that monetary policy should be used in conjunction with other policy tools to try to achieve the goals of the Employment Act and other important objectives. Discussion of monetary "targets" should not, however, be allowed to obscure the fact that the ultimate objective is a prosperous, expanding economy relatively free from inflationary pressures and providing a wide range of employment opportunities. Monetary policy can help contribute to the achievement of these objectives, but it would be patently unrealistic to suppose that policy actions in the financial sphere can always achieve a required effect upon such real variables as production, employment, growth in the capital stock, et cetera. Furthermore, there is no single most important financial variable, or set of financial variables, to which the monetary authorities can safely direct their exclusive attention. At different times and in different circumstances, the monetary authorities will find it advisable to seek to influence economic and financial activity in different ways. In our opinion, this makes advance specification of a single monetary guideline an undesirable step. Monetary research in recent years has clarified the nature of the transmission process by which an initial monetary effect on certain financial variables works through to ultimate target variables such as employment, output, prices, and the balance of payments. The Federal Reserve System has been studying these matters very closely in recent years and many of the questions have long been the subject of intensive academic inquiry. The current study which your Committee has undertaken will provide a very useful sampling of the range of opinion and controversy which still surrounds some of the unsettled questions in monetary theory and policy. 59 Because of the considerable progress that has been made in recent years in defining the methods and objectives of monetary policy, the Federal Reserve should now be able at any particular time to specify what financial variables it is seeking to influence and why. This is being done by the System to an increasing degree subject to necessary constraints on the timing of the release of information. It is our understanding that the System lias made a much greater, and largely successful, effort in recent years to develop quantitative measures of the impact and effect of alternative monetary actions. However, we are reluctant to suggest that the Federal Reserve should be encouraged, or directed, to concentrate upon a single quantitative guideline, or combination of guidelines, as H.R. 11 contemplates. Research study within and without the System points to the complexity of the interrelationships over time among financial and real variables in the U.S. economy. Serious damage could be done to the prospects for a successful stabilization policy if the System were forced to concentrate upon some single monetary "rule" or "guideline" since such guides may rapidly become inappropriate in a dynamic situation. Over the past decade, and particularly within recent years, there have been a number of important and far-reaching structural changes in the financial system and significant changes in investor behavior. To cite but a single example, there has been an increasing degree of responsiveness on the part of the public to changes in relative yield on alternative financial assets. In conjunction with successive increases in regulation Q ceilings, this has led to a blurring of the sharp distinction sometimes drawn between the money supply on a narrow and a broad definition. A monetary guideline phrased rigidly in terms of either definition could have led to an inappropriate monetary reaction at several times in the recent past. This is simply one manifestation of the difficulties encountered in attempting to establish, before the fact, a single standard by which monetary policy would be guided. In summary, the ultimate target is the productive performance of the economy itself, not the behavior of some financial variable or set of variables. The Federal Reserve can provide, and, we believe, is providing reasonably specific information on the immediate target variables it seeks to influence, and the presumed effects thereby exerted on the economy and the balance of payments. But the Federal Reserve cannot safely limit its attention in advance to any single monetary guideline or set of guidelines. Therefore, we oppose the suggestion that there should be an advance legislative, or executive, specification of the immediate, as opposed to ultimate, targets of monetary policy. Question I.If. Concerning debt management policy: Given the goals of the Employment Act, what can debt management do to help their implementation? (If you believe that debt management has no role to play in this matter, please explain why.) Answer—Treasury response. The influence of debt management policies on the economic and financial situation is primarily through alterations in the term structure of the public debt in private hands, a process which typically proceeds by smiall steps at any time. The total amount of public debt in private hands, of course, is determined more by fiscal and monetary policy than by debt management decisions. These debt management decisions can, however, have a significant marginal influence on the ownership distribution of the public debt. BO Beyond this, at favorable times, debt management decisions may exert an important catalytic effect on financial markets. To illustrate the slow process of altering the term structure of the public debt, it may be noted that 10 advance refunding operations over a period of nearly 5 years were required to lengthen the average maturity of the privately held marketable debt from 4 years 6 months in September 1960 to 5 years 9 months in June 1965. From June 1965 to January 1968, a period of -approximately years during which the Treasury was unable to issue long-term securities because of the 4 percent limitation on bond coupons, the average maturity of the privately held debt fell to 4 years 4 months, a reduction of 1 year 5 months. These figures, and consideration of the typical size of Treasury debt management operations in terms of the overall amount of marketable public debt outstanding in private hands—usually 3 to 4 percent—suggest that it is unlikely that alterations in the maturity structure of the debt can ordinarily be brought about rapidly enough to have a major short-run influence on the liquidity of private investors, and, consequently, on their economic decisions. It seems reasonable to believe that alterations in the term structure of the debt usually have only modest effects on interest rate patterns or on the flow of funds. Nevertheless, debt management operations can, if carefully coordinated with other policy instruments, assist in the achievement of economic and financial goals. It has often been argued that issuing long-term Treasury securities in periods of economic slack absorbs available long-term funds, prevents long-term interest rates from declining as rapidly or as much as they might otherwise decline, and, consequently, interferes with the course of the economic recovery. Conversely, it is argued that debt management should give primary stress to long issues at times of inflation. Assessments of these arguments among economists tend to vary substantially. As a practicalmatter, however, the issuance of moderate amounts of long-term Treasury securities in such circumstances is not likely to have significant adverse effects. In such periods, long-term investment funds are often temporarily placed in short-term instruments either to avoid being locked up indefinitely at low levels of long-term interest rates or because there is no demand for such funds. The offering of Treasury long-term securities, thus, may simply tap funds which would not otherwise be in the long-term market. Also, possible effects on long-term interest rates from an additional supply of long-term Treasury securities are likely to be swamped by overall downward pressures on interest rates resulting from monetary policy actions and from the relative excess of supply compared to the demand for funds in all maturity areas of the market. Moreover, in periods of strong economic activity when interest rates are higher, the existing statutory interest rate ceiling is likely to prevent really long-term issues—a situation that has been particularly evident since May 1965. Hence, any rule that would confine Treasury issues of longer term (or, at least, intermediate term) maturities to periods of economic strength might in actuality preclude the Treasury from issuing such securities at all and would thus result in a massive deterioration in the term structure of the debt. 61 The report of the Commission on Budget Concepts focused attention on the total financing of the Federal budget as against the part financed by Treasury operations. In this broader context the Treasury Department, acting to monitor the timing and pricing of new issues, can contribute to the maintenance of orderly marketing conditions and a degree of continuity and stability in financial markets which otherwise might be missing. The significance of a broader definition of debt management is underlined by the fact that private holdings of marketable Treasury debt declined by almost $3 billion from the end of fiscal year 1964 through fiscal year 1968 despite substantial Federal deficits during these years. In the same period, private holdings of agency issues, including participation certificates and securities of the home loan banks and land banks, increased by over $16 billion. Mention should also be made of the economic and financial significance of the large volume of Federal loan guarantee and insurance activity. In addition to the large volume of direct agency issues and other Federal credit activities which give rise to a variety of Treasurybacked securities, the number and diversity of these operations—each with its own program and cash-flow problems—require careful planning by the Treasury in attempting to coordinate their market impact. An illustrative special circumstance occurred in 1966 when extraordinarily heavy private credit demands were pushing against an increasing degree of monetary restraint and a threat of financial crisis appeared to be developing. Debt management operations in the broad sense contributed at that time significantly to the improvement in the financial situation which followed the announcement of the President's anti-inflationary program on September 8. These measures included substantial reductions in Federal credit program activity and the contemplated offerings of participation certificates or agency financing. There is little question that financial markets can be catalyzed by debt management decisions, either favorably at particular times or unfavorably if offerings are not made with due regard to circumstances, including the preferences of investors and other market factors. To cite a recent example of a favorable influence: in the August 1968 refinancing, debt management operations appear to have facilitated and perhaps accelerated desirable adjustments to new conditions. The pricing of the new security in August indicated the peak in rates had passed and that a lower level of rates had become appropriate. Financing in the intermediate area, moreover, provided an opportunity to take advantage of favorable demand factors without placing undue pressures on flows of funds in either the long-term or short-term market areas. In recent years, considerable attention has also been paid to structuring the maturity distribution of debt offerings in a manner that would help minimize potentially adverse financial flows. Thus, in the early 1960's emphasis on adding to the supply of short-term securities and the resultant effects on short-term interest rates helped to contain the outflow of short-term capital abroad. More recently, debt management has been especially concerned with avoiding "dismtermediation" that could unfavorably affect the net flow of funds to the thrift institutions and, hence, the availability of funds for housing. 62 Question 1.5. Concerning open market operations: A. H.R. 11 makes no provision whatever for conducting open market operations for so-called defensive or road-clearing purposes; that is, to counteract seasonal and other transient factors affecting money market and credit conditions. Do you see any merit in using open market operations for defensive purposes or should they be used only to facilitate achievement of the President's economic program and the goals of the Employment Act? What risks and costs, if any, must be faced and paid if open market transactions are used to counteract transient influences? B. Do you believe that monetary policy can be effectively implemented solely by open market operations? C. For what purposes, if any, should (a) rediscounting, (&) changes in reserve requirements, and (c) regulation Q be used? How might H.R. 11 be amended to implement your recommendations? In responding to these questions, a single answer appears appropriate. Answer—Treasury response. In 1950 the House Banking and Currency Committee's Subcommittee on Monetary, Credit, and Fiscal Policies stated in its report: It appears to us impossible to prescribe by legislation highly specific rules to guide the determination of monetary and debt management policies, for it is impossible to foresee all situations that may arise in the future. The wisest course for Congress to follow in this case is to lay down general objectives, to indicate the general order of importance to be attached to these various objectives, and to leave more specific decisions and actions to the judgment of the monetary and debt management officials * • » (pp. 27 and 28 of the subcommittee report). We believe that the same considerations regarding the conduct of monetary policies are still relevant at this time. Circumstances may arise in which one or another of the instruments of monetary policy may be most appropriate for dealing with the economic and financial situation. For this reason, we believe the Congress should not want to direct the Federal Reserve System to use particular instruments only in prescribed circumstances, or to seek to reduce the number of policy tools now at the System's disposal. As regards so-called defensive open market operations, we believe that these are of definite assistance in carrying out the purposes of the Employment Act and, indeed, are essential to the proper conduct of monetary policy. These operations contribute substantially to the smooth functioning of our money markets and financial system which is of key importance to stable economic growth. Moreover, in the absence of such operations, the proper conduct of monetary policy could be severely complicated, since policymakers could find it extremely difficult to distinguish clearly between the effects of their actions that further basic policy objectives and those that affect transitory monetary market factors. Furthermore, a failure to undertake defensive operations could well lead to excessive money market fluctuations and cumulating speculation that could interfere with the achievement of monetary policy goals. The above observations should, of course, not be taken to imply that there is no room for improvement in monetary policy instruments— including, among other things, the possible development of devices 63 that would permit smoothing of money market conditions with relatively less reliance on defensive open-market operations. Research into the possibilities for such improvement is being actively carried on by the Federal Reserve System itself as well as by other students of monetary policy. Some of the results of this research—notably the proposals recently made within the Federal Reserve for changes m the discount mechanism—are currently being examined by various interested parties as well as the Federal Reserve Board itself and are scheduled to be subjected to further scrutiny by the Congress. I do not believe that I should comment on such specific proposals until there has been an opportunity for careful further study. Question i,5.D. Do you see any merit in requiring the Federal Reserve Board to make detailed quarterly reports to the Congress on past and prospective actions and policies? Are there any risks and costs in this procedure? In what ways, if any, would you modify the reporting provisions? ~What information do you believe should be included in such reports as you recommend the Federal Reserve submit to the Congress? Answer—Treasury response. We believe that the System already makes proper disclosure of its past policies. We would, however, see no inherent objection to a requirement that formal reports be made to the Congress on a regular quarterly basis as suggested by the question. It is much more difficult, on the other hand, to visualize advantages that might be gained if the System should attempt publicly to forecast its future actions and policies. Such forecasts would need to be extremely tentative, since the Federal Reserve System does not control administration and congressional actions on fiscal policy, nor can it be expected to foresee many other autonomous events that may subsequently require policy changes. Even in such a highly tentative form, however, the forecasts could have substantial disadvantages as already noted in the answer to question 1.1 above. They are likely to be regarded as firmer than intended, and could well create anticipatory actions by participants in the private financial markets which could have adverse financial and economic consequences. This, in turn, would make it far more difficult for monetary policy to respond rapidly and flexibly to evolving changes in economic and financial conditions. Impairment of monetary policy flexibility, even inadvertantly, would be a matter of considerable concern, since the responsiveness of Federal Reserve policy to events is a major advantage of the monetary policy tool and an essential ingredient of the proper functioning of economic stabilization policies. Question I.5.E. What costs and benefits would accrue if representatives of the Congress, the Treasury, and the CEA were observers at Open Market Committee meetings? Answer—Treasury response. The Treasury Department believes that its relations with the Federal Reserve System are on a basis which leads to a continuing, meaningful exchange of views on economic and financial developments. For this reason the Department feels that it would not benefit from sitting as an observer at FOMC meetings. The Department would also be reluctant to see any change in the conduct of Open Market Committee meetings which might have the effect of limiting the frank exchange of views among members of the Open Market Committee and could impair the traditional independence 64 of the System within the Government. It believes that the performance of the Federal Reserve System is best judged by the results of its policy actions, and notably the effects on the economy as such. Question II. H.R. 11 provides for the following structural changes in the Federal Reserve System: 1. Retiring Federal Reserve bank stock; 2. Reducing the number of members of the Federal Reserve Board to five and their terms of office to no longer than 5 years; 3. Making the term of the Chairman of the Board coterminous with that of the President of the United States; Jp. An audit for each fiscal year of the Federal Reserve Board and the Federal Reserve banks and their branches by the Comptroller General of the United States; and 5. Funds to operate the Federal Reserve System to be appropriated by the Congress of the United States. Please comment freely on these several provisions. In particular it would be most helpful if you would indicate any risks involved in adopting these provisions and discuss whether their adoption woidd facilitate the grand aim of H.R. 11, which is to provide for coordination by the President of monetary and fiscal policies. Answer—Treasury response: General comment.—In approaching the general subject of possible structural changes in the Federal Reserve System, it is appropriate to recall the following passages from the testimony on similar legislative proposals regarding the Federal Reserve that former Secretary Dillon gave to the Subcommittee on Domestic Finance of the House Committee on Banking and Currency in February 1964: This committee is dealing with a living institution—an institution that has demostrated its capacity to innovate, to experiment, and to adapt itself to a very wide range of circumstances. But in this process of change, it has never lost certain characteristics—an established tradition of independent judgment; a mixture of regional participation in policymaking with ultimate central control that is unique in our Government; an ability to attract highly qualified officials and staff; and a reputation for operating efficiently and impartially. The structure that has resulted does not fit easily into the framework of standard tables of organization. Policy responsibility is widely dispersed and coordination depends in part on informal working relationships built up over the years. Vestigial elements of an earlier conception of private participation in central banking policies—elements that are more symbolic than real today—are still visible. But change without clear purpose can be dangerous too. If there are persuasive reasons for particular proposals—if it can be shown that ownership of Federal Reserve bank stock by member banks has biased Federal Reserve policy decisions, or if budgetary or auditing practices have been loose, to take two examples—by all means, this committee should act. But I doubt the advisability of taking action simply for the sake of achieving symmetry with other Government agencies, particularly if there was danger that such action might impair a long tradition of regional participation and efficient service of which I believe the country can be proud. These considerations, in my view, are fully applicable to the specific proposals cited under question II. If the United States had to create a brand new central bank today, the specific features that any one of us might favor would not, in all likelihood, coincide precisely with the existing structure. But this structure is one that is based on an evolution of over 50 years, and that is on the whole working remarkably well. Hence, 1 do not believe that changes should be made unless 65 it can be shown that they are clearly needed and would result in significant net benefits. In particular, it is highly important that no steps be taken which might diminish public confidence in the efficiency and integrity of our monetary management—a confidence which is itself one of the essential preconditions for the achievement of the goals of the Employment Act of 1946. Question II J. Retiring Federal Reserve bank stock. Answer—Treasury response. In terms of the actual operations of the Federal Reserve System and the formulation and execution of monetary policy, it makes no real difference whether Federal Reserve bank stock is retired or not. The ownership of stock by member banks does not, as such, give these banks any right or ability whatever to control the Federal Reserve banks or determine Federal Reserve policies. Even the right of member banks to participate in the election of Federal Reserve bank directors is not directly tied to the ownership of stock; if the stock were retired, means could undoubtedly be found to retain essentially the same system for electing directors as exists at present; or, conversely, changes in the procedures for electing directors could be made without retiring the stock. In contrast to private firms, moreover, Federal Reserve banks do not require capital stock as a financial underpinning for their operations. The case for retaining or retiring Federal Reserve stock thus basically hinges on the presumed intangible or psychological advantages or disadvantages of Federal Reserve stock ownership. In opposition to such ownership, it has been argued that it tends to convey an impression to the public—however unjustified this may be—that Federal Reserve banks are in fact dominated by private banks. Those who take this view argue that if the System is to make use of devices that are of a largely symbolic nature, these should primarily stress the public service character of the Federal Reserve. Proponents of retaining stock ownership, on the other hand, feel that this device has positive advantages in terms of giving member banks a greater sense of participation in the System and in eliciting their interest and cooperation. We do not have a strong view regarding the relative weight that might be given to these opposing considerations. However, given the relatively smooth functioning of the Federal Reserve System under present arrangements, it would appear that changes in the stock ownership device should only be instituted if it can be demonstrated that the arguments in favor of such action are compelling. Question IL2. Reducing the number of members of the Federal Reserve Board to five and their terms of office to no longer than 5 years. Answer—Treasury response. While some reduction in the size of the Board's membership and the length of terms might prove useful, specific proposals in this area need to take careful account of the benefits which accrue under the present system as a result of broad membership and the encouragement of careful deliberation removed from political pressures. A reduction in the length of terms from 14 years to as short a period as 5 years might, in particular, carry greater risks of subjecting Board members to pressures of this kind than would be desirable. 6(j Question U.S. Making the term of the Chairman of the Board coterminous with that of the President of the United States. Answer—Treasury response. Adoption of this proposal would be desirable. It was proposed to the Congress by President Kennedy in 1962, is favored by the Board of Governors itself, and has been explicitly endorsed by Chairman Martin on a number of occasions. Making the term of the Board Chairman (and also of the Vice Chairman) coterminous with that of the President should help assure that an incoming President would have full cooperation in the formulation and execution of financial policy. A high degree of cooperation and understanding has been developed between the Federal Reserve Board and the executive branch through informal working arrangements in recent years. It may be better, however, to make explicit provision for Presidential selection of the Chairman (and Vice Chairman) rather than to assume that a cooperative working arrangement could always be established easily and promptly at the beginning of a presidential term. The Board itself would continue to be chosen under the existing arrangements which have worked well and provided a necessary immunity from political pressures. Question 114• An audit for each fiscal year of the Federal Reserve Board and the Federal Reserve banks and their branches by the Comptroller General of the United States. Answer—Treasury response. While it would be the prerogative of the Congress to order a GAO audit of the Federal Reserve if it so desired, it is not evident that such a step would be either necessary or desirable. Under the present arrangements, the Federal Reserve banks are audited by a highly competent staff of the Board of Governors, while the Board itself is audited by independent public accounting firms of topflight reputation. These audits, furthermore, are made available to the Banking and Currency Committees of both Houses of Congress, and are thus subject to detailed congressional scrutiny. Unless it can be demonstrated that there are significant abuses which have arisen under the present auditing system, there is no compelling operational case to institute a GAO audit. We have no indications that such abuses exist or that any occasional problems that might arise would not be adequately corrected under the present auditing procedures. It might also be noted that institution of a GAO audit would involve added budgetary expense and extra training of auditing personnel. It is sometimes argued that even if the above-cited points are entirely correct, a GAO audit procedure might still be desirable as a symbolic measure, to assure the public that congressional scrutiny of the System's operations is fully adequate. In weighing this argument, however, the Congress will also need to consider potentially adverse "symbolic" effects that could result from institution of the audit. Thus, such a measure might widely be regarded as increasing the possibilities for reducing the independence of the System within the Government, and as possibly leading to undesirable interferences with policies. While it may in theory be possible to prescribe that the audits would have to be conducted strictly on the basis of standards and policy guidelines set forth by the Board of Governors itself, very careful consideration would have to be given to the risk that, in practice, the existence of a GAO audit could at times broaden into a review of monetary policies and tend to impinge on policymaking as such. 67 Question II,5. Funds to operate the Federal Reserve System to be appropriated by the Congress of the United States. Answer—Treasury response. It w^ould not be desirable to make the Federal Reserve subject to the regular congressional appropriations process. There is every evidence that the Federal Reserve is managed prudently and efficiently; thus there is no clear need for the proposal. Adoption of the proposal would almost certainly lead to a major reduction in the existing degree of Federal Reserve independence within the Government and in its insulation from day-to-day political pressures. It would also tend to introduce unnecessary operational rigidities that might diminish the System's ability to respond very promptly and flexibly to various domestic and international contingencies. While the role of the Federal Reserve within the Government is in many ways unique, it should be noted that the Congress has also exempted the other major bank supervisory authorities—that is, the FDIC and the Comptroller of the Currency—from the regular appropriations process. Question III. Your analysis of monetary developments, since 1964, including policy-induced changes and their effects on economic activity, is invited. Answer—Treasury response. During roughly the first half of the current expansion, monetary policy was consistently expansionary. From 1961 through 1964, most long-term interest rates were relatively stable and mortgage rates actually declined. There were regular annual increases in the total and nonborrowed reserves of the banking system in the to 41/^-percent range and net borrowed reserves remained positive. Short-term interest rates rose, but this partly reflected the effect of policies designed to keep U.S. money market rates in reasonable alinement with key foreign rates. From 1961 through 1964, commercial bank credit expanded steadily at about 8 percent a year. Growth in the money supply, narrowly defined, averaged a little above 3 percent annually. Increases in regulation Q ceiling rates and an expanding economy led to large and continuing increases in time deposits. As a result, the money supply plus time deposits grew fairly steadily at roughly 8 percent a year. While there was some modification of monetary policy in the interests of the balance of payments during this period, the general picture is one of relative monetary ease in support of the continuing domestic expansion. Monetary expansion continued in the first half of 1965, although in the face of relatively heavy credit demand member bank borrowings increased and net borrowed reserves turned negative for the first time in the expansion. Late in the first quarter of 1965, the Federal Reserve moved toward firmer conditions in the money market in an effort "to reinforce the voluntary foreign credit restraint program and avoid the emergence of inflationary pressures." Growth in both total and nonborrowed reserves remained sizable during the first half of 1965 and bank credit growth picked up to about a 10%-percent annual rate. Long-term interest rates remained relatively stable while short-term interest rates moved up to a new plateau following the November 1964 increase in the discount rate and the subsequent policy move toward firmer money market conditions. 68 From mid-1965 monetary policy began to operate in a different environment. An economy nearing full employment was also faced with the requirements of an expanding defense effort. Interest rates began to rise, initially in response to expectational factors. While there was no overt move toward monetary restraint until December, growth in total and nonborrowed reserves slackened after mid-1965. In early December 1965 the Federal Reserve increased the discount rate from 4 to 4y2 percent and raised the regulation Q ceilings. (As is well known, the administration was essentially in agreement with the direction of the move. It did, however, object to its timing, which came just before the period when budgetary and fiscal descisions were reached.) After the Federal Reserve action in 1965, the policy focus shifted to fiscal measures and the budget program. Growth in total demand was brought into reasonable correspondence with growth in productive capacity by the second quarter of 1966, and the pace of expansion— as reflected in quarterly increments in gross national product—became more moderate than in late 1965 and early 1966. During much of 1966 monetary restraint was primarily reflected in sharply rising interest rates and a drastic curtailment of mortgage credit. On the other hand, business loan and bank credit growth were not easily curtailed. In the first 8 months of 1966, bank loans to business grew at nearly a 20-percent annual rate, only a little below the rate in the second half of 1965. In retrospect, it appears that the December 1965 increase in regulation Q may have provided the commercial banking system with more latitude to compete ratewise for funds, primarily through the issuance of CD's, than was ideal during a period of monetary restraint. Serious financial strains and imbalances developed during the course of 1966. These primarily took the form of selective pressures on productive capacity and a growing imbalance in credit flows. By late summer, interest rates had reached their highest levels in four decades the housing industry was depressed, and steps had to be taken to insure the continued orderly functioning of financial markets. With the announcement of the President's September 8 anti-inflationary program and the benefit of subsequent steps taken by the Congress and the financial regulatory agencies, pressures on financial markets were relieved and a concerted easing of interest rates was set in motion. The financial environment improved steadily throughout the balance of the year, aided by a moderate shift toward monetary ease set in motion by the Federal Reserve during1 the autumn. During most of 1967, monetary policy was generally expansionary in terms of growth in such measures as bank credit, money supply, and reserves. Despite the slackening in the pace of economic activity in early 1967, private financial demands were heavy throughout the entire year. As an aftermath of the credit squeeze of 1966, efforts were made throughout the private sector to rebuild liquidity and in some cases to make advance provision for possible future credit needs. Furthermore, there was general belief in the business and financial community that the slowdown in the economy was likely to be temporary in duration and would be followed by a period of more rapid expansion. As a result, interest rates dipped only temporarily in early 1967 when the pace of economic expansion slowed and then rose during the balance of the year. 69 Monetary policy began a move back toward a more restrictive posture late in 1967 with a one-half point rise in the discount rate to 4 ^ percent following the devaluation of sterling. As the outlook for fiscal restraint remained uncertain, monetary policy was tightened further in 1968 with the discount rate increased to 5 percent in March and to 5y2 percent in April. There was general agreement on the need for the application of restraint and monetary policy was for practical purposes, the only available instrument. Fiscal policy was temporarily immobilized during this period by congressional inaction on the President's fiscal recommendations. Eventual enactment of the President's tax program at mid-1968 reactivated fiscal policy and greatly increased the degree of fiscal restraint being impjosed on the economy. The application of fiscal restraint and the shift in expectations it brought about soon led to a significant easing in interest rates. At mid-August the Federal Reserve Board approved a ^-percent reduction in the discount rate, primarily as a technical action to bring the discount rate more into line with prevailing money market rates. REPLY OF THE MEMBERS OF THE COUNCIL OF ECONOMIC ADVISERS T H E CHAIRMAN OF THE COUNCIL OF ECONOMIC ADVISEES, Washington, D.C., November 22,1968. H o n . WRIGHT PATMAN, Chairman, Subcommittee on Domestic Finance, Committee on Banking and Currency, Washington, B.C. DEAR MR. CHAIRMAN: I am enclosing the Council's responses to the list of questions on monetary policy and the structure of the Federal Reserve you sent to us last July. The Council members received independent requests for their views as members of the economics profession. We have confined our efforts, however, to the single joint response enclosed here. We might indicate that we look forward with interest to your intended hearings on H.R. 11. While we have reservations about some of the proposals in that particular bill, as indicated in our responses, we do think that some of the proposed reforms might be helpful, and we have added one or two suggestions of our Own. Moreover, an updated exploration of views on the workings of the monetary policy process should prove useful to all financial economists both in official positions as well as in academic life. Sincerely, ARTHUR M . OKUN. STATEMENT OF THE COUNCIL OF ECONOMIC ADVISEES TO QUESTIONS SUBMITTED BY HOUSE BANKING AND CURRENCY COMMITTEE Question 1.1. Do you believe that a program coordinating fiscal, debt management and monetary policies should be set forth at the beginning of each year for the purpose of achieving the goals of the Employment Act, or alternatively should we treat monetary and fiscal policies as independent rrmtually exclusive stabilization policies? CEA response. In order to achieve to the greatest extent possible the several, sometimes partially conflicting goals of economic policy—including high employment, reasonable price stability, vigorous growth, and a satisfactory balance-of-payments position—it is clear that all available policy instruments must be used together in a carefully coordinated manner. No one to our knowledge would seriously argue in favor of total separation of fiscal, debt management, and monetary policies as suggested by the second alternative posed in the question. Real coordination, however, does not merely involve the formulation of a program once a year, as implied by the first alternative in the question. Indeed, a once-for-all program formulated at the beginning of the year could well be a hindrance to the achievement of meaningful coordination of economic policies. What is required is a working together of all the relevant agencies in a continuously evolving joint effort to achieve the Nation's objectives in the light of constantly changing circumstances. (70) 71 The Council takes some pride in having helped during the past 8 years to institutionalize a good part of the policy coordination process. Explicit economic programs have, of course, made up an important part of the President's annual messages to the Congress on the state of the Union, the budget, and his Economic Report, the latter supplemented in more detail by CEA's annual report. But these messages have represented only a part of the process. Meetings, involving the Secretary of the Treasury, the Director of the Bureau of the Budget, the Chairman of CEA and, on occasion, the Chairman of the Board of Governors of the Federal Reserve System, have been held from time to time with the President, and informal dialog among these and other agencies—both at an official and staff level—has gone on constantly. The purpose has been continually to advise the President on the Nation's economic progress, and to evaluate and recommend new programs and policy actions as they appear to be needed. This continuous dialog has provided the real basis for effective coordination of policies. We want to emphasize two points about the coordination process. First, coordination should not be taken to mean that all policy instruments must necessarily be moving in the same direction. On the contrary, movement of one instrument toward, say, restraint may permit another instrument to move toward expansion. Second, we believe that any particular policy program should, in its general formulation, treat the various instruments of policy in different degrees of detail. In particular, fiscal policy can and should be spelled out rather completely in the program, but unnecessary precision should be avoided in specifying the roles assigned to monetary and debt management policy. This is simply prudent planning, designed to preserve as many policy options as possible so that ammunition is available to respond if actual economic developments should depart from forecasted trends. Fiscal policy decisions, in our opinion, should be made only at discrete intervals in order to promote a general course for the economy during the period ahead. This reflects a recognition of both the blunt ness of fiscal policy—it does not lend itself readily to frequent marginal adjustments—and the practical difficulties of turning it on and off quickly. The extended delays in enacting the 1964 and 1968 tax bills underscore the fact that taxes cannot be speedily adjusted in either direction under existing procedures, and significant changes in expenditure programs are also not easily accomplished. For these reasons, we think it is appropriate to settle as many issues as possible about fiscal policy at the time the annual budget and economic program are formulated. We should be prepared to alter these decisions if major unforeseen circumstances develop. But we would expect—more often than not—to live with those decisions until the following year's regular budget review. Monetary and debt management policies, on the other hand, are by nature considerably more flexible. General directions can and should be formulated in advance. But we believe they should always be considered as only tentative and provisional, based on the assumption that the planned fiscal policy and other developments will unfold as anticipated. Despite all the progress that has been and continues to be 72 made in forecasting the behavior of the private economy, projections still are sometimes wide of the mark; moreover, fiscal policy commonly turns out to be different from anticipations at the time of the January budget. When errors are made in forecasting or in projecting future fiscal policy, the inherent flexibility of monetary policy is very useful: It can be used either to probe the new situation, standing ready to pull back if the signs prove to have been misread, or to make a wholesale swing away from earlier conceived directions. An overly detailed specification in the original program that might diminish this flexibility would, in our opinion, be inappropriate. A masterpiece of coordination in an initial program might end up in a nightmare if it precluded a continuing adjustment to events as they actually unfold. Question /J. If you believe a program should be specified, do you believe that the President should be responsible for drawing up this program, or alternatively should such responsibility be dispersed between the Federal Reserve System and agencies responsible to the President? (Please note that informal consulting arrangements can be made as desired whether responsibility is assigned to the President or divided between the President and the Federal Reserve. The concern here is with the assignment of formal responsibility for drawing up the economic program.) CEA response. We believe that ultimate responsibility for achieving the Nation's economic objectives rests jointly with the President and the Congress, as the elected representatives of the people. Together, they eventually must settle on a broad economic program and see that the actions needed to carry it out are taken. The nature of the system, of course, assigns to the President the tasks of initial formulation of such programs. We believe this is altogether appropriate. The competing views of all must at some point be reconciled into a cohesive program and, in our opinion, responsibility for this reconciliation should rest in the highest elected office. As noted in our answer to the preceding question, we believe that the annual budget message and the other key Presidential messages formulating economic programs should contain detailed specification of all fiscal policy recommendations. The President's responsibilities for these messages are firmly established in the laws of the land. Of course, the programs are only recommendations, since the Congress ultimately bears the responsibility for enacting the enabling legislation. Some discussion of the role of monetary policy should also be included in the President's economic program. This discussion, as is true of the rest of the message, should reflect the President's considered view of what is best for the Nation in the current circumstances, with the Federal Reserve and other knowledgeable agencies giving freely of their advice in helping to formulate the program. But the key point, as explained in the preceding answer, is that the role assume for monetary policy should be provisional and couched only in general terms, so that the monetary authorities are not inhibited in responding to events as they actually unfold. The President can and should continue to make his views known on monetary policy issues as significant questions arise, and in our opinion, the Federal Reserve should give careful consideration to these views in its decisions. Ultimately, however, the Federal Reserve is answerable for its actions to the Congress. We believe that this divi- 73 sion of responsibilities is workable and indeed has generally worked satisfactorily. We have some suggestions to make in our response to the questions about reorganizing the Federal Reserve which are intended to assure continued coordination of policy between the Federal Reserve and the executive branch of the Government. Question 1.3. Concerning monetary policy guidelines: A. Should monetary policy be used to try to achieve the goals of the Employment Act via intervention of money supply (defined as desired) as provided in H.R. 11, or alternatively should H.R. 11 be amended to make some other variable or variables the immediate target of monetary policy; for example, interest rates, bank credit, liquidity, high-powered or base-money, total bank reserves, excess reserves and free reserves? Please define the target variable or combination of variables recommended and state the reason for your choice. {If desired, recommend a target variable or vari,ables not listed here.) It would be most helpful if, in providing the reasons for your choice, you list the actions the Federal Reserve should take to control the target variable (or variables) and also explain the link between your recommended target of monetary policy and the goals of the economy as defined by the Employment Act. B. Should the guidelines of monetary policy be specified in terms of some index of past, present, or future economic activity, or alternatively in terms of the target variable''s value or growth? For example, should the President's 1969 program for achieving the goals of the Employment Act be formulated to require consistency with some set of overall indicators of economic activity, or alternatively so that your target variable attains a certain vdkie or growth regardless of the economic winds? Please indicate the reasons for your preference. C. For only those persons who recommend that some index of economic activity be used to guide the monetary authorities in controlling the target variable: Should we use a leading {forward looking), lagging {backward looking) or coincident indicator of economic activity? It would be most helpful also if you ivould identify the index you would like to see used and specify how the target variable should be related to this index. D. For only those persons who recommend, that the guidelines be put in terms of the target variable's value or growth: Should the same guidelines be used each year irito the foreseeable future, or alternatively, should new guidelines be issued at the beginning of each year conditioned on expected private investment, Government spending, taxes, etc.? Please indicate the reasons for your preference. E. For only those persons who recommend that the guidelines be put in terms of the target variable''s value or growth and who also recommend that the same guidelines be used year after year into the foreseeable future: What band of values or range of growth do you recommend? {By way of clarification,, a band of values appears appropriate if your target variable is, say, free reserves, whereas a range of growth is appropriate if it is, say, money supply.) 21-570—68 6 74 F. For all those persons recommending that the guidelines be put in terms of the target variable's value or groioth (regardless of whether you recommend using the same guidelines year after year or revising them each year in light of expected private investment and fiscal policy): Under what circumstances, if any, should the monetary authorities be permitted during the year to adjust the target variable so that it exceeds or falls short of the band of values or range of growth defined by the guidelines issued at the beginning of the year? CEA response. We believe that flexible, discretionary monetary policy has made an important contribution to the achievement of the Nation's economic objectives and that it can continue to make such a contribution in the future. We do not, however, believe that it is possible to select any single one-dimensional guide for the conduct of monetary policy that will be satisfactory in all circumstances. Indeed, questions 1.3. A - F seem to suggest a much tighter connection between monetary variables and the Nation's ultimate economic goals than we believe actually exists. It is our view that in seeking guides for monetary policy the Federal Reserve should look primarily to those major measures of our overall economic performance that economic policy ultimately hopes to influence. These include total output^ together with its rate of growth and its relation to productive capacity; employment and unemployment; the behavior of prices and wages; and the Nation's balance-of-payments position. Since it is well known that monetary policy affects these major targets of economic policy only after some lag, we believe the Federal Reserve must base its policies not on the most recently recorded values of these target variables but on forecasts of their values extending several quarters into the future. Since there is commonly some uncertainty concerning the behavior of Federal expenditures and taxes, forecasts of fiscal policy as well as of the behavior of private demand are required. Such forecasts should be revised frequently as new data relating to the performance of the economy in the recent past become available. The forecasts will, of course, be conditional, based on an assumed monetary policy to be followed by the Federal Reserve, and the System should be prepared to adjust its policy as changes in the outlook seem to require. All sectors of the economy and components of aggregate demand are not equally affected by monetary and credit conditions. In its conduct of monetary policy, the Federal Reserve should, therefore, consider the probable impacts of its actions on specific sectors. In particular, it seems clear that residential construction is strongly affected by monetary policy, in large part because of peculiarities in the institutional arrangements for financing homebuilding. For that reason, in the conduct of monetary policy it is especially important to consider the probable effects on housing activity. In addition, there may also be, under some conditions, disproportionate effects on debt-financed spending on schools, highways, and other public facilities by State and local governments. It is our view that the Federal Reserve should operate by influencing directly those variables that will, with a lag, affect the future values of the target variables it is attempting to influence. This means, 75 in our opinion, that it should focus primarily on interest rates and the availability of credit. For example, it must attempt to judge whether interest rates in the short-term open market are such as to generate the flows of funds through thrift institutions that are needed to support the mortgage commitments and mortgage loans required to achieve appropriate levels of housing activity in the future periods of its forecast. The Federal Reserve clearly cannot control independently both interest rates and the stock of money, since the two are linked together. On the one hand, it can focus primarily on influencing interest rates in order to obtain the flows of credit to key sectors of the economy that are conducive to an appropriate level of overall economic activity, allowing the money supply to be whatever it has to be to achieve these results. Or, alternatively, it can focus primarily on controlling the money supply, allowing interest rates to take on the values that are consistent with the money supply so determined. If relationships in the financial sector were fixed and unvarying, it would make little difference which approach was taken. But it seems clear that this is not the case. There are frequent innovations in finance, and also evolutionary changes in behavior in the private sector such as the increase that has occurred in the last few years in the sensitivity of investors to relative changes in interest rates on different types of financial assets. Moreover, during any short-run period, there can be marked changes in expectations which significantly affect investors' choices among financial assets. In such a changing financial environment, the level and rate of growth of the money supply required to achieve the desired behavior of interest rates and credit availability may change considerably from one situation to another. Since such empirical and theoretical evidence as is available strongly indicates that it is interest rates and credit availability rather than the money supply per se that affect spending decisions, it seems wiser for the Federal Reserve to concentrate primarily on control of interest rates and credit conditions, letting the money supply adapt itself. We realize that there are some economists who believe that there is a very close connection between the money supply and GNP and that monetary policy should therefore attempt single-mindedly to control the money supply. (Among those who hold this view there is some dispute about the proper definition of the money supply—some would include only demand deposits and currency while others would also include time deposits.) We do not, however, share this emphasis on the overriding importance of the money supply (however defined). The fact is that there is no simple and apparent relation between the money supply and GNP. Moreover, we see no plausible reason why there should be such a close relationship. It should be understood that the Federal Reserve does not give people money—indeed, it is incapable of changing the public's wealth or net worth (except to the relatively minor extent that it causes changes in the market value of existing debt claims thereby generating capital gains or losses). Federal Reserve operations change the composition of the public's balance sheet by inducing people voluntarily to exchange one asset for another or to increase or decrease both their assets and their liabilities by equal amounts. Thus, these operations affect a wide variety of the public's financial assets and liabilities. Movements—that is, flows—of all of 84 these assets and liabilities can have repercussions on real economic activity, and thus must be monitored carefully in the conduct of monetary policy. Out of the myriad of items in the public's balance sheet, we can see no logical reason for attaching overriding importance to one particular entry defined as the money supply. Question H. Concerning debt management policy: Given the goals of the Employment Act, what can debt management do to help their implementation? (If you believe that debt management has no role to play in this matter, please explain why.) CEA response. Debt management policy plays some role in helping us to achieve our economic goals. But we believe its role is somewhat more marginal than the roles played by fiscal and monetary policy. In broad terms, fiscal policy determines among other things, how much debt there is to be financed, and monetary policy determines what portion of the debt is absorbed by the central bank and what portion is absorbed by the public. In comparison with these rather basic issues, we think of debt management—the exact timing, maturity, and other terms of financing—as being of second order importance. Within its limited sphere of influence, one might distinguish between long- and short-run effects of debt management. In the long run, debt management makes its mark by influencing the term structure of the debt held by the public. The full implications of alternative term structures of debt are not yet well understood. But we believe the term structure of debt has some influence on the structure of interest rates and ultimately on spending-saving decisions by various sectors of the economy. Moreover, a suitably balanced debt structure can help to provide a financial environment in which monetary policy works more effectively. The long-run nature of the effects of debt management operations on the structure of the debt bears emphasizing. The largest single debt management operation in recent years involved an exchange of less than 6 percent of the total amount of marketable Federal debt outstanding in private hands, and it changed the average maturity of the total marketable debt by less than 5% months. Most operations have been considerably smaller in size. To produce a major change in the term structure of the debt would require a whole series of fairly sizable operations. In the short run, debt management operations can at times contribute to the achievement of certain of our objectives by taking advantage of rigidities in financial markets and also by working on market psychology. In the early 1960's, increases in the supply of short-term securities outstanding helped to hold up short-term interest rates, thereby reducing the capital outflows that were adversely affecting the U.S. balance of payments. In the fall of 1966, rigid controls on the size and timing of offerings of Federal agency securities helped to restore confidence in financial markets following the near panic situation that developed during the summer. More recently, offerings have been scheduled in such a way as to minimize direct competition with savings flows to the thrift institutions so that these institutions would not be unduly limited in the funds they have available for making new mortgage loans. 77 Question IS.A. H.R. 11 makes no provision whatever for conducting open market operations for so-called defensive or road-clearing purposes, that is to counteract seasonal and other transient factors a,ffecting money market and credit conditions. Bo you see any merit in using open market operations for defensive purposes or should they be used only to f militate <achievement of the President's economic program and the goals of the Employment Act? What risks and costs, if any, must be faced and paid if open market transactions are used to counteract transient influences? CEA response. We believe that so-called defensive open market operations are an appropriate part of Federal Reserve activities and that the System should continue to engage in such operations. Of course, as a practical matter, defensive open market operations are not readily distinguishable from offensive operations. But granted that such a distinction is conceptually possible, we feel that both types of operations are needed to provide a smoothly functioning monetary system that can both easily accommodate the multitude of daily financcial transactions that are necessary for our Nation's commerce and business and also transmit efficiently the monetary policy forces aimed at moving the economy closer to our national goals. The purpose of defensive open market operations is to counter the effects of short-run swings in factors that would otherwise generate either excessive tightness or excessive ease in financial marlsets. If swings in these other factors were self-canceling, defensive open market operations would not be necessary. But the other factors are not that well behaved. Huge individual financial transactions may get bunched into a particular day or week, the public's demand for currency may suddenly spurt, check collection schedules sometimes are interrupted by weather, and international developments can cause a sudden surge in gold or deposit flows. Left uncountered, these developments would lead to erratic fluctuations in short-term interest rates as demands for funds varied in relation to the available supply. Perhaps this would not be disastrous, but it would introduce unnecessary complications in the conduct of business which can be easily avoided by using the Federal Reserve's open market operations to even out flows of funds in the market. In general, those who favor the elimination of defensive open market operations believe the Federal Reserve should seek single-mindedly to control some well-defined quantitative index of monetary conditions, such as the stock of money. For reasons indicated in our earlier answers, we consider this an overly simplistic approach to monetary policy. Under the approach we favor, which places much more emphasis on interest rates and credit availability as guides to policy, it is difficult to distinguish sharply between defensive operations and other kinds. Nor is there a need to make such a distinction, since the objective of policy is to move credit conditions smoothly in directions that will contribute to the achievement of our economic goals. 78 Questions I\5.B and 1.5.C.: B. Do you believe that monetary policy can be effectively and efficiently implemented solely by open market operations? C. For w\hat purposes, if any, should (a) rediscounting, (b) changes in reserve requirements, and (c) regulation Q be used? How might H.R. 11 be amended to implement yowr recommendations? CEA response. Our response considers both of these questions together. As we have indicated in a previous answer, we believe the objective of monetary policy should be to influence the cost and availability of credit in ways conducive to economic stability. We further believe that the chief means of influencing credit conditions should be through regulation of the supply of reserves available to the commercial banking system for credit creation. Since open market operations are the most flexible and effective tool for expanding or contracting bank reserves, we believe they should—and do—constitute the primary instrument of monetary policy. The Federal Reserve discount window is best viewed as a safety valve which enables banks—at a price—to escape pressures occurring during periods of tightening credit when these pressures may inadvertently become unduly concentrated on particular banks. By providing relief directly to the banks that are most in difficulty, the discount mechanism permits the global pressures caused by open market operations to be brought about more aggressively than would otherwise be possible and thereby makes monetary policy more effective. We would like to see the discount rate changed somewhat more frequently and routinely than has customarily been the case in order to keep it in a more consistent relationship to short-term market interest rates. At the same time we recognize that the discount rate is on occasion a useful signal of the Federal Reserve's intentions—especially to the international financial community at times of serious balanceof-payments difficulties—and we would therefore not favor the entire elimination of discretionary changes in the rate. Changes in reserve requirements are said to have advantages which may at times make them superior to open market operations as a tool for conducting monetary policy. One alleged advantage is that reserve requirement changes provide a definite signal to all observers that policy has changed. Another is that reserve requirement changes affect all member banks immediately in contrast to open market operations whose effects tend to show up first in the money centers and only gradually spread to outlying banks. While the evidence in support of these supposed advantages is somewhat limited, we believe that reserve requirement changes may on occasion be useful as a tool of monetary policy. On the other hand, we feel that frequent changes in reserve requirements would be undesirable; and, indeed, changes have been made quite infrequently in recent years. 79 The powers entrusted to the Federal Reserve under regulation Q enter the monetary policy process at a somewhat different point from the other instruments discussed above. Regulation Q has no direct effect on bank reserve availability. Rather it may be viewed as the second blade of a pair of scissors, cutting off bank competition for time deposits whenever a squeeze on reserve availability through open market operations or one of the other instruments pushes interest rates up near or beyond the regulation Q ceilings. In an ideal world, we would not favor the use of administrative ceilings to prevent healthy competition for funds among financial institutions. But the experience of the past several years makes it amply clear that certain of our financial institutions—particularly mutual savings banks and savings and loan associations—encounter serious problems when interest rates rise to too high a level. Experience also demonstrates that pressure on these institutions and on the housing sector which they are so important in fiancing can, under some circumstances, be relieved by skillful adjustment of the regulation Q ceilings. Several of the financial reform measures enacted by the Congress during the past 2 years have helped give these financial institutions a little more flexbility to live in a high interest rate world. But as long as their funds are invested mostly in long-term, rather illiquid assets bearing interest rates characteristic of several years ago, it appears that there will continue to be a need for at least a standby authority to set interest rate ceilings. Question I.5.D. Do you see any merit in requiring the Federal Reserve Board to make detailed quarterly reports to the Congress on past and prospective actions and policies? Are there any risks and costs in this procedure? In what ways, if any, icould you modify the reporting provision? What information do you believe should be included in such reports as you recommend the Federal Reserve submit to the Congress? CEA response. The Federal Reserve already makes numerous public reports of its actions, the explicit reasons for them, and its view of the general economic background underlying them. The interested observer can also piece together a reasonably good story of his own about what the Fed has been doing by following the weekly and monthly banking statistics. We feel that the combination of these reports and statistics is adequate to meet most legitimate needs. If the Congress, in its role as overseer of the Fed, should see the need for still more information, however, we see no reason to object. Full and frank reviews of recent actions and the reasons for them can improve understanding and ultimately bring us a step closer to our basic economic goals. But we would strongly caution against attempts to force the Federal Reserve to spell out in detail what its current policy stance is and what actions it plans to take in the future. Thus, we are completely opposed to that part of the proposal in H.R. 11 requiring details on "prospective actions and policies." Attempts to pin the System down on prospetcive actions can only inhibit its flexibility in dealing with actual situations as they develop. As indicated in our response to question 1.1, this flexibility of monetary policy is something that we feel should be preserved to the greatest extent possible. 80 Question I.5.E. What costs and benefits would accrue if representatives of the Congress, the Treasury, and the CEA were observers at Open Market Committee meetings? CEA response. We see little benefit from having CEA or other outside observers attend Open Market Committee meetings. Policymakers throughout the administration and the Congress have always reserved the right to deliberate in private, and we feel that the Federal Reserve has that same right. The presence of outside observers might work to decrease the candor and independence with wThich views are expressed in the FOMC. The suggestion that a CEA representative should be present at an Open Market Committee meeting seems to imply that we would then be in a better position {a) to press our own views and (b) to learn what Federal Reserve policy actually is. Actually, however, the spirit of cooperation that has been built up during the past 8 years has given us adequate opportunity to make our views known and to hear the views of others, and it is doubtful whether either we or the Federal Reserve would benefit further in these respects by our attending FOMC meetings. Question II. Appraisal of the structure of the Federal Reserve. H.R. 11 provides for the following structural changes in the Federal Reserve System: 1. Retiring Federal Reserve bank stock; 2. Reducing the number of members of the Federal Reserve Board to five and their terms of office to no longer than 5 years; 3. Making the term of the Chairman of the Board coterminous with that of the President of the United States; 4. An audit for each fiscal year of the Federal Reserve Board and the Federal Reserve banks and their branches by the Comptroller General of the United States; and 5. Funds to operate the Federal Reserve System to be appropriated by the Congress of the United States. Please comment freely on these several provisions. In particular, it would be most helpful if you would indicate any risks involved in adopting these provisions and discuss whether their adoption would facilitate the grand aim of H.R. 11, which is to provide for coordination by the President of monetary and'fiscalpolicies. CEA response. If one were starting from scratch, one would probably propose a structure for the Federal Reserve System substantially different from the present one. However, in evaluating proposals for reform and reorganization of the System under present conditions, its historical evolution and its effectiveness in performing its functions must be taken into account. We believe that, on the whole, the Federal Reserve has performed effectively in recent years in adapting monetary policy to the changing domestic economy and our balance-of-payments situation. Since an effectively functioning institution is more important than a logical organization chart, we believe there is a need for caution in recommending drastic changes in Federal Reserve organization. Our detailed comments on the proposals contained in H.R. 11, which follow, reflect this view. II. 1. Retiring Federal Reserve stock.—We believe that it is somewhat anomalous for a public institution such as the Federal Reserve to be "owned" by private stockholders. At the same time, however, we SI do not see that this anomaly has caused any difficulty with respect to determination of System policy. Thus, we see no compelling reason for eliminating the stock ownership. The risks involved in retiring the stock seem small, though there is always the possibility that confidence in the Federal Reserve could be weakened if the action were construed to imply a fundamental change in control of the System. On balance, we would favor this reform, although we do not believe the issue is very important. II. 2. Reducing Board membership to five and terms of office to no more than 5 yecvrs.—We note that H.R. 11 includes a provision abolishing the Federal Open Market Committee. Although no reference is made to this provision in the questionnaire, it is, in our opinion, the most important change in Federal Reserve structure contained in H.R. 11. Since the proposed abolition of the FOMC has a bearing on the question of Board membership, we consider the two provisions jointly. Taken by itself—that is, assuming retention of the FOMC—the proposal to reduce the number of Board members to five seems unwise because of its implications for the balance of power between Board members and Reserve bank presidents. Under present arrangements the seven members of the Board can make their common views prevail in the FOMC, since only five of the bank presidents vote in that Committee at any one time. We believe that this balance is appropriate and should be preserved, so that monetary policy, at least in principle, can be determined by presidentially appointed officials. This assurance would be removed if the Board were reduced to five members, unless there were simultaneous change in the structure of the FOMC. Leaving aside the key question of balance within the FOMC, we see advantages in having a smaller Board. We believe that a Board of five might be somewhat more effective than one of seven. With respect to length of term, we accept the philosophy in the Federal Reserve Act that Board appointees should have terms long enough to insulate them from political pressures. But we also believe that the present 14-year term is longer than necessary for this purpose and also so long that it limits in an undesirable way the turnover of views and ideas. If the present seven-man Board is to be retained, a term of 7 years would strike a more appropriate balance among the various objectives. On the other hand, if the Board were to be reduced to five members, we believe a term of 10 years would be appropriate, rather than the 5-year maximum term contemplated in H.R. 11. Turning to the key question of abolition of the FOMC, we encounter conflicting considerations. A proposal to abolish the FOMC and turn all the Federal Reserve's monetary policy powers over to a fivemember Board was made in 1961 by the highly regarded Commission on Money and Credit. The Commission's rationale for this recommendation was that monetary policy should be in the hands only of officials who are appointed by the President and confirmed by the Senate. We svmpathize very strongly with this objective. Moreover, the present FOMC, consisting of 12 members, is a somewhat cumbersome administrative body, a fact which requires a high degree of diplomatic skill on the part of the Chairman in achieving the consensus of views necessary to conduct an effective and coherent monetary policy. Thus, 82 there are sound arguments for eliminating the FOMC and concentrating power in the Board. But there are also risks involved in making drastic changes in the organization of an institution that has, on the whole, performed its functions satisfactorily. In some instances, Reserve bank presidents have made very important contributions to the formulation of monetary policy. Furthermore, the Reserve bank presidents have often played a useful role in the collection and presentation of information concerning economic developments in their regions and in the administration of nationwide banking and credit policies in those regions. If the FOMC were abolished, with the Reserve bank presidents acting only in an advisory capacity with no actual vote in policy formulation, it seems certain that the stature of the office and of the personnel occupying it would be sharply reduced, perhaps with adverse effects on the relations between the Reserve banks and their regional communities. That is, abolition of the FOMC would probably produce a drastic change in the character of the Federal Reserve System, with results that are rather difficult to predict. For this reason, we are hesitant to recommend abolition of the FOMC, even though we can see some advantages in it. As an alternative to abolition of the FOMC, we believe the Congress should consider making Reserve bank presidents subject to Presidential appointment and Senate confirmation. This would put the Reserve bank presidents in the same category as other officials with major responsibility for national economic policy. II. 3. Making the term of the Chairman of the Board coterminous with that of the President of the United States.—We fully support this proposal. We note that Chairman Martin himself has repeatedly supported such a provision and that on April 17, 1962, President Kennedy submitted to the Congress a message making a similar recommendation. We believe that enactment of this proposal would help provide the basis for increased trust between the President and the principal officer responsible for monetary policy. The Chairman would be better able to participate in the councils of the executive branch and the Nation would be bettter assured of effective coordination of economic policy. We note that H.R. 11 proposes to give to the Chairman the power to designate a Vice Chairman. We would prefer to leave this power with the President, as the Federal Reserve Act presently provides. We would, however, recommend a proposal making the Vice Chairman's term in that office also coterminous with that of the President, in line with recommended change in the term of the Chairman. We might also note that the current dating of Board terms is not very compatible with an attempt to make the term of Chairman coterminous with that of the President. Under present law, terms of Board members expire on January 31 of each even year. Thus, a new President taking office on January 20 of an odd year might have to wait as long as a full year for a vacancy to open up on the Board so that he could appoint the man of his choice as a member and Chairman. This problem would be reduced if at least one member's term expired each year, as contemplated both in H.R. 11 and in our recommendation for 7-year terms in our response to the preceding question. If terms continue to expire every other year, however, we would suggest changing the law so that the expiration dates fall in odd rather than even years. 83 A somewhat related problem that could inhibit the President's choice of a new Chairman arises from the present geographical and occupational limitations on Board membership—especially the restriction that no more than one member may be from any Federal Reserve district. This could prevent the President from securing the best qualified Chairman if his choice should happen to come from a district already represented on the Board. We suggest that the Congress consider an amendment that would eliminate these restrictions entirely, or at least liberalize them. 11.4. Animal audits of the Board, the Reserve banks, and their branches by the Comptroller General.—We believe that the auditing procedures presented in effect in the Federal Reserve System are satisfactory from the viewpoint of assuring that the Federal Reserve operates with efficiency and economy and in the public interest. Those procedures involve a complete examination annually of each Reserve bank and branch by the Board's staff of examiners, an examination of one of the Reserve banks each year by a commercial auditing firm, and a continuous audit of each bank by a resident auditor, responsible to the bank's board of directors. These examinations check not only the bank's financial condition, but also its discharge of all responsibilities and its compliance with law and regulations. In addition, the Board itself is audited each year by independent public accounting firms. Reports of these various examinations are available to the appropriate committees of the Congress. 11.5. Congressional appropriation of fu/nds to operate the Federal Reserve.—We doubt the need for this provision. As indicated in our previous answer, we believe that the Federal Reserve is operated efficiently and economically. Moreover, virtually all of the System's earnings above operating expenses are already paid over to the Treasury of the United States. In fiscal year 1968 alone, this payment was almost $2.1 billion. A significant reduction in the net expense of the System to the taxpayers presumably could be achieved only by curtailing operations, which we believe would be unwise. Question III. Your analysis of monetary development, since 196J^ including induced changes and their effects on economic activity, is invited. CEA response. The Council's 1968 annual report has already summarized our views of monetary developments up through the end of 1967. We confine our comments, therefore, to developments thus far this year. Monetary and financial developments in 1968 fit broadly into two periods. Interest rates climbed sharply in the early months of the year, as monetary policy tightened in defense of the dollar and to curb mounting inflationary pressures at home while enactment of the proposed tax surcharge continued to be delayed. Although there were temporary interruptions in the upward trend, by mid-May most rates had climbed one-half to a full percentage point from their early 1968 lows. High quality corporate borrowers were paying more than 7 percent for funds and 3-month Treasury bills commanded a rate as high as 5.90 percent. During this period, interest rates reached peaks higher than those attained in the widely heralded monetary crunch of 1966. The breakup in late May of the logjam on the tax increase and its ultimate enactment and imposition brought a marked easing of pressures and fears throughout financial markets. And with fiscal policy finally assuming a more proper role, monetary policy was able to relax 84 somewhat the restraint imposed earlier, giving a definite signal in this direction with a cut in the Federal Reserve discount rate initiated on August 16. Market interest rates reflected these developments, dipping in early August to levels below their lows set early in the year. This downtrend has been somewhat reversed in more recent months, and in early November the prevailing level of rates wTas high by historical standards. The degree of monetary restraint imposed in the first several months of the year was quite severe. At a time when a burgeoning economy was sharply stimulating private demands for credit and when the Federal Government was also a heavy net borrower in contrast to its usual seasonal repayment of debt, increasingly restrictive steps by the monetary authorities effectively slowed the amount of total credit creation, particularly by commercial banks. These steps had actually begun with the increase in the Federal Reserve discount rate from 4 to 4y2 percent following the devaluation of sterling in November 1967. This was followed by two further increases in the spring of 1968, bringing the rate to 5y2 percent by late April, its highest level since the 1920's. Meanwhile the Federal Reserve had also increased reserve requirements against member banks' deposits, and had steadily tightened its open market policy. The result was that growth of total bank credit slowed from a Hy 2 percent rate during 1967 to a 6y2 percent rate in the first 6 months of 1968. Credit demands were strong but banks simply could not meet them, as interest rate ceilings established under regulation Q made it increasingly difficult for the banks to attract new deposits to support their lending operations. The lower level of market interest rates prevailing in more recent months has restored banks' ability to compete effectively for time deposits. And acquistion of these and other funds has supported rapid growth in bank credit since mid-year. Special factors accounted for much of this gain, but it is clear that the easing in monetary policy was filtering through to increased availability of credit. Flows of funds to and from other financial intermediaries appear also to have improved somewhat recently compared with experience during the spring. Indeed, after reports of increasingly severe shortages of mortgage funds in the late spring, which helped to bring a sharp curtailment in home building activity, the more recent signs suggest that home builders and buyers have not had any undue difficulty obtaining mortgages, although still at high interest rates. The one financial variable that has been rather at odds with the general picture described above is the narrowly defined money supply. Thus after growing at a relatively moderate rate in the first 3 months of the year, growth of the money supply accelerated very sharply during the April-July period. In large part, this seems to reflect a slow adjustment by the private sector of the economy to an unusually large payout of Government deposits during this period. Rising transactions needs associated with the rapidly growing economy and a heavy volume of securities market transactions may also have been a factor. As we interpret it, this surge in the money supply was not indicative of an early easing in monetary policy. Nor do we believe that return to more normal money growth in subsequent months reflects a tightening in monetary policy compared with its posture during the spring and early summer. Developments this year point up the inherent dangers in focusing exclusively on so narrow a financial variable as the money supply. STATEMENTS OF RESPONDENTS ON QUESTIONNAIRE CONCERNING H.R. 11 STATEMENT OF E. SHEKMAN ADAMS, FIRST NATIONAL CITY BANK I . MONETARY POLICY GUIDELINES AND OPEN M A R K E T OPERATIONS Coordination of fiscal, debt management and monetary policies is obviously desirable, provided it helps to achieve wise policies. It does not follow, however, that a "program" to accomplish coordination should be announced at the beginning of each year. The word "program" suggests a degree of inflexibility that would be highly undesirable in economic policymaking. This applies particularly to monetary policy. Your questions about monetary policy guidelines seem to imply that monetary policy should be conducted according to some sort of formula. One of the great virtues of monetary policy is that it is flexible and can be adjusted to changing conditions. I do not think that anyone can intelligently set targets for a whole year in advance for any of the variables influenced by monetary policy. It is useful, nevertheless, to have some framework for thinking about monetary policy. I find it helpful to think in terms of the ways in which monetary policy affects economic activity. Monetary controls are effective and have significance because they affect expenditures made by individuals, by businesses, or by governments. These effects upon spending are mostly indirect and are brought about through the influence of monetary policy on credit conditions. They reflect the reactions of the community to the credit conditions which the monetary authorities are able to influence; namely, (1) interest rates, the cost of credit, (2) the availability of credit, which is reflected chiefly in the lending and investment policies of various suppliers of credit, and (3) changes in the money supply, especially those which reflect changes in bank credit. All of these three factors need to be taken into account. Under particular circumstances, one or another of them may be of much greater importance than the others. But none of them should ever be ignored. This is one reason why monetary policy cannot be effectively and efficiently implemented solely by open-market operations. These operations affect primarily the volume of bank credit. The monetary authorities need additional controls to exert the influence they should be able to exert on interest rates and the availability of credit. It also explains why I would not favor amending the Employment Act to make specific reference to the growth of the money supply. I would fear that such a change might encourage paying too much attention at times to this one factor and not enough attention to interest rates and credit availability. (85) 86 It would be far more constructive to amend the Employment Act to make specific reference to price stability as a major goal of public economic policy. You ask about debt management policy. I think that the usefulness of public debt management as a means of shortrun economic management is quite limited. In theory, debt management could be used to combat inflationary or deflationary swing? in the economy, but this is seldom true in practice. For instance, during a boom, countercyclical policy would call for the issuance of long-term bonds by the Treasury to curb capital spending. As a practical matter, however, one cannot expect the Treasury to do much long-term financing when interest rates are at historically high levels and when nonbank investors have no desire to add to their holdings of Treasury securities. I I . STRUCTURE OF T H E FEDERAL RESERVE I share the view of the great majority of monetary economists that it is definitely in the public interest to protect the Federal Reserve from greater political pressure. Most of the structural changed proposed by H.R. 11 are designed to undermine the present degree of semi-independence of the Federal Reserve within the framework of government and are therefore undesirable. One exception is that I think there may be merit in making the term of the Chairman of the Federal Reserve Board coterminous with that of the President of the United States. I have no judgment on the advisability of reducing the number of the members of the Board. I I I . RECENT MONETARY POLICY The major monetary developments of recent years have stemmed largely from the unprecedented expansion of the U.S. economy accompanied by inflationary policies on the part of the Federal Government. Large budgetary deficits, incurred when the economy was operating close to capacity, have been largely responsible for creating inflationary pressures throughout the economy, including the reactivation of the wage-price spiral which the Government has done little to restrain. We are now in the midst of an inflationary boom, the outcome of which cannot be predicted. Similarly, the Government's failure to deal effectively with the U.S. balance-of-payments problem has brought the American dollar into serious jeopardy, and the end of this story has not been written yet either. The monetary authorities have been acutely aware of these developments and have taken them into account in formulating their policies. On the other hand, it has been apparent that monetary policy could not achieve price stability and balance-of-payments equilibrium singlehanded. As Allan Sproul once observed, we cannot expect monetary policy to offset all the unwise policies in the rest of the economy. Realizing this, the monetary authorities presumably felt that the least bad alternative for them most of the time was to permit monetary expansion to continue at a rather rapid pace, probably at a faster pace than they would really have liked. In the spring of 1966, the Reserve authorities apparently reached the conclusion that the situation was worsening to such an extent that a restrictive policy was called for. The "credit crunch" that followed 87 again demonstrated that, if it is used boldly, monetary policy can be a powerful brake on the economy. However, policy became so restrictive that it threatened to create a chaotic situation in the financial markets, and this again demonstrated the fact that, as a practical matter, there are real limitations on the extent to which this brake can be applied in the real world. However, the main lesson of recent monetary developments relates to the matter of the coordination of monetary policy and other public economic policies. The fiscal policies and other policies of the Federal Government have had an inflationary impact on the economy. Although the monetary authorities have not had the power to correct this situation, they nave done their best to exert a constructive influence—which is more than can be said for many officials in Government. Whether they should have done more or less than they did is naturally a question for debate among the Monday morning quarterbacks. But this question is not really too important. What is important is that the monetary authorities have sought consistently to act in the public interest, whereas fiscal and other governmental policies have been unwise because they have reflected the pressures of political expediency. The obvious lesson of this is that if we seek to achieve greater coordination of monetary and fiscal policies, our aim should be to coordinate fiscal policy with monetary policy, not the other way around. In short, the authors of H.R. 11 are plainly concerned with a problem of great significance to our economic well-being. However, they are approaching it from the wrong end and with the wrong assumptions. The problem lies not with monetary policy but with fiscal policy. The need is not to destroy the semi-independence of monetary policy, but rather to improve the organization and procedures that will help to produce better fiscal policies. If the Congress would turn its attention to this problem, it could make a major contribution to the future growth and stability of the American economy. STATEMENT OF CAUL T. ARLT, UNIVEESITY OF ILLINOIS Re No. 1.—The idea of a program coordinating fiscal, debt management, and monetary policies is an appealing one. To set forth this program at the beginning of the year would involve a careful specification of the goals to be achieved. This is no easy task in view of the plurality of goal variables with all their inherent conflicts. I am assuming, of course, that the "goals of the Employment Act" would include the more recently acquired objective of achieving a better balance in the international payments position of the United States. I believe there is merit in a program of coordination in that it would require the Federal Reserve to "take a position" based on its understanding of the monetary mechanism, its reading of the economic indicators, and its evaluation of the influence of nonmonetary policy forces on the goals to be achieved. I should add, however, that if the monetary authorities are to be forced into a more formalized program of coordination, they assume an impossible burden if they must coordinate with the type of fiscal policy experienced within the last few years. Much of the criticism of the Federal Reserve with respect to allegedly inappropriate growth rates of the money supply should be analyzed in the perspective of the Federal Reserve attempts to cope 88 with the prolonged deliberations associated with efforts to institute fiscal restraint. Re No. 2.—I believe the President should be assigned the formal responsibility for drawing up the economic program. Re No. 3A.—Monetary policy should employ some intermediate target variable as it works to achieve the goals of the economy. There is need for some quantitative measure to indicate the thrust of monetary policy and one that is predictably linked to measures of spending and income. Unfortunately, what that measure may be is still a matter of dispute among economists. Arguments continue over the relative feasibility of such guides as interest rates, credit volume, some reserve measure, or the money supply. In my own thinking such measures as interest rates or bank net reserve positions (free or net borrowed) are poor indicators of the thrust of monetary policy. Because these reflect both credit demand and supply forces it is difficult to derive from them the contribution of the monetary authorities. In a related vein, it may be said that the monetary authorities have very little control over interest rates and the net reserve positions of commercial banks. If proposed legislation specifies a target variable, I would urge that it use either the growth rate of the money stock or the growth rate of the monetary base as the more appropriate measures of what the Federal Reserve is doing. But I would also submit, since economists are not in agreement with respect to the "best" guidepost, that legislators proceed cautiously in their specification or financial targets and avoid imposing hard and fast rules on the monetary authorities. Of the two target variables, money stock changes and monetary base changes, I prefer the monetary base. The supply of the monetary base is substantially under the complete control oi the Federal Reserve System. Recen/t studies have shown that movements in Federal Reserve credit determine most of the movements of the monetary base. Although member bank borrowing from Reserve banks and changes in the gold stock are not under the direct control of the monetary authorities, one may assume that open market operations may be used to offset short-term changes in these and other accounts in order to achieve a desired level of the monetary base. The demand for the monetary base consists of the demand of commercial banks for excess reserves and required reserves and the demand of the nonbank public for currency. Banks' demand for required reserves is a derived demand reflecting the demands for private demand deposits, Government demand deposits, net interbank deposits, and time deposits. Changes in the monetary base have an important influence on output, employment, and prices through an adjustment process in which banks and the nonbank public adjust their holdings of real and financial assets so as to bring the amount demanded of the monetary base equal to the amount supplied. In this process, economic activity, prices of real assets, and interest rates are changed. Empirical studies appear to show a relatively close relationship between changes in the monetary base and changes in the money supply. In the short run, however, changes in the money supply often reflect movements in Government demand deposits or movements between demand and time deposits. For this reason I tend to lean toward the monetary base rather than the money supply as the best available guide to monetary management. 89 Re No. 3 B and F.—It would be helpful if the Federal Reserve at the beginning of the year specified a desired rate of growth of the monetary base. The particular rate of growth selected would reflect the consensus hammered out in the coordinated program drawn up by the President. To repeat earlier parts of this statement, I would insist that this specified rate of growth of the monetary base be considered as a benchmark and not a binding prescription imposed on the monetary authorities. We are attempting to achieve a plurality of goals with inherent conflicts and we must recognize that during the year the importance attached to particular goals may change. Furthermore we are using a target variable for monetary policy which may or may not be the most appropriate measure or indicator. We do not yet know enough about the strength and predictability of any of the possible financial variables suggested as intermediate guides to policy to impose a -fixed course of action on the Federal Reserve. Accordingly I would urge that the Federal Reserve be permitted to deviate from the specified rate of growth if such action were accompanied by a detailed report to Congress explaining the rationale underlying its policy decisions. The distinctive advantage of specifying at the beginning of the year a particluar rate of growth of some financial variable is that the public gets a better understanding of the strategy employed by the monetary authority whenever the target growth rate is changed. In short, my position is that the monetary authority enjoys a wide area of discretion, but tied to that discretion is the responsibility for more detailed communication with the public. Re No. —We have not reached the point where debt management may be used as an important stabilization tool. The most we can hope for is the development of the neutral approach in which Treasury debt offerings become more "regularized." I would also urge the removal of the ^-percent interest rate ceiling on Treasury bonds to permit greater flexibility in debt management. It would be my hope that improved and more regular financings by the Treasury could then be achieved without requiring the Federal Reserve policy of "an even keel" during the period of financing. Re No. 5A.—I believe that money market facilities are adequate enough and the participants sophisticated enough to adjust to many of the money market changes now being cushioned by Federal Reserve "defensive" operations. Furthermore, it appears that Federal Reserve emphasis on money market stabilization or "money market strategy" has often led to unintended changes in such variables as total reserves and the money stock which, I believe, are more closely linked with the goals of spending and employment. If policy is to be defined quantitively in terms of a longer run target such as the desired rate of growth of the monetary base or money stock, the monetary authorities would, of necessity, reduce the scope of their operations designed to influence shorter run money market variables. Re No. 5 B and C.—Under most circumstances I would favor placing complete reliance on open-market operations. Reserve requirements I would not change except during war emergencies and then it would be more feasible to take off the limits to reserve requirement increases. The discount window as currently administered is not very 21—570—-68 7 90 effective. If current proposals for change, particularly with respect to "automatic drawing rights" and more frequent changes in the discount rate, are implemented, the discount window might prove to be an effective supplement to open-market operations. Regulation Q interest ceilings should be removed. Far too many distortions in the flows of funds are produced in financial markets where some interest rates are held by law or administrative decree, while others are allowed to fluctuate freely. The experience of 1966 is a case in point. Re No. 5 D and E.—I see merit in more detailed reporting to Congress by the monetary authorities, particularly with respect to past actions and policies. Prospective actions and policies should be indicated only in general terms and, as explained earlier in this statement, I would not want the monetary authorities locked in by a prescribed rate of growth of a target variable. In keeping with my belief in a more complete disclosure of the rationale underlying Federal Reserve actions and policies, I see merit in the proposal to have selected observers at the Open Market Committee meetings. Some procedures would have to be adopted to prevent indiscriminate revelations of FOMC deliberations and actions. Re Part II on structure.—In keeping with the idea that the Federal Reserve be less independent of the President's office and more independent of the banking community, I am in full support of proposals No. 3 and No. 1. ^ As long as the Board must assume its numerous supervisory functions in addition to its monetary policy function, I see no merit in reducing Board membership to five, nor do I see any advantage in limiting terms to 5 years. I am strongly opposed to propositions 4- and 5. These proposals would contribute nothing to the President's program and at the same time would violate what I regard as a healthy independence of the central bank within Government. Re III comments on recent monetary policy.—In retrospect it appears that the Federal Reserve was too drastic in its restraint from April to^ October 1966 after having been too expansive in early 1966 despite intentions to restrain. Had the Board's monetary strategy been geared to a target rate of growth of the monetary base or the money supply the economy might have been spared the sharp changes that developed in 1966. The expansive policy of the monetary authorities in the first half of 1967 was appropriate in view of the marked slowing up in the pace of economic activity. The second half of 1967 is another story. In the face of growing demands and rising spending, monetary policy was too expansive. Although the Federal Reserve was aware of the expansiveness of its monetary management, it avoided restraint because of the constraints of "even keel," the fear of renewed disintermediation, the influence of impending tax legislation, and the concern over the position of the British pound. Underlying these concerns was the fear that interest rates might rise too high if monetary policy were to swing over to restraint. The developments in 1967 pointed up the difficulties of obtaining needed fiscal restraint as well as demonstrating the problems of achieving stabilization while attempting to realize an intermediate interest 91 rate objective. The need for coordinated policy was never more apparent. The substantial rate of growth of the money supply in July, August, and part of September of this year now appears to have been excessive, although at the time the monetary growth probably reflected the generally pessimistic forecasts of a marked slowdown in economic activity. The timing and impact of the fiscal restraint package enacted in June now seem to have been miscalculated in view of the continuing vigor of total spending. STATEMENT OF JOSEPH ASCHHEIM, GEORGE WASHINGTON UNIVERSITY I. QUESTIONS ON MONETARY-POLICY GUIDELINES AND OPEN-MARKET OPERATIONS 1. Question No. 1 limits the respondent's choice to that between the two alternatives stated in the question. Yet these two alternatives do not exhaust the full range of possible arrangements for the conduct of fiscal, debt management, and monetary policies. This writer, for one, regards neither of the two alternatives stated in the question as desirable. Consider the first alternative—that is, that a program coordinating fiscal, debt management, and monetary policies should be set forth at the beginning of each year for the purpose of achieving the goals of the Employment Act. In conformity with the constitutional separation of powers in the United States, the monetary authority is a creature of the legislative branch, whereas the fiscal and debt-management authorities are components of the executive branch. The rationale for such a separation of powers is that the money-creating function and the money-spending function should not be vested in the same branch of Government, so as to remove the temptation of the spending branch to inflate. Such temptation is enhanced when both functions are vested in the same branch of Government. Now, to have the President, as provided by H.R. 11, include in his program guidelines concerning monetary policy is to contravene the separation of powers indicated above. Guidelines for the conduct of monetary policy should not be charged to the Chief Executive's responsibility when it is not the executive branch that is charged with the money-creating function. Consider now the second alternative—that is, that we should treat monetary andfiscalpolicies as independent, mutually exclusive, stabilization policies. This alternative is a strawman. Obviously monetary and fiscal policies are not mutually exclusive, but rather complementary, policies. To interpret the notion of the independence of the central bank as implying the exclusiveness of monetary policy is absurd. Monetary policy must at all times be conducted with reference to the fiscal policy extent, or else economic stabilization will be undermined instead of enhanced. Thus, the relevant question is not whether there should or should not be coordination of monetary and fiscal policies. The objective of economic stabilization makes coordination indispensable. Instead, the question is: What kind of coordination should there be ? Should the 92 coordination be that which would be brought about through the executive branch setting forth the guidelines for monetary policy as well as conducting fiscal policy, or should there be another kind of coordination? Having already responded in connection with the first alternative that the constitutional separation of powers in the United States calls for another kind of coordination, we now turn to the suggestion of another kind. A third alternative, one that overcomes the drawbacks of each of the first alternatives, is the following. In conformity with the constitutional separation of powers, guidelines for the conduct of monetary policy should be laid down not annually by the Chief Executive, but more broadly by the legislative branch. In turn, the central bank, in pursuit of the congressionally given guidelines, wTould informally but constantly be expected to coordinate its monetary policy with the fiscal policy conducted by the executive branch. The monetary authority being the creature of the Congress it is the responsibility of the Congress to lay down guidelines that will direct the conduct of monetary policy toward economic stabilization. That responsibility has thus far not been fully discharged by the Congress. How the Congress can fulfill this responsibility will be suggested in answer to question 3 below. In line with my answer to question 1 above, I believe that the Employment Act of 1946 should remain intact in its provision for the President's economic program. 3.A. The money supply, the level of interest rates, and the term structure of interest rates should be stated in H.R. 11 as the target variables of monetary policy. Specifically, the Federal Reserve System should be directed to vary the money supply and to influence the level and term structure of interest rates so as to promote the attainment of the goals of the Employment Act. The money supply, defined as currency plus demand deposits, constitutes the stock of generalized purchasing power in the economy. The size of this stock is amenable to central bank control with a high degree of precision. Variations in this stock are a strategic factor in economic fluctuations. In contrast, the level and term structure of interest rates are not amenable to central bank control with a high degree of precision. They are, however, also important in determining the volume of economic activity. Yet there does not exist a unique or stable relationship between the size (or rate of change) of the money supply and the level (or rate of change) of interest rates or term structure of interest rates. Nevertheless, the level and term structure of interest rates are subject to considerable central bank influence via the weapon of openmarket operations amid a large and widely distributed Government debt. Such influence can be exerted in order to contribute to economic stabilization. It involves using open-market operations in two dimensions: (1) net absorption or release of the cash reserve base, thereby varying the money supply; and (2) swapping operations that can leave the money supply unchanged but alter the term structure of Government debt. 3.B. The guidelines should not be specified either in terms of some particular index or in terms of the target variable's value or growth. Instead, the Employment Act of 1946, applicable to the entire U.S. Government, should be amended in its goals to read, "maximum em- 93 ployment, production, and purchasing power consistent with reasonable price-level stability." In turn, the Federal Reserve Act should be amended to provide congressional guidelines to the Federal Reserve System. To specify these in terms of some index of economic activity or the target variable's value or growth would be to curtail unduly the monetary authority's range of discretion that is necessary over time in pursuit of the goals of the Employment Act, as amended. There does not exist a unique relationship between "maximum employment, production, and purchasing power consistent with reasonable price-level stability" on the one hand, and any one index of economic activity or any one value or growth rate of the target variable on the other hand. Consequently, the Federal Reserve System should be afforded the discretion to vary the money supply and to influence the level and term structure of interest rates as it deems necessary for economic stabilization. In practice, the FRS has not only sought to contribute to economic stabilization in the sense of enhancing reasonable price-level stability while attempting to counteract cyclical economic fluctuations. The FRS has at the same time been engaged in (a) counterseasonal offsetting operations, and (b) lending to member banks at a rediscount rate that is intermittently a subsidy rate. Neither of these two additional activities of the FRS is necessary for the economic-stabilization role; indeed, both distract the FRS from focusing on the sufficiently complex task of harmonizing its own economic-stabilization effort with that of the fiscal authority. The private financial sector can be expected to look after its own seasonality and member banks can be expected to rely on the rest of the private economy for obtaining loanable funds without subsidy from the FRS. Accordingly, the congressional guidelines for the FRS to be written into the Federal Reserve Act should specify that the FRS is to conduct monetary policy aiming at economic stabilization without subsidizing commercial banks and without engaging in defensive, that is, counterseasonal, operations. Within those constraints, the FRS would be free to exercise its discretion in varying the money supply and influencing the level and term structure of interest rates consistent with the goals of the Employment Act, as amended. To help the implementation of the goals of the Employment Act, debt management can be conducted in such a wray as to avoid interference with the conduct of monetary policy. This noninterference approach to debt management vis-a-vis monetary policy implies that debt management will be geared to the aim of minimizing the interest burden of Government debt, given the conduct of monetary policy by the FRS. 5.A., 5.B., and 5.0. See the answer to 3.B above. 5.D. Detailed quarterly reports to the Congress are too frequent to be consistent with the exercise of discretion in the conduct of monetary policy. On the other hand, annual reports seem to be too infrequent to be timely. Semiannual reports would, therefore, be most appropriate* 5.E. The costs of having observers at Open Market Committee meeting are at least two. Firstly, such an arrangement detracts from the free and full discussion that Federal Reserve officials might otherwise engage in, but would avoid whenever they individually or collectively,, w7ould be apt to lose face by admission or mistakes made. Secondly, it 94 would be difficult to prevent leakage of Federal Reserve decisions to unauthorized individuals or even the public with various observers present at meetings charged with important profit-and-loss implications as Open Market Committee meetings are. The benefits of such an arrangement would be to increase advance information about Federal Reserve decisions on the part of officials who are observers. I consider the costs as substantially outweighing the benefits. I I . APPRAISAL OF T H E STRUCTURE OF T H E FEDERAL RESERVE I favor structural changes (1) through (4) for reasons that led to their suggestion as implemented in H.R. 11. In contrast, once structural change (4) has been enacted, structural change (5) seems to me to be a redundant complication of the task of monetary policy. Once it is provided that the FRS is audited by the Comptroller General each fiscal year, subjecting the System to the congressional appropriation process, only encumbers the conduct of the System's work without enhancing its honesty or trustworthiness. I I I . COMMENTS ON REGENT MONETARY POLICY The subject of this section, monetary developments since 1964, is too broad and far ranging to be dealt with in the context of the above comments on H.R. 11. STATEMENT OP GEORGE I . BACH, STANFORD UNIVERSITY This is in response to your letter of July 9, requesting my comments on numerous aspects of H.R. 11 on which hearings will be held this autumn. I have organized my answers to correspond to the questions sent with your letter. 1-1 ana 2. The Government should be concerned continuously with the coordination of fiscal, debt, and monetary policies, looking toward the achievement of the goals of the Employment Act of 1946. The effects of monetary and of fiscal policies cannot, realistically, be considered in isolation. Since in fact monetary and fiscal policies both affect the level of income, employment, and prices, it is important that they be made with full recognition of these joint effects. 1-2. It seems to me appropriate that the President should at the beginning of each year state in his Economic Report broad plans for the achievement of the goals of the Employment Act. In substance, he now does so. It would oe appropriate for him to be somewhat more specific about the implications for monetary policy of the major economic proposals he makes at the beginning of each year if he wishes to do so. IntlLat event, as I presume is the practice now, he would presumably want to confer with the officials of the Federal Reserve System, or ask his Council of Economic Advisers to do so, before deciding on his proposals. I see no advantage in trying to assign to the President a sharper responsibility than this. This is true because neither the President nor any other economic analyst can hope to spell out in detail a year in advance what would be the most desirable monetary policy actions—unless one were to substitute a specific legislative "rule" for monetary policy, in which case suggestions from the President might be superfluous. 95 I-3-A. Monetary policy should be used to help achieve the goals of the Employment Act via control of the money supply and through other channels. The basic objectives of monetary policy are presumably stable economic growth and high-level employment of men and machines without substantial price inflation. Extensive research over the past decade suggests that, if one were required to choose one intermediate variable on which the Federal Reserve should concentrate, the money supply (defined as currency and demand deposits) would be a reasonable selection. On the other hand, the evidence is not clear that this should be made the sole immediate target of monetary policy. A roughly stable growth rate in the money stock seems to be highly correlated with a roughly stable growth rate of the real economy, but there are numerous exceptions. Unless we can be sure that by stabilizing the rate of growth of the money stock we would also be making the maximum contribution to stabilizing the growth rate of the economy, it would be unwise to prescribe such a "guideline" or "rule" as the exclusive target of monetary policy. We cannot be sure of this casual relationship on the basis of existing research findings. Research results do seem convincing that the Federal Reserve should pay substantial attention to the growth rate of the money stock, and that there is a general presumption in favor of a relatively stable growth in the money stock (of perhaps 2 to 6 percent per annum). But more evidence is needed to justify placing exclusive reliance on this guide to action. First, we aren't clear as to whether this narrow definition of money is superior to a broader definition, that includes time deposits at commercial banks and possibly at savings and loan institutions. Second, use of the money stock as a sole target suffers from the weakness that this target is not exclusively under the control of the Federal Reserve, though the Fed can exercise rough control over the money stock if it is willing to let interest rates fluctuate widely. As long as any target (such as interest rates or the money stock) is partly under the control of market forces, it is a dangerous and imperfect guide to Federal Reserve policy and to the evaluation of that policy, since we are never sure whether target changes are the result of Federal Reserve action or market forces. To meet this criterion, the best intermediate target would be the "reserve base" (unborrowed reserves plus currency in the hands of the public). This target is fully under the control of the Federal Reserve, and on that score it would be a preferable target to the money stock. Broadly, it would provide the same results, and I believe that the Federal Reserve should pay substantial attention to the growth rate in the reserve base, as to the money stock and to other important variables. For the same reasons, exclusive reliance on interest rates as an immediate target of monetary policy is extremely dangerous, since interest rates are determined only partly by Federal Reserve action, and partly by market forces. I I - l - D . It seems to me appropriate for the Congress to provide more specific directives for the Federal Reserve. Such a directive might specify growth in the money stock and in the reserve base as important indicators in the formations of monetary policy. I do not, however, believe that Congress should specify either of these, or any other intermediate target, as the exclusive guide to monetary policy. There are too many uncertainties about the linkage between monetary actions U6 and the real economy to justify exclusive reliance on one target now. As a practical matter, it is clear that the Federal Reserve does now pay substantial attention to these variables, so it does not seem to me urgent that such a statement be added to the Employment Act of 1946 or to special legislation governing the Fed. This is in spite of the fact that I do support a strong presumption that 2-to-6-percent annual growth in the reserve base or the money stock will ordinarily contribute importantly to stable economic growth. The Fed should certainly be free to deviate from such a presumption if special circumstances arise. 1-4. The use of debt management to help implement the Employment Act of 1946 is appropriate, though not a device of very great importance. The evidence to date fails to support the argument that shifts in the composition of the debt arising from conscious government policy play a major role in controlling the economy's growth rate. On the other hand, I see no reason why this policy device should not be used insofar as it can make an effective contribution. I-5-A. As indicated above, I do not believe that H.R. 11 should be adopted insofar as it directs the FOMC to conduct open operations in accordance with "the programs and policies of the President." It would be difficult, if not impossible, for the President to specify in advance for a wThole year what the FOMC should do through open market operations ; for him to try to do so would serve no good purpose. In giving any directive to the Fed, Congress or the President should recognize the importance of short-run money market conditions as one consideration in the implementation of monetary policy. In my judgment, the Federal Reserve has been unduly concerned many times in the past with short-term money market conditions. Care should be taken that such considerations not be allowed to dominate long-run monetary policy. The recent announcement of changes in the discount procedure marks an important step toward placing more reliance on the market to make its own short-run adjustments. However, seasonal variations, variations in float, short-term government financing requirements, and the like, are important enough to justify careful attention to them on a day-to-day basis. Pending a more complete understanding of how the markets now act and would act under less Federal Reserve intervention, it would be dangerous to remove completely such market conditions from considerations in making monetary policy. I-5-B and C. Open market operations seem to me the most important channel for the Federal Reserve to influence money markets and the growth of the real economy. However, I see nothing to be gained through removing the Fed's power to change reserve requirements, and I favor more extensive use of rediscounting as a device to permit individual banks to adjust their reserve positions. I think that, as indicated above, the recent Federal Reserve discount proposals are a step in the right direction; I would move even further toward reliance on individual bank discounting. I do not believe that regulation Q, or comparable ceiling individual rates, are desirable policy. The Fed should rely more heavily on quantitative measures, mainly open market operations. However, the abrupt elimination of direct controls and rate ceilings might be disruptive. Thus, the Federal Reserve and other supervisory agencies should move as rapidly as is feasible to raise such rate ceilings to the point where 97 they have little impact, thus gradually removing them from active use except under exceptional circumstances. I-5-D. I see no reason why the Federal Reserve Board should not be asked by Congress to submit regular quarterly reports on the actions it has taken, while recognizing that such reports should not be expected for a matter of some weeks or possibly a couple of months after the end of the quarter. I oppose any requirement that would make the Federal Reserve report its prospective actions to the Congress. Public announcement of such advance plans would make the implementation of stabilizing monetary policy extremely difficult. For the Federal Reserve to tie its hands in this way in advance of unforeseen developments would seem foolhardy under present circumstances. The present reports of the FOMC, received some 3 months after the action is taken, seem generally appropriate to me. They might be more detailed and more clearly relate the actions taken to policy goals, but to try to enforce more detailed quantitative as well as qualitative reporting would be of dubious value, pending the results of further research on the entire subject covered by H.R. 11. I-5-E. I see no important advantages to be gained from having representatives of the Congress, Treasury, or CEA as observers at open market committee meetings. This reflects my judgments that there are substantial advantages to be had from a Federal Reserve which has substantial "independence" within the Government. To make the Federal Reserve completely independent of the executive branch and Congress would be foolish and pointless in a democratic government like ours. But to make the Federal Reserve completely subservient to the President would lose some real advantages that the Nation now gains from having the Federal Reserve as a buffer between the day-to-day swings of public and political processes and the longer range goals of monetary policy. I have presented my views on this matter, including a detailed analysis of the problem, in testimony before this committee ("The Federal Reserve System After Fifty Years," vol. 2, 1964, pp. 1387-1398). II-1-5. I have presented my views at length on these matters before this committee in the 1964 hearings, "The Federal Reserve System After Fifty Years" indicated above. Briefly: 1. I see little to be gained from retiring Federal Reserve bank stock at this time. If the Federal Reserve were being established now, clearly there should not be such stock owned by the commercial banks. On the other hand, it has now become an accepted part of the system and does no apparent harm. This is not an issue that would justify stirring up a major controversy now. 2. If the structure of the Federal Reserve were to be re-formed, I wTould favor a reduction in the number of Board members to five, and shortening of the stated term of office. A five-man board with a 10-year term of office would be an appropriate compromise between the desire to keep the Board insulated from short-term political pressures and also sensitive to changing public views reflected by both Congress and the administration. 3. I strongly favor making the term of the chairman of the Federal Reserve Board coterminus with that of the President of the United States. To saddle a President with a Reserve Chairman in whom he does not have confidence is likely to lessen the influence of the Federal 98 Reserve rather than to increase it. As a practical matter, the benefits from a semi-independent Federal Reserve like ours come mainly in assuring that the points of view of the monetary authorities is strongly stated and duly considered in governmental policy formation and execution. Thus, it is highly important that both the President and the Congress respect and feel comfortable with the Chairman of the Federal Reserve Board, if he is to participate effectively in influencing governmental macroeconomic policy as well as overseeing narrower money market actions of the Federal Reserve itself. 4 and 5. Since I believe that there are substantial benefits from maintaining a semi-independent Federal Reserve along the general lines we now have, I oppose placing the Federal Reserve under annual congressional appropriations or providing for an audit of the Federal Reserve by the Comptroller General of the United States. As a practical matter, to place the Federal Reserve under these two rules would be to put it closely under the control of Congress and to subject it to short-run, almost day-to-day, intervention and control by the Congress. The evidence indicates that the Federal Reserve currently is effectively audited by an outside auditor and that it exercises commendable care in the expenditure of funds. The likely savings to the public from these two steps would be minute; the likely cost would be great through eliminating the degree of independence which the Federal Reserve now has from short-run political pressures. The Congress is free at any time, under the present arrangement, to intervene in Federal Reserve operations and to call the Federal Reserve to account. No more direct control seems to me needed or appropriate. STATEMENT OP MARTIN BRONPENBRENNER, CARNEGIE-MELLON UNIVERSITY (PRO-TEMPORE) VANDERBILT UNIVERSITY 1. I find it more than usually difficult to reply to your most recent questionnaire to economists, dated July 9 of this year. This difficulty is not only due to the searching character of your questions, but involves my incomplete sympathy with the Employment Act of 1946, which you appear to take as given and propose to strengthen from the monetary side. In my view, this laudably intentioned statute, taken seriously and literally, opens the door to unlimited cost-push inflation by collusive bargaining between business and labor, with price and wage increases chasing each others' tails in spiral fashion. This is because monetary and fiscal agencies would be obligated to "validate" by expansive policies each sucecssive round of wage and price increases, all in the name of maintaining full employment and output, and maximizing the economic growth rate. Whatever the deficiencies of Federal Reserve monetary management in the years since 1946, it has deserved primary credit for preventing any such "Latin America, here we come" type of runaway inflation. 2. My personnel monetary-policy view, spelled out most fully in the Journal of Law and Economics (1965) is that the monetary authority should so regulate the money supply that in each period (month or quarter) it grows at a rate equal to: The estimated growth rate of the full-employment labor force in that period, plus The estimated growth rate of man-hour productivity in that period, minus The estimated growth rate of monetary velocity in that period. 99 In symbols: (dM/M) = (dL/L) + (d^A) - ( d Y / Y ) . This formula can be derived from the equation of exchange (MV = P Y ) by relating national income ( Y ) to the labor force (L) and labor productivity (7R), (Y = LTT), and by holding the price level (P) constant (dP = O). It makes no difference which detailed definition of the money stock (M) one uses, provided only that the definition of the income velocity of circulation (V) is consistent with our definition of M. 3. This rule should be followed as closely as may be by the monetary authorities, with unavoidable errors in one period compensated by adjustments in the subsequent one rather than being permitted to cumulate. The mechanism of following this rule should be primarily open market operations, and secondarily variations of commercialbank reserve requirements. (In my opinion, the present upper limit on the commercial-bank reserve ratio is too low, and should be either raised substantially or replaced by a limitation on the permitted rate of increase per year.) We should also reconsider imposition on commercial banks of variable "secondary reserves" of Federal debt securities, as has been advocated many times. 4. The monetary rule, and its anticipated effects, should serve as guides for policy recommendations by other public agencies, including both the Congress and the members of the Washington administrative "Quadriad" (Treasury Department, Council of Economic Advisers, Bureau of the Budget) more directly concerned with taxation, public expenditure, debt-management, and employment problems. You will notice that the rule says nothing of the foreign exchanges. My belief is that, like commodity markets, they should be left free, subject only to limitations on daily (or possibly also longer-period) rates of change in either direction. I should also propose removal of the existing prohibition of interest payments on bank deposits, or of the existing legal maxima on rates paid on time deposits, certificates of deposit, savings and loan shares, and similar credit instruments. 5. The details of Federal Reserve System structure embodied in H.R. 11 seem, if you will pardon my saying so, matters of subsidiary importance. I should, instead, be interested in procedures for identifying and disciplining members of the Board of Governors, or subsidiary staff members, responsible for egregious and continued breaches of the proposed monetary rule. 6. Should experience indicate that this rule poses insurmountable estimation problems or disorderly interest-rate gyrations, or should collusive-bargainers be able to "strike" against it effectively over long periods, we may consider suspensions, modifications, or return to "discretion," but we should not assume the worst in advance. Furthermore, we should realize both the necessity of threatened unemployment and excess-capacity to keep cost pushers under control and the initial implausibility of such threats unless actual unemployment and excesscapacity are permitted after bargained wages and administered prices rise. 7. My criticism of post-1964 monetary policy is twofold. Most importantly : The Federal Reserve System has permitted the monetary growth rate (dM/M) to fluctuate between wide limits, first letting inflation proceed almost unchecked and then causing near-panic con- 100 ditions by sudden applications of monetary brakes. (The resulting rises in interest rates and declines of credit availability, called crunches could have been avoided at least cost by slower monetary expansion in the first place.) My less important criticism, at least for the short run: The long-period or average growth rate of the money supply has been somewhat too high, and interest rates somewhat too low for pricelevel stability. 8. I am aware of wide divergencies between the positions outlined above and the current "conventional wisdom" within my profession. Perhaps two closing statements are in order: ( 1 ) 1 should not be read to imply that "only money matters," and (2) I see no dichotomy between monetary and fiscal policy. We need not choose between them, and both can work in harmony. Among those economic authorities with whom I find myself most nearly (although not completely) in agreement are Karl B runner, of Ohio State and my colleague, Allan Meltzer, of Carnegie-Mellon (both of whom have worked with you and your committee), Milton Friedman, of Chicago, and E. S. Shaw, of Stanford. Submitted with respect transcending any disagreement. STATEMENT OF KARL BRUNNER, OHIO STATE UNIVERSITY R E P L Y TO Q U E S T I O N N A I R E ON H . R . 11 I. 1 AND 2 Two conditions are both necessary and sufficient for a meaningful "coordination" of monetary policy, fiscal policy, and debt management policy. The first condition involves an adequate choice of objectives and a sufficiently clear and stable concensus concerning the relative weight assigned to the objectives selected. The second condition pertains to an adequate knowledge of the economic process linking policy actions and objectives. Neither condition has been satisfied by our policymaking institutions. The authorities neglected to acquire an adequate knowledge of monetary processes linking policy and the behavior of bank credit, interest rates, and money supply. This neglect obstructs rational monetary policies and causes serious misinterpretation by the authorities of their own policy. In the absence of any reliable knowledge about the broad properties of monetary processes any requirement to coordinate policies remains quite useless. There is little advantage in coordination executed in the context of serious misinterpretations concerning the structure of monetary processes. The recent trend in policymakers' choice of objectives poses another problem for meaningful coordination. Policymakers appear inclined to extend and complicate the range of objectives. Moreover, they also appear inclined to modify quite rapidly the relative weight of various objectives or constraints. In such contexts every policy mix actually pursued can be easily justified to be optimally designed and properly coordinated. For every policy mix there exists a set of objectives and a conception of monetary processes which permits a policymaker to claim optimality of existing policies relative to such conceptions and selected objectives and in the absence of a comparatively stable consensus concerning objectives and in the absence of validated conceptions about monetary processes the requirement of coordination is premature and useless. 101 I. 3.A If the authorities possessed perfect knowledge about the structure of monetary processes their policies could be directly adjusted in response to the desired state of ultimate goals summarized by the Employment Act. Our imperfect knowledge and the lag in the accrual of information concerning the state of the economy make it advisable to guide monetary policies in response to an intermediate target intercalated between the instruments used for policy actions and the ultimate goals. The money supply (inclusively or exclusively) still appears at the present stage to provide the most reliable shortrun target of monetary policy. The transmission of monetary impulses to the pace of economic activity is mediated by a relative price process affecting the whole range of assests and liabilities. The impulse reaches the demand for current output via the substitution relations existing between the holding of assets and the purchase of their services, or the substitutions between existing and newly produced assets. In the context of this transmission process monetary policy is not restricted to channels operating essentially via investment expenditures, neither is it dependent on the relative importance of borrowing costs or the sizable occurrence of credit transactions. The Federal Reserve authorities could execute the target policy in the following manner. (a) First, the authorities determine an acceptable range for the growth rate of the money stock over the next 6 months (see I. 3.B for further remarks on this point). ( i ) The authorities assess the expected movement of proximate determinants of the money stock (i.e. of currency ratio, time deposit ratio, adjusted reserve ratio, and Treasury deposit ratio) for the next 2 or 3 months. (e) With the assessment of the proximate determinants available, the authorities determine the growth rate of the base for the next 2 or 3 months required for the average growth in the money stock determined in thefirststep. (d) The assessment of proximate determinants should be reconsidered every month and consequently also the required growth rate of the base. One last aspect needs emphasis at this stage. The dispute concerning the optimal choice of intermediate targets remains quite unsettled. A good part of the discussion bearing on these and related issues was unfortunately not designed to settle the pending issues. Such discussions could become substantially more constructive if every participant would specify the analytical and evidential results which will dispose him to accept or reject any specific traget proposals, including his own. These conditions wrould reveal more sharply the existence or1 absence of an adequate analysis in support of a particular proposal. I . 3. B AND D Under the present circumstances broad indicators of economic activity closely associated with our ultimate goals offer poor guidance for the continuous adjustment of policy. It was stated above that the growth rate of the money stock is the most useful tjarget at present. It would be inappropriate however, in the context of fixed exchange 102 rates, to impose a rigid constraint on the required growth rate of the money stock. Monetary growth should be maintained within a band (say 2-6 percent p.a. for the exclusive money stock) without any sharp reversals and counterreversals following in close succession as in the recent past. Adjustment of monetary policy to the conditions of the balance of payments does not require the pronounced short run instability exhibited by past policies. A 6-month target between 2 percent and 6 percent without radical changes between successive 6-month periods should be sufficient to cope with balance-of-payments problems. Frequent or decisive changes in the target should be justified by the Federal Reserve authorities by a detailed analysis submitted to the appropriate congressional committees. The procedure described is loose and flexible enough to permit operation over several years. A gradual adjustment with growing experience will unavoidably occur. I. 3F The Federal Reserve authorities should have the power to change the growth rate within the band described above. Similarly, they should be given the right to move on exceptional occasions outside the band. In the latter case and in case the authorities change the target by more than 1 percent between any two adjacent 12-month periods a detailed report and analysis justifying the decision would be required. The reports submitted should be subjected to hearings by appropriate congressional committees. This procedure imposes some restrictions on Federal Reserve policy and also generates pressures to acquire better validated conceptions which are exposed to critical examination. I. 5.A The criteria guiding the Federal Reserve's defensive operations dominated on many occasions its policy conception and reinforced the misinterpretations of policy. The constraints on the required growth of the base sketched above would sufficiently attenuate at the moment the concern for defensive operations. Further restrictions appear unnecessary at the moment. I . 5. B A N D O Neither academic literature nor Government documents have ever established a case for the existence of reserve requirements or the Federal Reserve's power to change the requirement ratios. Similarly, no relevant analysis or evidence has ever been presented on behalf of regulation Q. And the case for discount policy rests essentially on strictly political considerations. From the point of view of a rational monetary policy designed to shape a stable movement in economic activity, open market policy is the only instrument required for the authorities. All the other instruments were dominantly used for political purposes, or purposes of income distribution or allocative purposes. The use of monetary policy instruments for purposes other than monetary stabilization only aggravates the problems of confronting the Federal Reserve authorities. 103 I . 5. D AND E The requirement of quarterly reports would contribute to the development of a greater sense of intellectual responsibility on the part of the Federal Reserve authorities. The reports could be an excellent device compelling the authorities to acknowledge their responsibility to execute policy on the basis of whatever systematic knowledge is at their disposal. The reports should require a description of their recent policy including a detailed and specific justification for the interpretations advanced. The reports should also explain the recent movements in the money stock and the role of policy in the observed behavior. Moreover, the reports might usefully include projections of the money stock and bank credit and describe the policy required to realize such projections. I I . 1 TO 5 The first proposal has little bearing on the conditions for a rational monetary policy and the last proposal does not promise suitable pressures for a rational longrun policy. Proposals 2 and 3 simplify the Federal Reserve's organizational structure and should be welcomed. Proposal 4 might obstruct an allocation of resources by the Federal Reserve System which cannot be justified in terms of monetary policy considerations and the conditions required to prepare and execute a rational policy. n i Four aspects of oue monetary policy pursued since 1964 should be recognized. (a) The misinterpretation of policy conveyed to the public and the press in the late fall of 1965. The increase in the discount rate was generally interpreted as a move toward a more restrictive policy. Policy became actually more expansionary until May/June 1966. (5) A sharp reversal occurred around May/June 1966. The break in policy was sudden and substantial. This reversal in policy was the single most important factor contributing to the minirecession of 1967. (o) A counterreversal occurred in November/December 1966. This counterreversial was at least as sudden and pronounced as the previous reversal and prevented the retardation beyond the scope of a minirecession. Policy during the year 1967 followed one of the most expansionary courses on record, and contributed to the substantial accelerations in economic activity and the price movements. (d) Monetary policy continued in 1968 to apply substantial thrust to the economy. Until August 1968, monetary policy has not contributed to any significant retardation. But the aibsence of any further accelerations in current monetary policy generates a state where the consequences of last year's accelerations exert a slightly retarding effect. Even without a sharp deflationary break in monetary policy we should expect a moderate retardation in our economy this winter. 104 STATEMENT) OF MEYER L. BURSTEIN, WARWICK UNIVERSITY AND ASPEN, COLO. ANSWERS TO "QUESTIONS ON MONETARY POLICY GUIDELINES A N D OPEN MARKET OPERATIONS" 1.1. Surely monetary and fiscal policies should not be treated as independent. 2. H.R. 11 concerns the President's recommendations. Obviously the President should alone be responsible for his own recommendations. The question verges on larger questions of distribution of monetary powers between the Executive and the Federal Reserve. Ideally such powers should, I think, be concentrated in the former. Political realities appear to favor the present arrangements. 3.a. My views on the theoretical aspects of this problem are fully expressed in two books. M. L. Burstem, Money (Cambridge, Mass.: Schenkman Publishing Co., Inc.; 1963) and M.L. Burstein, Economic Theory: Equilibrium <& Change (London: John Wiley & Sons, Ltd.; 1968), esp. ch. 13. I would argue for a minimum of specificity in H.R. 11 and perhaps to that extent am unsympathetic with H.R. 11 itself. Thus the quoted language would be improved, I think, by eliminating the words, "including the money supply as defined by him." Turning to the question itself, I can think of no sensible reason for concern with money supply for its own sake. So naturally I would be more interested in policies focusing on such variables as interest rates and credit availability, affected by monetary policies as they are, than in policies focused on "M" purely and simply. The complexity of the underlying analytical and practical problems is such that no specific language should appear in the bill: it is enough to state that the President should give views on monetary policy. Elementary considerations of legal draftsmanship as well as those of economic theory support this conclusion. 3.b. In this context past performance is interesting only to the extent that it permits prediction of future events. And, since econometrics is so crude a technique dealing with so difficult a subject, no specificity should exist on connection with these indexes. 3.c. I think that the question is rather futile. See my answer to 3.b. 3.d. Obviously one must be extremely flexible about this sort of thing. Under no circumstances would wTe wish to give Government functionaries incentives to cook their statistics in order to support one or another rigid theory which they have become identified with. 3.e. No answer. 3.f. I would make no mandatory limitations. I am positively opposed to simplistic "rules" for monetary policies. I surely am opposed to putting authorities into "statistical" straitjackets. My concerns along these lines are heightened by international considerations. The ideology which appears to underlie H.R. 11 included floating exchange rates. But, so long as we do not have floating exchange rates, BOP considerations must loom large in official calculations and will from time to time lead to substantial departures from paths suggested by internal considerations only. 4. Debt management can, I think, play a limited part, a distinctly limited part, in implementing these policy goals. Only massive debt management operations carried on over short intervals could have 105 much impact. These are empirical judgements. There is sound theoretical authority for debt-management operations to have some effects. 5.a. Let me begin by stating my strong opposition to the language stated in 5.1 think it important that the deliberations of the Federal Eeserve Board not be published and am unhappy about the degree of publication which already has taken place. Such publicity is inconsistent with discretionary policies, and discretionary policies are favored by me. (Cf. my answer to 3.f.) Of course, open market operations inevitably will be used for defensive purposes from time to time: the authorities cannot identify the forces against wThich they are operating until rather long after the fact. Nor can I categorically oppose openmarket operations in this connection. Still I favor Federal Reserve discounts and advances as the preponderant defensive device as did the Mitchell committee: open-market operations are a crude procedure to control forces which tend to operate unevenly, both geographically and otherwise; the "size" of the defensive operations usually is open ended while open-market operations are difficult to fine tune to that extent. 5.b. No. My answer to 5.a goes far to support this answer. Obviously the relationship of FR discount rate to market rates is considerably important: large open-market sales would be less effective if discount rate were permissive for example. Of course, we must distinguish between open-market operations designed simply to accomplish a certain change in the monetary base from others in which the operator is instructed to deal freely at specified bill rates: the latter instance has effects not greatly different from policies geared to bank rate in the traditional British fashion for example. 5.c. I have indicated that I regard rediscounting as a legitimate and significant central-banking device. And reserve-requirement changes can, from time to time, be useful, noting that greater selectivity of impact effects is possible through these. I do not esteem regulation Q in this connection. I would not refer to these matters in H.R. 11. 5.d. I make clear in 5.a that I am opposed to requiring the Federal Reserve to make such reports. The upshot would find the Federal Reserve more conscious of the political implications of their actions than now is the case and would tend to polarize attitudes: as in the United Nations, the political consequences of backing down tend to become amplified under the glare of publicity. 5.e. None. This humiliating suggestion would cause the Board simply to meet in each other's homes in secret. II.l. No comment. 2. I favor this. An incompetent member now is permitted too long a tenure if indeed competence ever has been critical for reappointment. 3. No, I am opposed. This would put the Federal Reserve into the heart of politics, leading up to a worse system than at present: there would be no real independence of the Federal Reserve but there would be considerable administrative and other confusion. 4. Why? 5. Machievelli said, "either embrace men or annihilate them." Contumacious and petty measures such as this would not destroy the powers of the Federal Reserve but would poison the atmosphere of monetary policymaking. III. This is too large a matter for treatment in this format. 21-570—68 8 106 STATEMENT OF PHILLIP CAGAN, COLUMBIA UNIVERSITY I . ON MONETARY POLICY GUIDELINES AND O P E N - M A R K E T OPERATIONS 1. Do not use fiscal policy for stabilization. It is too slow and cumbersome. 2. A program a year ahead is too far ahead to plan and is, therefore, impracticable. Policy changes need to be flexible. Only set general policy for the year, provisional guides. 3. Concerning monetary policy guidelines: A. Goal: Currency outside banks, plus demand and time deposits. Target: Monetary base. B. It is best to keep an even rate of growth to avoid disruptions—with only slight variations. C. Nocomment. D. Guidelines should be reviewed periodically—cannot look a year ahead. E. The guidelines should be determined by price trends. F. Flexibility of money supply growth outside the guidelines should be allowed to some extent if recession develops or price trends change. 4. Concerning debt management policy. Avoid rocking the boat. Keep average maturity relatively constant. 5. Concerning open-market operations: A. I agree with ignoring transient influences in conducting open-market operations, but we must give marketing institutions time to adjust to this change of Fed behavior whereby a money supply growth rule is followed. B. Monetary policy can be implemented solely by openmarket operations. C. Abolish rediscounting; do not change requirements; abolish regulation Q. D. There is no need for a report. Avoid wasteful paperwork. If policy were a certain rate of monetary growth, intended rate could be announced. E. I see no merit in having observers at open-market committee meetings. Acrimony and indecision would result. Avoid dispersing decisionmaking power. II. No comment. III. Concerning recent monetary policy, there has been too much fluctuation in the money supply. This has been disruptive to the economy. STATEMENT OP GREGORY C. CHOW, IBM RESEARCH CENTE1 I . ON MONETARY POLICY GUIDELINES A N D OPEN M A R K E T OPERATIONS 1. I believe that a program coordinating fiscal, debt management, and monetary policies should be set forth at the beginning of each year for the purpose of achieving the goals of the Employment Act. 2. I believe that the President should be responsible for drawing up this program. 107 3. Concerning monetary policy guidelines: A. I believe that monetary policy should be used via intervention of money supply (defined as currency plus demand deposits adjusted) alone. Interest rates, like other prices, should not be directly manipulated but left to the determination by market forces. High-power money cannot serve as a target, but only as an instrument, since it affects the level of economic activity only indirectly through its effect, among the effects of other factors, on the quantity of money. The Federal Reserve may be given much discretion in choosing the means of controlling the target variable (the supply of money). It is fully recognized that the level of economic activity is governed by other factors than the supply of money (current, past, and even expected in the future), but controlling this variable can diminish economic instability and promote economic growth. B. The guidelines of monetary policy should be specified in terms of the target variable's value or growth, rather than some index of past, present, or future economy activity, because we do not yet know precisely enough the dynamic relationships between the target variable and future indexes of economic activity, and because too much manipulation of money supply would by itself create uncertainty in the economic world, thus leading to economic instability. C. Not relevant for my position. D. The same guidelines should be used each year into the foreseeable future, again because of the reasons stated in B above. E. I recommend 3.5 to 4.5 percent per year for the range of growth of money supply (currently plus demand deposits adjusted) for the following reasons. Empirical studies of the past seven decades, including my own, have shown that the demand for money in constant dollars is proportional to real GNP, given the rate of interest, and decreases by about 0.75 percent for a 1-percent increase in the rate of interest. These findings are also consistent with postwar experience. From 1950 to 1967, real GNP grew at an average annual rate of 3.7 percent, while the rate of interest increased at an average rate of 4.1 percent, thus accounting for about 3.7-(.75) 4.1 or 0.6 percent increase in the demand for money per year. Since the supply of money increased at an average annual rate of 2.6 percent, the excess of supply over demand, at about 2 percent per year, is sufficient to explain the rise in price (at an average annual rate of 1.9 percent by the Consumer Price Index, or 2.2 percent by th GNP deflator) during the same period. Thus, if the rate of interest were to be kept from rising, the supply of money should be increased at the same rate as real GNP. From the experience of the last 5 years (1962-67), real GNP was capable of growing at an annual rate of 4.7 percent. Therefore, the growth of the stock of money at a rate of 4 percent can be absorbed by the growth of real GNP without causing inflation and rising interest rates—witness the period from 1962 to early 1965, when the stock of money was rising at about 4 percent per year and both the rate of interest and the price level were fairly stable. 108 A question still remains. Although a 4 percent growth in money supply and in real GNP is consistent with stable price and stable interest rate, can one exclude the possibility of rising price, to be compensated for by rising interest rate ? This possibility is unlikely if the rise in the rate of interest in the past has been due to insufficient money supply, or to the expectation effect of inflation resulting from an excess of money supply in certain periods. Both of these causes will be weakened by the introduction of the policy here recommended. Note, however, that insofar as the rate of interest is also affected by other factors, it may experience a rising (or falling) trend independently of the monetary policy here recommended, and should this happen, the demand of money would fall (or rise), thus requiring a smaller (or greater) increase in money supply. F. From the analysis just presented, I recommend that the range of 3.5-4.5 percent for the rate of growth in money supply be adhered to for at least 5 years. After that, the range may be adjusted for another 5 years only if there should be strong evidence for a serious inflationary or deflationary trend observed during the first 5 years. Under no circumstances should monetary authorities be permitted during the year to adjust the target variable outside the range issued at the beginning of the year. 4. Concerning debt management policy: I believe that debt management has little role to play in this matter. Debt management here presumably means managing the composition of Government debt, not its total which is the result of past Government deficits and surpluses. I share the view of the majority of economists that the quantity of a certain form of assets, namely, money, has more influence on economic activities and especially on the price level than does the composition of one type of assets, namely, Government debt. 5. Concerning open market operations: A. I do not recommend using open market operations to counteract seasonal and other transient factors affecting money market and credit conditions. B. I believe that monetary policy can be effectively and efficiently implemented solely by open market operations. C. For the purpose of stabilizing the rate of growth of money stock, I do not see that changes in (a) rediscount rate and (b) reserve requirements can accomplish any more than what open market operations can. I am not in favor of regulation Q, or any Government attempt to control the rate of interest in the market. D. I see no compelling reason for requiring the Federal Eeserve Board to make detailed quarterly reports to the Congress on past and prospective actions and policies. As long as the President shall transmit to the Congress by January of each year a program including the growth of the money supply, it is not essential for the Federal Eeserve Board to make detailed quarterly reports to the Congress. Such a requirement may facilitate control of the Federal Eeserve Board by the Congress, but if the President is to coordinate monetary and fiscal policies under the amended Employment Act of 1946, it is his responsibility to insure proper execution of these policies. The real question is .how the President can fulfill his responsibility if the Federal Eeserve Board reports directly to the Congress. 109 E. It is unnecessary to require representatives of the Congress to serve as observers at Open Market Committee meetings because, once given the responsibility and the rule of conduct, the Open Market Committee should be given a free hand to discharge its responsibility. If the committee should consider it beneficial to have observers from the Treasuiy and the CEA (or from the Congress), it could invite them on its own. II. APPRAISAL OF THE STRUCTURE OF THE FEDERAL RESERVE 1. I have no strong feeling about the retiring of Federal Reserve bank stock; 2. I am in favor of reducing the number of members of the Federal Reserve Board to five and their terms of office to no longer than 5 years; 3. I am in favor of making the term of the Chairman of the Board coterminous with that of the President of the United States; 4. I am in favor of an audit for each fiscal year of the Federal Reserve Board and the Federal Reserve banks and their branches by the Comptroller General of the United States; 5. As in II.l above, I have no strong feeling about the appropriation of funds by Congress to operate the Federal Reserve System. The proposal under H.R. 11 seems superior to the existing arrangement, but the latter is not the main defect of the Federal Reserve System today. N I . COMMENTS ON RECENT MONETARY POLICY I would not wish to attribute changes in economic activity since 1964 to specific monetary policies, because I believe that the assignment of cause and effect cannot properly be made by simply citing the movements up and down of a few economic variables, especially when the history is so short and recent. I would also warn against readily accepting the criticisms of the Federal Reserve which are based on such citing of movements between a few economic variables. For example, whether the slow monetary growth in 1966 was the main cause of the mini-recession early in 1967 can be answered only by a much more elaborate analysis than the presentation of these facts alone. STATEMENT OF CARL F. CHRIST, THE JOHNS HOPKINS UNIVERSITY DEIAR ME. PATMAN : I am honored by your request for my views concerning H.R. 11. The pressure of other commitments has prevented me from writing a detailed reply to your thoughtful questions. However, I believe that my testimony before the Joint Economic Committee on May 9, 1968, will give you a good deal of information about my opinions concerning the proper relationship between the Congress and its creature, the Federal Reserve System. (The testimony follows:) Mr. CHRIST. I am very glad to be here today, Senator Proxmire, to contribute what I can and also to learn from the committee and my fellow witnesses. The central questions before us today are whether the Federal Reserve (a) can and (b) should cause the stock of money to increase 110 fairly steadily at a rate of about 3 to 5 percent a year, and (c) what circumstances, if any, would justify a higher or a lower rate of growth of the stock of money. The main objectives of monetary policy are full employment and a stable price level. At the outset we have to admit that we cannot hold the Federal Reserve responsible for everything that happens in the economy. In the first place, there are other factors on the scene, and the Federal Reserve cannot accurately forecast what they will all do. In the second place, the effects of Federal Reserve policy are not all felt immediately; they are spread out over a period of variable length, but at least several months. These two facts mean that the Federal Reserve often cannot know what is the proper action to take today, in order to offset some disturbance that will happen next week and whose effect will be felt next month or next quarter. But even granted perfect prediction, we could not hold the Federal Reserve responsible for everything, for there are times when a choice must be made between two conflicting aims, and even the Federal Reserve cannot have both. For example, suppose—not unrealistically—that the Treasury, acting under instructions from the Congress, undertakes a large increase in spending, and that the Congress does not increase tax rates—when I wrote this, the Congress didn't look as though it was going to increase tax rates and I am very pleased that it now looks as though this may happen. The obvious result would be a large increase in the budget deficit, if there were an increase in expenditures with no increase in tax rates. The Treasury would have to finance this deficit by offering new U.S. Government securities for sale. What will happen? Consider two possibilities. First, the Federal Reserve could assist in the financing by buying and holding whatever portion of the new securities is not taken up by private investors. In that case, the stock of money would increase, because part of the money that the Treasury spends would be created when the Federal Reserve buys new Treasury securities. Or, take the second possibility, the Federal Reserve could decline to assist in the financing; that is, buy none of the new Treasury securities offered. In that case, the Treasury would have to offer better terms to the private market; that is, higher interest rates, in order to induce the private market to buy all the securities offered. Then the stock of money would not increase, but interest rates would increase. Thus, the Federal Reserve has a choice, when faced with a Treasury deficit; the Federal Reserve can increase the money stock while maintaining interest rates about the same, or hold the money stock fixed while permitting interest rates to go up. Of course, one could imagine a policy somewhere between these two, permitting some increases in both the money stock and in interest rates. But the Federal Reserve cannot stabilize both the money stock and interest rates in this situation when there is a large deficit. Similarly, when faced with a Treasury surplus, the Federal Reserve has a choice between stabilizing the money stock while interest rates fall, or stabilizing interest rates while the money stock falls, but cannot stabilize both. Ill It is pretty clear that the Federal Reserve can control the stock of money within narrow limits. I mean they can make the stock of money, come within plus or minus one-half percent of any desired level, 99 weeks out of 100. By the way, the money stock concept I am using is the Federal Reserve's own: currency and demand deposits. It is certain that a policy of increasing the money stock at 4 percent a year, or between 3 and 5 percent a year, would not be the best possible Federal Reserve policy, if we knew everything about how the economy operates. But we don't know that, and therefore, we don't know what the best possible policy is. I would like to argue first that, given our present knowledge, we will probably have better monetary policy if the Federal Reserve sees to it that, during every calendar quarter, the increase of the money stock is at a seasonally adjusted annual rate of between 2 and 6 percent, better I mean than we would have if the Federal Reserve follows policies like those of the past. I would like to argue second that the Federal Reserve ought not to change this rate of change abruptly, from a 2-percent annual rate in one quarter to a 6-percent annual rate in the next quarter, or vice versa. Third, it is more important to stabilize the rate of growth of the money supply than to stabilize interest rates, whenever the Federal Reserve must make a choice. For the long run, a 4-percent annual growth rate in the stock of money is about right. Real GNP has been growing at 3.9 percent a year since 1948—when one might say the economy had returned to normal after World War II. At roughly constant interest rates, which we have not had within the last 20 years, a roughly constant price level, the demand for money grows roughly in proportion to real GNP. If the money stock grows much faster than 4 percent a year, say 8 percent or more, then aggregate demand is induced to grow much faster than capacity. When demand catches up and overtakes capacity, there is upward pressure on the price level. If the money stock grows much slower than 4 percent a year, say it doesn't grow at all, or even declines, then aggregate demand is induced to fall rapidly behind capacity. When this happens, we have deflation, downward pressure on prices, and unemployment. During 1941-45, the money stock grew at 22 percent a year; everyone agrees that this was far too fast for stability. During the depressions of 1921 and 1929-33, and all the recessions since 1921—they were in 1924, 1927, 1938, 1949, 1954, 1958, 1961—the money stock actually declined in absolute terms, which in my opinion should not be permitted. I think that is a very important criticism of Federal Reserve policy in the past, that they have permitted the stock of money to decline during depressions. The evidence so far is not persuasive in favor of the claim that small variations in the rate of growth of the money supply cause business cycles. But it is clear that an actual decline in the money stock, or a prolonged period of little or no growth, aggravates any recession that is in progress or that might develop. Similarly, a prolonged period of rapid growth in the money stock aggravates any overheating that is in progress or that might develop. 112 Furthermore, rapid changes in the rate of growth of money stock are themselves a disturbing factor. That is why I would like to see the Federal Eeserve keep the rate of growth of the money stock fairly steady, between 2 and 6 percent a year, and to vary this rate of growth only gradually. It should be pointed out that if the Congress were to require the Federal Eeserve to follow any such rule, the Congress would thereby restrict its own freedom of choice in some situations. Consider again the case in which the Congress provides for a large increase in expenditure with no increase in tax rates, so that a large deficit develops. If the Federal Eeserve is prohibited from increasing the money stock at a rate greater than 6 percent a year, say via a congressional rule, then a large share of the deficit would have to be financed by the sale of Treasury securities to the private sector, thus driving interest rates very high, and not completely preventing inflation either—an undesirable situation. Notice that, if the Federal Eeserve is required to keep the money stock from growing faster than 6 percent a year, and if the Congress increases expenditures greatly, then the Congress has only the following choices open: to endure high interest rates and some inflation, or to increase tax rates, or some combination of these two. The basic alternatives among which the Nation must choose may be seen more clearly if looked at from another angle. There are three important ways in which the Treasury's expenditures may be financed: (1) by taxation, (2) by increasing the stock of money, and (3) by increasing the amount of Government debt in private hands (that is, by borrowing from the private sector). By choosing the level of Government expenditure and the level of taxes, the Congress determines the amount of the Government budget deficit, or surplus. Let's suppose there is a deficit. Then, it must be financed by some combination of increasing the stock of money, and increasing the amount of Government debt in private hands. The most important function of the Federal Eeserve is to control how this deficit financing is to be divided between increasing the stock of money and increasing the amount of privately held Government debt. This the Federal Eeserve does chiefly by deciding what amount of Treasury securities to buy and hold (thus increasing the money stock), and what amount—that is offered by the Treasury—not to buy, thus requiring private holdings of the Government debt to increase. I have been speaking of a deficit, but if there is a budget surplus the opposite choice is open to the Federal Eeserve, decrease either the money stock or the private holdings of Government debt. Just as the Congress has the authority to fix Government expenditures and taxes, and thus to fix the budget deficit, so the Congress has the authority to decide how much of the deficit should be financed by increasing the money stock, and how much of it should be financed by borrowing from the private sector. I have suggested that the Federal Eeserve ought to make the stock of money grow at a rate between 2 and 6 percent a year. But the foregoing discussion makes it clear that such a policy will not work well unless the Congress keeps the budget deficit or surplus within suitably narrow limits, so that the amounts of Government securities dumped on the private market by a budget deficit are not too large, and 113 conversely so that the amounts of Government securities taken out of private hands by a budget surplus are not too large. When I say the budget deficit or surplus should be kept within suitable limits, I mean a range something like a deficit of from $15 to $17 billion on the one hand to a surplus of $10 or $12 billion on the other hand. In this sense, fiscal policy, which determines the size of the budget deficit, and monetary policy, which determines the stock of money, ought to be in harmony. The congress is the only authority that can make them so. Treasury and Federal Reserve actions can be substitutes for each other with respect to aggregate demand. For example, the Treasury alone can stimulate aggregate demand by selling new securities to the private sector and using the proceeds to buy goods and services for Government programs. Or the Federal Reserve alone can stimulate aggregate demand by buying securities for the private sector in the open market, thus increasing the stock of money. But the effects of the two methods upon interest rates are different. When the Treasury buys goods financed by borrowing from the private sector, interest rates are bid up; when the Federal Reserve buys securities in the open market, securities prices are bid up and interest rates are pushed down. The Federal Reserve can counteract the aggregate-demand effect of this Treasury action, or in the interest-rate effect, but not both. Treasury and Federal Reserve action can be substitutes for each other when a certain effect on aggregate demand is desired, or when a certain effect on the general level of interest rates is desired. But when there is a desired level of aggregate demand, and a desired level of interest rates, then cooperation between the Treasury and the Federal Reserve is required. It is extremely important to realize that the policies required of the Treasury and the Federal Reserve to achieve the domestic objectives of full employment and stable prices will sometimes conflict with the achievement of balance-of-payments equilibrium at a given exchange rate. This conflict has persisted in the United States for several years, programs 3 or 4 years. It may still be with us even if the present buoyant business temper moderates. In the face of such a conflict, we have several choices. Since we have gold and foreign exchange reserves, we can continue in deficit on our balance of payments, but only until the reserves are gone. Our other choices, among which wre may choose now, but among which wTe must choose when our reserves are gone, are these: reduce Government spending and lending abroad; impose restrictions on private foreign trade and capital movements; impose a recession on the domestic economy to dampen private import demand and possibly increase exports; or seek a new exchange-rate level where equilibrium is possible. The last of these alternatives, in my view, is the best. It is encouraging to see the development of econometric models of the U.S. economy, in greater sophistication and detail. I believe that they hold promise of teaching us ever more about our economy and how it operates and responds to public policy. In spite of substantial improvements in the past generation, I am sorry to say that I know of no model that I would now trust with the task of formulating stabilization policy for the United States. 114 In summary, my answers to the questions before us are these: First, the Federal Reserve can control the stock of money very closely. Second, I believe it would be an improvement if the Federal Reserve would increase the money stock each calendar quarter at a seasonally adjusted annual rate of between 2 and 6 percent. Third, the Federal Reserve should adjust the rate of growth of the money stock within these limits, making only gradual changes in the rate of growth, and raising or lowering that rate of growth in accordance with its best judgment as to whether economic conditions are—or soon will be—too bouyant or to slack. Fourth,, this policy will work best if the Congress will keep the budget deficit or surplus from being very large, and from changing very rapidly. There is the end of my opening statement, Senator Proxmire. I have an appendix of tables at the end of the prepared statement that might be useful Chairman P R O X M I R E . Without objection it will be printed in the record in full. Mr. C H R I S T . Thank you very much. (The appendix tables follow:) APPENDIX TABLES TABLE 1—DECLINES IN THE U.S. MONEY STOCK (DEMAND DEPOSITS AND CURRENCY, SEASONALLY ADJUSTED) DURING DEPRESSIONS AND RECESSIONS SINCE 1921 Percentage decline on the money stock during recession Month during which the money stock reached its peak March 1920 December 1922 September 1925 October 1929 March 1937 January 1948 July 1953 July 1957 July 1959 Number of months before the money stock regained its previous peak level 15.0 2.0 3.0 33.0 6.0 2.0 .2 1.0 3.0 Source: M. Friedman and A. Schwartz, " A Monetary History of the United States." pp. 709-715, and Bulletin, June 1964, pp. 682-690. 53 10 26 79 20 27 9 9 27 Federal Reserve TABLE 2.—RATE OF CHANGE OF T H E U.S. MONEY STOCK ( D E M A N D DEPOSITS A N D CURRENCY, SEASONALLY ADJUSTED) ANNUAL PERCENTAGE GROWTH RATES FOR CALENDAR YEARS AND QUARTERS, 1956-68 Rate for rolonHar Odiciiual year Year 1953 1954 1955 1956 1957 1958 1959 I960.... 1961 1962 1963 1964 1965 1966 1967 1968 - il.l 2.7 2.2 1 1.3 1 —. 7 3.8 1.6 i —. 6 3.0 1.4 3.8 4.1 4.7 2.2 16.3 Rate for calendar quarter 1 2 3 U.9 11.2 4.0 11.5 i.O U.8 4.0 1-2.8 2.6 U.7 3.8 2.9 2.5 5.8 16.3 4.2 U.6 2.2 2.4 1.9 i.O 5.6 2.5 i -2.3 2.8 1.5 4.3 3.9 3.5 3.3 17.2 10.3 3.1 U.8 1.6 1-.3 3.2 i -.3 2.9 2.5 i - l . l 2.9 16.2 5.7 i —. 2 16.8 4 10.6 4.2 1.6 2.1 1-2.6 4.6 i -3.9 i.O 4.2 4.4 4.0 3.3 16.8 i —. 2 5.1 i Denotes a rate of change outside the range from 2 percent to 6 percent a year. Source: Federal Reserve data for monthly averages of daily figures. Each rate is calculated from the difference between the last month of the period (year or quarter) and the last month of the preceding period. 115 TABLE 3.—AVERAGE ANNUAL GROWTH RATES OF SELECTED INDICATORS FOR THE U.S. ECONOMY OVER THE PERIOD FROM 1948 TO 1967 [In percent] Total 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. Price level (GNP deflator) Population GNP in money terms GNP in real terms U.S. Government debt privately held Time deposits (commercial banks) Money stock (currency plus demand deposits) Money stock plus time deposits U.S. Government debt privately held, in real terms Time deposits in real terms Money stock, in real terms Money stock plus time deposits, in real terms Velocity of money (GNP divided by the money stock) Interest rate (Aaa bonds) 2.1 1.6 6.0 3.9 .7 8.7 2.4 4.6 —1.4 6.6 .3 2.5 3.6 3.6 Per capita 4.4 2.3 -.9 7.1 .8 3.0 -3.0 5.0 —1.3 .9 Source: Federal Reserve Bulletin, and Economic Reports of the President, 1968. Chairman PROXMIRE. Thank you, Professor Christ. STATEMENT OF JACOB COHEN, UNIVERSITY OF PITTSBURGH I. REPLIES TO QUESTIONS ON MONETARY POLICY GUIDELINES AND OPEN MARKET OPERATIONS 1.1 am in favor of a coordinated projection of fiscal, debt management, and monetary policies. 2. For purposes of centralized coordination, the President should be responsible for drawing up the program. 3A. Recent discussions of "indicators" and "targets" have the virtues of encouraging research on the linkages between monetary policy and real output. At the same time, however, incomplete knowledge about these linkages means that reliance on simplistic approaches—single indicators or targets—run the risk of failure. An implicit assumption in such research is that monetary policy must work through general quantitative controls rather than through selective controls. Whether this is based on likely effectiveness or doubts as to the political feasibility of controls are questions not found discussed in current monetary debates. From the standpoint of objective economic analysis it is helpful to distinguish between the "economics" and "politics" of economic policy. While the linkages going from monetary policy to expenditures have not been satisfactorily worked out, the linkages going "backward" from expenditures to monetary policy are more certain. Expenditures by the various sectors of the economy must be financed out of current income, out of borrowing, out of dissaving (sale of existing physical assets, sale of financial assets, reductions in money holdings). While these sources of funds may not be sufficient conditions for an increase in spending— nonfinancial sectors in the economy initially have to make decisions to spend—yet these decisions are contingent upon sources of funds. The analysis of monetary policy should put more emphasis on the final linkage—the necessary conditions for expenditure. Income flows are not directly affected by monetary policy. On the other hand, financial sources of funds (borrowing, financial dissaving) are the concern of monetary policy. We suggest that the appropriate 116 target for monetary policy is the volume and composition of financial (credit) flows. While possibly this can be regarded as a "single" target, it is sufficiently comprehensive to be free of the limitations of other targets currently being advocated. The contemporary debate on targets centers on the use of the money stock and on interest rates. Interest rates are an unsatisfactory target because the relationship between interest rates and financial flows is not a negative one. For interest rates to be a satisfactory target variable, increases in interest rates should indicate that financial flows are being restrained and decreases that financial flows are being encouraged. A study of the statistical data (quarterly data 1952-67, seasonally adjusted and unadjusted) reveals that the corporate bond rate and other money market rates and net funds raised by private domestic nonfinancial sectors have moved in the same rather than in opposite directions.1 If we correlate expenditures on consumer durables, residential construction by the household sector with both household personal savings and interest rates and similarly correlate corporate business expenditures on inventories and plant and equipment with business gross saving and interest rates, the same positive relationships appear between expenditures and interest rates. Underlying these positive correlations is the strong demand for credit in the postwar period. If we conceive of financial markets in terms of supply and demand curves for credit, these correlations suggest that the demand side of the market has shown sharper fluctuations than the supply side. The demand for credit has been the dynamic element responsible for both fluctuations in interest rates and credit flows. Unless higher interest rates originate on the supply side, interest rates will prove to be a misleading target for monetary policy. Some positive correlation can be found in the seasonally unadjusted data between changes in the money stock (defined as net demand deposits, foreign deposits plus currency outside banks) and net funds raised by private domestic nonfinancial sectors. This relationship is illusory, however because changes in the private sector's holdings of money are a component of net sources of credit which in turn equal funds raised by private sectors. If we subtract changes in the private sector's holdings of money from net funds raised, the relationship becomes significantly negative. If we expand the concept of funds raised to include financial dissaving by the household and corporate business sectors the negative relationship between these flows is further strengthened. In view of these relationships, the money stock, like interest rates, fails to offer a satisfactory proxy for the behavior of financial flows. Bank credit represents the category of financial flows with the closest linkage to monetary action. As evidenced in the experience of 1966 sharp variations in the rate of growth of member bank reserves together with maintenance or reduction of ceilings under regulation Q can succeed in reducing the rate of increase in bank loans. It is questionable however, whether the efforts of the Federal Eeserve can be regarded as an unqualified success in view of the ensuing "liquidity crisis." 11 am Indebted to Philip Wiest for running the regressions underlying these paragraphs* 117 More attention should be given to the possibilities of direct control of bank credit. Because borrowers from banks have limited ability to substitute one source of funds for another, control of specific categories of bank credit should be effective in controlling specific categories of expenditures. This empirical relationship between bank loans and expenditures was recognized in the fall of 1966 when the Federal Reserve attempted to control the boom in business capital outlays by controlling business loans. If the Federal Reserve had the power to impose special reserve requirements against bank business loans the expansion in capital outlays could have been controlled earlier in the boom without precipitating instability in finanicial markets, particularly the market for municipals. Consumer credit controls, special reserve requirements against bank loans, secondary reserve requirements against deposit liabilities—these are devices which have a current or potential usefulness. The regulation of an economy's expenditures by such specific controls has less analytical appeal than the development of "pushbutton" controls or attempts to put the economy on "automatic pilot." Yet there is ample intellectual challenge in evaluating (a) the past effectiveness of specific controls in the United States and Western Europe, (b) the selective effects of general controls^ (c) the ability to "finetune" selective controls and (d) the transitional problems associated with their introduction. Whether we are less economically free when we are subject to such controls as compared with the effects of inflation is a related philosophical issue. B. While we suggest total financial flows in the economy as the appropriate target, we would be opposed to guidelines which call for a fixed rate of growth in financial flows. The composition of financial flows and their linkages with real spending and financial spending should temper any decision as to the appropriate rate of growth. For example, a distinction would have to be made between a sector's borrowing or financial dissaving which financed real expenditures and borrowing or financial dissaving which represented simply a shift in portfolio composition. In the latter case, the extent of "financial overlay"—the ratio of financial flows to real expenditures would be changing and this would have to be considered. Secondly, a given volume of financial flow may be increasingly directed into certain "bottleneck" areas of spending and thus exert an inflationary effect even though totalflowTsremained relatively constant. Rather than gearing policy to the volume of financial flows the monetary authorities should attempt to estimate the interplay of real and financial flows in the economy. By projecting a "grid" of real and financial transactions for the major sectors of the economy the monetary authorities would be better able to determine the optimal rate of increase infinancialflows. 4. In times of depressed economic activity debt management can minimize the Federal sector's competition for funds by selling of debt to the banking system. If sold to the central bank (the law permitting) , variations in reserve requirements could control private bank credit expansion in subsequent inflationary periods. If sold to the commercial banks, supplementary reserve requirements which could be satisfied by commercial banks' holding government securities would prevent commercial banks from shifting from investments in governments into more profitable private loans. 118 5A. Open-market operations should be used to counteract transient factors (including seasonal factors) which otherwise would cause short-run instability in interest rates. While the economy could possibly tolerate an increase in short-run fluctuations in interest rates, the increased uncertainty as to interest rates could mean higher average interest rates with adverse effects on real investment. Open market operations conducted with a view to offsetting transient movements in bank reserves may conceivably exert their own unsettling effects on interest rates particularly in the Government securities market. Presumably, however, the instability in interest rate movements would be even greater without such intervention. Possibly the revised use of the discount window as proposed by the Federal Reserve can provide an adjustment mechanism without requiring open-market transactions, thus avoiding interest-rate effects. B. Rather than depending on the uncertain linkages running from open-market operations to expenditures on the GNP, reliance should also be placed on the control of credit by selective means, such as special reserve requirements against deposit liabilities or certain categories of bank assets, or downpayment and maturity requirements in the case of consumer and mortgage credit. C.(a) Rediscounting should be retained as a safety valve enabling banks to make short-run adjustments in their reserves as the result of seasonal or other transient factors. It is difficult to think of any useful purpose being served by discretionary changes in the discount rate. Recent Federal Reserve proposals to give commercial banks a basic borrowing privilege with the discount rate moving with a "cluster" of money market rates is a step in the right direction. (b) More attention should be given to making variations in reserve requirements a "two-way street." While increases are deemed a bluntedged instrument, such drastic action may sometimes be necessary. With the likelihood that Federal borrowing will increase substantially in the near future, the direct or indirect sale of debt to the Fed could economize on Treasury interest payments. Increases in reserve requirements under such circumstances would control credit expansion based on the associated increases in bank reserves. (c) A major factor in shifts in funds between commercial banks and savings institutions in recent years has been the upward adjustment of interest rates on time deposits under regulation Q. The pre-1960 situation when time deposit rates were not competitive has much to recommend it. The crunch of 1966 could possibly have been avoided had successive increases in ceiling rates not taken place. H.R. 11 could be amended on page 10 line 16 after "monetary affairs" to state "including discount policy, reserve requirements, administration of regulation Q and the provision of stand-by powers to impose selective credit controls." D. The idea of reporting is a good one except that it would be more consistent with the role of other departments or agencies involved in economic policy for the Federal Reserve to report directly to the President. If reports are made quarterly the requirements should be for rather general reports because of the possible tieing up of the resources of the Federal Reserve in preparing such reports. The problem of timelags in the availability of data would also be an argument for rather general quarterly reports. 119 More detail could be provided in an annual report. Such reports would have great usefulness if formulated in a flow-of-funds framework which forecast the likely expenditures and sources of finance of the major sectors of the economy. The Federal Reserve has made internal use of the flow-of-funds accounts for projection purposes. Their forecasts have never been published nor a step-by-step explanation given of their derivation. Such projections would make explicit the anticipated financial flows accompanying real spending plans and would offer advance warning as to likely pressure points in the economy. The kinds of financial flows that need restraining could thus be singled out. E. Since H.R. 11 calls for the abolition of the Federal Open Market Committee, the intent of this question is not clear. If open-market authority is entrusted to the Board, it should be at the discretion of the Board as to what other agencies of Government participated in their deliberations. Informal consultations with other departments or executive agencies or with Congress are of course possible and have been customary in the past. I I . APPRAISAL OF T H E STRUCTURE OF T H E FEDERAL RESERVE I would be in favor of all of the structural changes proposed with the exception of 5-year terms for members of the Board. Overlapping 10-year terms would provide better continuity to the Board. Making the term of the Chairman of the Board coterminous with that of the President is an important step toward coordination of monetary and fiscal policies under the President. M. COMMENTS ON MONETARY POLICY SINCE 1964 The outstanding impression provided by monetary developments since 1964 is that we are in an era of inflationary pressures generated by the competition of rising public expenditures with rising (and possibly induced) private outlays. It is too must to expect that conventional "hands-off" types of general monetary controls can succeed in curbing inflation. These must be supplemented by controls which directly affect the sources of finance of "trouble-making" private expenditures. STATEMENT OF ROBERT L. CROUCH, UNIVERSITY OF CALIFORNIA AT SANTA BARBARA I . QUESTIONS ON MONETARY POLICY GUIDELINES AND OPEN MARKET OPERATIONS The economic authorities have three basic shots in their stabilizing locker; debt management, fiscal, and monetary policy. Of these, the former is distinctly less powerful than the latter two. It is residual to the fiscal policy adopted. Given a certain budget deficit or surplus, new securities have to be issued or old securities have to be retired. The only decision is, which securities ? Potentially, this gives the economic authorities some discretion over the structure of interest rates. If, in the face of a budget deficit, they increase the issue of short-term securities relative to long-term securities, they will raise short rates 120 relative to long rates; and vice versa. The impact of changes in the structure of interest rates on aggregate economic activity is quite limited, however. Consequently, the responsible authorities should pursue what might be described as a balanced portfolio approach to the problem. They should supply those securities which, in their judgment, the market is most readily prepared to absorb. Under no circumstances should they subordinate their monetary policy to the exigencies of debt management or fiscal policy. For example, the Federal Reserve should not be pressured into purchasing Government securities in an attempt, which would prove fruitless in the long run anyway, to hold down the interest burden of the Government debt. The implications of the economic authorities' fiscal decision has inevitable repercussions on the level of Government debt. The monetary authorities should not feel circumscribed in their policy by changes in the debt one way or another. In particular, they should not feel obliged to bail out the fiscal authorities from the repercussions of their actions which, in the operationally most relevant case of a budget deficit, is higher interest rates. A given fiscal policy implies a certain debt. That should be financed most "efficiently," that is, at the lowest cost but without subordinating monetary policy to this task. This merely entails tailoring the individual securities issued to what the market will absorb most readily. The two major economic stabilization techniques are, then, monetary and fiscal policy. The institutional context within which the economy operates makes it far easier to employ monetary policy in the appropriate manner for stabilization purposes than fiscal policy. Since 1960 we have had two occasions in which fiscal policy has been consciously employed for stabilization purposes. The tax cut of 1964 aimed at inducing economic expansion and the tax increase and expenditure cuts of 1968 aimed at inhibiting the unrestrained expansion underway at that time. Even casual familiarity with the events at these times indicates how difficult it is to operate fiscal policy in an efficient manner. It is primarily a question of what is called the "action lag." This is the time between the need for action being recognized and the action itself being taken. The wholly appropriate congressional control of taxes and expenditure decisions inevitably means that the time between the need for action being recognized and the appropriate actions themselves being implemented is a lengthy one. Deliberation, debate, and decision on such weighty issues are inevitably time consuming. But during this process it is entirely possible, indeed probable, that economic circumstances change to such an extent that the decision which eventually emerges is inappropriate to the then prevailing economic circumstances. The economic authorities' timing, which is of the essence where stabilization policy is concerned, is then more likely to be off than on. Fiscal timing apart, the quantity of fiscal action is also more likely to be wrong than right. Again, with congressional control of our fiscal decisions, the eventual policy recommendations are inevitably a compromise. There is no guarantee whatsoever that it will be quantitatively appropriate. At best one can expect qualitatively correct decisions. That is to say, decisions which imply an expansionary fiscal policy when economic expansion is called for and vice versa. Given the institutional format in which the fiscal game is played, it is only good fortune when the fiscal decisions are quanti- 121 tatively appropriate for stabilization purposes. I assume, in an imperfect world, that Congress adequately reflects our social preferences vis-a-vis the public/private division of our economic activity. Debt management policy is relatively unimportant and fiscal policy is relatively inefficient as an economic stabilization device. What of monetary policy ? Fortunately, it can rise to the occasion. There is no doubt that an appropriate monetary policy is capable of creating an economic environment in which the aspirations enshrined in the Employment Act of 1946 would be fulfilled. A competitive and predominantly free enterprise economy is quite capable of generating employment for all those capable and willing to work as long as it is not prevented from doing so by the introduction of arbitrary and inappropirate monetary disturbances into the system. The fact is that economic contractions and unrestrained expansion do not just happen. They are caused—and caused by the monetary authorities implementing erroneous polices. In recent years, there have been three easily identifiable such instances. Two unnecessary economic contractions or slowdowns and one unrestrained, unsustainable expansion. In 1960 and again in 1966 the Federal Reserve caused the money supply to contract. The inevitable result was that, soon after, the rate of growth in real gross national product fell almost to zero, employment fell, unemployment rose, and industrial production declined. These events were not the inevitable and unavoidable consequence of a continuously adjusting free economy. They were the direct and avoidable, consequence of the Federal Reserve contracting the money supply. By way of contrast, the latter half of 1967 and in 1968 to date, the Federal Reserve has been pursuing a wildly overexpansionary monetary policy. A rise in prices of at least 4 percent is now inevitable. There is no way this can be avoided. And if the present policies continue to be pursued, prices will continue to rise at this rate. In the last 8 years, then, the ill-conceived policies of the Federal Reserve have subjected the economy to two bouts of unnecessary contraction and one bout of unnecessary overexpansion. We need to create a monetary environment in which the self-generating growth potential of a competitive, free enterprise economy can bring forth its fruit. Such an environment can easily be created. The Federal Reserve should be bound by law to insure that the supply of money (currency plus demand deposits) should be increased by at least 3 percent per annum and by no more than 5 percent per annum. If a wider definition of money was to be adopted (say currency plus demand and time deposits), the maximum might be raised to 6 percent per annum. The imposition of such a rule on the monetary authorities in place of their current unlimited capacity for discretionary action would free us from both deflation and inflation in the future. I would recommend, then, that monetary policy and fiscal policy be kept distinct and separate. The rule of monetary policy suggested above would be sufficient to guarantee full employment and continued growth without inflation in a free, competitive economy. Fiscal policy should be eschewed as a stabilization device. It should merely reflect the community's own decision on the balance they wish to establish between private and public goods. Fiscal decisions would then affect the mix of income (that is to say, the extent to which the gross national prod21-570—68 9 122 uct is comprised of private consumption and investment compared to public consumption and investment) and not the level of income. A budget deficit, for example, would lead to an increase in the public component of gross national products at the expense of the private component (due to the rise in interest rates which would occur); and vice versa if it w7as decided to reduce our collective consumption of public goods and a budget surplus was run. Debt management policy should be reduced to the purely technical function of marketing the debt implied by a deficit (or redeeming the debt that a surplus would allow) most cheaply; without, of course, undermining the monetary rule suggested above. Question 2 The appropriate division of responsibility to institute the economic reform suggested above is quite simple. (1) The monetary rule should be laid down by act of Congress or, at the very least, by a resolution expressing the sense of Congress. (2) Fiscal policy would continue to be, as at present, the outcome of the deliberations of, and debate between, both Houses of Congress and the Executive. (3) Debt management would remain the responsibility of the Treasury in consultation with the Federal Reserve. Question 3A As stated in answer to question 1, the immediate target of monetary policy should be the achievement of growth in the supply of money (narrowly defined) at a rate between 3 and 5 percent per annum. This would be sufficient to guarantee full employment, continual economic growth, and stability in the level of overall prices. There is no reason to complicate this goal by diverting the Federal Reserve's attention to other, subsidiary, variables such as bank credit, liquidity, free reserves, interest rates, and so forth. Indeed, a large measure of our past and present troubles have been, and are, directly due to the Federal Reserve's focusing its attention on misleading targets. In particular, it pays entirely too much attention to the level of interest rates. Usually, it is the Federal Reserve's myopic concentration on this variable which imposes unnecessary gyrations on the economy. The contemporary (summer 1968) situation is a case in point. Monetary, or nominal interest rates are at relatively high levels. The Federal Reserve interprets this as indicative of monetary stringency. Rut, in actual fact it is the result of a too easy monetary policy. During this period, the Federal Reserve has been increasing the money supply at between 8 and 10 percent per annum. This makes inflation of the order of 4 or 5 percent per annum inevitable. Consequently, to compensate for this anticipated inflation, lenders are only willing to lend at the present high nominal interest rates. Paradoxically to some, perhaps, nominal interest rates continuing at high levels indicates a too easy monetary policy and not the reverse. Failing to understand this, the Federal Reserve is attempting to lower interest rates by expanding the money supply even more rapidly. This may be effective in the short run but it is self-defeating m the long run since the anticipated price increases that will ensue as inflation takes hold will feed back to the interest rate and lead to higher levels still. During contractions the Federal Reserve is misled by changes in interest rates, too. In contractions, nominal interest rates fall. The 123 Federal Reserve is prone to interpret this as indicative of monetary ease in spite of the fact that it is invariaJbly causing the supply of money to decrease at the same time. Consequently, exactly when it should be inducing an increase in the supply of money it is doing exactly the opposite because it erroneously regards falling nominal interest rates as self-evidently expansionary. The ultimate goals of domestic economic policy are full employment, economic growth, and overall price stability. In a free enterprise, competitive economy, a money supply continuously expanding at between 3 and 5 percent per annum is both necessary and sufficient to achieve these ends. This, then, should certainly be the proximate target of monetary policy. All other targets should be subordinated to this end. If, as I believe it should, a fixed rule of monetary expansion is imposed on the Federal Reserve, there are certain institutional reforms that should be introduced to ease the Federal Reserve's implementation of this policy. At the moment, it has three instruments of monetary control at its disposal; open-market operations; changes in bank reserve requirement ratios; and changes in its discount rate. In their present form, the latter two are counterproductive. The existing reserve requirement ratios observed by commercial banks are a patchwork hitsorical compromise. At present, they differ among banks according to their geographic location and whether or not those banks belong to the Federal Reserve System. In addition, the reserves required against time deposits are lower than those required against demand deposits. This means that changes in the money supply occur as a result of shifts in reserves among banks and between the two classes of bank deposit. This means that, irrespective of the Federal Reserve's capacity to determine the total of reserves, the Federal Reserve's control of the supply of money is undetermined. I do not wish to exaggerate the significance of these matters, but it does seem that a more reliable control of the supply of money would be established if (1) uniform reserve requirements were applied to all member banks, (2) the same reserve requirement ratio was applied to both demand and time deposits, and (3) all commercial banks were compelled to become members of the Federal Reserve System. The latter reform might be implemented through a strict judicial interpretation of the "currency clause" in the Constitution. Having established uniform reserve requirement ratios for all banks against both classes of deposits, the Federal Reserve's existing power to make variations in these reserve requirement ratios should be revoked. There may have been some justification for such a power in the past, but there is no longer. There is no monetary event that changes in reserve requirements ratios can achieve that cannot be done better through open-market operations. The operation of the discount rate mechanism in its present form also leaves much to be desired. The original purpose of the discounting privilege was to provide for a lender of last resort in the monetary system to whom recourse might be made in times of financial stress. It was designed to perform the function of a safety valve. It has, however, developed into a semipermanent leak since it is usually set below market rates of interest. This means that it is normally profitable for banks to borrow from the Federal Reserve and lend the funds on 124 the private market. Consequently, there is frequently a state of excess demand at the discount window. The borrowed funds which the Federal Reserve chooses to make available are, therefore, rationed among the competing claimants by nonprice rationing techniques. Such administrative discretion should have no place in the monetary mechanism. The appropriate reform is to make the discount rate a penal rate recourse to which would only be made in minimum amounts for the minimum possible period while the bank in question makes appropriate adjustments in the scale of its operations as rapidly as possible. While always being maintained above market interest rates, the discount rate should be allowed to vary with them. At the moment, discretionary changes in the discount rate are frequently misleadingly interpreted. The latest example occurred with the reduction in the discount rate in August 1968. This was widely heralded as a move toward monetary ease following, by implication, a period of monetary restraint. In actual fact, the Federal Reserve was allowing the supply of money to increase at an annual rate of about 11 percent per annum at this time. This is the antithesis of a tight money policy. If the Federal Reserve was subjected to the constraint of adherence to a monetary rule and the discount rate was pegged at a constant differential above market rates and allowed to fluctuate with them, the Federal Reserve's control of the money supply would be made more perfect and both its, and the public's, attention diverted from a myopic concentration on interest rate changes. The reforms suggested above to the reserve requirement and discount rate mechanisms, leaves the burden of the implementation of continuous monetary expansion according to an announced rule solely to open-market operations. They will prove adequate to the task. Question SB My answer to this question will be apparent from the context of the discussion of the previous questions. Briefly, the immediate target variable of monetary policy snould be an annual rate of increase in the money supply (narrowly defined) between 3 and 5 percent. This rule should be adhered to regardless of the so-called economic winds. In fact, as I have said earlier, economic winds do not happen they are caused. Adherence to the rule would allow calm and orderly economic change to occur without periodic buffetings being imposed on the economic system by destabilizing blasts of alternating hot and cold air emanating from the Federal Reserve. Question 3C Not applicable. Question 3D The same rule of monetary growth should be adhered to year after year in all except abnormal circumstances. Abnormal circumstances should be subject to strict interpretation and require congressional action. One has in mind, for example, a declaration of war as a reason for modifying the rule. Small changes in the behavior of the goal variables of domestic economic policy (low unemployment, the rate of economic growth, and stability in the overall level of prices), should not be allowed to induce abrogation of the rule of monetary expansion. 125 In a competitive, free enterprise economy continuously adjusting to fresh stimuli, such small changes are only to be expected. They must be accepted. The situation will be exacerbated, and not ameliorated, by attempts to fine tune them out. In a dynamic economy one must be prepared for adjustment and change to new circumstances. The crucial thing is to create an environment in which such manifestations of departure from equilibrium work themselves out quickly. Such an environment would be created by the suggested monetary rule. The rule should certainly not oe tampered with in the light of conjectural estimates about anticipated investment, Government spending, taxes, and so forth. In a fully employed, growing economy such changes would only affect the mix of income and not its level. And so they should. They reflect individual decisions or democratically expressed collective decisions and the achievement of them would only be frustrated by discretionary action allegedly designed to offset their effects. The monetary authority should not be empowered to offset the attainment of our individual and collective decisions. On the contrary it should be restrained from so doing. This is what the monetary rule is designed to do. Question 3E The Federal Reserve should be mandated by act of Congress or congressional resolution to increase the supply of money, defined as the sum of currency plus demand deposits, at no less than 3 percent per annum and no more than 5 percent per annum. Question 3F See answer to question 3D. Question ^ The short answer to this question is, very little, At best, debt management policy can affect the structure, as opposed to the general level, of interest rates. However, recent research has shown that even their capacity to bring about such changes in the structure is strictly limited. Consequently, the Treasury and Federal Reserve should not concern themselves with this issue. The Treasury, in consultation with the Federal Reserve, should, instead, so tailor their supplies (redemptions) of securities as to minimize the interest cost associated with the national debt. This means ordering their debt issue (or redemption) program to accord with the relative preferences of the market. Question 5A The appropriate technique to employ to combat seasonal and transient factors affecting money markets is the so-called sale under repurchase agreement. This provides "money with strings", as the saying goes. In other words, it eases periods of seasonal and transient stringency without diluting the Federal Reserve's permanent open market posture, which, if a monetary rule and the other technical reforms mentioned in answer to question 3A are adopted, would be that of a persistent net purchaser of securities on the open market in amounts designed to implement the appropriate continuous growth in the money supply. Question 5B Yes. See answer to question 3A. Question 50 See answer to question 3A. Question 5D If the Federal Reserve was constrained by Congress to the monetary rule that has been suggested, there is little information that the Federal Reserve could report to Congress that Congress would find interesting. No harm, though, would be done by introducing this formal accountability of the Federal Reserve to the Congress. On those abnormal occasions when Congress frees the Federal Reserve to abrogate the rule, Congress doubtless would require a detailed account of the Federal Reserve's activities in the unusual circumstances. Question 5E If, as I recommend, a monetary rule is imposed by Congress, the proceedings of the Open Market Committee would be dull and supremely uninteresting. Little benefit would accrue to outside official observers and no harm would be done if they were absent. My own belief is that few would wish to attend such dull proceedings more than once. I I . APPRAISAL OF T H E STRUCTURE OF T H E FEDERAL RESERVE The recommendations under this section are designed to reduce the independence of the Federal Reserve and make it subservient to the wishes of elected officials. This is to be commended as long as it does not imply that monetary policy becomes a political football. This could not, of course, occur if the monetary rule that has been suggested was adopted. In such circumstances, it would be wholly appropriate to subordinate the independence of the Federal Reserve and force upon it the simple technical function of providing continuous monetary growth. I I I . COMMENTS ON MONETARY POLICY SINCE 1964 This period is almost perfectly designed to illustrate the difficulties into which an independent Federal Reserve empowered to take discretionary monetary action can get the economy into. From June 1964 to April 1965, the money supply increased at 4 percent per annum. This was in conformity with the rule that has been suggested. Practically all the economic indicators were favorable; real income was high and rising, unemployment was low and falling and prices were relatively stable. The Federal Reserve, however, did not leave well enough alone. In April 1965 it began to accelerate the rate at which it was increasing the money supply so that in the ensuing year from that date the money supply increased by 6 percent. The inevitable followed. Prices which had been rising at just over 2 percent per annum, soon began to rise at 3.3 percent per annum. The Federal Reserve's too-easy money policy was generating an unsustainable expansion. It therefore reversed itself. But instead of adjusting carefully back to a more reasonable rate of monetary expansion, it over reacted. Beginning in April 1966, the money supply was actually made to decline slightly. The pre- 127 dictable outcome followed. After a short lag, a recession set in. Between the last quarter of 1966 and the second quarter of 1967, industrial production fell by 3.6 percent, the rate of increase in real gross national product fell to a meager 1.1 percent and unemployment rose. Inflation was effectively eliminated as evidenced by the reduction in the rate of price increase to just over 2 percent. But reestablishment of this relative stability in overall prices had been bought at the cost of lost jobs and production. Having overreacted in one direction, the Federal Reserve soon proceeded to overreact in the opposite direction. Instead of increasing its rate of monetary expansion to a level capable of producing a sustainable expansion, it began flooding the economy with money. By April 1968, this had reached unprecedented levels. Since that date the Federal Reserve has allowed the money supply to increase at more than 10 percent. Although it worked up to this orgy of overexpansion gradually, the reemergence of inflationary forces was already apparent in the second half of 1967. Prices were rising then at about 4 percent per annum. The budget was in chronic deficit at this time and this was identified as the villain of the piece. The cry went up for a tax increase and an expenditure cut. After due debate, a compromise program of fiscal restraint was passed. This seems to have induced in the Federal Reserve a sense of total abrogation of its responsibilities. Persuaded that the fiscal reversal would take the heat off the economy it entered into a period (which is still, unfortunately, in progress at the time of writing) of ludicrous overexpansion. Involving, as it does, an episode of monetary overexpansion of historic proportions, the danger is that when the inevitable inflation ensues and the Federal Reserve realizes that it has unshakeable responsibilities for the economy's continued good health which cannot be shrugged off onto the fiscal authorities, the Federal Reserve will overreact. A monetary contraction together with a tight fiscal policy contains the seeds of serious economic disorder. The immediate policy problem as of August 1968 is, given the more restrained fiscal position, to get the rate of increase in the money supply down to a sustainable level slowly. The Federal Reserve must reverse itself, but it must do it in a sober manner. STATEMENT OF JOHN M. CULBERTSON, UNIVERSITY OF WISCONSIN GENERAL STATEMENT The urgent need to improve the control of stabilization policy in the United States does not, in my view, involve primarily a matter of improving "coordination" of monetary, fiscal, and debt management policy. Improvement of control over stabilization policy seems to require not increased centralization of undefined or "discretionary" policies but a more basic restructuring of the control apparatus to reduce the uncertainty of policy, free policy formation from the biases exhibited in the past, and make the formation of policy systematically responsive to the available economic knowledge. A justification of this diagnosis, evaluation of recent stabilization policy, and an outline of a program to bring stabilization policy under more effective control is developed at length in my recent book, Macroeconomic Theory and Stabilization Policy (New York: McGraw-Hill Book Co., 1968). 128 Rather than now taking the actions defined by H.R, 11, I should prefer giving priority to institution of a systematic program for evaluation of monetary, fiscal, and debt management policies according to performance criteria defined by existing economic theory. Such clarification of the performance of stabilization policy seems a necessary prerequisite to definition of or acceptance of the basic reforms required to bring about a reliably controlled policy system. "Coordination" appears to be a euphemism for increased centralization of power over stabilization policy by increasing the control over monetrary policy by the administration. Considering the manner in which administrations of both parties have used fiscal policy and debt management policy—and the identifiable political biases m policy formation—it seems reasonable to fear that giving the administration control also over monetary policy may result in a performance worse than that of the recent past, and will increase the hazard of a catastrophically bad performance of stabilization policy. REPLY TO QUESTIONNAIRE ON H . R . 11 I 1-2. An annual promulgation of stabilization policies is likely to promote inflexibility of stabilization policies, make nominal stabilization policies subject to political biases, and be on balance harmful. The conception of the Employment Act of 1946 that the Government should at the beginning of the year determine the prospective gap in total demand and fill it through policy actions is unrealistic, being responsive to the static, stagnationist view of the economy prevailing at the time of its enactment, which has not been supported by subsequent events. What seems required, rather, is to govern policy by defined strategies or decision rules making them continuously responsive to the changing economic situation. Such defined programs for policy would reduce the discretionary element in policy formation and require consideration of political aspects of policy control within a new set of terms of reference. Discretionary formation of stabilization policy either by the administration or by independent agencies such as the Federal "Reserve seems to give rise to definable biases and nonoptimal policies. 3. Definition of strategies or decision rules for monetary policy and other branches of stabilization policy should rest upon a rational process of strategy formation on the basis of economic theory. This is not a suitable topic for offhand opinions or judgments. One class of defective decision rules is those defined in terms of variables that can lead to cumulative errors of policy because of lack of allowance for feedbacks affecting the target variables. This class of case is illustrated by the traditional credit approach to monetary policy followed by the Federal Reserve. To illustrate the hazard of cumulative error, the Federal Reserve judgment as to "proper" interest rates and credit conditions errs in setting them too high, leading to a softening economic situation or recession, leading to reduction in demand for credit and declines in interest rates (and "easing" of credit conditions), leading the Federal Reserve further to reduce the provision of bank reserves and monetary growth, leading to further economic weakness, and so on. 129 Decision rules defined in terms of the money supply (or in terms of bank reserves if accurate allowance is made for changes in bank demand for reserves and other reserve factors that it is desired to offset) can avoid this hazard of cumulative error from misestimated or ignored feedback, since they relate directly to the politically controlled independent variable in question. Given enough knowledge regarding the economic system, a decision rule stated in terms of credit variables could be defined that is equivalent to any rule stated in terms of the money supply, but existing knowledge does not suffice to permit this to be done with confidence. Thus, under existing conditions of limited knowledge, criteria defined in terms of the money supply are much the less hazardous. They also avoid the evasive or politically biased characterization of nominal policies by the Federal Reserve and the administration based upon shifting, nonquantitative, or obscure credit criteria. 4. The problem of defining alternative debt management strategies and choosing the one that is optimal under existing limited knowledge quite parallels the problems of control of monetary and fiscal policy. Any assertion that "fiscal policy has no role to play in this matter'5 could only derive from the proposition that the behavior of total demand is invariant with respect to any conceivable conduct of debt management. Existing knowledge does not suffice to support such a proposition. 5. Again, the crucial problem is to achieve agreement on a reasonable decision rule, strategy, or monetary-policy program, presumably one defined in terms of the money supply (or equivalently in terms of bank reserves). So long as such a decision rule is adhered to, temporary smoothing operations by the Federal Reserve probably do little harm. Building some allowance for seasonal changes in demand for money into the decision rule seems appropriate in the present state of knowledge. The instruments of Federal Reserve monetary policy other than open market operations are probably superfluous and ought to be placed on a standby basis or eliminated. This involves some complications in member-nonmember bank relations, and other matters. If the Federal Reserve operates under some defined strategy, quarterly reports and presence of outsiders at meetings would be superfluous (although reports to Congress giving an official justification should perhaps be required in the case of any discretionary departure from the standard policy program). On the other hand, lacking a defined policy program or set of performance criteria for monetary policy, it is not clear that frequent reports or presence of outsiders in meetings wTill result in improved policy actions. It may increase the hazard of a catastrophically bad set of policies—as in the classic case of an administration with an excessive fiscal deficit pressuring the Federal Reserve to provide bank reserves to hold down interest rates. If we are to attack the subject in terms of a catch phrase, "checks and balances" seems as relevant as "coordination." i i The crucial problem seems to me to be bringing monetary policy (and public information regarding it) under an effective control system. Adopting any defined program of monetary policy would imply 130 the reasonableness of some simplification of the Federal Reserve administration. The changes thus indicated, however, would be quite different from the ones that would be called for under an approach involving retention of discretionary monetary policy with the power effectively transferred from the Federal Reserve to the administration. Making the proposed administrative changes at this time, I fear, would only muddy the waters with reference to the really important problem. in The recent record is strikingly consistent with the interpretation that variation in the rate of growth rate of the money supply (reflecting, at the margin, Federal Reserve actions regarding the provision of bank reserves) has been a major cause of changes in the rate of growth of total demand. In this period, the record obviously could have been worse than it was, for runaway inflation and recessions were avoided. However, it appears that a preferable performance could have been achieved with a readily definable monetary policy. Perhaps more importantly, the existing control system appears potentially subject to cumulating errors on a dangerous scale, these in recent years being avoided only late in the game and seemingly in a somewhat accidental manner. Deviations of the growth of total demand from a path that might reasonably have been defined as a target seem attributable in least in part to the below-normal monetary growth of 1959-62, to the excessively rapid monetary growth from the spring to 1965 to the spring of 1966, to the abrupt halting of monetary growth during the rest of 1966, and to the extremely rapid growth rate of the money supply since the beginning of 1967. This erratic and seemingly somewhat accidental monetary policy cannot be justified as optimal in the light of existing knowledge, nor can the control system from which it derives be characterized otherwise than as hazardous. STATEMENT OE PAUL DAVIDSON, RUTGERS, THE STATE UNIVERSITY I . INTRODUCTION Any objective inquiry into improving the economic effects of the monetary policy of a central bank should begin with (1) a statement of objectives of such policy and (2) a discussion of means that can achieve these goals. The four most often mentioned practical goals of monetary policy are— (1) To prevent inflation; (2) To encourage full employment; (3) To encourage sustained rapid economic growth; (4) To counteract balance-of-payment deficits. In framing these objectives I have deliberately worded objectives No. 1 and No. 4 in negative or obstructive formats, while No. 2 and No. 3 utilize more positive wording. My rationale for this is to emphasize that active pursuit of objectives No. 1 and/or No. 4 by traditional 131 monetary methods will normally obstruct the achievements of objectives No. 2 iand No. 3. Accordingly, it is my view that monetary policy should be oriented solely toward achieving full employment and economic growth. This does not mean, of course, that monetary policy should operate in a vacuum. Nor does it mean that money and monetary policy cannot have some impact on the general price level or the balance of payments. What I wish to recommend is the coordination of monetary and fiscal policy with an incomes and foreign trade price policy so that the four objectives listed above can be approached simultaneously. Mere coordination of monetary and fiscal policy, while a step in the right direction, will not be the administrative panacea for reaching these objectives under present institutions—even if accurate forecasts of future events could be achieved. Although discretionary control over the money supply is essential if we are to obtain full employment and sustained economic growth, any attempt to utilize changes in the money supply as the primary tool to restrict general price increases or to cure balance-of-payments deficits will, under our present market-oriented system, insure unemployment while severely hampering growth. I I . FULL EMPLOYMENT AND ECONOMIC GROWTH Full employment and economic growth with their promise of unprecedented prosperity, could presently provide a higher standard of living for all Americans. Full employment and growth could mean the rapid elimination of poverty in the United States. Full employment and growth could bring about increased social stability as group antagonisms dimish with rising income levels. Full employment and growth could improve our position in the cold war not only by strengthening our defenses, while simultaneously increasing our aid to the uncommitted countries, but it would also demonstrate to the world the vitality of a market economy in providing for the economic and social ladvancement of its citizens and its friends. With all these obvious advantages that accrue to a fully employed economy, surely full employment and economic growth must be the primary economic goals of our society. Yet, except for the military escalation in Vietnam operating in tandem with the 1964 tax cut, American economic policymakers, Republicans ;and Democrats, cabinet members and central bankers alike have, for more than a decade, pursued a course designed to prevent the achievement of full employment. The policymakers, of course, are not malevolent but they have been trapped in a conflict of goals which dilutes our fervor for maximum output. Several years ago, Secretary of Labor Willard Wirtz posed the problem very graphically when he said: "You sometimes get the feeling, sitting where I do that there is a shell game going on in the discussion of this particular [unemployment] problem, and that the shells are marked 'inflation,' 'unbalanced budget,' and 'unfavorable trade balance' * * * every suggestion which is made to iadvance the purpose of full employment is met by one or another of these arguments, and very often by all three." 132 Why do we participate in such a game, the outcome of which cheats us out of full employment and rapid growth ? The game goes on and on because the winners (and the game is rigged so that we know who the winners will be) outvote the losers at the polls. But majority rule ought not to be tyranny. Majority rule is neither right nor proper here because we have failed to guarantee an inalienable right to the minority—the right to a job and a respectable level of income. Up to now we have failed to create an economic environment in which democratic rule yields the optimum result within a monetary, market-oriented economy. Such an economic environment can be created, however, with an appropriate battery of monetary, fiscal, and income policies. Until such coordinated policies are developed, this shell game will continue and as the late President Kennedy lamented, we will continue to content "ourselves with pious statements about the wastes of our human resoures." Ever since the 1930's, economists have realized the recessions can be avoided and full employment can be achieved by fiscal policies such as tax cuts or increased governmental spending and/or expansion in the money supply. Moreover, if the economy begins from a position of less than full employment, policies that stimulate increased economic activity simultaneously reduce unemployment, and stimulate investment and growth; for one of the most important messages of the "Keynesian" revolution in economics was the complementarity of consumption and investment in recession. Thus we learned that it is possible to have more butter, more plant and equipment, and more guns, too, if only we had the courage to pursue certain fiscal and monetary policies. Although there continues to be a debate among economists as to whether, as the Chicago school succinctly asserts, "money matters," that is, a questioning of the relative efficacy of expansionary fiscal compared to monetary policies, most economists now agree that expansion of market aggregate demand is a requirement for continuous full employment and economic growth in peacetime. What has been often overlooked in this professional controversy over whether "money matters," is that an increase in market demand means not merely an increase in wants but also an increase in the ability to pay for goods and services. An increase in the ability to pay, in a modern market-oriented, monetary economy, must involve an increase in the supply of money before the increased demand can be made operational in the marketplace. This fundamental notion that an easy-money policy is a prerequisite to expansion and growth is, as I have tried to demonstrate in a number of articles (1), (2), (3), an essential concept necessary to the understanding of the mechanism underlying the traditional Keynesian policy prescriptions for economic expansion. As John Maynard Keynes wrote more than 30 years ago: The banks hold the key in the transition from a lower to higher scale of activity * * ». The investment market can become congested through a shortage of cash. It can never become congested through a shortage of savings. This is the most fundamental of my conclusions in this field (6, pp. 668-669). Or again: A heavy demand for investment can exhaust the market and be held up by the lack of financial facilities on reasonable terms. It is, to an important extent, the "financial" facilities which regulate the pace of new investments * * * too 133 great a press of uncompleted investment decisions is quite capable of exhausting the available finance, if the banking system is unwilling to increase the supply of money * * *. The control of finance is, indeed, a potent, though sometimes dangerous, method for regulating the rate of investment (though much more potent when used as a curib than as a stimulus) (5, p. 248). Easy-money policies are a necessary but not a sufficient condition for stimulating economic growth,. If the desire for new investment goods is weak because of poor profit opportunities, then easily obtainable finance will not, by itself, do the trick. If, on the other hand, the desire for investment is strong among businessmen, the banking system and the monetary authority can play an essential role in providing funds on terms which the investors deem attractive. It is at the level of financing investment projects that the money supply plays an essential role in stimulating economic growth in a monetized market economy, once the investment desire is present in the economy. Fiscal policy, on the other hand, may develop latent investment demand either by increasing profit opportunities by augmenting consumer or Government demands in the marketplace or by increasing after-tax profits on existing market demands by use of subsidies, tax credits, or profit-tax cuts. Nevertheless, unless investors can obtain funds, they cannot place orders for additional investment goods no matter what level of profits are expected to be earned on these potential investments. Since m modern, money economies with a developed banking system, the money market may not "clear"; that is, there may be an unsatisfied fringe of borrowers (particularly when business is active), aggregate demand may be deficient merely because there is a shortage of money. Accordingly, fiscal policy may be a necessary; but it is not by itself a sufficient condition for full employment and economic growth. In a monetary economy, it is finance (i.e., increases in the money supply) which provides the energy fuel that permits the investment tail to wag the gross national product dog. It is obvious, therefore, that the necessary and sufficient conditions for full employment require the coordination of fiscal and monetary policy. To the extent that H.R. 11 has as one of its major objectives "to improve the coordination of monetary,fiscal,and economic policy," it must be warmly supported. Nevertheless, coordination of monetary and fiscal policy is not the panacea for our economic problems. In the absence of a coordinated "incomes policy" to prevent inflation and a foreign trade policy to correct balance-of-payments deficits, a coordinated fiscal and monetary policy may be required to deal with these latter issues—a task which they are not equipped to efficiently handle. Accordingly, before providing my conclusions on H.R. ll's detailed recommendations for coordination, I should like to discuss the inflation and balance-of-payments questions. III. INFLATION The 1964 tax cut was the first major measure taken by Congress for the expressed purpose of expanding aggregate market demand in order to move toward full employment. This action plus the subsequent military expenditure expansion as hostilities in Vietnam increased brought the United States close to full employment and rapid economic growth 134 for the first time in more than a decade. But with this achievement came the usual corollary of a free market economy—rising prices. No one is against full employment per se. Moreover, if one begins in a recessionary period, full employment and rapid economic growth are complementary objectives which simultaneously can be achieved by a judicious mix of proper monetary and fiscal policy. It is the increasing inflationary effects as unemployment declines which constitutes a basic conflict and which induces policymakers to adopt measures designed to restrain aggregate demand, and hence hopefully restrict price increases by creating slackness in labor and product markets. This fear of inflation is not new; however, the fear of massive unemployment which was generated in the great depression, as well as the hot and cold wars which followed, overrode the objections to inflation and made possible the expansionist policies in the forties and early fifties. But almost a third of a century has passed since the great depression and for many citizens these terrible years are as remote as the ravages of the Civil War. Continuing inflation in the forties and early fifties increased our fear of rising prices, while the continuing prosperity has dulled, for most white urban workers at least, the fear of unemployment. At the present time, inflation and not unemployment appears to be the most likely source of economic dislocation, although it is my firm belief, that much of the riots of the urban ghetto community and the problems of the rural poor reflect the continuing unemployment and underemployment problems in those sectors of the economy. A truly fully employed economy would not only raise the level of real income for the entire community, but it would open up job opportunities for members of many minority groups, so that, in general, the average level of real income of these minorities would rise more rapidly than the national average. Under present institutional arrangements, however, the rate of inflation that would accompany sustained full employment would severly damage (1) the real income of those citizens on relatively fixed money incomes, the so-called rentier groups—the retired, the disabled unemployed, widowrs, orphans, mothers with abandoned children, and even some white collar workers, certain government employees such as policemen, teachers, etc., and (2) the real wrealth of middle- and upper-income groups who held their wealth in the form of savings accounts, bonds, and other fixed sum obligations. Moreover, even organized labor would find inflation galling in that it would mean that collectively gained money-wage advances turned out not to be as sizable an increase in economic welfare as they would have been with stable prices. Management, on the other hand, might find the increased truculence of labor (both organized and unorganized) under sustained full employment exceedingly difficult to deal with. The inflationary pressures would also create problems in export markets and encourage foreigners to compete domestically. The resulting political winds, which were correctly foreseen 25 years ago by Kalecki, have produced a "political trade cycle," where, as the level of unemployment declines and prices rise, rentier and other interests combine to pressure government to return to the orthodox policy of cutting down budget deficits and restrictive monetary policies. Thus it is not surprising that first the Federal Reserve Board, 135 and later the administration began to advocate restrictive policies before full employment had been reached, much less sustained. These restrictive policies, whether coordinated or not, ultimately place the major burden of fighting inflation often on those citizens least capable of bearing it—a group which may be called the LIFO workers—the last hired in prosperity, the first out in recession. This group includes young people just entering the labor force, unskilled workers primarily located in ghetto areas, and even older workers nearing retirement ages (unless protected by seniority rules). Equity, it seems to me, requires that we redistribute this burden more broadly. Of course, it is not irrational for the rentier and other groups to bring political pressure to stop inflation since they can suffer absolute (or at least relative) economic losses as prices rise. Though they may favor full employment and economic growth in the abstract they are forced by their economic self-interest to push for the only antiinflationary policies available—restrictive monetary and/or fiscal policies. As a consequence, no matter under whom or how well monetary and fiscal policies are coordinated, we will be unable, for political reasons, to achieve full employment and sustained economic growth until a viable economic policy designed to sever the existing connection between rapidly rising prices and low levels of unemployment is introduced and coordinated with monetary and fiscal policies. In order to understand what general type of policy is required, it is essential to explicitly define some basic economic concepts and principles. Although economists have ofttimes demonstrated excessive taxonomic dexterity in categorizing "causes" of inflation, we can avoid many semantic problems by taking recourses to a few simple economic concepts. It is neither rising prices of nonreproducible goods such as rare paintings or sculptures, nor the prices of securities listed on the New York Stock Exchange, nor even the prices of reproducible nonconsumer goods like aircraft carriers, which are the main focus of public concern in discussions of inflation. Inflation becomes a major cause of public interest only when it is the market prices of reproducible goods that bulk significantly large in consumers' budgets that are continuously increasing. Keeping this pragmatic view of the public concern about inflation in mind, the problem can be readily analyzed by concentrating on what economists call the "flow-supply price of goods," where the latter is defined as that price "which is sufficient and just sufficient to make it worthwhile for people to set themselves to produce the aggregate amount" [8, p. 373] of output. Our emphasis on supply prices should not be interpreted as supporting the myopic view that demand factors cannot affect price; nevertheless if the supply price for any given quantity of reproducible goods does not alter, then no matter how far the market price may be momentarily displaced from that supply price, the price of future output will subsequently return.1 1 If only nonreproducible goods such as works of arts by dead artists were rising, no major public policy problem would arise. iThis latter case would be an example of a pure demand-price inflation and could readily be analyzed primarily by concentrating on changes in demand factors. 144 Supply prices can increase for three main reasons: (1) diminishing returns, (2) increasing profit margins, and (3) increasing money wages (relative to productivity increments).2 For more than a century, economists have taught that every expansion of output and employment will normally involve increasing costs and increasing supply prices because of the law of diminishing returns. Diminishing returns, it is held, is inevitable—even if all labor and capital inputs in the production process were equally efficient—because of the scarcity of some input such as raw materials or managerial talent. Actually, however, economic expansion will lead to increasing costs (and prices) not only because of the classical law of diminishing returns but also because labor and capital inputs are really not equally efficient. Expansion of output in our economy often involves the hiring of less-skilled workers, and the utilization of older, less-efficient standby equipment and therefore adds to diminishing returns. Thus, as long as unemployment is declining, diminishing returns inflation will be an inevitable and unavoidable consequence of further expansion. The severity of diminishing returns inflation will vary with the level of unemployment. When the rate of unemployment is high (say about 5 percent), idle capacity will exist in most firms, so that diminishing returns are likely to be relatively unimportant. As full employment is approached, however, an increasing number of firms will experience increasing costs, and diminishing returns inflation will become more important. Although in the short-run diminishing returns inflation is an inevitable consequence of every expansion in employment, in the long run, improvements in technology, Government-sponsored training and educational programs, and increases in capital equipment per worker can offset this price rise. The second type of inflation will occur when businessmen (particularly in our more concentrated industries) come to believe that the market demand for their product has changed sufficiently so that it is possible for them to increase the markup of prices relative to costs. If managers in many industries increase their profit margins, we will experience a profits inflation as the supply or offer prices rise. Third, every increase in money-wage rates, which is not offset by productivity increases will increase costs, and if profit margins are maintained, increase supply prices. Consequently, we can expect that increases in money-wages induce price increases. This phenomenon is often referred to as wage-price inflation. Since as unemployment levels decline it is easier for workers to obtain (collectively and individually) more liberal wage increases, we may expect wage inflation to become more pronounced as employment rises; although wage inflation can occur even without expansion, if labor is able to secure increases which exceed productivity increments. Historically, rises in the price level has been due to some combination of these three inflationary forces. Thus, changes in the price level are ultimately related to changes in money wage rates, changes in profit margins, and diminishing returns. 2 If imports are an important component of the output of most reproducible goods, then rising import prices can affect the flow supply price. For the United States, I do not believe this is a significant problem and hence I have omitted it from the discussion. 137 Every significant expansion in economic activity will induce some price increases because of diminishing returns. With rising prices, workers will, at a minimum, seek cost-of-living wage increases. Moreover, as pools of unemployment dry up, workers will be more impenitent in their total wage demands. Managers will be more willing to grant wage increases m a rising market, for they are more certain that they will be able to pass the higher labor costs on in higher prices. Also, management will find that as they hire more workers to meet the rising demands for their products, the cost of searching out and training the remaining unemployed will increase; consequently, they will often attempt to bid away workers from other employers rather than to recruit from the remaining unemployed. In addition, if management believes that the growth in demand is sufficiently strong they will increase profit margins and increase the inflationary tendency. Finally, legislators may find that the legal minimum wage becomes substandard as inflation occurs, and therefore, in a humanitarian spirit, they may raise the legal minimum. All these factors feed back on each other to create mounting wage-price pressures for as long as the economic expansion is permitted to continue. Since the rate of diminishing returns, the rate of increase in moneywage rates, and changes in profit margins are normally closely related to decreasing unemployment levels, our present anti-inflationary policies are oriented to maintain a sufficiently high unemployment rate to control the impact of changes in these factors on price levels. Any monetary and/or fiscal policy aimed at preventing all price increases before full employment is reached, can be successful only if they perpetuate sufficient unemployment. All expansions in economic activity, whether they are initiated by increasing Government's demands for goods and services or by an increase in demand by the private sector tend to bring about some price increases. It should be obvious, however, that any increase in aggregate demand would induce changes in the supply price of reproducible goods, if there is no change in the money wage rate (relative to productivity) or gross profit margins, only to the extent that diminishing returns are present. Moreover, this diminishing returns associated price rise would be a once-and-for-all rise associated with increasing real costs of expansion due to lower productivity. Installation of new equipment and training programs would help offset any price rise due to this aspect. If, on the other hand, there is an increase in money wages in excess of productivity, whether demand is unchanged or not, the resulting supply price will be higher except if gross profit margins decreased proportionately. Similarly, increases in gross profit margins can induce price increases. Consequently, in the real world of changing levels of aggregate demand (usually at less than full employment) an incomes policy which controls both the money wage and profit margins will provide more stability in the purchasing power of money than a policy which permits "free" collective bargaining and unrestricted pricing practices. Although some economists have attacked such a policy as undesirable because it would not permit markets to optimally allocate resources, I believe that such a criticism is for all practical purposes, irrelevant. First of all, these critics implicitly assume that present 21-570—68 10 138 resource markets are efficient allocators. There is, however, evidence that indicates that existing labor markets are not very good allocators under existing free collective bargaining arrangements [9, ch. 5]. More importantly, resource allocation merely requires changes in relative prices and not in the general price level. Different variants of income policy have been suggested which would permit these relative price changes while restricting a general price increase [7] [9, ch. 6]. Secondly, any possible loss in social welfare due to possible resource misallocation, in our economy, will be small relative to the welfare loss resulting from our continuing failure to maintain full employment and growth. As long as there are several million unemployed who are willing and able to work, I think that an economy that continuously utilizes these resources is less wasteful than a system which requires millions to be perpetually "on the dole" (a system which ultimately must foster social antagonisms) in order to maintain reasonable price stability via monetary and fiscal policies alone. In sum, there is no monetary or fiscal policy which can provide sufficient conditions to insure price stability, without wrecking any chance of sustaining full employment and economic growth. Hence there is an urgent need to develop a viable incomes policy. An incomes policy obviously requires that the public interest be taken into account at the wage bargaining table and when management is making its pricing decisions. This policy must be considered a necessary supplement to monetary and fiscal policies which would guarantee continuous full employment. In return for this guarantee of full employment and optimum production levels, labor would be required to restrict its wage demands to, at most, rises in productivity, while business must hold profit margains constant. The administrative details of implementing such a policy could take a variety of forms. The British, for example, have established restrictions on wage, salaries, and dividend increases. A National Board for Prices and Incomes was established which can require notifications of increases in prices and pay and can legally delay implementation of these increases if the Board finds them unjustifiable and if voluntary compliance to holding the price-pay levels cannot be obtained. In a larger economy, such as ours, we may prefer a somewhat different arrangement that that adopted by the British. In any event, collective bargaining or pricing decisions which do not take the public interest into account should no longer be tolerated. If, in fact, we could go even further and keep both money wages and gross margins constant, then with technological progress, price levels would decline. This would allow all consumers, including renters, to share in the gains of technology. This ideal variant of an income policy (which is less likely to be politically acceptable) would provide the greatest degree of fairness; for as long as some groups in society have their income fixed in money terms, then equity should require that all remuneration be somewhat fixed in money terms. The desirability of instituting a full employment policy in coordination with an incomes policy is clear. The problem is to find a political leader who will advocate these policies which will be, at least initially, unpopular. (Many people might find themselves liking the results of such a policy, once they got over the shock of it.) Who will come forth 139 to demand a simultaneous full employment and an incomes policy \ Is there anyone in our society who will provide the political impetus that will convince most of us to pay this required tariff to sustain full employment ? Obviously no one has yet appeared on the political scene. No one will speak against the status quo and for the LIFO workers (who are usually the young, the uneducated, the migrants, and the members of minority groups, who are often disenfranchised by race, age, education, and residential requirements). Many "liberal" groups are not ready to admit that unions ought to be restrained in the public interest, while "conservatives" do not desire to see managerial pricing decisions limited by the public interest. i v . BALANCE-OF-PAYMENTS DEFICITS I have held the payments problem for last for two reasons: (1) The magnitude of the payments problem for the United States is small in comparison to the previously discussed subjects; and (2) it is my personal belief that the United States should not allow foreigners to control its domestic economic policies; accordingly, methods for dealing with payments deficit should have a relatively lower priority. The traditional monetary policy approach for eliminating a payments deficit is tight money—a policy specifically aimed at (1) stanching net short-term capital outflows, and simultaneously (2) inducing slack demand at home, thus encouraging industries with exportable products to search for new markets abroad, while domestic demand for imports decline. If such a policy is successful, although our balance-ofpayments position will improve, the recessionary effects makes it socially undesirable. An increase in exports relative to imports is the obvious cure for a payments deficit. This can be accomplished without creating unemployment (or even devaluation) via an alteration in the domestic price level relative to the foreign price level. A prominent English economist, Sir Roy Harrod, has recently shown that an incomes policy could not only be used to control the price level at home, but it could be used simultaneously to alter the export price level relative to import prices in order to improve the balance of payments [4]. Hence, it would appear that an incomes policy could be designed to concomitantly prevent inflation and eliminate payments deficits, thus freeing monetary and fiscal policy to concentrate on achieving full employment and growth. Moreover, the utilization of an incomes policy, which allows export prices to alter slowly relative to import prices, would tend to eliminate the need to alter exchange rates and thus reduce the possible capital gains incentive for speculation against so-called "key currencies." V. CONCLUSIONS AND RECOMMENDATIONS Having developed my position at length, I believe I can now succinctly present my major conclusions and recommendations to the committee. (1) A coordinated monetary, fiscal, and incomes policy should be a major objective of economic policymakers. Since fiscal policy and incomes policy are, by their very nature, likely to reside in the execu- 140 tive branch of the Government, it seems practical to give responsibility for coordinating monetary policy with these other policies to the administration. To disperse power over these various policies would be almost to guarantee that economic policies would, at times, be at cross purposes. It is obvious that the brake on an automobile is a check on the accelerator, but no one seriously suggests that one passenger in the auto should work the accelerator and another the brake pedal. By analogy, we cannot afford separate passengers to independently operate monetary, fiscal, and other economic policies. Nevertheless, it is of limited value to only coordinate control of the brake and accelerator pedals, while the steering wheel of money wages and profit margins are left to be driven by an "invisible hand." As long as unbridled wage and price decisions are permitted, disastrous crashes can be avoided only by utilizing the brake pedal almost continuously and/or constraining the accelerator pedal to permit very slow forward movements. (2) The major instrument of monetary policy should be the money supply and its prime target should be to provide sufficient finance to bring the unemployment rate down, say to 3 percent or less. As long as money markets do not automatically "clear," the expected rate of return (adjusted for risk) on new investment projects can be significantly greater than the rate of interest. Consequently, a reduction in the rate of interest may not stimulate additional investment purchases as credit rationing limits the number of entrepreneurs who can obtain finance in order to make operational their demand for capital goods. Furthermore, when there is an unsatisfied fringe of borrowers there is no way of knowing whether those investments projects which are being financed are more productive than those projects which cannot obtain funds. Consequently, control over interest rates rather than over the supply of money may result in misallocating resources in the investment goods industries. The monetary authority must, therefore, exercise its role via primarily the money supply and not rely on interest rate changes alone to do the job. (3) Although monetary, fiscal, and incomes policies should be coordinated, it must be recognized that the first two should be oriented primarily to achieving full employment and growth and should not be concerned with price level problems per se. An incomes policy, on the* other hand, should have primary responsibility for controlling our domestic price level and its relationship to import prices. (4) If rapid economic growth is to be sustained, the money supply must increase in anticipation of the output growTth. In an uncertain world, where expectations are volatile and often unpredictable, the relationship between the required increase in the money supply and theincrease in the economy's wealth is much too complex to be handled by any simple rule. Money clearly matters in the process of economic growth in a monetary economy, but a simple rule can be no substitutefor wise management of the money supply. Accordingly, the money managers cannot fix their gaze to any onestatistical index—although they should always keep global statisticssuch as the unemployment rate and the rate of growth of gross national product in view. Nevertheless, disaggregative statistics on unemployment rates for particular groups and regional gross product 141 growth must also be utilized in suggesting a desirable coordinated fiscal, monetary and incomes policy. Price indexes, for reasons I have already elaborated on, should be of secondary importance for the money managers. (5) Although it would be possible to achieve monetary policy solely via open market operations (as long as the public owned a significant amount of Government bonds), I see little reason for restricting the Fed solely to this tool. If the objectives are clearly recognized, then the Fed ought to be given as much flexibility as possible in choosing the method of achieving these objectives, since no two particular cases will be identical in all respects. (6) Reducing the number of members of the Federal Reserve Board should not necessarily be an objective. What is desired is better educated members who understand the interrelationships of monetary, fiscal, and incomes policies, not fewer members. I do not believe it is essential that members need know the intricacies and mechanics of the banking system any more than members of the Council of Economic Advisers need know the labyrinthine relationships among governmental bureaus. (7) If monetary policy is coordinated with the other economic policies of the administration then I see no merit in having the Fed making separate reports—separate from The Economic Report of the President—to Congress. If monetary policy is left uncoordinated, then a requirement for separate quarterly reports by the Fed not only has little merit, but such a requirement might be detrimental if it opened the Federal Reserve Board to more political pressure to pursue, what I have labeled above, "political trade cycle" policies. (8) Coordination would necessarily involve representatives of the Treasury and CEA at open market committee meetings, and, I would hope these representatives would be participants and not merely interested onlookers. (9) As far as appraisal of the structure of the Federal Reserve is concerned, I believe that it follows from my strong advocacy of coordination that (a) the Chairman of the Board's term be coterminous wTit,h the President of the United States, and ( i ) since the Federal Reserve is an instrument of the public and not of the member banks, there is no necessity to maintain the fiction of private ownership. Accordingly, the Federal Reserve bank stock should be retired. (10) Since a central bank by its very nature as the monetary authority does not need a cushion of "undistributed profits," I see no reason why the Federal Reserve should not pay all its earnings over to the Treasury, while funds to operate the System would be appropriated by normal legislative means. Certainly, if the Chairman of the Federal Reserve had to submit a budget request to the President—as does the Secretary of the Treasury and the Chairman of the CEA—coordination of policy would be facilitated. (11) The term of members of the Federal Reserve Board depends, in part, on what individuals are likely to be appointed as members. If members are to be selected primarily from the banking community and are expected to return to this sector after a single term, then I believe the longer term the better, for a long term frees the members from having their own future economic self-interest affect their decisions. If, on the other hand, one anticipates selecting them from the academic 142 field—such as is now done for CEA members—then a term similar to Cabinet members seems desirable if coordination is going to be efficiently accomplished. In any case the choice of 5 years rather than, say, 4 years, as H.R. 11 provides, strikes me as strangely incongruous with political realities. (12) It follows from my analysis in section III above, that the Federal Reserve's policies of the last 3 years have been socially undesirable. The continued rise in the consumer price level during the past few years is indicative of the failure of monetary policy to contain the inflationary pressures, while the continued high unemployment rate in the ghettos must, at least in part, be associated with these policies. Ultimately, policymakers must recognize that labor and management in our system share responsibility with the monetary and fiscal authorities for the maintenance of price level stability, full employment, and economic growth. An incomes policy is an essential consort to a sound monetary policy. Until this notion is accepted, modern market-oriented systems such as ours will continue to follow erratic paths of economic growth. REFERENCES 1. P. Davidson, "Keynes's Finance Motive," Oxford Economic Papers, March 1965, 17, pp. 47-65. 2. , "The Importance of the Demand for Finance," Oxford Economic Papers, July 1967,19, pp. 245-253. 3. , "Money, Portfolio Balance, Capital Accumulation, and Economic Growth," Econometrica, April 1968,86, pp. 291-321. 4. R. F. Harrod, Reforming the World's Money (London 1965). 5. J. M. Keynes, "Alternative Theories of the Rate of Interest," Economic Journal, December 1937, k7, PP. 663-9. 6. , "The Ex-Ante Theory of the Rate of Interest," Economic Journal, December 1937, 47, pp. 663-9. 7. A. P. Lerner, "Employment Theory and Employment Policy," American Economic Review Papers and Proceedings, May 1967,57, pp. 1-18. 8. A. Marshall, Principles of Economics, 8th edition (New York, 1950). 9. S. Weintraub, Some Aspects of Wage Theory and Policy (Philadelphia, 1963). STATEMENT OP WILLIAM G. DEWALD, OHIO STATE UNIVERSITY R E P L Y TO QUESTIONNAIRE ON H . R . 11 i. 1. Yes. An economic policy program would be useful. Monetary and fiscal policies are not independent in their effects, at least not in the short run. Thus, planning and coordination are essential to avoid policy actions that have the wrong overall effects in magnitude or even direction. It should be noted that coordination offers no assurance that appropriate policy actions would be taken. 2. The President should be responsible for the national economic policy program. He has the broadest responsibility, though the ultimate power for national policy is shared with Congress. Agencies reporting to the President carry out the administration's spending and taxing policies. The Treasury and the Federal Reserve share authority for monetary and debt management policies. All responsible authorities should be consulted in shaping the Nation's policy program, but the President should coordinate it. Independent authorities such as the 143 Federal Reserve should be encouraged to exercise a right of public dissent and even independent action as an expression of Congress check on the President's power. But Congress should require an explanation of Federal Reserve actions, where possible, before they are taken. 3. It is a difficult problem to find proper guidelines and measures of the stance of monetary policy. H.R. 11 specifies the money supply as the appropriate guide. The money supply is not only affected by policy actions, but also affects basic objectives. Interest rates are an alternative guide. A problem is that changes in the money supply or interest rates are brought about not only by policy actions, but by nonpolicy factors. Whether an increase in the money supply or a decrease in interest rates is a reflection of an expansionary policy depends on the nature of the economy and what action, if any, was taken. The proper indicator of the expected effect of policy on goals would be a dated sum of predicted effects of each policy instrument. Predicted effects may be arrived at by simple extrapolation or by complex statistical techniques—but whatever the approach, it should be subjected to critical evaluation. Which particular target policymakers use is less important than explicit accounting for the effects of actions on the indicator and on ultimate goals. Interest rates, free reserves, and other largely equivalent measures of money market pressures are miselading indicators where policymakers ignore nonpolicy effects on them and where there is a shift in the relationship between such variables and goals. An example is the decrease in interest rates induced by a decline in the demand for credit to finance spending at the advent of recession. Contractive policy actions have often been taken that prevented interest rates from falling as far or fast as they would otherwise because policymakers have misinterpreted the source of easing in interest rates and other measures of money market pressures. Comparable criticisms can be made of the money supply and related magnitudes as indicators of policy. For example, a financial crisis would have the effect of increasing the demand for money and raising interest rates. This in itself would tend to induce an increase in the supply of deposits and money even if there were no policy actions. In fact, the quantity of money might increase despite contractionary policy actions, but less than would have been the case if there had been no policy actions. There are serious shortcomings of any intermediate variable as an indicator of the stance of policy where the structure of the economy is not explicitly taken into account. In fact, it is necessary to do this just to define meaningful quantities. I am inclined to measure the overall stance of the major instruments of monetary policy by what can be called maximum money—the amount of Government-issued money 1 divided by the average required reserve ratio for commercial banks. Such a magnitude is almost altogether subject to control by the monetary authorities. It is a constraint which importantly limits the expansion of bank deposits and bank credit. Maximum money can be given the interpretation of a policy-controlled constraint that affects the supply function of money and is largely independent of demand factors. Admittedly, maximum money or any alternative indicator of policy is at best an approxima1 A close relative of Milton Friedman's high "powered money" and Karl Brunner and Allan Meltzer's "monetary base." 144 tion to an ideal—the effect of each policy instrument on economic welfare. Policymakers have often not only misjudged the magnitude but also the direction of the effect of their actions. Increasing maximum money would unambiguously be expansionary; reducing it, contractionary. Hence, directional errors, at least in terms of immediate effects, would be avoided by reference to maximum money as a monetary policy guide. If there were no other monetary policy actions and the amount of maximum money were increased, then the policy stance in terms of its ultimate effects would unambiguously be expansionary even if noncontrolled factors should induce a contraction in the actual quantity of money or increase in market rates of interest. If, as I believe, there are significant effects of monetary policy actions in the short run, then the appropriate policy would be to increase maximum money at less than its long period average growth rate during periods of expected inflation and at more than average during periods of expected deflation and unemployment. However, if lags in effects are long and variable, then the proper policy would be to increase maximum money steadily, unless economic performance deviated a great deal from objectives. Such a prescription for policy is made without regard to other policies than open-market operations and required reserve ratios. If there were other actions their effects would also have to be appraised. It is reasonable that no other actions should be taken or perhaps all in the same direction, at least until it is possible to predict the magnitude of their effects with considerable accuracy. 4. Monetary policy involves changes in government demand obligations. Debt management includes policies that affect the supply of all government debt. It is reasonable to expect that large changes in the relative amounts of short- to long-term government debt would have important effects on the economy, more important indeed than small changes in government demand obligations. There are economists who would deny this proposition because in their view only government demand obligations are capable of affecting the economy. At the other extreme are others who argue that it doesn't make any difference whether government debt is issued in the form of long-term bonds or demand obligations. In my view over sufficient time the economy will adjust to whatever supplies of various maturities of government debt are outstanding by the substitution of private securities. But debt management does have shortrun stabilization potential. Proper utilization of debt management is to lengthen the average maturity of the government debt including reduction in demand debt during periods of excessive spending. In the opposite circumstances it is appropriate to shorten the average maturity of government debt and to increase demand debt in order to stimulate spending. The historical record has often shown perverse debt management policies from this point of view. During the 1930's there was substantial maturity lengthening in government debt. Government support policy during the inflation of the Second World War and its aftermath had the effect of shortening the maturity of debt, making debts of all maturities essentially short-term claims on the government. In the period since the Federal Reserve-Treasury Accord in 1951, the average debt maturity has tended to lengthen during periods of economic contraction and to shorten during periods of expansion. 145 The demand for government securities of long term naturally rises during periods of contraction as investors find weakening alternatives to government debt. It is precisely at such times that the Treasury should press short-term securities and government demand debt on the market to force investors to turn to issues of private securities. 5. (a) It is reasonable that the Federal Reserve conduct open market operations to prevent shocks to thefinancialsystem that are introduced by the government itself—for example the effect of Treasury tax collections and spending. But seasonal variability in interest rates and money market conditions in general may also reflect seasonal cost factors in the economy that should be permitted to direct resources toward that period of time when they may be employed most efficiently. This is generally recognized with respect to cyclical variation. But it is not at all clear that open market operations to prevent seasonal variability in interest rates contribute to economic welfare. The burden of proof should be on the Federal Reserve to demonstrate how deseasonalizing open market operations serve a useful purpose. ( i ) Monetary policy could be effectively implemented solely by open market operations without causing windfall gains and losses to banks as is a necessary result when there are changes in reserve requirements. (c) Changes in other instruments of monetary policy should be introduced wherever there is a strong case that they can add to the efficiency of the financial system. Changes in reserve requirements are appropriate to affect bank profits and the competitive position of commercial banks relative to competing financial institutions. Changes for other purposes should be eschewed. The availability of discounting from the Federal Reserve is not necessarily as essential to an efficiently operating system. Nevertheless, this central banking service is typically available not only in the United States but elsewhere. It may provide a low cost way by which the financial system can adjust to shocks. But adjustments will occur whether discounting privilege is available or not. The unanswered question is whether the adjustment by discounting would be cheaper than alternatives. Given that there is a discounting arrangement, changes in the discount rate can have an effect on the economy, though probably a small one. It is reasonable that the discount rate should more or less parallel changes in market conditions. The rate might be automatically changed by fixing it at a certain interest differential above short term government securities rates or possibly the Federal funds rate. (d) I think it is very appropriate that the Federal Reserve should report to the Congress quarterly about prospective actions and policies and their likely effects. There are risks in this procedure—risks to the Federal Reserve that its limited ability to predict and to explain why it does what it does will become a matter of public record. As indicated above, I would modify the provisions of H.R. 11 to require the Federal Reserve to explain the likely effects of proposed regulations. Reports should include information pertinent to the explanation of the effects of the actions of the monetary authority. The Federal Reserve should indicate clearly the nature of the regulation, what the purpose is of the change in the regulation, what will be the likely effects on banks and on others. This year the Federal Reserve is to introduce major changes in the definition of legal reserves and reserve requirements. 146 In some ways these changes establish an arrangement comparable to that that exists in Canada—an arrangement in some ways designed to maximize the destabilizing influence of the Bank of Canada on the financial system.2 There is no public record of w^hat the Federal Reserve thinks it will accomplish by the new regulations. It has requested major banks to supply opinions of likely effects on their operations. But what of the public interest ? Congress created the Federal Reserve. But it didn't create another legislative authority to introduce major structural changes in banking. Major changes in the structure of banking deserve a careful hearing before Congress before they are introduced. Accordingly I would suggest that the H.R. 11 be amended to require reports from the Federal Reserve on prosipective changes in banking regulations and their likely effects. (e) Should the meetings of the Open Market Committee be opened to representatives of Congress, Treasury, and the Council of Economic Advisors? I do not think so. The Federal Reserve is an agency of Government charged with particular responsibilities. It is perfectly reasonable for Congress to demand that the Federal Reserve explain precisely why it does what it does and what it proposes to do, but I can see no particular value in having outside observers at Open Market Committee meetings. Presumably committees permit individuals to take strong positions in argument and to get educated. Presence of outside observers might kill an aspect of the deliberative process. I think it appropriate that the minutes of the Open Market Committee be made public but that it deliberate in privacy. II. APPRAISAL OF THE STRUCTURE OF THE FEDERAL RESERVE 1. A certificate of membership is a satisfactory substitute for capital stock in the Federal Reserve banks. It is appropriate to eliminate Federal Reserve bank stock to demonstrate the public nature of the Federal Reserve. I would support this provision. 2. Reducing the number of members of the Federal Reserve Board to five from seven would not have any particular cost. The quality of people selected for the Board would be about the same. The committee would still be large enough that it need not be dominated by a chairman. Hence, why have seven people when five could do the job? Cutting terms of office to no longer than 5 years might lead to an undesirable turnover of the Board but probably not. The opportunity to reappoint Board members would allow for development of very able central bankers if they could be recognized. That is the key problem, not the number of members of the Board. I would support this proposal. 3. Making the term of the Chairman of the Board of Governors coterminous with the President's term is a reasonable proposal and I support it. It would permit coordination of overall economic policy. 4. The proposal for outside audits of Federal Reserve is also reasonable. Congress does not appropriate funds to operate the Federal Reserve under the present arrangements. But Congress has the ultimate responsibility for Federal Reserve operations and should have budgetary control. 2 H. G. Johnson and J. W. Wilder, "Lags in the Effects of Monetary Policy in Canada," Royal Commission on Banking and Finance. November 1962, p. 141. 147 5. I support the proposal that Congress appropriate funds for Federal Reserve operations. This would put some teeth in congressional controls on the Federal Reserve and reduce the independence of the Federal Reserve to introduce changes in policy and regulations without congressional sanction. III. RECENT MONETARY POLICY Monetary policy has been unduly expansionary in the last year and a half. If the economy slows in the future and monetary policy actions follow the pattern of earlier downturns, the stance of policy can be expected to become unduly contractionary. Monetary policy is inherently a highly flexible instrument of government stabilization policies, but it has often been and continues to be misdirected by overlooking its own actions and to be myopic by not looking far enough at the effects of its actions. Recent changes in the structure of reserve requirements and borrowing from the Federal Reserve are massive in their likely effects on the economy. Congress should demand an explanation, albeit after the new regulations have been imposed, and require that no future changes be made without legislative approval. STATEMENT OF JAMES S. EARLEY, UNIVERSITY OF CALIFORNIA, RIVERSIDE, CALIF. My statement follows the series of questions that accompanied the request for a statement of viewrs. I. MONETARY POLICY GUIDELINES AND OPEN MARKET OPERATIONS 1. I emphatically believe that monetary andfiscalpolicies should not be independent or mutually exclusive stabilization policies. Coordination is vital, and in my judgment not sufficiently provided at present. My views as to how this could best be secured are set forth later. A coordinated program set forth at the beginning of each year, as proposed in the bill, might be useful, but there would need to be flexibility to take care of uncertain and unanticipated developments. My later remarks will clarify this position also. 2. I believe that the President should be responsible for formulating and coordinating programs in the monetary and fiscal fields, with consultative arrangements with the Federal Reserve System. 3. Monetary policy guidelines: A. I emphatically believe that the goals of the Employment Act should not be sought via primary dependence on the regulation of the money supply. In my judgment the money supply, however defined, is a false and dangerous guideline. I agree with Governor Mitchell and the staff of the Federal Reserve Board, and with most other experts in this field, that no one variable is sufficient guidance and that excessive concentration on any single variable will be seriously misleading. Although knowledge and techniques of monetary and credit management need to be improved, I feel that the various criteria used by the staff of the Board of Governors are intelligent ones 148 in the light of our present knowledge. If, however, a specific group of variables were to be given more attention, I believe they should be data on the flow of funds in the various credit markets, together with the fundamental data of employment conditions and price behavior. B. I feel that the fundamental objective of monetary policy should be cast in terms of the maintenance of high employment and reasonable price stability. If one could assume that the various policies of Government—monetary, fiscal, and others—were appropriately coordinated, then I would favor specifying some maximum level of unemployment or general level of utilization of economic resources as a "target variable." Again, however, I think that following any single "target variable" or growth rate thereof, regardless of the economic winds, would be a mistake. C. If indexes of economic activity were to be used to guide coordinated monetary and fiscal policy, I would favor using leading indicators rather than coincident or lagging ones. Although I distrust formulas, I feel that the agencies doing economic forecasting and charged with responsibility for formulating policy should pay very close attention to the "leading indicators" that have beenidentined by the National Bureau of Economic Research. 4. Debt management policy: I believe that debt management can have a supportive although not a large role in reaching the goals of the Employment Act. It is usually appropriate to issue short-term Government securities when stimulus to financial markets is desired, and some benefit may come from concentrating borrowing in the longer maturities in times of economic overheating. But there are disadvantages in pushing either of these policies too far, and I believe that the main dependence should be on monetary and fiscal policy rather than debt management. 5. I approve of the requirement in H.R. 11 that the Federal Reserve System should conduct open market transactions in accordance with the programs and policies of the President pursuant to the Employment Act, but I disapprove of the requirement that the Federal Reserve Board must submit quarterly reports to the Congress, "stating, in comprehensive detail, its past and prospective actions and policies * * *." I do feel that the Board should be more responsive than heretofore to the goals of the Employment Act and the President's economic program, and periodic reports might be useful but I do not feel the Board can reasonably be asked to be this specific, prompt, and anticipatory in its reports. Monetary and credit regulation requires a certain degree of uncertainty with respect to future actions m an uncertain and changing world, and detailed reports on prospective actions would be harmful, in my judgment. A. I believe that the Federal Reserve System should continue to have freedom to use open-market operations for "defensive" and "road clearing" purposes. Such operations normally facilitate rather than interfere with the achievement of the fundamental goals of the Employment Act. B. I emphatically do not believe that monetary policy can be effectively and efficiently implemented solely by open-market operations. For many reasons the rediscounting power is important to sound credit management under our banking system. I agree with the recently is- 149 sued proposals for recasting the rediscount mechanism, as set forth by a system committee of the Board of Governors. This program would actually increase the role of rediscounting and of the rediscount rate in our monetary regulation. I believe this would be wise. C. (a) Rediscounting.—My views on this are set forth above. (b) Changes in reserve requireinents.—These may be useful in some circumstances, although I believe that rediscounting and open market operations are normally the more effective set of instruments. (c) Regulation Q.—Under present conditions the Reserve Board's power over maximum deposit interest rates is desirable. The flow of funds between banks and other depositary institutions is an important determinant of the influence that money and credit exert on economic activity. If properly used, the regulation of the deposit rates being paid by financial institutions may help achieve monetary objectives. The present system by which this regulation is divided among the Federal Reserve Board, the FDIC and the Federal Home Loan Bank Board is, however, an awkward and potentially dangerous one, and I think firmer coordination should be secured in this matter. In the absence of effective coordination, it might be better to let deposit interest rates respond more freely to market forces, so long as there is effective control of other monetary and credit conditions. D. My views on Federal Reserve Board reports to Congress have been stated above. I see some advantage in periodic reports so long as they do not require great detail or specify precise future actions. E. I would favor representatives of Congress, the Treasury and the CEA being observers in Federal Reserve Open Market Committee meetings. In fact, as explained later, I feel the Open Market Committee itself should be reconstituted. II. STRUCTURE OF T H E FEDERAL RESERVE SYSTEM If, as is stated, "the grand aim of H.R. 11 * * * is to provide for coordination by the President of monetary and fiscal policies," I feel that some but not all of the listed structural changes in the Federal Reserve System are advisable: 1. I have no objection to the Federal Reserve banks being changed from bank-owned to Government-owned institutions, as provided in the bill, but I do not see that this change is vital to reach the stated objective of H.R. 11. 2. I do not favor reducing the number of members of the Federal Reserve Board. There is a great deal of work to be done by the Board, and the present number of members is not excessive. In view of the desirability of experience and continuity, I also question whether reducing the term of office to 5 years is advisable. On the other hand, the present term of 14 years is probably unduly long. Perhaps 7 years w^ould be a good compromise. This would permit appointment of a new member at least every year. 3. I favor making the term of the Chairman of the Board coterminous with that of the President. This does not imply that there should be a new Chairman each time there is a new President, but it would help make clear the ultimate responsibility of the President for the functioning of the Federal Reserve System. 150 4. I see no important purpose that would be served by an annual audit of the Board and the Federal Reserve banks by the Comptroller General of the United States. I believe that the internal auditing procedures of the System are adequate and on the whole preferable. 5. I would not favor making the expenditures of the Federal Reserve System subject to congressional appropriation. I think congressional appropriation would make transitory political pressures greater than they should be. Monetary policy and central bank operations are extremely complex, and must be carried out professionally. Although the ultimate responsibility of the Federal Reserve to Congress and the Nation should be made clear, the System should not be subjected to great political heat. If the responsibility and authority of the administration over Federal Reserve policies were made clear this, along with congressional power to amend the Federal Reserve Act, would be sufficient political influence, in my judgment. I I I . COMMENTS ON RECENT MONETARY POLICY Much of the criticism of the Federal Reserve System made by economists has been of its purported failure to take sufficiently vigorous action to combat the periods of underemployment and sluggish growth during the later 1950's and the very early years of the 1960's. More recently there has been criticism that the System has not vigorously combatted the "creeping inflation" of 1965-68. It is in connection with these criticisms that some economists have criticized the System for not paying sufficient attention to changes in the quantity of money. I share some of the criticism of the System with respect to the earlier periods of underemployment. Partly because of international difficulties, but also, I believe, because of a real bias of the Board membership of that time toward "avoiding inflation at all costs", the System did not in my view carry on a sufficiently vigorous expansionary monetary policy during several of the postwar recessions. The fact that during some (but not all) of those periods the stock of money, narrowly defined, actually shrank slightly was not, however, the main cause of the trouble. Other more sensible criteria would lead to the same conclusion. Within the last 2 or 3 years the System has been criticized for letting the money stock expand too rapidly during several intervals, including parts of 1967 and 1968. More significant of the failure to curb undue expansion, however, was the fact that bank business loans expanded even more rapidly than the money stock in most of these periods. The tendency toward rising prices and other signs of economic overheating were other more reliable warnings of the difficulties besetting the System than the behavior of the money stock. But we should not make the Federal Reserve, much less the money stock, a whipping boy. Within the last year, for example, the System has again been criticized because prices have risen seriously while the money stock also expanded considerably. Those who argue that the Federal Reserve should have taken a more contractionary policy than it did, are saying in effect that interest rates should have been permitted to go even higher, and credit become even tighter, than they have been in recent months. The critics also overlook the fact that the main in- 151 fiationary causes of inflation in 1967-68 have been large Government expenditures and deficits. Federal Reserve authorities, along with others, called repeatedly for greater fiscal restraint. It is a real question whether the System, unsupported by fiscal policy, should have been expected to go much further than it did. The main lesson of late 1965 and of 1967-68 is that improved coordination of monetary and fiscal policy is needed. It would be a tragedy if the attention of the administration, Congress, the Federal Reserve officials, and the general public was diverted from this lesson by the simple-minded notion that some mechanistic control over the quantity of money would be a remedy for this basic need. Other provisions of the bill I agree with the objective of permitting all FDIC-insured banks to be members of the Federal Reserve System. In fact, I would favor making this membership compulsory. Other comments 1. I believe that the Open Market Committee of the System should be reconstituted to include the Secretary of the Treasury and the Chairman of the Council of Economic Advisers or their designates. I also would recommend that the number of presidents of the Federal Reserve banks on the committee should be reduced to three. I see no objection and some advantage, however, in having the presidents of all 12 Reserve banks or their designates be present at FOMC meetings. 2. I think that serious consideration should be given to setting up a new body having coordinating responsibilities and power in the monetary, credit and fiscal fields. Suitable membership for such a body might be the Secretary of the Treasury, the Chairman of the Council of Economic Advisers, the Chairman of the Federal Reserve Board, and possibly the chairman of the congressional Joint Committee on the Economic Report. STATEMENT OF OTTO ECKSTEIN, HARVAKB UNIVERSITY 1. Fiscal and monetary policies, including debt management, must be considered simultaneously and planned in consistent fashion. The two policies can defeat each other if they pursue opposite objectives, and excessive reliance on one instrument or the other can produce instability in financial markets. 2. The preparation of the monetary component of the fiscal-monetary policy plan for each year must reflect the actual distribution of responsibility. So long as the statutes give considerable discretionary independence to the Federal Reserve System, that agency must prepare the program. It probably wTould be useful to have a mutual review of the drafts of the respective reports, although final responsibility must rest with the agencies that issue the reports. 3.A. The challenge posed to the traditional approach to monetary policy by the Chicago school of monetary theory has resulted in a useful dialog which comes at an opportune time. Over the last 20 years, economics has moved in the direction of more precise quantitative analysis. The financial aspects of the economic system have also become increasingly subject to quantification, although this development is more recent. As a result, the times are almost ripe for a quantitative 152 approach to monetary policy. Fiscal policy today rests on analyses of the major macroeconomic magnitudes, such as the gross national product, the unemployment rate, the price level, and the balance-of-payments deficit. Increasingly, econometric models are used to work out first approximations of the impact of alternative policies on the performance of the economic system. Monetary policy is on the verge of a similar development. I do not believe that the scientific tasks are complete, however. There is no one target variable by wThich monetary policy can be measured. On the one hand, there are sufficient technical "bugs" in each of the measures that have been advanced to preclude their use. For example, the "money supply", whether broadly or narrowly defined, is subject in the short run, to large swings caused by irrelevant factors, such as changes in Treasury deposits, tax collection dates, etc. Some of these factors are partially corrected through the seasonal adjustment procedures; but such correction cannot be perfect because these technical elements do not follow a rigid seasonal pattern. For example, the rate of increase of the money supply has been subject to data revision a year or two after the event because certain seasonal factors were overlooked in the initial estimates. To wholly tie the management of monetary conditions of the American economy to such slender statistical reeds strikes me as farfetched. On the other hand, a thorough studv would probably show that the optimal measure of monetary policy is not one single number, but a pattern of numbers reflecting the several dimensions such as money supply, the state of liquidity of the various components of the financial system, the total lending capacity of the banking system, and the level and structure of interest rates. The impact of the economy on the financial variables must also be identified. Just as the impact of fiscal policy could not be identified properly until we had estimates of the full employment surplus or deficit of the budget, we cannot measure monetary policy without correction for underemployment or overutilization. 3.B. I do not believe that we are ready to define precise guidelines for monetary policy in terms of any index of economic activity or of the monetary target variables. If the range of a guideline is set very, very broadly, and if the guideline is not mandatory but simply demanding of explanation if violated, then its use could be adopted more readily. In my judgment, if guidelines are adopted, they must be related to the performance of the economy, not to any index of money or credit statistics. It seems to me inevitable that monetary policy must be based on forecasts of the economy, particularly forecasts of the likely performance of the economy in relation to the major objectives of the society. Of course, rational policy does not rely more on forecasts than necessary, uses as brief a forecast as possible, and preserves as much flexibility as possible to respond to changes in actual conditions. But to tie policy to monetary targets puts the problem on the wrong track. 3.C. I believe that the National Bureau of Economic Research classification of leading, lagging, or coincident indicators has been superseded by later econometric work. In its day it was a useful approach to business cycle analysis. Today any reasonable econometric model, and there are several, incorporates these notions into particular 153 equations, uses the leading indicator evidence systematically and in rational, quantitative fashion. Econometric models, leaving to judgment those matters which inevitably must remain so ( such as the outlook on defense spending, the likelihood of tax changes, etc.) are the best approach to forecasting and to the planning of policy. To be sure, the models must be used with sophistication; changes in economic structure, the presence of errors, and limitations of the data must be recognized, and heavy weight must be given to the actual evidence of coincident developments. The monetary target variables must be defined in terms that can be related to the economic model. If the money supply for, example, is used as the target variable, then the econometric model must use the money supply as a key variable reflecting monetary influences. Indeed, the suitability of a target variable must in part be determined by its ability to make itself felt in econometric models. Vague, general, longrun associations are not sufficient to choose a particular target variable over others because the empirical associations are consistent with many alternative economic relationships. 4. Debt management is part of the monetary policy of government. Because interest cost is a genuine cost of government—despite its quaint classification as a transfer payment in the national income accounts— our Government cannot be insensitive to interest cost. Thus, it would not be appropriate to make the stabilization objective the sole objective of debt management. On the other hand, the Treasury does have a special obligation to modulate its actions so that they are broadly consistent with general economic policy; its responsibility is considerably greater than that of even the largest private borrowers. 5A. The management of seasonal credit flowTs was certainly a primary goal of monetary policy at the time that the Federal Reserve System wTas established. We have no subsequent experience with a monetary system that does not contain a central bank managing money to offset seasonal swings. We do not know how successfully the credit system would adapt to the removal of this form of management, No doubt, alternative private arrangements would develop. We would not have an annual money panic at the time of the seasonal surge in bank loans tofinanceChristmas retail business. But I am not familiar with any attempt to spell out what these private institutions would be like. Nor have I seen any studies measuring the social cost of public seasonal credit management. Until there is evidence that there are major social costs and that the private alternatives have been thought through, I see no reason to engage in this economic experiment of considerable risk. 5. B, C, D, and E : I have not studied these questions sufficiently to reach my own conclusions. II. I have no firm views on the proper structure of the Federal Reserve System. It does seem to me that the term of numbers of the Federal Reserve Board is too long, and that the term of the Chairman should be coterminous with that of the President of the United States. I also have questions about the composition of the Federal Open Market Committee and of the role of regional bank presidents. The regional bank presidents are not selected mainly for their periodic responsibilities for national economic policy. It is not clear to me why the Federal Reserve Board itself is not also the Open Market Committee; no doubt there are historical or practical reasons. The argument of 21-570—68 11 154 regional presentation does not seem to me to carry as much weight today as in an earlier economy. I do not wish to submit a detailed review of monetary policy since 1964. On the whole, policy has promoted the general economic objectives of full employment, price stability, balance-of-payments equilibrium, and economic growth. One can quarrel with details of timing and the extent of some immense credit for managing a flexible monetary policy during a costly war financed mainly by borrowing. STATEMENT OF DAVID I. FAND, WAYNE STATE UNIVERSITY I . QUESTIONS ON MONETARY POLICY GUIDELINES AND OPEN M A R K E T OPERATIONS 1. Do you believe that a^program coordinating fiscal, debt management, and monetary policies should be set forth at the beginning of ecwh year for the purpose of achieving the goals of the Employment Act, or, alternatively, should we treat monetary arid fiscal policies as independent mutually exclusive stabilization policies ? 2. If you believe a program should be specified, do you believe that the President should be responsible for drawing up this program,, or, alternatively, should such responsibility be dispersed between the Federal Reserve System and agencies responsible to the President? (Please note that informal consulting arrangements can be made as desired whether responsibility is assigned to the President or divided between the President and the Federal Reserve. The concern here is with the assignment of formal responsibility for drawing up the economic program.) Answer 1.1.2. In my opinion, coordinating of fiscal debt management and monetary policies at the beginning of each year would be desirable. It seems to me that too much emphasis has been placed recently on the stabilization potential of shortrun changes in fiscal policy. It is far from clear that the evidence does in fact support the stabilization properties that are attributed to fiscal policy actions. I would also suggest that monetary policy should be used to stabilize aggregate demand, and not to bring about abrupt and substantial changes in policy, as in 1966. 1.3. Concerning monetary policy guidelines: A. Should monetary policy be used to try to achieve the goals of the Employment Act via intervention of money supply (defined as desired) as provided in H.R. 11, or, alternatively, should II.R. 11 be amended to make some other variable or variables the immediate target of monetary policy; for example, interest rates, bank credit, liquidity, high-powered or base money, total bank reserves, excess reserves, ana free reserves? Please define the target variable or combination of variables recommended and state the reasons for your choice. (If desired, recommend a target variable or variables not listed here.) It ivould be most helpful if, in providing the reasons for yowr choice, you list the actions the Federal Reserve should take to control the target variable (or variables) and also explain the link betyo.een your recommended target of monetary policy and the goals of the economy as defined by the Employment Act. 155 B. Should the guidelines of monetary policy be specified in terms of some index of past, present, or future economic activity, or, alternatively, m fezm? of the target variable's or growth ? For example, should the President's 1969 program for achieving the goals of the Employment Act be formulated to require consistency with some set of overall indicators of economic activity, alternatively, so that your target variable attains a certain value or growth regardless of the economic winds? Please indicate the reasons for your preference. £7. For only those persons who recommend that some index of economic activity be used to guide the monetary authorities in controlling the target variable: Should we use a leading (forward looking), lagging (backivard looking), or coincident indicator of economic activity? It would be most helpful, also, if you would identify the index you would like to see used and specify how the target variable should be related to this index. I). For only those persons who recommend that the guidelines be put in terms of the target variable's value of growth: Should the same guidelines be used each year into the foreseeable future, or alternatively, should new guidelines be issued at the beginning of each year conditioned on expected private investment, Government spending, taxes, and so forth? Please indicate the reasons for your preference. E. For only those persons who recommend that the guidelines be put in terms of the target variable's value or growth and icho also recommend that the mme guidelines be used year after year into the foreseeable future: What band of values or range of growth do you recommend? (By way of clarification, a band of vahoes appears appropriate if your target variable is, say, free reserves, whereas a range of growth is appropriate if it is, say, money supply.) F. For all those persons recommending that the guidelines be put in terms of the target variable's value or growth (regardless of whether you recommend using the same guidelines year after year or revising them each year in light of expected private investment and fiscal policy) : Under what circumstances, if any, should the monetary authorities be permitted during the year to adjust the target variable so that it exceeds or falls short of the band of values or range of growth defined by the guidelines issued at the beginning of the year? Answer 1.3. A-F. It is very hard to legislate guidelines that could be followed by the monetary authorities in all circumstances. In general, if a rule or guideline is developed it should be in terms of the money stock or of changes in the money stock, as these are among the most important variables that the Reserve authorities can directly influence. It is also desirable to develop a stabilization program that does not require too many short-run changes, as such changes may, at times, become an independent source of instability The recent suggestion of the Joint Economic Committee that the Federal Reserve should try to keep variations in the money growth rate in a 2- to 6-percent range each year has much to recommend it. As we develop more experience with this approach it may be possible to develop a better guide. 156 74- Concerning debt management policy: Given the goals of the Employment Act, what can debt management do to help their implementation? {If you believe that debt management has no role to play in this matter, please explain why.) Answer. 1.4. In my opinion the stabilization potential of debt management has been overrated. Debt management properly defined involves essentially a swapping operation, and its overall effects on aggregate demand are a mixture of several effects. There is therefore some question as to whether countercyclical debt management is worth the cost. I would not, however, go to the other extreme and argue that debt management should be primarily concerned with minimizing the interest outlay on the debt. Probably the most sensible policy would be to try to get a stable debt structure, which would indirectly contribute to the stabilizing role of monetary policy. 1.5. Concerning open market operations: H.R. 11 requires that the FOMC conduct open market transactions "in accordance with the programs and policies of the President pursuant to the Employment Act of WlfG^ And in this connection, H.R. 11 provides that "The Federal Reserve Board shall submit a quarterly report to the Congress, stating in comprehensive detail, its past and prospective actions and policies under this section and otherwise with respect to monetary affairs, and indicating specifically how such actions and policies facilitate the economic program of the President A. H.R. 11 makes no provision whatever for conducting open market operations for so-called "defensive" or "road-clearing" purposes, that is to counteract seasonal and other transient factors affecting money market and credit conditions. Do you see any merit in using open market operations for defensive purposes or should they be used only to facilitate achievement of the President's economic program and the goals of the Employment Act? What risks and costs, if any, must be faced and paid if open market transactions are used to counteract transient influences? Answer I.5.A. The case for using open market operations for defensive purposes is not entirely clear to me. It is true that seasonal and other factors could prove to be disruptive in the financial markets. But defensive measures may also interfere with market processes and make it difficult to recognize more basic f orces. At the present time we may be oversupplied with techniques and weapons for such defensive operations, and it may be desirable to examine these issues to determine whether some of these operations can be dispensed with. I.5.B. Do you believe that monetary policy can be effectively and efficiently implemented solely by open market operations? C. For what purposes, if any, should {a) rediscounting, {b) changes in reserve requirements, and {c) regulation Q be used? How might H.R. 11 be amended to implement your recommendations? Answer 1.5. B. and C. Monetary policy could be effectively implemented solely by open market operations. It is not, however, clear that this would necessarily be the most efficient way to operate the central bank. The other measure mentioned in (c) may, therefore, at times provide some extra flexibility for the authorities. But here again it appears that we may have too many weapons. Why do we need both a discount window, and a provision for borrowing at the penalty rate. 157 1.5.D. Do you see any merit in requiring the Federal Reserve Board to make detailed quarterly reports to the Congress on past and prospective actions and policies? Are there any risks amd costs in this procedure? In what ways, if any, would you modify the reporting provision? What information do you believe should be included in such reports as you recommend the Federal Reserve submit to the Congress? Answer 1.5 .D. I think it might be useful for the Federal Reserve Board to report to Congress on their policy actions. I am not clear on whether one could expect them to disclose prospective actions and policies unless we have a definite rule, and we eliminate all discretion. 1.5.E. What costs and benefits would accrue if representatives of the Congress, the Treasury, and the SEA were observers at Open Market Committee meetings? Answer I.5.E. The benefits of having more officials at Open Market Committee meetings is that it reduces the likelihood of a serious error. At the same time, it also makes it more difficult to arrive at a decision. On balance, a smaller group may be more desirable. II. APPRAISAL OF T H E STRUCTURE OF TIIE FEDERAL RESERVE H.R. 11 provides for the following structural changes in the Federal Reserve System: 1. Retiring Federal Reserve bank stock; 2. Reducing the number of members of the Federal Reserve Board to five and their terms of office to no longer than 5 years; 3. Making the term of the Chairman of the Board coterminous with that of the President of the United States ; It. An audit for each fiscal year of the Federal Reserve Board and the Federal Reserve banks and their branches by the Comptroller General of the United States; and 5. Funds to operate the Federal Reserve System to be appropriated by the Congress of the United States. Please comment freely on these several provisions. In particular, it would be most helpful if you would indicate any risks iwoolved in adopting these provisions and discuss whether their adoption would facilitate the grand aim of H.R. 11, which is to provide for coordination by the President of monetary and fiscal policies. Answer II. 1-5. I think that these provisions emphasize the fact that the central bank is not a private institution but a Government agency. They are, in my opinion, desirable measures. III. COMMENTS ON RECENT MONETARY POLICY Your analysis of monetary developments, since 1964, including policy induced changes and their effects on economic activity, is invited. Comments III. The implementation of monetary policy since 1965 has been defective, even though Federal Reserve authorities diagnosed the situation correctly arid took extraordinary measures to correct the overall posture, when they discovered their errors. The authorities misinterpreted interest rate movements in early 1966, and therefore could not clearly assess the impact of their actions. They apparently failed to distinguish between movements in nominal and real rates, and between nominal balances and real balances, and like many others 158 were oversold on the stabilization potential of short-run fiscal changes. Their efforts to restrain later on in 1966 was therefore both abrupt and severe, and generated a small crisis in the money markets—the so-called credit crunch. Subsequent to the crunch they again overreacted to prevent the so-called minirecession of early 1967. In retrospect, it is evident that the Reserve officials overestimated the impact of an increase in the (nominal) discount rate, and underestimated the fact of extraordinary changes in the growth rate of the money stock. The failure to recognize the diverging paths of nominal and real interest rates when prices are rising (or falling) has caused errors in policy in the past. Hopefully, this lesson of the 1965-68 period will not be forgotten soon. STATEMENT OF WILLIAM FELLNER, YALE UNIVERSITY (1) When declaring that the attainment of "maximum employment, production and purchasing power" is an essential objective of American economic policy, the Employment Act of 1946 uses terminology that obviously requires interpretation. For example, our recent unemployment rates—official estimates of persons who at the time of successive recent surveys were looking for a job but by then had not found one—are in the neighborhood of 3y2 percent with seasonal adjustment (3 percent without such adjustment). For some time now the duration of unemployment has been 4 weeks or less for well over one-half of the unemployed. But the duration has been 15 weeks or more for a small proportion of the unemployed (during the year 1967 this proportion was about 15 percent) ; and the incidence of unemployment has been different on different sections of the population (distinctly lower on married men than on other members of the labor force; distinctly lower on whites than on Negroes, etc.). Many types of labor are in short supply. For some time consumer prices have now been rising at a yearly rate of about 4 percent. New wages settlements are said to involve 6 to 7 percent increases. The language of the Employment Act is not particularly helpful in providing guidance in such a situation. Perhaps there is by now reasonably general agreement among policymakers that of late we should have played safer against inflation even at the expense of the current unemployment rate in the foregoing sense. But this conviction can hardly be derived from a literal interpretation of the language of the Employment Act which formulates "maximum employment" as the goal of employment policy. (2) On what to me seems the only reasonable interpretation of an ambiguous text, the Employment Act tells us that policymakers should aim for approximating full employment as closely as they can without sacrificing other essential objectives. The proper balancing of objectives must be left to the authorized policymakers of each period. It is inconceivable that an act should specify all relevant objectives, and should attach weights to these. The Employment Act performs the function of serving as a reminder that major policies bear importantly on the employment level, and that sacrifices at the expense of the level of employment are justifiable only if very important other objectives are at stake. The act, 159 as it now stands, serves as an effective reminder of this to the Federa] Reserve as well as to other policy agencies. (3) In the long run the conflict between employment objectives and other significant policy objectives is much smaller than over periods of shorter duration. This is because raising the employment level by inflationary means involves reliance on "money illusion," that is, reliance on the inclination of individuals and of representatives of private groups to be satisfied with the attainment of money-income targets even if it turns out that given money-income targets correspond to smaller real incomes than had been assumed. Money illusion is rarely complete: it shows in failure fully to adjust the moneyincome targets to changes in the price level. Temporarily, policies based on money illusion can appear to be more or less successful, though at the expense of the stability of the price level. But money illusion fades away gradually, and the possibility of achieving gams in the employment level by inflationary methods gradually disappears. An inflationary policy of forced high employment must sooner or later be stopped. The measures by which this can be done inevitably reduces employment for a while below the level at which it could have been kept without inflation. Whether one does or does not regard the achievement of high employment and the avoidance of inflation as conflicting objectives depends therefore to a considerable extent on whether one takes a short or a long view of the matter. For reasons not all of which need to be described in derogatory terms different people assign different weights to short- versus long-run considerations. However, the public is realizing now that long-run considerations have recently received much too little weight. (4) The practices of the recent past have been far from ideal. Any fruitful analysis of past deficiencies suggests guidelines for the future. In this sense we shuold indeed be looking for guidelines. On the other hand, I do not believe that the shortcomings of Federal Reserve policy could be remedied by subjecting the System to some specific formula. The reason why this does not seem promising to me is that in matters of such complexity no act or amendment could be sufficiently specific (sufficiently unambiguous) to prevent the adoption of unfortunate policies based on ill-advised interpretations of the text. A text so rigid as to leave no room for "interpretation" would be unacceptable because the details of the situations in which future decisions will have to be made are unpredictable. At the end of these comments I will formulate a proposition that could, I think, serve as a somewhat flexible guideline to monetary policy (see point 6 below). (5) In several phases of the recent past the Federal Reserve has shown more concern with putting a brake on inflation than have other groups of policymakers. The policies of the Federal Reserve have not been very successful, but this is a different question on which I will comment presently. I feel opposed to further integrating the Federal Reserve System with the Government in the conventional sense because I feel that the Federal Reserve, as it is now constituted, could more readily afford to take a long view of policy matters (and to accept temporary unpopularity) than can many other policy agencies. I think Federal Reserve appointments should become less rather than J 60 more "political." I will add that I see no advantage in exchanging the stock of the Federal Reserve banks for certificates of membership. (6) While the Federal Reserve has shown concern with inflationdangers, it has in recent years not been effective in fighting these dangers, not even as effective as it could have been given the fiscal policies of the past 3 years. The Federal Reserve has rightly been criticized for letting the economy become oversupplied with liquid assets—particularly with money in the broader sense of currency plus demand deposits plus time deposits—though the past 3 years do not constitute a period of uninterrupted oversupply. It follows from what I said earlier that I do not favor attempts to formulate rigid rules for the rate at which the money supply (in the relevant broader sense) should be expanded in order to avoid the dangers of deflation without creating those of inflation. But it does seem obvious to me that a policy purporting to be one of anti-inflationary restraints miscarries if it leads to an increase of the money supply at a rate far in excess of the increase in output, and does so under circumstances in which there exists no reason to assume that at the given price level the public has a correspondingly increased demand for money balances. And yet this is what was typically happening in recent years. I therefore consider it important that in the future Federal Reserve policy should be much more mindful of the money supply (in the broader sense) than it has been so far. It should be mindful of the adverse consequences of an excessive or of an insufficient rate of increase of the money supply; and wThile the terms "excessive" and "insufficient" must be interpreted in view of somewhat crude estimates of the public's money demand at prospective levels of output, of prices, and of other economic variables, this fact does not by anv means render the foregoing statement empty (because estimates can be quite unreasonable, especially if they are merely implied). The Federal Reserve should in the future be able and willing to justify changes in the money supply in terms of reasonable assumptions concerning the effect of the money supplv on acceptable objectives. Congressional committees could exert a favorable influence on the Federal Reserve in this regard, though I believe that our past policies would have been worse, rather than better, if the Federal Reserve had been made part of the executive branch of the Government. A policy that oversupplies the economy with monev in order to prevent interest rates from rising to "undesirable levels" is doomed to become self-defeating even by its own standards, because the inflation which it causes makes high money rates of interest correspond to low real rates of interest. Hence, such a policy is apt to lead to very hito;h money rates. This, too, follows from a proposition formulated earlier in these comments: policies that work onlv as long as the public is significantly influenced by money illusion will sooner or later backfire. STATEMENT OF LEO FISHMAN, WEST VIRGINIA UNIVERSITY T H E C A S E FOE N A T I O N A L I Z I N G T H E FEDERAL RESERVE S Y S T E M The stated purpose of H.R. 11 is to make the Federal Reserve Svstem resnonsive to the best interests of the people of the United States and to improve the coordination of monetarv. fiscal, nnd economic poliov. With this purpose I fully concur. To this end H.R. 11 provides for 161 explicitly assigning power over basic national monetary policy to the President of the United States. Passage of H.R. 11 would thus explicitly invalidate the assumption of autonomy of the Federal Reserve authorities with respect to basic national monetary policy. In recent years many well-informed citizens as well as members of Congress, other public officials, and professional economists have come to recognize that the two most important sets of tools that can be used in implementing public economic policy in the United States are the tools of fiscal policy and the tools of monetary policy. Somewhat less well known is the fact that each of these sets of tools has its own advantages and disadvantages. At times the desired ends may best be achieved by placing primary or even sole reliance on monetary policy. In other situations it may be best to rely solely or largely on fiscal policy. If fiscal policy and monetary policy are used at cross purposes, each cannot fail to counteract, at least to some extent, the effect of the other. Moreover, even if they are both ostensibly directed toward the same general ends, neither monetary policy nor fiscal policy can be used in optimum fashion unless they are adequately coordinated with each other. An essential feature of the Employment Act of 1946 is that it assigns to the President, as Chief Executive officer of the Federal Government, the primary responsibility for coordinating all "plans, functions, and resources" of the Federal Government for the purpose of promoting "maximum employment, production, and purchasing power." Statements made at various times during the debates preceding passage of the act indicate clearly that this feature of the act was not the result of careless drafting or lack of forethought. Perusal of the debates in Congress preceding passage of the Employment Act of 1946 should be sufficient to resolve any lingering doubts on this matter. It was the deliberate intent of Congress to strengthen the role of the President with respect to the determination and implementation of national economic policy. When the legislation had been revised for the last time and the Senate was about to vote on it, Senator James H. Murray pointed out that the bill made it clear that "the basic responsibility for developing the employment program within the executive branch is that of the President * * *. The effect of this act," he continued, "is to underscore the responsibility of the President as the elected representative of the entire country, and as head of the executive branch of the Government." It is impossible for the President to discharge the responsibilities assigned to him in the Employment Act of 1946 unless he exercises the power to coordinate national monetary policy with national fiscal policy. In fact, during the debates in Congress preceding passage of the Employment Act of 1946 it was observed that monetary policy would be used by the President to promote the purposes of the legislation. On the other hand, no reference was made in these debates to the powers of the Federal Reserve authorities, nor was any mention made of their right to exercise their powers independently of the President. In recent years, however, the Board of Governors of the Federal Reserve System has claimed (and on several occasions has exercised) complete autonomy with respect to monetary policy. On more than one occasion William McChesney Martin, Chairman of the Board of 162 Governors of the Federal Reserve System, has stated before congressional committees that monetary policy to achieve broad national goals is determined by the Federal Reserve System. On at least two occasions monetary policies were adopted by the Federal Reserve authorities despite objections expressed by the President and his advisers. In April 1956, during Dwight D. Eisenhower's administration, the Board of Governors raised the discount rate although the Chairman of the Council of Economic Advisers and members of the Cabinet had held that such action would not be consistent with other Government policies designed to achieve the goals of the Employment Act of 1946. Similarly, in December 1965, the Board of Governors raised the discount rate, although President Johnson had recently indicated that he considered such a change ill advised and ill timed. Chairman Martin does not often concede that monetary policy is also determined independently of the Congress, but this is actually the case. As Senator George W. Malone remarked to Chairman Martin in 1957 when Martin appeared before the Senate Committee on Finance, "Congress has not one iota of authority, except the authority to change the [Federal Reserve] act * * *." Senator Malone also observed, "Congress has nothing to do with [the administration of monetary policy] * * *. We can talk to you, but we cannot do anything through it. Your judgment cannot be questioned for anything done under that act, unless we amend it." There is evidence that some dissatisfaction has existed in Congress over the assumption of independence by the Federal Reserve authorities and also over the moneary policies they have followed. In its annual reports, the Joint Economic Committee has repeatedly expressed disapproval of both the basic nature of monetary policy and the failure of the Federal Reserve authorities to coordinate monetary policy with the economic policy of the administration. In the 1966 report, for example, the Joint Economic Committee declared that it was "seriously concerned about the conduct of monetary policy in this country." The committee also stated, "While the rest of the executive branch was coordinating activities and plans preparatory to submitting them to Congress in January, the Federal Reserve w7ent its own way." As long as the Board of Governors continues to assert and to exercise complete autonomy in matters pertaining to national economic policy, it is possible for U.S. monetary policy and U.S. fiscal policy to be oriented toward different and incompatible sets of goals. It is impossible for the President to coordinate all "plans, functions, and resources" of the Federal Government for the purpose of promoting "maximum employment, production, and purchasing power." Proponents of complete autonomy of the Board of Governors of the Federal Reserve System in matters pertaining to monetary policy have claimed that their point of view is soundly based on tradition and historical precedent, on judicial decisions, and on legislation enacted by Congress. In other statements on this issue published within the past few years I have demonstrated that these claims are not valid. (See, for example, my article, "The White House and the Fed," which appeared in the July/August 1966 issue of Challenge.) I have accordingly argued that if the President is to discharge the responsibilities assigned to him in the Employment Act of 1946, he must exercise the power to 163 coordinate national monetary policy with national fiscal policy; that the basis for such exercise of power by the President already exists; and that the passage of new legislation is not necessary. Nevertheless, the issue has not yet been resolved. Moreover, inasmuch as the structure of the Federal Reserve System and its relation to the Federal Government are unique, there is some question of what steps the President might take to bring about such a change and to enforce his power to coordinate monetary and fiscal policies if he were challenged by the Federal Reserve authorities. Accordingly, there would be a definite advantage in the passage of legislation, such as H.R. 11, dealing explicitly with these matters. Three provisions of H.R. 11 are designed specifically to correct the present situation. One of these requires the President to include monetary policy guidelines in his annual economic report. Another stipulates that open market operations and other tools of monetary policy "shall be conducted in accordance with the programs and policies of the President pursuant to the Employment Act of 1946 and other provisions of law." The third calls upon the Federal Reserve Board (which would replace the present Board of Governors of the Federal Reserve System) to submit a quarterly report to the Congress stating its past and prospective monetary policy actions and indicating "specifically how such actions and policies facilitate the economic program of the President." These provisions of H.R. 11 should have the desired effect. Their language is clear and explicit, particularly when considered within the context of the statutes they amend, and the other supporting provisions of H.R. 11. To strengthen the coordinate relationship of monetary and fiscal policy under the direction of the President, H.R. 11 provides for substantial changes in the structure and financing of the Federal Reserve System. In effect the Federal Reserve System, which at present is owned by the member banks, would be nationalized. Stock in the Federal Reserve banks now held by the member banks would be retired; all interest, discounts, assessments, and fees received by Federal Reserve banks would be paid to the United States Treasury; operations of the Federal Reserve banks and the Federal Reserve Board (which would also replace the Federal Open Market Committee) would be financed with funds appropriated by Congress. It is likely that these features of H.R. 11 will provoke considerab/e controversy for reasons not directly related to the main purpose of the bill. It may, therefore, be useful to consider some of the direct effects of these provisions and to anticipate some of the arguments that mav be offered against them. Nationalization of any type of economic activity in the United States is typically resisted and feared. There is a strong preference for private ownership and control. An attempt to extent public ownership—and especially national ownership—to any type of economic activity is generally opposed not only on its own merits, but also because it is viewed as an opening wedge for other similar encroachments on free competitive enterprise. In this case, however, such fears are without foundation. The Federal Reserve System, as noted above, is unique with respect to both the organization and its existing relation to the Federal Government. The Federal Reserve banks are certainly not free, competitive enter- 164 prises. The Federal Reserve System was organized and has for several decades been functioning not for profit, but to influence credit conditions, to meet the needs of commerce and industry, and for various other purposes related to the satisfactory functioning of the monetary and economic system of the country. Although each of the 12 Federal Reserve banks is owned by the member banks in its district, the relationship is purely formal and involves virtually no power with respect to determination of important policy decisions or control of the level or disposition of earnings. The important controls and influences, to the extent that they are not specified in Federal legislation, emanate largely from the Board of Governors and the Federal Open Market Committee. Nationalization of the Federal Reserve System thus cannot legitimately be regarded as an encroachment on free competitive enterprise. From a purely financial point of view, nationalization of the Federal Reserve System in and of itself should have relatively little effect. Since the stock in the Federal Reserve banks presently owned by the member banks would be redeemed at par, the member banks would experience no direct gains or losses. And since the member banks cannot presently receive more than a 6-percent return on the par value of the stock they hold in their Federal Reserve bank, they can probably earn at least as great a return by loaning or investing the funds they receive when the stock is retired. H.R. 11 provides that future earnings of Federal Reserve banks will be paid directly to the United States Treasury. Relatively large revenues are derived from the operation of the Federal Reserve banks, mainly in the form of interest payments on U.S. Government securities. But under existing arrangements the Treasury already receives over 90 percent of the net earnings (before payments to the U.S. Treasury) of the Federal Reserve banks, since dividends payable to member banks are limited to 6 percent as indicated above. Clearly the magnitude of the possible increase in Treasury receipts is not sufficient in and of itself to justify nationalization of the Federal Reserve System, nor is this the reason why proponents of H.R. 11 favor nationalization. One possibility that cannot be completely disregarded is that some State member banks might discontinue membership in the Federal Reserve System if the Federal Reserve System were nationalized. Any such defections, however, would be based largely on psychological considerations, rather than on any substantive change in the operations of the member banks or in their functional relationship with the Federal Reserve bank of their district. If it should appear that large-scale defections might occur, incentives of one kind or another might be offered to State banks to maintain their membership. With minor adjustments, the recently proposed plan to make Federal Reserve bank credit more readily available to commercial banks might serve this purpose. It should also be noted, however, that if open-market operations are used as the principal tool of monetary policy, the effectiveness of monetary policy is not limited by the number of member banks or by the volume of member banks' assets. 165 Other organizational changes provided for in H.R. 11 also appear to be consistent with the main purpose of the bill. Abolition of the Federal Open Market Committee, for example, would virtually eliminate the influence of Federal Reserve bank presidents on national monetary policy. This is as it should be. These presidents are appointed by the board of directors of their respective Federal Reserve banks, which are owned by the member banks in their district. There is no reason why they should play an important role in the determination of national economic policy, nor is there any true statutory basis for their exercise of such a role. When the 1935 amendment to the Federal Reserve Act was passed, Congress did not anticipate that open-market operations would be used to achieve broad national economic goals. Indeed at that time Congress specifically refused to grant to the Federal Reserve System any mandate to increase its powers in such a way as to influence the general level of economic activity. The Federal Reserve System is authorized by the Federal Reserve Act to use the tools of monetary p o l i c y to cope with seasonal and other transient factors affecting money market and credit conditions. The fiscal autonomy of the Federal Reserve System has enabled it to assert its independence of the Congress. This independence would be terminated by H.R. 11 by virtue of the requirement that the Federal Reserve System operate and administer its affairs with funds appropriated by the Congress. Regular appearances of its officials before congressional committees authorized to inquire into the financial and fiscal affairs of the Federal Reserve System will assure full disclosure and publicity to the details of the operations of the Federal Reserve System. This process wTill also help to make the Federal Reserve System more responsive to the will of democratically elected officials of the Federal Government. My support of H.R. 11 is not based on any sharp dissatisfaction with the manner in which monetary policy has continuously been administered under the present structure of the Federal Reserve System. Monetary policy has become a principal method for achieving the goals of the Employment Act of 1946. The postwar prosperity, the small number and minor character of the postwar recessions, and the record long economic expansion that began in February 1961 have in no small measure been made possibly by sound and judicious use of monetary policy. But the time has come for a change. Views and policies of a political nature are frequently espoused by the monetaiy authorities of the Federal Reserve System. They have become prominent public figures. Increasingly these officials find themselves in the center of political controversy as they expound and defend their monetary policies and the goals they hope to achieve. Monetary experts and technicians must play a supporting role in the determination of monetary policy. But the essential political decisions involved in monetary policy determination should be the responsibility of officials who are elected by the people or who are responsible to elected officials. H.R. 11 will serve to accomplish this change. 166 STATEMENT OF WILLIAM J. FRAZER, JR., UNIVERSITY OF FLORIDA1 A bill (H.R. 11) before the House Committee on Banking and Currency seeks to amend both the Employment Act and the Federal Reserve Act with the view to making monetary policy more responsive to the need to achieve the goals of the Employment Act. One proposed amendment would bring monetary policy and debt management explicitly into the Employment Act for the first time. It would require the President, in presenting his economic program to the Congress, to make recommendations on fiscal and debt management policies and guidelines concerning monetary policy, including the growth of the money supply. Other amendments provide for changes in the structure of the Federal Reserve, mainly with respect to its policymaking functions. An objective is to improve the coordination of monetary, fiscal, and economic policy generally. The legislation comes after years of study, both of the present structure of the Federal Reserve as a policymaking organization and of guidelines or rules for the conduct of monetary policy. It gives rise to a number of questions about the coordination of policies, the appropriate target (s) for policy, the relationship between the target (s) and business activity, the form in which target values should be stated, the role of "defensive" and "dynamic" operations, the necessary tools, and so on. Section I below deals with the specific questions that have been raised by the Domestic Finance Subcommittee, pursuant to holding hearings on the legislation; and section II is an appraisal of the structure of the Federal Reserve in the light of the legislation, related suggestions, and the background of study preceding it. Strands of two familiar controversies about the making of decisions with respect to Federal Reserve policy run throughout the present paper; notably, the one over whether decisionmaking should be centralized within a single agency or governmental body or whether the Federal Reserve should be relatively independent within the framework of government; and the one over whether the Federal Reserve should follqwT a strict rule or exercise discretion in effecting changes in monetary policy. The most common argument in defense of the independence of the Federal Reserve in the framework of government is the quasi-judiciary one, as defined later. Others are introduced, however, relating to compensating errors and a proposed educational function for the Federal Reserve Board. The latter arguments in support of some form of independence are said to be economic, as distinct from pragmatic ones. In view of the provision in the legislation for quarterly reporting by the Board, and related guideline and rule suggestions, the arguments for some form of independence on the part of the Federal Reserve are not necessarily in conflict with provisions for changing the structure of the Federal Reserve. Nevertheless, with the view to achieving one of the objectives of the legislation—namely, making monetary policy more responsive—a case is made for tying 1 A number of individuals read and commented on the initial draft of the present paper; namely, Profs. Frederick O. Goddard, George B. Hurff, Charles A. Matthews, James G. Richardson, and Miss Lahoma Riederer, of the University of Florida, and Prof. William P. Yohe, of Duke University. All are absolved from any responsibility for the commissions in, and the omissions from, the paper. 167 Federal Reserve policy more directly to the need to refine measures of and to attain national economic goals, all via newly defined reporting procedures and a modified form of rules proposal. In particular, the Federal Reserve should attain within limits, as suggested by the Joint Economic Committee, a rate of change in the money stock (variously defined), and should, in addition, achieve an average growth rate in the money stock over longer periods of time, since the limits are set to begin with to allow for errors and some defensive and countercyclical maneuvering. Provisions for deviations from the guidelines are suggested. Deviations should be permitted when empirically verifiable explanations, as outlined later, can be given in the various reports for doing so. In addition to the strands of controversy, there are throughout the present paper critical appraisals of a banking view as distinct from a modified monetarist's view. The banking view is said to be characterized by a preoccupation with banking mechanics, an emphasis on ties to the money and credit markets, and an emphasis on the prospect of influencing the achievement of national economic goals directly through changes in the tone of the money and credit markets ana degrees of credit ease or tightness. The "monetarist's" label has been used to apply to those who emphasize mainly the relationship between rates of change in the money stock and the national economic goals. Even so, the present modified monetarist's view may be said to be characterized by an emphasis on interrelationships between changes in the structure of interest rates, changes in rates of change in the stocks of bank credit and money, and national economic goals. With reference to the strands of controversy and the banking and monetarist's views, the Federal Reserve may be said to have traditionally been alined in defense of independence within the framework of government, in defense of the exercise of discretion in policy matters, and with the banking view. Its position in all three of these instances may be said to have related to a form of mysticism, indeed, a mystique as described later, not, of course, with respect to every policymaker or every bank in the System but with respect to the policymakers as a group and the System as a whole. One's position on monetary rules and the Federal Reserve's exercise of discretion is closely related to a view of economic knowledge (or lack of it). Also, the original structure of the Federal Reserve System as a policymaking organization, and the structural changes effected by the Banking Acts of 1933 and 1935, are said to have been related to problems and economic knowledge of the times, all in relation to prevailing views about the centralization of power. In view of these immediately foregoing interrelationships, section III below relates a review of aspects of recent monetary policy to the banking view ascribed to the Federal Reserve, and section IV is an overall view of notions about rules and economic knowledge in relation to the Federal Reserve as a policymaking organization. Some of whatever may be original in this paper centers about the way in which various elements from earlier controversies are combined, along with analytical notions, in an appraisal of the structure of the Federal Reserve as a policymaking organization. In summary, arguments and analysis disclose and support the need for, and the desirability of, the following: making monetary policy a more serv- 168 iceable instrument of overall economic policy; a form of guidelines, as suggested by H.R. 11, by the Joint Economic Committee and as presently modified; structural changes in the Federal Reserve, partially as provided by the present legislation; and reporting procedures, also as partially provided by H.R. 11. The original Federal Reserve Act may be said to have given proper allowance for the banking view, given the economic knowledge and the characteristics of the economy of its time, including a relatively close link between commercial bank loans to business firms and expenditures by such firms. With the passage of time, however, at least two things have happened to render the structure of the Federal Reserve as a policymaking organization and the banking view inappropriate: (1) the simple tie between bank lending and expenditures has been broken; and (2) the emphasis in attaining national economic goals has shifted to the more abstract plane of the interrelationships between interest rates and rates of change in stocks of credit and money and in the flow of income. Now to make monetary policy more responsive, changes in the structure of the Federal Reserve are called for. The need for these changes is indicated by the decline in the relevance of the banking view and in the related role of judgment about the satisfaction of credit needs. The proposal for a single policymaking board is supported, subject to the introduction of certain policy guidelines to allow for the altered nature of the policymaking function, and subject to the retention of elements of the System's original regional (or Federal) character. The guidelines and the latter provision are defended on economic grounds. I. MONETARY POLICY GUIDELINES AND OPEN MARKET OPERATIONS Under the Employment Act of 1946 the President is responsible for transmitting to the Congress each year an economic report setting forth a program for achieving national economic goals. The reports have dealt with about the same topics over the years, with varying degrees of emphasis on the public and private sectors, the need for legislation to deal with the President's program, and on the respective goals—for example, economic stability without inflation, economic growth, full employment—and later poverty. The reports, too, at least as early as 1954, have recognized the importance of the Federal Reserve's control over credit in maintaining economic stability. The discussions on this subject were usually in general terms of credit case or tightness, sometimes with emphasis on special aspects of the economy such as housing. One characteristic of the Act and another of the reports (including the President's report or letter of transmittal and his Council's supporting report), however, are presently of special interest, notably: (1) the act has been viewed as giving expression to interest on the part of the Government in the aspects of economic life outside the sphere of credit and monetary policies; and (2) as the years have passed, more elaborate statistical information has been included in the reports, and the standards of achievement have been continuously on the rise. The original framers of the act—with the leadership of Congressman Wright Patman in the House and others in the Senate—were doubtlessly wise in proposing high standards of performance and 169 omitting requirements in the form of fixed quantitative targets. Now, even so, the rising standards of performance and our enhanced ability to cope with economic problems seem to require some form of more specific statements, quantitative in character, particularly about credit, monetary and interrelated policies. As provided in H.R. 11, the President's Economic Report is now to include recommendations on debt management and monetary policy guidelines, and there are additional questions about the Federal Reserve's traditional independence within the framework of government as well as about their reporting on credit and monetary policy. Recent hearings and a report by the Joint Economic Committee also deal with standards for guiding monetary actions [27; and 28].2 This section, consequently, deals with some of the questions raised by H.R. 11 and related materials. 1.1.—Do you believe that a program of coordinating fiscal, debt management and monetary policies should be set forth at the beginning of each year for the purpose of achieving the goals of the Employment Act, or alternatively should we treat monetary and fiscal policies as independent mutually exclusive stabilization policies? The recommendations for a fiscal policy must allow for monetary policy,3 and vice versa. For example, shaping the revenues and expenditures of the Government so as to affect the flow of income and the levels of employment and prices affects the amount of bank credit and money needed to achieve maximum employment, production, and purchasing power. Such a fiscal policy would also affect interest rates. A tax credit for capital expenditures by businesses, such as initially adopted in the United States in 1962, would in particular affect interest rates. The tax credit seeks to alter the rate of return (or the flow of returns) on additional capital expenditures with the view to inducing a larger flow of expenditures and in so doing it affects the rate of interest [14, pp. 9-11], defined either as some abstract rate or as a market rate. The rate of return on additional plant and equipment and the rate of interest are closely related. In fact, changes in the rate of return on capital expenditures possibly have more direct effects on the rate of interest than changes in bank credit [17, pp. 212-213] .4 This prospect is, in addition, related to the view that the volume of bank loans to manufacturing firms is more largely determined by conditions on the demand side of the market for bank loans than by conditions on the supply side [15, pp. 77-78]. 2 Numbers in brackets refer to references listed at the end of this paper. Sometimes page numbers are given in addition. References to the various sections and subsections of the paper appear in parentheses : e.g., (sec. I) or (sec. I.S.A). 8 The term "monetary policy" is used to mean a variety of things. Sometimes it is used synonymously with the term "Federal Reserve policy," particularly with respect to credit conditions (as indicated by, say, a rate of interest), the money stock and bank credit (that is the stock of bank loans and investments). At other times—particularly since "the revival of belief in the potency of monetary policy" in the 1950's, and 1960's [20, pp. 2^3]—"monetary policy" means policy with respect to the rate of growth in the money stock (defined as currency plus adjusted demand deposits, or as some broader measure, for example including time deposits). The present question uses the term monetary policy in the first of the preceding senses, but a; sharp distinction in meanings becomes important in dealing with some of the questions as in the present instance. * This point is related to the possible view that banks, bankers, and the banker-dominated Federal Reserve (36, pp. 35-36) have unusual influence on the level of interest rates (28, p. 22) as distinct, say, from the fiscal and tax policies of Congress. There was no doubt a time historically when the availability of credit and interest rates on funds for business expenditures were primarily and arbitrarily set by bankers, but the Congress today shares a greater part of the responsibility for high or low interest rates, the inflationary element in interest rates (19 and 23), and for prices than at a former time. There are the tax policies, as well as a host of others, including those with respect to a sustainable level of unemployment (20, pp. 7 - 1 1 ) , and some "natural" level of structural unemployment (i.e., unemployment due to the mismatching of job skills and job vacancies). Programs concerning job retraining and minimum wages are involved. 21-570—68 12 170 Despite the interrelationships between Federal Reserve and fiscal policies, coordination of the policies should be sought on an informal basis at least, and by having the Federal Reserve report explanations for its policies to the Joint Economic Committee, as has been suggested as a constraint 011 discretionary policy [28, p. 17]. There is a case for the independence of the Federal Reserve as a policymaking organization as it may be revitalized (sec. II). The Federal Reserve's function, as redefined in H.R. 11 and this paper, could be viewed as special and distinct, without there being excessive conflict in the uses of the diverse Federal Reserve and fiscal policy instruments and in the economic objectives of the Federal Reserve and the agencies of the executive branch of the Government. 1.2.—If you believe a program should be specified, do you believe that the President should be responsible for drawing up this program, or alternatively, should such responsibility be dispersed between the Federal Reserve System and agencies responsible to the President? The responsibility for drawing up programs concerning economic policies should be a function of the respective agencies. The Council of Economic Advisers should be responsible for coordinating programs of agencies in the executive branch, for presenting a general economic forecast, and for expressing its views on rates of change in bank credit and the money stock. The Federal Reserve should be responsible at least for increasing the money stock during any quarter at an annual rate, say, of not less than 2 percent or more than 6 percent (27, p. 230, and 28, pp. 16-17), subject to other qualifications given later (sees. 1.3. and I.3.D toF). As is presently the case, the Federal Reserve System should be accountable to the Congress (and the Joint Economic Committee in particular) for the achievement of national economic goals. The executive branch of the Government, too, subscribes to national economic goals, but due to its essentially political character its relation to the Congress will be more tenuous than that of the Federal Reserve, even a revitalized Federal Reserve. The Congress is responsible for specifying the national economic goals to which all agencies of the Government subscribe. Any major departure from the goals as defined by past interpretations, such as a long-term goal of faster economic growth (that is, a higher rate of change in gross national product per capita in constant dollars) should be approved by the Joint Economic Committee, and possibly by congressional statute. As further emphasized subsequently (Sec. I.3.B), once a specific target value such as the rate of change in the money stock is set, values for other variables are determined, given fiscal (or fiscal and tax) policy and the structural characteristics of the economy. These structural aspects have influence either by remaining unchanged or being changed. They would include the degree of the inadequacy of job skills for jobs, minimum wages, factors affecting the degree of competition in product and labor markets (18, pp. 337-351), Government subsidies on housing, interest rate ceilings on FHA-insured and VA-guaranteed mortgages (18, pp. 404-412), the extent of support for mortgages in secondary mortgage markets, and so on. All of these aspects of the economy are virtually beyond the purview of the Fed- 171 eral Reserve, although the effects of its policies will depend on them. These must be considered. Nevertheless, monetary and credit policies generally viewed can be frustrated directly in proportion to the Federal Reserve's efforts to give weight to the special effects resulting from the structural characteristics of the economy.5 1.3.A.—Should monetary policy be used to try to achieve goals of the Employment Act via intervention of money supply (defined as desired) as provided in H.R. 11, or alternatively, should H.Ii. 11 be amended to make some other variable or variables the immediate target of monetary policy f As Mitchell has emphasized (27, p. 120), "in our dynamic economy, no single variable—whether it be the money stock, money plus time deposits, bank credit, total credit, free reserves, interest rates, or what have you—always serves adequately as an exclusive guide for monetary policy and its effects on the economy." Even so, when the Board of Governors and the Federal Open Market Committee are left free to select one variable and then another and to express policy in any of a wide variety of measures, and when a policy in the System is arrived at through a consensus of opinions about policy (first, of Federal Reserve economists making recommendations and, then, by the policydetermining authorities), all as distinct from the reasons for the policy, then ignorance is compounded by ignorance. The result of arriving at a policy by consensus about the policy rather than the underlying reasons is a policy that cannot be explained by a proper use of language. The written language comes to be used to conceal meaning and thwart communication, while giving the appearance of dealing with fundamental truths.6 The joint committee is apparently correct in the view (28, p. 12) that "the Federal Reserve does not appear to have developed a set of priorities for its own guidance." 7 Moreover, they have not sought to develop the explanations underlying their policies in any empirically verifiable form; and the Federal Reserve's effort on a large scale proj5 Common illustrations of the effect of the Federal Reserve's policy (viewed in terms of interest rates) involves instances in which some temporarily invariable ceiling exists on the interest rate payable. Governor Mitchell, for example, cites as one of the best examples of the effect of policy the postponement of a revenue bond issue of the Chesapeake Bay Bridge (27, p. 131). He said, "if the level of interest rates was raised, it (the project) could not be financed." Well, this is a perversion of the focus of monetary policy in the presumed context of a relatively free enterprise economy, and it is not an illustration of the effects of policy as it might apply, say, to manufacturing corporations (e.g., 15). The justification for interest rates as a general credit control device would seem to require more than examples from the welfare and Government-oriented sectors of the economy. The tendency to invoke such examples in defense of discretionary monetary policy is itself an example of two things : (1) the extreme to which the defense of discretionary policy may be carried ; and (2) the need for some stabilizing constraint on such policy. 6 This is why Representative Reuss has trouble with the language in which monetary policy is discussed (27, pp. 229-23i3). This, too, is why the Joint Economic Committee reports on the minutes of the Federal Open Market Committee as follows (28, p. 10) : The minutes made available are couched in the most general, nonquantitative monetary and stabilization terms. They have tried to indicate a considerable reliance on institution and mystique in shaping actions rather than giving Congress, or observers of monetary affairs, a full opportunity to follow the developing and sometimes conflicting concepts or reasons which have influenced decisions. 7 The improvement of decision rules or even their possibility is not discussed in the Federal Reserve System's writings. Ironically, Christian comments to me that his findings (7) suggest "that a tacit, inarticulable, or unconscious set of decision rules were followed by the monetary authorities over the study period." Possibly the absence of references to rules in 'System research and the recurring possibility of the presentee of an unconscious set come about because even the discussion of rules suggests an increasing obsolescence of the tools of the trade of the discretionary authorities. The interests of officials can be advanced with subtlety and power. In some instances these become competitive with some and complimentary with other types of research. 172 ect, the Fed-MIT model, is just further evidence of a Federal Eeserve "mystique" 8 noted in a committee report (28, p. 12). A good bit of discussion has centered about defining the proper indicator of Federal Reserve policy (14, pp. 16-25; and 26, pp. 91-102). At the one extreme there has been the view that the true indicator cannot be influenced by any factor outside of the control of the Federal Reserve. This narrows the responsibility for control to Federal Reserve policy actions—that is, open market transactions and changes in discount rates and reserve requirements. At a different level, there has been a tendency to emphasize, as Dewald and Gibson do (11), that regularity exists in noncontrolled factors affecting member bank reserves and that their behavior can be adequately predicted so as to give the Federal Reserve control over member bank reserves. Going a step further, there appears to be sufficient agreement that the Federal Reserve can (as distinct from should) control the money stock.9 At still another level of Federal Reserve responsibility there has been a tendency to emphasize (1) credit flows or interest rates (27, p. 168), (2) the change in the relationship between prevailing and prospective rates (14, pp. 35-39), and (3) even changes in the structure of rates (15, pp. 73-74). Gaines suggests (27, p. 168) that interest rates are subject to the direct influence of the Federal Reserve in an operational sense; "while changes in the money supply are influenced by Federal Reserve policy, the influence tends to be at a second remove rather than at a direct point of entry of the central bank into the economic process." In the case involving the change in the relationship between present and future rates the Federal Reserve is even made responsible for the level of the income velocity of money. Changes in business conditions, as indicated by changes in the velocity are said to be brought to the level of Federal Reserve operations (14). This is highly abstract. A potentially useful concept is simply set forth without elucidation. Now changes in rates of change in bank credit, member bank deposit liabilities or the credit proxy (27, p. 132), and in the money stock, and changes in interest rates and the term structure of rates are all interrelated. The present point, however, is this: as one moves from 8 Characteristic of the mystique is the apparent assumption that knowledge about the effects of policy exists and that it is embodied in judgment about the need for a particular policy. Those invoking the mystique as a substitute for knowledge have often seemed to present as their best defense, (1) an acquaintance wih a frustrating array of facts and details and (2) the promise of future research to confirm the validity of their view. The traditional tendency to invoke increasingly complicated detail as a disguise for knowledge has apparently contributed to the Federal Reserve's support of the "Fed-MIT," "special purpose," "large-scale," "econometric" model ( 1 0 ; and 14, p. 6). The model has been often cited as progress in the right direction (27, pp. 190, 2 0 0 - 2 0 1 ; and 28, p. l;5). The research support apparently resulted from a misconception of either monetary policy or the special purpose model in its exploratory stages. Principal difficulties with the model, apart from statistical ones commonly mentioned, concern (1) the emphasis on linkages as "causal" sequences and (2) the prospect of varying a controlled variable so as to achieve a specific target value in another so-called "dependent" variable. 9 The pro-rules economists of course accept this (27, pp. 77-118) and Guy E. Noyes of Morgan Guaranty expresses a possibly widely held view as follows (27, pp. 181-182) : It is not a question of whether banks adjust their demand deposit liabilities promptly to changes in reserve availability, but only how they do it. In Rhort, it is theoretically irrefutable that the Federal Reserve can, within a matter of weeks, force the banking system, and the economy, to accept any moderate change in the money stock it chooses. It is not quite correct to say the Fed can make the money supply whatever it chooses, because large changes in short periods would create some institutional problems—but no one is talking about large abrupt changes anyway. So this qualification has no practical significance. Tilford C. Gaines says (27, p. 200), " I think that if the state of knowledge is not yet sufficient for them to provide a more sophisticated framework, that money stock would be an acceptable first approximation." 173 changes in the rate of change in the money stock to changes in the structure of interest rates, one is at the same time ascribing an increasing amount of responsibility to the Federal Reserve. The phenomena they work with become increasingly complicated. At the highest level of responsibility, we are holding the Federal Reserve responsible for offsetting shocks to the economy from outside factors, for dealing with shifts in expectations, and so on. As a first approximation to minimizing chaos, however, the course as suggested by the Joint Economic Committee is clear (28, p. 11): The policymaking officials of the Federal Reserve System (or a reconstituted group as suggested in H.R. 11) should maintain on a quarter-byquarter comparison, an appropriate normal range of increase in the money stock seasonally adjusted, say, of from 2 to 6 percent per annum,10 subject to some qualifications as noted now and later (sec. I.3.D to F). The present qualifications are as follows: (1) the policymaking officials should also be responsible for providing empirically verifiable explanations for changes in the money stock extending above the upper bound and below the lower bound; 11 and (2) the Federal Reserve officials should be encouraged to pursue sophisticated policies as they develop the capacity and understanding for carrying out such policies. As mentioned elsewhere (15, pp. 68-69) : If there is knowledge underlying deviations from simple rules such as given growth rates in bank credit and the money stock, then presumably it can be put in empirically verifiable form and verified, given large research staffs and the modern computer. If such knowledge of policy and its effects cannot be demonstrated, then there would seem to be little room l e f t f o r judgment. In such a case, adherence to a "neutral position" or simple rule would seem to be the best course. I.3.B.—Should the guidelines of monetary policy be specified in terms of some index of past, present or future economic activity, or alternatively in terms of the target variable s value or growth? The terms "economic activity" and "business conditions" may be used interchangeably. The former suggests activity such as hours of work, the number of people working, and so on. Economic activity may be constant and coincide with given levels of employment, unemployment (and, therefore, with a constant percentage of unemployment), prices, and rates of interest, as well as with constant rates of increase in output, income, and the money stock (and, therefore, with a eon10 As the Joint Economic Committee has indicated (28, p. 17) "there is no intention to make the 2- to 6-percent range a permanent and unchanging one." They note a variety of factors that may affect the range. Selden mentions the desirability of change in any monetary rule (27, p. 98). 11 An empirically verifiable explanation would be one containing statements about economic relationships between variables. They should be of a form that could be refuted or affirmed by certain tests. The statements and accompanying discussion should be of sufficient substance to permit testing in several ways: (1) by reference to empirical data and results from statistical analyses of such data ; (2) by testing against known alternative explanations to determine the best one ; and (3) by testing for logical consistency within the explanation as well as for consistency with other explanations on other occasions. Such complicated testing is important in social studies as distinct from laboratory sciences for several reasons: (1) because of the inability to control certain variables while others are operating; (2) because of the wide variety of interrelationships between economic variables; and (3) because of the shortcomings of certain statistical methods when applied to noncontrolled experiments (i.e., experiments in which other things are not controlled when the effects of a given variable are being considered). The Joint Economic Committee, or other potentially appropriate committees of the Congress probably should not contend directly with the foregoing sort of explanation. Nevertheless, there could be a shorter explanatory statement for the committee and a more formal underlying statement. The availability of the latter to monetary economists and other interested citizens would contribute to the soundness of the explanation. The socalled explanations to the Congress would not of course change radically, just by voting a statutory requirement; there would still be a good bit of "Federal-Reserve-ese" for some time. A statutory requirement, nevertheless, would stimulate movement in the right direction. 174 stant ratio of income to money). Thus "economic activity," interest rates, and the velocity ratio may all decline when output and income are increasing, only at slower rates. This is the w7ay some economists use the term "economic activity." 12 Although one must recognize the difficulty in identifying turning points in business conditions as a practical matter, commonly used ones are those of the National Bureau of Economic Research.13 In view of the foregoing, guidelines for Federal Reserve policy can be specified in terms of levels for interest rates, prices, the ratio of income to money, and business conditions, on the one hand, and rates of change for the money stock, income, and so on, on the other. Specifying targets in terms of these levels and rates of change is not inconsistent. The tw^o sets of changes are not alternatives as the above question implies. For example, programs for achieving the goals under the Employment Act may be formulated to call for given rates of change in income, and given levels for employment (and unemployment as a percentage of the labor force) and economic activity. The levels and rates of change as set forth, however, are not independent (17, pp. 304-319). Once a specific target value is set for one of the variables, values for the others are implied, given fiscal policy and the structural characteristics of the economy. Changing the interrelationships between the variables would for the most part require structural changes in the economy. These would involve, for example, changing the level of structural unemployment (i.e., unemployment due to a mismatching of employee qualifications and job requirements) through educational programs, or changing housing demand through Government subsidies and interest-rate ceilings on FHAinsured and YA-guaranteed mortgages. These latter changes are of course outside of the purview of the Federal Reserve. 1.3.0.—Should we use a leading {forward-looking), lagging (backxoavd-looking) or coincidental indicator of economic activity? Forecasters get relatively good scores if they can recognize turning points in business conditions once or shortly after they occur, as Fels emphasizes in reporting on the problem of forecasting and recognizing business cycle peaks and troughs (13, pp. 3-48) . He also notes (13, pp. 47-48) weak evidence "that users of the NBER indicators approach actually have done better than their fellows." One may, consequently, view the matter of forecasting separately from that of selecting a desirable indicator of economic activity. Even so, indicators selected for the purpose at hand should have their turning points coincide roughly with turning points in business conditions as reported by the NBER. These indicators also should relate fairly directly to the national economic goals for employment and incomes, on the one hand, and to Federal Reserve influence in the operational sense, on the other. There has been controversy over what we mean by "levels of business" in the context of analyses of the relationship between the rate of growth of the money supply and the level of economic activity (e.g., 14, note 29, p. 14). More recently and in a similar context, statements in Joint Economic Committee hearings by Wallich (27, p. 17 and 20) and Davis (27, p. 310) attribute incorrect meaning to Milton Friedman's use of the terms "business activity," "economic activity," and "levels of business." Therefore, I have sought to be explicit about the use of the term "economic activity." 13 The National Bureau's dates for peaks and troughs are so widely accepted that Fels uses them in scoring the accuracy o-f forecasters (13). In other words, he has the forecasters attempting to forecast or identify what the National Bureau will subsequently date as a cyclical peak or trough. To change a Keynesian metaphor slightly, the matter is analogous to forecasting the winner of a beauty contest: you must select not necessarily the prettiest, but the one you think will be selected by the judges to be the prettiest. (On the dating of turning points specifically, see ref. 13, note 2, pp. 3 - 4 . ) The suggested indicators are as follows: the ratio of income to the money stock (i.e., the velocity of money ratio (14, pp. 1-41)) ; the interest rate on some long-term debt instrument (e.g., Moody's Aaa corporate bond rate or that on long-term governments) ; and the spread between the yields (as rates) on long- and short-term bonds (15, pp. 66-101). These measures possess in high degree the attributes listed above, especially after some smoothing and allowance for "noise," and allowance for the role of judgment and imperfections in the NBER's technique. Moreover, the velocity-interest rate association is one of the strongest that exists in economics (17). The relationship holds for the economy generally as well as for the key business and consumer sectors. So we have closely related variables as well as variables that relate to economic activity generally, and Federal Reserve operations and national economic goals in particular. Abstracting from the seasonal and defensive type operations of the Federal Reserve (37, p. 8), the velocity-interest rate association is sufficiently strong to imply an income target once targets for the money supply and a long-term rate of interest are given. In relating the target money stock variable or the target interest rate variable to the velocity ratio, one should not think in terms of lags in causal effects or in terms of invariant distributed lag patterns, as have become so common in the monetary research of the 1960's (e.g., 27, p. 222; and 14, pp. 25-39). Friedman, for example, has measured the average lag time between peak rates of change in the money stock and the peaks m business conditions, on the one hand, and that between trough rates of change in the money stock and troughs in business conditions, on the other. But these are averages and they likely vary with the duration of the cycle. Furthermore, m reviewing results from analyses of so-called distributed lags in the relationships between time series, one is confronted with a constant lagtime and a fixed distributed lag pattern, whereas neither the lag time nor the pattern is invariant over time.14 Now the foregoing points about the inadequacies of notions about lagged "effects" and constant distributed lagged patterns may be illustrated in two different ways. Recall, to begin with, two very different periods in monetary history: (1) the classical pushing on a string, 1937-38 period (8, pp. 26-32; and 34); and the intensive capital boom ending in 1966 (15, pp. 72-93). In the one period, the economy was in deep depression by post-World War II standards, excess reserves of banks were substantial, and some negative yields were reported on Treasury bills (8, pp. 29-30). In the boom ending in 1966, the reverse conditions prevailed. We would not, I suggest, expect a given increment in reserves or a given rate of increase to have the same effect on the money supply in these two instances or for the two effects to be distributed in the same patterns. Indeed, the Board of Governors thought that the link between reserves and the money supply was very weak at the time of the 1937-38 recession. 14 To be sure, recent research reveals just such instability in so-called lag coefficients. In testing for the stability of regression coefficients, for example, Christian concludes as follows (7. p. 477) : "both the irregularity of response to inflation and the instability of the coefficient of the lagged dependent variable further suggest that the distributed lag formulation of the linear model is unreliable. It has also been demonstrated that the single-period regression equation is substantially less efficient than the moving regression in obtaining information about the behavior of the monetary authority." 176 The Federal Reserve has much more control under moderately stable or high-employment conditions than under the extreme conditions of 1937 and 1938, or so it would seem. The lagged patterns are not fixed. There is a premium on maintaining economic stability, both because of welfare considerations and because of the greater control we have over the economy. Additional evidence of the inadequacies of notions about causal sequences and lagged effects in the relationship between aggregate time series is provided by an examination of characteristics of the corporate manufacturing sector (16). In one traditional instance, it has been common to view roughly parallel movements in bank loans and business inventories and to conclude that the funds from the loans were being used to purchase the inventories. As further research has revealed, however, firms differ in their financial structures and holdings of inventories vary disproportionately with their reliance on bank loans, all as they increase in asset size. Over time somefirmsare buying inventories to a greater extent than others and some are borrowing funds from banks, all in such a way that roughly parallel changes occur in the time series. Other examples of the inadequacies of analyses that treat relationships between aggregate time series as "casual" and of a "fixed" distributed lag type could be cited (e.g., 15, pp. 66-104). However, my summary answer to the question of how the target variable (s) should be related to the index of business conditions (namely, the velocity ratio) is to beware of notions of "causal" and fixed distributed lags. Instead, rely on the concepts of cyclical and secular changes, such that varying patterns between the series unfold over time. We should conceive of entire cyclical and secular phases and of the possibility of achieving sustainable rates of growth in stock and flow variables, with the view to eliminating economic instability. This is in lieu of thinking about a controlled variable that you change by a specific amount to obtain, after a fixed time, a certain pattern of effects, other things being equal in the sense that they remain unchanged. The other things are not unchanging in our going economy, even though the statistical method in wide use (i.e., the classical, least squares, regression method) has built into its computational mechanism the assumption that they are.15 There are, finally, cyclical and secular patterns in the aggregate time series, and some of the respective series must reflect some responses to the same factors, as well as aspects of the financial structure of firms and other structural aspects of the economy. This is particularly true if, as suggested above, the aggregate series are not causally related in a strong and direct way. The apparent support for the view that some time series share in common responses to the same changes in the setting is one reason why expectations in monetary analysis deserve some emphasis (e.g., 14 and 15). It is also a basis for agreement with those who wish to reduce the wide swings in the rate of change in the money stock as a means of stabilizing the economy, until we understand better the factors affecting expectations.16 15 These last sentences may be compared with footnotes 7 and 10 above. 18 References to expectations in the May 1968 hearings of the Joint Economic Committee are instructive! Mr. MITCHELL. Some of the monetary lags are short. The effect on expectations ia immediate * * *. Mr. MITCHELL. Well, this gets to be kind of troublesome. A lot of the meaning, the influence of monetary action is on expectations. Chairman PROXMIRE. Here is what I think is the kind of thing he (Professor Christ) is getting at. He is pointing out that you did have this very hard to understand and explain 177 1.3. D to F.—Should the same guidelines be used each year into the foreseeable future, or alternatively, should new guidelines be issued at the beginning of each year conditioned on expected private investment, Government spending, taxes, etc J Under what circumstances may the guidelines be changed? The rate of change in the money stock has been mentioned as the guideline for Federal Reserve operations. Also, an appropriate normal range of increase of from 2- to 6-percent per annum, after seasonal adjustment, has been mentioned (sec. 1.3.A.). This range should be sufficient (1) to allow for Federal Reserve's inability to achieve a target value within the range during any quarter and (2) to permit some countercyclical maneuvering, preferably with minimum rates occurring in expansion phases of business activity and maximum rates falling in recession phases. I would, however, further recommend an average rate of change per annum for the longer period, since there should be no question of the Federal Reserve's ability to achieve it. The 2- to 6-percent range would seem to imply a secular growth rate of about 4-percent per annum, depending on whether the Federal Reserve sought above or below average values most frequently. The 4-percent average rate is on the high side, assuming a narrowly defined money stock (i.e., currency plus demand deposits adjusted) and judging from secular changes in the post-World War II period. The Federal Reserve and possibly the Council of Economic Advisers, therefore, may wish to suggest a lower stock and an accompanying revision of the limits with the spread in percentage points remaining unchanged. The principal justification for changing the secular rate would be a revision of the outlook for the long run. Certainly, this should be relatively stable, since it should be relatively free of the effects of transitory influences. But revisions may be permitted in the average growth rate at intervals, again with well-founded explanations for doing so. These revisions could correspond to the average length of post-World War II business cycles. As stated earlier (sec. 1.3.A.), the Federal Reserve would still be free to vary the rate of change in the money stock beyond the maximum rate or below the minimum rate provided that an empirically verifiable reason can be given for doing so. An infrequent change in the average growth rate for the money stock is preferred for a combination of reasons, namely: one objective is to try to stabilize unwarranted changes in policy, and a distinction between transitory and more pervasive influences should be possible. If the guidelines can be changed frequently (i.e., the average rate and the 2- to 6-percent range) without well-founded explanations for doing so, then the guidelines as such are of no use at all. situation that occurred last year (i.e., 1967), in which the money supply was increasing rapidly and the price of money (i.e., the rate of interest) was going up at the same time. Interest rates were high, although the money supply was increasing. It is hard to understand. He argued, and the other economists seemed to agree, that one reason is because the Fed was expected, to continue in the future to increase money supply at a rapid rate. This was inflationary, and because under these circumstances the economic reactions to the expectation of inflation is to follow policies that tend to drive up the interest rates, people are less likely to lend money if they expect it is going to have a much lesser value in the future. They are going to ask for higher rates before they do lend it. They argued, therefore, that if the Federal Reserve were committed to a policy of not increasing the money supply at a more rapid rate than 6 percent per year, that would not have that kind of expectations, and interest rates would be inclined to be lower. (See 27, pp. 131, 133, and 140, respectively.) 178 I.Given the goals of the Employment Act, what can debt management do to help their implementation? In the management of the Federal debt, the Treasury exerts several possible influences. These in turn have the possibility of helping in the implementation of the goals of the Employment Act. The first set of possible influences concerns changes in the composition of the debt, both with respect to the Treasury's savings bond program and with respect to the time-to-maturity distribution of the marketable portion of the debt. The savings bond program (including advertising and promotional aspects) has the possibility of influencing saving by the public out of current income. This could imply some direct effect on expenditures and therefore on the average of prices for current output, the rate of increase in gross national product and the level of employment. Most of the potential influence of the savings bond program is likely on changes in the composition of individual savings and secondarily on increases in the rate of change in saving out of current income. The effect on current income is unlikely to be noticeable except during periods of national emergency when appeals may be made to patriotism. Changes in the time-to-maturity composition of the marketable debt (i.e., in proportions of the marketable debt according to short-, intermediate-, and long-term debt instruments) have the possibility of exerting two effects: (1) contributing to the potential effectiveness of monetary policy (defined with respect to interest rates), and (2) "twisting" of the term structure of interest rates or the yield curve (i.e., varying the spread between the yields on long- and short-dated Government securities, and at the same time varying the slope of the line resulting from the fitting to a scatter of points consisting of yields and corresponding dates for the maturity of different issues of Government securities). In the first instance, increasing the proportion of the debt in the long maturity sector affects the liquidity of some financial institutions at the time of rising interest rates. As is well known, a given rise in interest rates across all maturities is accompanied by a greater price decline on long-term issues; and over the cycle of yields on "default free" securities, the yields on short-term securities vary more than those on long-term issues. The holders of long maturities in effect get frozen in or locked in to some extent during the transition from lower to higher rates.17 Thus, the larger the proportions of longterm issues the greater the potential for locking in the securities and reducing liquidity. A difficulty encountered in relying on this effect is that the commercial banks, significant holders of governments, have a 37 Evidence of a lock-in effect at commercial banks has been reported as a result of analyses of cross-section data for the 1965-66 period (29). The effect depends on reluctance to realize losses on marketable securities. Kane finds, in particular, that banks' unwillingness to take such losses varies inversely with their capital position. Underlying the reluctance is the notion "that selling securities below their book value impairs reported bank capital and the opinion that it is unwise for a bank to allow its reported capital to be impaired." As reported by Kane, "this concern for the preservation of the accounting value of bank capital has traditionally been rooted in: (1) bankers' fear of misinterpretation and criticism by stockholders, depositors, examiners, and colleagues in the banking fraternity ; (2) bankers' desire to minimize interference from regulatory restrictions tied to the size of the capital account (such as) maximum loans to one borrower maximum mortgage holdings, etc." The <favailability" doctrine—as the doctrine surrounding the lock-in effect of credit policy was called)—was revived in the 1950's. Its principal defender at that time was Robert V. Roosa, who for many years was associated with the Federal Reserve Bank of New York and later with the U.S. Treasury. It was an attempt to explain the effectiveness of credit policy in a world in which the central bank is unable or unwilling to bring about substantial interest rate changes and in which rising interest rates were thought to have little influence on capital expenditures anyway (18, pp. 631-634). 179 preference for business loans when credit-worthy business firms seek them, as during extended periods of rising interest rates (15, pp. 77-78, 91-93).18 The lock-in effect is more than offset, probably by the very forces giving rise to the higher interest rates to begin with. Even so, the longer the average maturity of the marketable debt the better, at least with respect to stabilization during an expansion phase of business conditions. In the second instance—that of twisting of the yield structure—the twist is supposed to come about as a result of a change in the relative supply conditions affecting the respective maturities of securities. The change in maturity composition of the Federal debt, however, may be minor in relation to developments in the private sector. Meiselman suggests that it is minor (32). Another reason for not expecting effects from altered supply conditions is that expectations about future rates of interest play a prominent role in determining the structure of interest rates. The expectations effects can potentially overwhelm the supply effects from the changes in debt composition. The empirical evidence in support of the expectations theories is abundant, as reported in Meiselman's review of current research (32) and elsewhere (31). Kane and Malkiel report on a survey of expectations to determine the potential for twisting the yield curve (30, pp. 343-355). They find some dispersion of investor expectations at banks, nonfinancial corportions, and life insurance companies. They conclude from their April 1965 survey findings and from a review of hypotheses that there is some potential for twisting the yield curve by altering the relative, supplies of different maturities (as in the Federal Reserve's "operation twist" or "nudge" of the early 1960's), all in partial contrast with Meiselman's conclusions from surveying current research (32). Next, changes in the structure of rates have some effects, regardless of whether Federal debt management or twisting operations by the Federal Reserve have any. These effects have been dramatically described for savings and loan associations (e.g., 32). They result in part, howrever, because of the peculiar attributes of those institutions. They essentially borrow on short-term loans and specialize in the purchase of a single type of long-term security; namely, mortgages. Thus, given the fact that the spread between long- and short-term rates on "default free" securities varies fairly directly with the cycle of interest rates on "default free" securities, the savings and loan associations can find themselves in a weakened condition during an extended phase of rising interest rates. The cost of the funds they borrow rises, the returns on the mortgages purchased at low rates remain unchanged, and new, higher yielding mortgages constitute a small proportion of portfolios. Further, tight credit (i.e., rising interest rates) has a strong effect on home construction and the supply side for the highyielding mortgages for a combination of two reasons: (1) household income is a constraint on the household's ability to make payments; and (2) both monthly and down payments increase and prohibit some from financing. Ceilings on FHA-insured and VA-guaranteed mortgages also play a role at times (18, pp. 404-412). 18 Two arguments are usually advanced to oppose arguments for the lock-in effect. The«e are (1) that banks cannot refuse to decline loans to depositors without incurring their disfavor: and (2) as emphasized by Kane (29), "provisions of the Federal tax law treat bank losses on security sales on very favorable terms." 180 The effects of credit policy on home construction (and, therefore, national income) should not be exhibited as a good example of either the potential effects of debt management or of credit policy. In fact, we should not wish to destabilize the housing industry to stabilize the economy, and we may in fact wish to isolate the housing industry from changes in credit conditions entirely. A conclusion to be drawn about the term structure from studies over the last decade by independent analysts is as follows (32) : "The only dependable way to change the relationship between short- and longterm interest rates is to change the level of rates." But I go even further, as far as the stablization aspects of rate changes are concerned. Absolute levels of rates per se are not the proper measure to focus on in conducting a credit policy. Speculative analysis (14, 15) has led to the conclusion that the major interest rate relationship to focus on in the case of the important business sector of the economy and in the conduct of a stabilization policy is as follows: that between present rates of interest (or rates of return of additional capital expenditures by business firms) and the probability of a future rise in rates of interest (or rates of return on additional capital). I will not labor the analysis nor the support for it 19 in this compendium. The concluding points for section 1.%,. are these: (1) Federal debt management is unlikely to have much influence on the term structure of interest rates after initial market adjustments and the elimination of money market "noise;" and (2) changes in the level of interest rates per se are not the proper focus for a policy seeking to stabilize business conditions. I.5.A.—Do you see any merit in using open-market operations for defensive purposes or should they ~be used only to facilitate achievement of the President's economic program and the goals of the Employment Act? What risks and costs, if any, must be faced and paid if open-market transactions are used to counteract transient influences? Open-market operations should be used to achieve a given rate of growth in the money stock within limits (sees. 1.3.A. and I.3.D to F). As proposed, these limits currently allow for seasonal variations in the money stock and some countercyclical maneuvering by the Federal Reserve. The rates of change in the money stock, moreover, may be exetnded beyond the limits, when satisfactory explanations for doing so are given. All of the policies directed at achieving stabilization or secular changes, however, should be directed toward achieving the goals of the Employment Act, with due allowance for the President's economic program. The President's economic program is likely to be too vulnerable to partisan political considerations. Federal Reserve operations should not be tied to it directly. For example, the temptation for 10 In brief, focusing on the difference between long- and short-term rates, say. on governments, cyclical changes in business, and financial activities of manufacturing corporations may be explained in terms of probabilities of future increases in interest rates. The maximum yield spread over a cycle in interest rates (and business conditions) reflects the strongest subjective probability of a future rise in the rate of interest over that cycle. The maximum spread is said to reflect the most likely possibility of acquiring liquidity on favorable terms. The acquired liquidity is then used as a source of funds for effecting an increasing flow of expenditures as the expansion phase develops. Recession follows a minimum-yield spread and the least likely prospects for a future rise in rates. Developments in this phase of activity are not exactly the converse of those in the expansion nhase. Whether the expansion phase is an extended or brief one affects entire sets of developments. The object of policy, in such a context, is to stabilize swings in expectations about the "normal" or "average" rate of interest, viewed as a subjective probability. Empirical support is offered for this analysis (15, pp. 66-101) 181 a President to pursue, or effect the timing, on occasion, of an inflationary policy as a means of giving a temporary spurt to the economy is too great, even though such a policy may be subsequently and overall costly in terms of economic resources and unwarranted changes in the distribution of income among the participants in the economy. The rules or money stock proposal does not preclude the Federal Reserve from using other targets—for example, "changes in the level and structure of interest rates." It simply requires, as modified (sec. 7.-5. D to F), that the Board understands the interrelationships between the various targets and set forth meaningful explanations, especially for varying the rates of change in money stock beyond set limits. There are seasonal patterns in the money stock. These probably arise from "defensive," "seasonal," or "road clearing" activities of the Federal Reserve. Such activities would include the occasional underwriting of new issues of Government securities. Quite likely the seasonal and defensive operations also eliminate some of the potential seasonal and random changes in interest rates. Without further study and experimentation, these defensive operations of the Federal Reserve should not be eliminated. To be sure, as evidenced in part from published materials, the Federal Reserve's ( and particularly the New York bank's) understanding of defensive operations is quite superior to the System's understanding of the interrelationships between interest rates, the term structure of rates, the money stock, and national economic goals. This asymmetry in understanding is due to two interrelated factors: (1) The dominant "banking view" of the Federal Reserve (33, p. 96; 34, pp. 35-41; and sec. I I below) ; and (2) the more difficult nature of the task of fulfilling the "dynamic" responsibilities attributed to the Federal Reserve since the Banking Acts of 1933 and 1935 and the Employment Act of 1946. A problem that has existed with the Federal Reserve since the legislation of the early 1930's and the early post-World War II period, as we are beginning to understand, is as follows: The Federal Reserve focus, as characterized by the banker orientation of the 1913 act and the immediately succeeding decades, has never been broadened to accommodate the changes in responsibilities. This is true despite the fact that an earlier and fairly strong and direct tie between bank lending to business and business expenditures is irrevocably broken. The large firms with significant impact on national economic goals can and do organize and plan their activities so as to avoid strong dependence on banks (16). As an organization the Federal Reserve is banker oriented. Its organization, indeed, encourages a preoccupation with the mechanisms of banking, at times, at the expense of the consideration of the achievement of national economic goals. There is temptation to focus on these mechanics in the sense that one can demonstrate so much sound erudition with respect to them. They are quite appealing to the worldly philosopher. The current challenge to dealing imaginatively and soundly with relationships between the various intermediate targets and national economic goals is demanding—disproportionately so in the environment of the Federal Reserve as presently organized. I.5.B and 0.—Do you believe that monetary policy can be effectively and efficiently implemented solely by open market operations? For what purposes, if any, shoidd (a) rediscounting, (b) changes in reserve requirements, and (c) regulation Q be used? 182 The "dynamic" operations of the Federal Reserve—that is, those directed toward the achievement of national economic goals—can be conducted solely by open market operations. Even so, some extra efficiency might result in a fairly taut and stable economy from the use of announcement effects in offsetting temporary shocks to the economy (14, pp. 33-35). In this respect, continued control over discount rates and possibly over legal reserve requirements (as ratios) may be useful. Other means centering about reporting provisions could substitute (sec. I.5.D). The role of announcement effects in stabilizing the domestic economy, however, needs to be better understood than at present. In addition, their role would probably be altered by the adoption of H.R. II. Announcements about discount rate changes are useful, at times, under the present organization of international financial transactions. In the early years of the Federal Reserve, the discount rate and its related mechanism were viewed as being related primarily to the conduct of "defensive" operations. These were regarded as the main operations of the System in its early years and de jure until the Banking Act of 1933. No doubt, the discount mechanism is still important, however modified (12), especially in the conduct of defensive operations and in performing the System's function as "a lender of last resort." On the other hand, with market and communication facilities of the postWorld War II type, commercial banks can acquire extra reserves for overnight or for longer periods (18, pp. 37-39 and 286-298). The Federal funds market, trading in securities, and correspondent banking arrangements, all facilitate the movement of extra reserves between banks, and open market operations may supply reserves to the commercial banks as a group. Defensive operations could be efficiently carried out without the use of Federal Reserve discount window as a means of extending credit, after a period of transition to some new arrangement, including improvements in secondary markets for bank assets and liabilities. Federal Reserve authority to vary reserve requirements was introduced in the Banking Act of 1935. In the thinking of the period, the added credit control was viewed in two ways: (1) as a means of tightening the link between the level of bank reserves and changes in money and credit; (2) as a means of shifting some control from the regional banks, including the New York bank m particular, to the Board. Thus, in the light of current knowledge about the link between the level of bank reserves, bank credit, and the money stock (11; 27; 34), and in view of the provisions in H.R. 11 to centralize policymaking functions at the Board level anyway, there is no strong justification for varying the reserve requirements as far as credit and monetary policy are concerned. Variations in differentials in the requirements for broad classes of banks and deposit liabilities affect the relative profitability of both the classes of banks and the classes of deposits. One possibility is to emphasize their use as a regulatory device as distinct from a general credit and monetary control.20 Regulation Q, the authority under which the Federal Reserve regulates interest-rate ceilings on savings- and time-deposit type liabilities 20 Using changes in the differences between reserve requirements for different classes of deposits or for different classes of banks does not mean, of course, there are no effects on* say, the flow of credit and the money stock. These effects simply can be neutralized (offset) in part or entirely or they can be permitted to operate, with open market operations remaining the principal credit control device for purposes of carrying out credit and monetary policies. 183 of commercial banks, should not be used as another general credit control device, as it apparently came to be used in the 1965-66 period. Regulation Q should be eliminated, or changes in interest-rate ceilings under the regulation should be confined to the regulation of interestrate competition.21 One possibility is to make the rate ceilings floating ceilings (that is, tie the ceilings to direct variations in some average of rates on the securities or the savings-type instruments of institutions for which the regulation of competition is sought).22 There are too many ways to regulate credit simply and directly, such as through open market operations, without resorting to the gymnastics performed under regulation Q in the 1965-66 period. The regulation of deposit interest rates was originally introduced in the Banking Act of 1933 (25, ch. 2). The objective then and in the years prior to the 1965-66 period was the regulation of competition between selected financial institutions for savings. From 1936 to 1957 the rate ceilings under the regulation were invariable (and usually above the rates actually paid over most of the period). After 1957 some adjustments were required as interest rates continued in their secular post-World War II rise. For example, in late 1964 the maximum rate payable on time deposits of less than 90 days was raised from 1 to 4y2 percent and, following the introduction of negotiable certificates of deposit (CD's) in February 1961, commercial banks were able to compete at times for the noncash liquid balances of businessfirms,such as those held in the form of short-dated Treasury issues and commercial paper. Other rate ceilings and other classes of savings affected more directly savings and loan associations and savers' shares in them. As market rates moved upwTard in 1965, however, the rate ceiling of 4 ^ percent adversely affected the competitive position of the commercial bank's short-dated CD's. The ceiling on this class of bank liabilities was raised along w7ith the ceiling on time deposits generally to 5Y2 percent, and the growth of the CD holdings by business firms continued to increase. Commercial bank competition for funds continued to increase in 1966, as did market rates of interest generally. This time, as other market rates increased, the ceiling on CD's w^as not raised further. Apparently, it was viewed appropriate to allow some attrition in the CD's with the view to moderating the expansion of commercial banks' business loans (2, p. 278). In September Congress granted the Federal Reserve the temporary authority (later renewed and extended until September 1968) to set ceilings on different bases,23 including size, and a ceiling of 5 percent was set on CD's under $100,000. What started as a regulation of competition for savings was broadened in 1966 into a credit regulation. In our Federal Reserve vernacular, regulation Q became a credit control device. Continuing the above example, it works as follows: An increase in the yield on CD's in rela21 Haywood and Linke conclude from their study (2(5) that excessive interest-rate competition "was not and is not a sufficient problem to warrant continuous regulation of deposit interest rates." They recommend steps for the elimination of the regulation of interest rates on savings and time accounts. 22 Haywood and Linke indicate that early versions of the Banking Act of 1933 would have established interest-rate regulation such that the ceiling rates varied with market rates (25, ch. 3). 23 The emergency, interest-rate ceiling legislation of September 1966 is broader based than suggested in the text. The request for emergency legislation was supported by references to conditions in the housing market (25, ch. 4 ; and 27, p. 188), specifically the hope to get more funds to the savings and loan associations. (Note the discussion under sec. / A ) 184 tion to Treasury bills induces, say, large manufacturing corporations to exchange some Treasury bills for cash (i.e., demand deposits) and cash for new CD's. A net effect is a decrease in banks' demand deposit liabilities and an increase in time deposit liabilities. A further net effect, consequently, is a reduction in required reserves, and extra reserves to support further credit expansion, in view of a smaller reserve requirement for time deposits. A decline in the relative yield on CD's has the reverse effect. 1.5.D.—Do you see any merit in requiring the Federal Reserve Board to make detailed quarterly reports to the Congress on past and prospective actions and policiesf Are there any risks and costs in this procedure? In what ways, if any, would you modify the reporting provision? What information do you believe should be included in such reports as you recommend the Federal Reserve submit to the Congress? The Federal Reserve should make quarterly and annual reports to the Congress, possibly through the Joint Economic Committee. There could be, indeed, two sets of reports: (1) an annual report on the essentially banking, supervisory, and administrative aspects of the Federal Reserve, and (2) a quarterly report on those aspects of operations concerning defensive operations and mainly national economic goals. The first set of reports could be directed primarily at the Senate and House Banking and Currency Committees, and the second set at the Joint Economic Committee or all three committees. The first set of reports would mainly concern bank regulation, the performance of banks (banking services and so on), the banking structure (monopoly and competition in banking), and the budgetary and control aspects of the Federal Reserve System. The second set of reports would be the set concerning mainly economic stabilization, the goals of the Employment Act, and monetary and credit policies. This distinction between the reports would f acilitate the long-sought separation of purely administrative functions from the essential policy functions (e.g., 1, p. 346; and 9, pp. 87-88). Such a separation would go a long way toward improving communication with respect to Federal Reserve policy (sec. IJH.A). The major merits in having the Board make a report on future and past aspects of money and credit policies and their relationship to national economic goals are threefold: (1) the reporting would relieve some of the present efforts devoted to trying to second guess the Federal Reserve; (2) the reporting would contribute to narrowing the range of speculation about the Federal Reserve's plans and actions; and (3) the possible need to explain policies, actions, and inactions, would in itself go a long way toward assuring the pursuit of sound policy. In the first two instances, efforts to second guess the Federal Reserve engross a large part of the talents and resources of some very able people, not only in financial institutions but in large nonfinancial ones as well. However, even quarterly reporting and the publication of forecasts would not wholly eliminate speculation about Board plans and actions. There would still be speculation about the revisions in the policy and in the forecasts. In the third instance, the reporting and explanation requirements would provide for "an educational feedback" to the Federal Reserve. As a staff report, Subcommittee on Domestic Finance, concluded (38, p. 83): "Such a feedback is required 185 to assure that mistakes lead to critical reevaluations of operating objectives and methods. Without it, past errors are almost sure to be repeated in future years." The reporting provisions of H.R. 11 should be modified to minimize the reporting on the details of policy and to emphasize the explanations for the policies, particularly those concerning national economic goals. The need to set forth explanations in empirically verifiable form and to present supporting evidence was mentioned earlier (sec. 1.3.A). The report should include balance-of-payments information, and information on the domestic economy such as the level of prices, employment (and unemployment as a percent of the labor force), and the level and structure of interest rates, on the one hand, and rates of change in bank credit (including loans and investments), bank reserves, the money stock, deposit and negotiable-type liabilites of banks, and gross national product, on the other. In minimizing detail in its policy report, however, the Board should be required to do the following: (1) focus primarily on the interrelationships between the money stock and national economic goals as noted earlier (sec. I.3.A) ; (2) present only a general statement of defensive operations; (3) review the relationship between its credit and monetary policies and the economic program of the President (sec. I.5.B) : and (4) comment on structural features of the economy, especially those affecting the attainment of goals (sees. 13. and I.3.B). In particular, the provision in H.R. 11 about "stating in comprehensive detail" should be left open ended—that is, changed to emphasize "verifiable explanations" and the necessary supporting material. The Board should be left free to present what is called for to support its explanations, particularly for the need to vary rates of change in the money stock beyond stated bounds (sees. 1.3.A and I.3.D to F). An annual forecast should be expected every quarter, with attention being given annually in the report to forecasting secular trends. Reporting on prospective changes should take the form of a forecast, both with respect to policy measures and national economic goals. The policy measures in the forecast should at least include the money stock. Under an open-ended arrangement, the Board may also wish to include such interrelated measures as the level and structure of interest rates (as indicated by the spread between rates on long- and shortterm Governments) and the income velocity of money (sec. I.3.C). The national economic goals would at least be given absolute values in the forecasts.24 These may be accompanied by rates of change in stock and flow variables (e.g., the money stock and income), as distinct from interest rate, employment, and price-level variables. From the policy point of view, the forecasts should center about crucial policy variables and target values for national economic goals and the most immediately related measures. Conceivably there could be target values for national economic goals that would differ from the forecasts. For example, there may be price-level and employment targets toward which policy actions and inactions would be directed, but there may be structural forces at work such as those giving rise to cost-push inflation, all such that forecast price and employment values would differ from their target values. 24 The need to give quantitative content to the goals has long been assumed and in some instances recognized (1, pp. 285-305). 21-570—6S 13 m A danger in emphasizing detailed reports as distinct from verifiable explanations in the reporting provision of H.R. 11 is that the request can be met too easily. Congressmen and others might be confronted with computer output from a model with n number of equations and possibly an n number of variables, n being any number, probably a large one. The adoption of the reporting procedure as outlined for H.R. 11, as well as other related changes of H.R. 11, would result in some personnel changes, extending to a number of economists. Even so, the benefits outweigh the cost and risk of putting H.R. 11, with suggested changes, into effect. L5.E.—What costs and benefits would accrue if representatives of the Congress, the Treasury, and the CEA wrere observers at Open Market Committee meetings? Under H.R. 11 Open Market Committee meetings would be Board meetings. But assuming the Federal Reserve structure as a policymaking organization and its attributes in the mid-to-late 1960's, there would not likely be any advantage in having additional representatives present. On the one hand, the meetings, at 3- or 4-week intervals, have been too frequent for purposes of reconsidering policies, particularly those dealing with national economic goals, and, on the other, there have been insufficiently frequent meetings for developing a majority support for an explanation of policy changes. The committee meetings have served mainly as a forum for reviewing economic conditions and arriving at a consensus about policy changes, however vaguely stated, and for issuing instructions in a general form to the manager of the open market account. The attendance at meetings has been large, since those attending possibly include seven members of the Board, selected staff, 12 reserve bank presidents, advisers to the respective president, and advisers to the FOMC as a whole. They are essentially politcal meetings where consensus is sought on policy changes rather than the underlying reasons. The attendance is too large for a working session on fundamentals. The improvements in policy discussions under H.R. 11, with suggested modifications, would follow from three sources: (1) having the policymaking officials at one location for frequent working sessions; (2) having the officials report the targets for policy actions and their relationships to the achievement of national economic goals; and (3) having the policymaking officials responsible for explanations of changes in monetary policy. Reporting provisions with emphasis on empirically verifiable explanations (sec. 1.3.A) should serve to lessen the possibility that meetings of the Board would not degenerate into essentially political meetings, as outlined in the previous paragraph. IT. APPRAISAL OF T H E STRUCTURE OF T H E FEDERAL RESERVE H.R. 11 is a bill intended, among other things, to "improve the coordination of monetary, fiscal, and economic policy." It contains certain provisions to achieve this end. An additional objective—notably, to provide for coordination by the (President—is implied. The two provisions bearing most directly on this objective of coordination by the President are those (1) making the term of the Chairman of the Board coterminous with that of the President, and (2) shortening 187 the terms of office from the present 14 to 5 years. Combined with the transfer of current FOMC functions to the newly reconstituted Federal Reserve Board, these provisions presumably give the President some additional influence over monetary policy via appointments and thereby reduces the independence of the Board. Improving coordination of policies to achieve national economic goals, however, need not necessarily imply greater control by the President. Nor, for that matter, need reductions to five members and 5 year terms be viewed as giving the President too much direct control, given the minor changes suggested in H.R. 11 in this paper. The issue of the Board's independence from some presidential and other influences may be related to an educational function for board members and to the importance of focusing on the goals of the Employment Act. The common argument against independence is that "only the President can choose the monetary-fiscal policy mix appropriate to the given situation with some hope of seeing it implemented" (1, p. 282). The centralized authority can, in setting forth the mix, minimize small mistakes in arithmetic and logic, and otherwise avoid crosspurposes and duplication of efforts. Several common arguments in support of independence are (1) the quasi-judiciary one (i.e., the need for removal or protection from immediate pressures, often arriving from temporarily popular causes (35, pp. 59-62; 38, pp. 16, 23-24, 31-32; 9, pp. 85-86)), (2) the compensating-errors one (i.e., the need of an independent agency to compensate for the errors of other policymakers (14, p. 25)) , and, (3) as some would add, the fewer-but-biggererrors argument (i.e., the argument that centralized authority makes bigger mistakes because, for example, its opinions are taken too seriously and it seeks power and entrenchment). These arguments take various forms. Earlier (sec. 1.5.A), the temptation to the President to pursue short-run objectives, at the overall expense of the loss of idle resources was mentioned. In addition, Arthur Burns comments thusly (4, pp. 78-79) : "Centralization of economic authority in the office of the President has its intellectual appeal, but let us not overlook the protection against the risk of concentrated error that the economy now derives from the dispersal of power in our governmental scheme." Wallich views this type of argument as a pragmatic rather than a logical or economic one [39] ,25 Below, as elsewhere in this paper, the latter argument is transformed into an economic one and due regard is given to the public's right to ultimately control. The economic argument proceeds from the simple Wallich comments on coordination versus independence as follows (39, p. 30) : " I t is popular to say that the President should have power over monetary policy in order to coordinate the two instruments. That implies, however, something that is not at all true, namely, that the President has power over fiscal policy. He does not have such nower. He proposes, Congress disposes. It is not clear where the focus of fiscal policy is ; hut so long as the President does not have power over short-term fiscal policy, I' can see no logical reason to concentrate monetary power in his hands." His argument is simply a pragmatic one : "But the argument really is not a logical or an economic one. It is a pragmatic one. The people who like a softer monetary policy are for coordination ; the people who prefer a harder monetary policy favor Federal Reserve independence. Why is this? When there has been a conflict between the Federal Reserve and the administration, which has been a rare event, really, the pattern has always been very clear : the Federal Reserve is for the harder, and the administration is for the softer policy. Hence, the pragmatic answer is that to choose coordination means to choose a softer policy ; to choose independence is to choose a harder policy. I have a strong suspicion that if the recent trend in appointments to the Federal Reserve Board should continue, and if some day the Federal Reserve should be more liberal than the rest of the Government, some of my friends will discover the virtues of Federal Reserve independence and I shall be arguing for coordination." 26 188 fact that the current state of economic knowledge itself does not justify or support attempts to alter every possible wiggle in the economic indicators through control over credit and monetary variables (sec. IV). What is suggested is that the latter sort of policy may be pursued when empirically verifiable explanations are offered. The added potential influence of the public originates from two possible sources: (1) primarily their elected Representatives in Congress in combination with new reporting procedures (sec. 1.5.D) ; and (2) the possible addition of budgetary controls of the Congress under H.R. 11. The objective is appropriate control over the unaccountable use of independence, rather than its annihilation. The objective in appraising the adequacy of the structure of the Federal Reserve as a policymaking organization need not turn on the need for coordination through the President, so much as the need to more consistently pursue the goals of the Employment Act. The number of policymaking officials.—The most far reaching of the proposals to alter the Federal Reserve is that reducing the number of Board or, viewed differently, voting FOMC members. One's assessment of the most desirable number of Board members and the most appropriate structure for policymaking purposes may depend on one's view of the appropriate functions to be performed by the Board. As emphasis has shifted—first, from the essentially practical exercise of judgment about the adequacy of bank credit to suit the needs of commerce, industry, and agriculture to the impracticality of such judgments, and, then, to notions about explanations of relationships between policy measures and national economic goals—views about the functions of the Board might be expected to have changed. Preliminary questions arise: Is the Board simply a group of overseers for the Congress, with research directors and advisers the true policymaking if not voting figures? Is the Board's function as a policymaking group essentially similar to that of the Council of Economic Advisers, with the obvious differences that the former specializes in monetary and credit control and has the power to implement policies ? Much has been said about the need to enhance the prestige or status of the members of the Board. The Commission on Money and Credit emphasizes it in recommending a reduction in the number of Board members [9, p. 87]. However, a better way to enhance prestige might be to warrant it. Some have suggested that the prestige of the Council of Economic Advisers was enhanced by the type of issues and educational functions it engaged in during the early and mid-1960's. A favorite topic of J. A. Livingston, a columnist, about the time of the April 1967 appointment to the Board was that the Employment Act of 1946 had created the Council of Economic Advisers. He said: As the nation's Supreme Court of economics, the Council outranks the Board as the Supreme Court of credit. On another occasion he said: The influence of the Federal Reserve Board in American economic affairs has been declining since 1946. It has been layered in the Government hierarchy by the Council of Economic Advisers * * *. In assessing the functions of the Council of Economic Advisers, Walter Heller (Chairman, 1961-64) writes of economic education of, 189 by, and for Presidents [24, pp. 26-57]. He views the economic advisers' educational activity on behalf of the President, as: Simply a case of doing on a small scale—though with greater detail, depth, and diligence—what the President does on a grand scale, namely, communication, making the Government's economic policy and actions intelligible to the citizen, a process essential to democracy; and "broadening consensus, carrying the economic gospel not only to the uninformed but to the skeptic and the heathen. Heller points out that through public discussion, Council members broadened the depth of public dialog on economic problems and concepts, tested newT ideas, and prepared the public for the President's subsequent advocation and support of his economic programs. The Council apparently sought at times to create an environment in which certain issues could be discussed by political figures, without risking the political consequences of treading new ground and attempting to broaden public understanding, often of undramatic low7-stake issues. Now, five Governors may be too many, if one takes the view that its members are primarily a group of executive overseers, supervisors of the various activities in the Board's plant on Constitution Avenue. On the other hand, seven or 11 members may seem appropriate if one takes such views as the following: (1) That, as Senator Proxmire [27, p. 1] and others [5, pp. 322-368] emphasize, the Congress has given the responsibility for determining matters involving money under the Constitution, and that Congress has chosen to delegate the exercise of this authority to the Federal Reserve authorities under a trusteeship arrangement with the Federal Reserve as the servant, creature, and agent of Congress, (2) That the Board's functions in its own area are similar to those of the Council, with the important exception that the Board has the authority to implement policies, and (3) That some geographical representation of the policy level in monetary matters is desirable, all as envisioned by the Federal (regional) character of the System in the original Federal Reserve Act, congressional responsibility for money matters under the Constitution, and a proposed educational function for the Board. Some Board members have functioned frequently as discussants of timely economic issues, sometimes including issues before the Congress and sometimes including appeals to relatively parochial groups. The Board's independent position within the framework of Government places it in a position to aid in educating the public about monetary problems through pronouncements, speechmaking, and public appearances. In the 1960's, several areas stood out in which more public education would have helped in achieving national economic goals. These involved, mainly, areas in which conflicts resulted in special governmental objectives, and the use of money and credit tools as means of achieving the more immediate national economic goals, as well as such ultimate objectives as maintaining a relatively free enterprise economy and a reasonably "efficient system of world commerce, finance, and production." The areas in question included (1) the U.S. balance of international payments, (2) home construction, and (3) structural unemployment. Traditionally and for the foreseeable future, monetary policy measures and economic policies are fraught with controversy. This supports 190 the need for some geographical representation. Moreover, there are styles in economics—an eastern view and a western view, a predominant Harvard view and then a Chicago view. There is "occupational myopia," "tunnel vision," and so on. There are, as well, individual and regional preferences for certain public policies and the means of achieving them. To add compensating sources of stability two features have been emphasized: (1) the geographical one, and (2) an empirical one, the notion of presenting empirically verifiable explanations for policy changes (sec. 1.3.A). There is, as Heller points out [24, pp. 7-8], a greater consensus among economists today than in the 1930's. Taking a long view there has been a decline in "the warfare of rival ideologies." There has also been some refinement of standards with respect to empirical relevance. Emphasis on the need for potential support from empirical evidence may be viewed as the essential ingredient for developing an understanding of the interrelationships between policy measures and national economic goals. H.R. 11 relieves the Federal Reserve's regional bank presidents from policymaking functions. This is not undesirable, in view of the following: (1) emphasis on a separation of policymaking and administrative functions, (2) the recognition of the role of regional bank presidents as chief executive officers of operating banks, and (3) the recognition of geographical regions in Board appointments. The Federal Reserve System, including its regional banks, has developed a complex and useful information system for identifying changes in business conditions [2, pp. 282-288]. Of key importance in this network is the maintenance on the part of the regional banks of contacts with business economists, personnel and purchasing departments, and decisionmakers in business and economic communities. To be sure [2, p. 287], "from such contacts they [the regional banks] often can distill a sense of changing attitudes or intentions before the consequences are reflected in economic statistics." This kind of informational system probably in part explains w^hy the Federal Reserve has a relatively good record of identifying turning points in business conditions [13; and 14, pp. 25-28]. This system need not be lost by implementing the provisions of H.R. 11. Varying the weight to the various aspects of the foregoing discussion of the functions and the appropriate number of Board members, a board of from three or less to 11 members may be called for. A threemember board may follow from placing the weight primarily on the executive function. Shifting the weight to the educational and geographical aspects of membership would increase the number of members. The Chairman, in any case, is the chief executive officer. Among other executive functions, he votes to break the ties in the voting on policy matters. In coordinating the Board, moreover, the Chairman may give recognition to various functions and interests of members, as well as to the need to implement and defend specific credit and monetary policies. Special functions to be emphasized may include the essential banking functions (sec. 1.5.D), as well as the credit and monetary policy functions with interrelated degrees of emphasis on fiscal policy, structural-type economic problems (sees. 1.2 and I.3.B), the sectors of the domestic economy and international financial matters. To achieve the objectives of H.R. 11, given the above criteria, a relatively large board with emphasis on adherence to a form of rule is 191 preferable as noted earlier (sees. 1.3.A and LSI) to F). Even so, shorter terms than the present 1-1 year ones are desirable to permit the possibility of more frequent additions of freshness to the Board's deliberations. The term of appointments should depend in part on the number of members—for example, 5-year terms for a five-member board, although 10-year terms may be preferable for such a board to avoid a given President's appointing all members within a span of one term plus 1 year. In particular, the timing of appointments should be "staggered" as in the past to prevent a packing of the Board over a short period by any single President . Other provisions.—Other provisions of H.R. 11 call for (1) the retirement of Federal Reserve bank stock, (2) an audit of the Board and the regional banks and their branches, and (3) the removal of the Federal Reserve System's exemption from congressional appropriations procedures, all as have been discussed on other occasions over the years [5, pp. 353-391]. The first of these has indeed been wTidely discussed [9, p. 91]. The retirement of the stock would eliminate the symbol of the early view of the Federal Reserve as a joint private and public enterprise. It would also help remove the stigma of a private, banker-dominated organization controlling the money supply. The last two provisions have often been related to the question of the Federal Reserve's independence. An audit by the general accounting office seems appropriate [38, pp. 85-90; 5, pp. 353-368]. The need for removing the Federal Reserve's exemption from appropriational procedures is less clear. The flexibility enjoyed by the Federal Reserve under present procedures has some advantages. Its reputation for integrity and honesty among bankers and the business community is good, in part perhaps because of the role of certain individuals in the System, in part because of the association of the Chairman with the regained freedom of the Federal Reserve in credit and monetary matters following the Treasury-Federal Reserve accord of 1951, and in part perhaps because of appointments to the boards of directors of the regional banks. The simple possibility of being subjected to the congressional budgetary process probably in itself assures a good deal of prudence and care in the scrutiny of expenditures. On the other hand, there does appear to be a need for a reporting by the Board and by the regional banks on some common and systematic basis, particularly for expenditures on research and information gathering operations, public relations, educational programs, and the cost of producing various services. There is the possibility that funds are not distributed with sufficient regard for the size of the regional operations, the geographical regions themselves, and the need for a broader and more economical distribution of resources about the country. III. MONETARY POLICY, 1964-66 Much may be said about various aspects and effects of Federal Reserve policy since 1964, with emphasis (1) on home construction (sees. 14. and 1.5. B and C), (2) on the regulation of deposit interest rates (sec. 1.5. B and C), (3) on the wide variability in the rate of change in bank credit (4, pp. 23, 71, 81,111-112,119,122), and (4) on the less directly controlled sector comprising manufacturing corporations (15 and 16). This section focuses on the latter for several reasons: (1) ex- 192 penditures by manufacturing corporations are important in determining gross national product ; (2) the Employment Act's emphasis on achieving goals in a relatively free market economy would seem to require a justification for Federal Reserve policy with special regard to the less directly controlled areas of the economy; and (3) references to capital expenditures by business firms and bank loans to business were prominent in discussions of Federal Reserve policy in 1966. The 1964—66 period as a whole is focused upon (1) because it was the second of the major post-World War II booms in capital expenditures since the initial postwar adjustment, and (2) because weaknesses in the Federal Reserve's approach to economic stabilization and deficiencies in its use of controls may be illustrated with reference to the period. Weaknesses in the Federal Reserve's approach to economic stabilization.—The weaknesses in question are actually twofold: (1) there is insufficient recognition of the Board or FOMC as public educational bodies, and (2) under the present organization the policymakers as a group become preoccupied with the banking mechanism as such and apparently harbor preconceptions, harking back to an earlier banking period, about the role of bank loans to businesses, as illustrated later. In the first instance, in varying the rates of change in bank credit and the money stock in order to achieve certain goals, residual or side effects follow because of structural features of the domestic economy (sees. 1.2 and 1.3.B). The Federal Reserve has no control over this class of effects, in the sense that congressional actions to change the structure are outside of its special field of authority. Its function or responsibility then becomes a matter of reporting to the Congress and educating the public about the need for legislative action. But this is insufficiently recognized. In the period since 1964, stabilization polic}r may have been pursued as if it were to operate on expenditures and achieve employment goals, all independently of persistent U.S. deficits in the balance of payments, possible structural unemployment (i.e., unemployment due to an inadequate alinement of job skills with job vacancies), and effects on home construction. However, such a policy as outlined was not consciously pursued, judging from a review of the policy-oriented literature of the period. Quite possibly some amount of inflation may have been accepted as a means of temporarily reducing the percentage of employment, even at the possible cost of a loss of overall efficiency and employment of resources in the long run, and even though such a policy would have contributed to continued payments deficits, higher interest rates, and problems in the home construction industry. This touches on the so-called trade offs—that is, you trade-off a little of one undesirable development for another. If trade-offs are being made, then the Congress and the public should be informed and educated. H.R. 11 provides for better reporting (Sec. 1.5.D). Moreover, the policymaking officials of the Federal Reserve have tended to speak to some extent on the so-called structural problems in question. But this educational function needs to be more formally recognized in any restructuring of the Federal Reserve as a policymaking organization. The current provision in H.R. 11 for five governors offers little as- 193 surance that the respective districts of the country will adequately participate. Deficiencies in the Federal Reserved use of controls.—Illustrated below are (1) inadequacies of the Federal Reserve's view of its control over bank loans to business and over business expenditures, and (2) deficiencies in the Federal Reserve's use of controls. In general the main criticism is this: the Federal Reserve's preoccupation with the apparent security inherent in information gathering and with the appealing details of the banking mechanism and simple linkages predominates over the need for a broader conception of the developing situation. There are detailed criticisms, too. There are several principal elements in the ensuing illustrations. First, abstracting from money-market "noise" and defensive aspects of Federal Reserve operations (sees. 1.5. A and 1.5.D), increases in the rates of increase in bank credit (and therefore banks' deposit liabilities) contribute to a faster rise in business conditions and interest rates once an expansion phase of business conditions is unfolding [14]. Rates of change per annum in bank credit of 8.1, 8.5, and 13.8 percent occurred in 1963, 1964, and 1965, respectively; changes of 3.8, 4.3, and 4.9 percent occurred in the narrowly defined money stock during the respective years, reflecting in part drains into time deposits; and interest rates continued to rise. There wTas the $11 billion tax cut in 1964, and later increases in defense and other expenditures. The analytical conclusion is that by accelerating the rise in bank credit (and related bank liabilities), the Federal Reserve contributed to the ensuing boom and the accompanying increase in interest rates. As the market yield on 3-month Treasury bills increased from 3.64 to 4.08 percent during the year ended in November 1965, it become necessary to raise the discount rate from 4 to 41/2 percent in December of 1965, in part just to maintain the discount rate's relative position in the structure of rates. This was later described by one Board member as a public announcement of a shift to credit restraint [2, p. 277]. On the one hand, credit and the related monetary policies contribute to an expansion in business conditions and the accompanying increases in market rates of interest and, on the other, the Federal Reserve takes credit for a shift to credit restraint when an increase in the discount rate is necessitated by expanding business conditions and rising market rates of interest to begin wTith. This is peculiar. The several additional ingredients of the present illustration are as follows: (1) The noncash liquidity of manufacturing corporations may be indicated by their noncash liquid assets net of bank loans as a liability, both in relation to asset size; noncash liquid assets (say, governments, although the category may include negotiable CD's as well) may be liquidated in exchange for a reduction in bank loans, and an increase in bank loans may serve as a substitute for the liquidation of governments as a source of funds; and changes in governments and bank loans, together and in combination with changes in cash, form the principal means of adjusting liquidity, although these adjustments need not relate exclusively, or even primarily to the need for temporary funds [16, ch. 2]. (2) Corporate liquidity is related inversely with 194 business conditions and directly with the planning of capital expenditures [15, pp. 66-101]. (3) As firms increase in asset size (or by average firm size in the case of industry groups) their noncash forms of liquidity increase; in other words, as firms increase in asset size, noncash liquid assets (i.e., marketable assets such as governments, commercial paper, and negotiable CD's) increase significantly more and bank loans significantly less in relation to asset size [16, chs. 3 and 4]. (4) Commercial banks have a strong preference for extending "commercial" or business loans to credit-worthy businesses (as indicated, e.g., by their liquidity) when there is a strong demand for such loans [15, pp. 77-78]. Now the first three of the immediately preceding ingredients are related to the planning of the financing of capital expenditures by the mature industrial corporation. As described by Gralbraith [22], the mature corporation is related to the need to plan as a result of advanced technology, and by, among other things, the needs of the decisionmaking groups for some independence from outside disturbances. A major means of achieving this is the planning of financing. According to Galbraith, "no form of market uncertainty is so serious as that involving the terms and conditions on w7hich capital is raised." He says [22, p. 39], "apart from the normal disadvantages of uncertain price, there is danger that under certain circumstances no supply will be forthcoming at an acceptable price." Large firms plan financing to a greater extent than smaller ones and there is some evidence that these plans take account of expected patterns of cyclical development [15]. Liquidity is built up in advance of the bulk of capital expenditures, and worked down as the boom in expenditures progresses. The reduction takes two forms and the one may be traded off for the other. They include (1) a reduction in the marketable-type liquid assets in relation to size and (2) an increase in bank loans. One of the differential effects of not planning expenditures to a relatively high degree is shown in figure 1. Accepting liquidity as a constraint, the series in the figure for the 1955 to 1957 and the 1964 to 1966 capital booms suggest that liquidity contributes to differential patterns of capital outlays, with the thousand largest manufacturing corporations increasing their proportion of the total to over 80 percent. The reason the 1955 to 1957 and 1964 to 1966 patterns do not coincide with the 1958-60 expansion in business conditions is that the expansion must be sustained for some time before the reductions in liquidity become a strong constraining factor on either overall capital expenditures or different sets of expenditures. 195 — — — All Manufacturing — Durable Goods Percent FIGURE 1. CAPITAL EXPENDITURES BY LARGE (AND LARGE DURABLE G O O D S ) MANUFACTURING FIRMS AS PERCENTAGES OF CAPITAL EXPENDITURES BY ALL (AND ALL DURABLE G O O D S ) MANUFACTURING FIRMS N O T E : T h e p e r c e n t a g e s are computed from NICB estimates of expenditures b y the thousand largest m a n u f a c t u r i n g corporations (and durable goods m a n u f a c t u r i n g corporations among t h e m ) and from Commerce-SEC estimates of expenditures b y all m a n u f a c t u r i n g firms (and durable goods m a n u f a c t u r i n g firms among t h e m ) . Estimates are s e a s o n a l l y a d j u s t e d . 196 67 Biweekly FIGURE 2 , BANK LOANS (TO BUSINESS) AND INVESTMENTS BY WEEKLY REPORTING MEMBER B A N K S , 1964-1966 ^Changes (in percentages) are changes from the comparable year-ago dace, "^Data are biweekly averages for weekly figures. ^Revision in the series. allow for the revision. Percent changes, however, are adjusted to Source of data: Board of Governors of the Federal Reserve System. 197 The final ingredient—commercial banks' preferences for business loans—of this overall illustration is reflected in figure 2. Year-to-year changes (as percentages) in bank loans were increasing in the 1964-66 period until about August of 1966. Year-to-year changes (as percentages) in bank investments were declining, and indeed the 1965-66 changes were negative. Given the foregoing ingredients of this overall illustration, these patterns—both of the changes in capital expenditures and of changes in bank loans—are expected. They are expected when a boom proceeds temporarily at a level that cannot be sustained as a matter of secular change. In order to control and manipulate the changes, as distinct from trifling with them, the entire process must be kept in view. Federal Reserve actions under regulation Q in 1966, and their September 1 letter in particular are symptoms of a poorly conceived, poorly articulated, and poorly managed set of developments. Special' administratwe technique.—As noted earlier (sec. 1.5.B. and C), as business conditions continued to expand in September 1966, the ceiling interest rates on CD's under $100,000 was reduced to 5 percent and that on CD's of $100,000 and over was left at percent while market rates of interest continued to exert additional credit restraint via a runoff in CD's. Twice during the summer reserve requirements on member bank time deposits of over $5 million were raised by 1 percentage point. But these were belated actions coming late in a boom that had been encouraged by earlier credit and money supply changes. Corning as they did, with banks facing strong demand for business loans, they could only force further the liquidation of municipal securities by banks. Having thus contributed to the liquidation of municipals, the prospect of further increases in rates of interest, and unsettled financial market conditions, the Federal Reserve intervened in an attempt to exercise control over bank loans via special administrative technique [2, pp. 280-281, 288-290]. Questions are raised then both about the use of the technique itself and its likely differential impact on businessfirmsof different size. The technique for effecting control over bank lending was set forth in a special letter from the Federal Reserve banks to all their member banks in a letter. In August 1967 it was described in First National City Bank's Monthly Economic Letter as "the FRB's now famous September 1 letter to member banks." The technique was adopted and the letter sent with a view to encouraging moderation in the expansion of business loans and in the liquidation of investments. In the letter it was stated that the latter objective "will be kept in mind by the Federal Reserve banks in their extensions of credit to member banks through the discount window." Continuing, the letter said that "banks adjusting their position through loan curtailment may at times need a longer period of discount accommodation than would be required for the disposition of securities." The view was expressed that "a slower rate of business loan expansion is in the interest of the entire banking system and the economy as a whole." 198 The September 1 letter and the technique outlined in it appear as another example of the Federal Reserve's preoccupation with the details of the banking mechanism. In the long-run the administrative control's only likely effect would be for larger firms, with their strong propensity to plan, to reduce even further their reliance on bank loans as a potential source of funds. In the immediate situation of September 1966, the major possible effect of the letter was likely discriminatory with respect to intermediate to small firms. Although large firms were increasing their use of bank loans in 1966, the intermediate to small firms were in relation to their size, still more dependent on the loans. Thus, because of small firms' dependence and their more limited access to other means of effecting cash adjustments, possible effects of the attempt to control bank loans via administrative technique were discriminatory. IV. RULES ( " G U I D E L I N E S " ) , ECONOMIC KNOWLEDGE AND T H E FEDERAL RESERVE: A HISTORICAL PERSPECTIVE The history of rules to eliminate administrative discretion in the management of money is a long one. In some respects it may be seen as being broadly based in doctrine opposing the centralization of power generally. The original Federal Reserve Act and the notions surrounding it provide one such example. The System was to operate automatically according to certain rules and related notions with the view to expanding bank credit (and presumably money) to satisfy the needs of trade (and presumably income), and the monetary policy function of the System, such as it was, was decentralized. The rules and the automatic functioning of the System centered about the "real bills" doctrine, the gold-flows mechanism, and a simple quantity theory of money. The only credit control instrument at the disposal of the System in the early years was the discount rate, or more generally the discount mechanism, and changes in the rate were initiated at the regional bank level, as they are today, only the Board must approve. The System was to perform as a "lender of last resort" and provide additional liquidity to commercial paper originating at the commercial banks as they extended credit. This was done by the discounting of commercial paper ("real bills") or by advancing credit in a properly secured form. The paper was real in the sense that the credit originated in the extension of loans for the purchase of goods in shipment and inventories. On a temporary basis, as in the case of a seasonal defense against crisis, the Federal Reserve could serve as lender to the commercial banks and therefore satisfy a temporary need for credit and possibly avert the sort of recurring financial crisis of the pre-Federal Reserve (i.e., pre-1913) era. Moreover, a properly functioning gold-standard mechanism could serve as an overall regulator, both of the Federal Reserve System and the growth of bank credit (and, therefore, the money supply). As trade, credit, and the money supply expanded excessively, the price level would rise in relation to price levels abroad and set in motion an outflow of gold. The latter would serve to constrain the growth of bank reserves and the money supply (and thus the level of prices, via the simple quantity theory of money). A decline in the domestic price level in relation to foreign price levels had the 199 reverse effects. All the Federal Reserve had to do was to adhere to the rules. There was little room for discretion. All was to work automatically. Economic stability was to be achieved, since it was thought to be the result of the malfunctioning of the financial system to begin with. In the 1930's and later, economists came increasingly to emphasize the interrelationships between money and the financial and real good sectors in analyzing cyclical and other changes in output, employment, and prices. Paradoxical as it may seem, however, the growth of monetary analysis in the 1930's was accompanied by a widespread emphasis on the ineffectiveness of monetary policy under alleged liquidity trap conditions, such as may have prevailed in the 1930's. The renewal of widespread belief in the effectiveness of monetary policy depended largely on post-World War II research and the large amount of empirical work since the advent of the modern computer. A large portion of this depended on certain reactionary souls, mainly at the University of Chicago, who maintained a rather militant faith in the efficacy of money, even when it was placed by some in a secondary position in relation to fiscal policy [19]. As time passed in the 1920's, the Federal Reserve System, the domestic economy, and the gold-flows mechanism did not work exactly as envisioned. The Federal Reserve Act had permitted the regional banks to buy some securities in the open market as a means of obtaining earnings for operating expenses, and soon this opened the door to the prospect of exercising control over bank reserves via open market operations. The opportunity for the exercise of discretion by bank officials—particularly at the Federal Reserve Bank of New York, under the leadership of Benjamin Strong—was broadened. The range for discretionary administration was further broadened and formally recognized in the Banking Acts of 1933 and 1935, following the catastrophic economic collapse of the early 1930's. Power in the Federal Reserve shifted to the Board in Washington; general credit controls came to include open market operations and changes in reserve requirements, all as emphasized in money and banking texts today. Thus, observing the enhanced discretionary powers and the Federal Reserve's traditional emphasis on reserves and bank credit, observing an apparent loose link in the 1930's between bank reserves and the money stock, and embracing a quantity theory of money with emphasis on the level of prices—doing all of these things—Irving Fisher put forth his 100-percent reserve scheme in 1935.26 Henry Simons, also, advanced proposals for monetary reform in the 1930's. In addition, he vacillated between favoring a rule for the money stock or an instruction to the authorities to keep the price level stable [21]. And, finally, operating partly in the Fisher-Simon tradition and partly in the empirical tradition of the National Bureau of Economic Research, 26 Under the 100-percent reserve scheme, the Federal Reserve—or a special currency commission—would take over all of the assets of the commercial banks and the bainks would in turn be required to hold 100 percent reserves against demand deposits. In this way fractional reserve banking would be eliminated and open market purchases would increase the money supply directly without any subsequent multiplicative effects resulting from increments in reserves and fractional reserve requirements. The banks, of course, would be compensated ; they would charge their customers for service, or as later suggested by Milton Friedman, receive interest on their reserves. Also, the 100-percent plan, as noted by Bronfenbrenner (3), might require other changes in financial institutions, e.g., investment trusts "to supply funds to the traditional small-business customers of commercial banks." All of this, of course, was outside the mainstream of economic analysis in the 1930's. 200 observing the economic consequences of the exercise of discretion by Federal Reserve authorities, and studying wide variations in the rate of change in the money stock—doing all of these things—Milton Friedman and his followers of the post-World War II were—and currently are—strong advocates of the need for some monetary rule. Their advocacy and research in support of the need for some monetary rule, moreover, came to have widespread impact on economic thinking in the 1960's. Indeed, congressional hearings before the Senate-House Economic Committee in May 1968, dealt essentially with a rules proposal and the possible need for constraint on Federal Reserve variations in the rate of change in the money stock [27]. Later, William Proximire, chairman of the Economic Committee, spoke of the need for the Federal Reserve "to adopt a constant and moderate monetary policy." He noted that the committee recommends reports by the Federal Reserve if during any quarter the increase in the money stock is less than an annual rate of 2 percent or more than 6 percent. Surveying the record Proxmire said, "the Federal Reserve has a record of deepening almost every recession or depression we have suffered in the last 30 years by reducing the money supply." Continuing he said, "it has often excessively increased the money supply to fan the flames of inflation when the economy has been booming." Bronfenbrenner* s review of monetary rides.—Bronfenbrenner ranges widely over the rules-versus-discretion literature (3). In so doing, he refers to an irrelevant state reached by comparisons of economic performance under a fixed rule proposal, on the one hand, and discretionary policy, on the other. A main point he makes is that "even if, in ordinary times, a given rule performs no better than ordinary authorities, one may advocate it for the same reason be buys life insurance 'loaded' in the company's favor." Bronfenbrenner also advances a proposal similar to Friedman's but, at the same time, includes a package of guideposts for adjusting the money supply target. The proposal is that the money stock grows at the same rate as income, including a labor force and an average productivity component. To compensate for adjustments in the money stock that may be called for because of special developments, Bronfenbrenner introduces changes in velocity. Thus, before acting on the change in the rate of change in the money stock, the authorities can make corrections "for the effects of changing taste and financial innovations on the velocity of the monetary circulation." This, of course, broadens the policy discussion to include the whole host of analyses of the velocity-interest rates association [17]. Bronfenbrenner apparently thought that the introduction of guidelines for compensating changes in the money stock will make the rule proposal more acceptable, more relevant to the world we live in. In contrast, all of the invited participants in the May 1968 hearings on rules versus authority wTere opposed to the idea of guidelines for compensating changes, including in particular the guidelines suggested by Representative Reuss [27, pp. 229-231]. Moreover, as Selden noted, such secular changes as the post-World War II rise in the income velocity of money could readily be compensated for in the framework of advocates of rules proposals. He said, for example: "a simple procedure would be an annual review of the guidelines to determine whether they need revision." Continuing, he said, "the guidelines could 201 be adjusted gradually to take care of longrun changes in the demand for money/' Hearings.—The May 1968 hearings before the Joint Economic Committee were on the question of whether the Federal Reserve should confine changes in the growth rate for the money stock within limits such as 3 to 6 percent per year. Much of the discussion centered about suggestions by Representative Henry S. Reuss that were made available prior to the hearings. Reuss stated in the form of a rule simple limits for the rate of change in the money stock per annum. These the membership of the Joint Economic Committee generally subscribed to,27 but Reuss went further and also stated specific qualifications to the rule as a basis for gei:erating some discussion. Responses to the Reuss proposal and related discussion and interest on the part of the Joint Economic Committee have apparently brought out the following: the inability of the Federal Reserve to state any specific and consistent criteria for monetary and credit policy; a fairly widespread agreement among participants in monetary policy discussions over the desirability of having Congress set upper and lower limits to the growth rate per annum for the money stock, subject to the need to give explanations to the Joint Economic Committee for growth rates extending outside of the limits for any given quarter; and the undesirability of listing specific exceptions to the limits on the rate of growth of the money stock. Knowledge and the mystique.—A portion of the discussion of rules and discretion centers about knowledge or lack of it. Most participants to the discussion agree that monetary and banking mechanics and phenomena are complicated, but then part over the question of knowledge. Traditionally the rules proponents have said that we do not have sufficient knowledge of the effects and lags in effects of policy to successfully use it as the Federal Reserve has sought to use it. They conclude that in the absence of such knowledge some simple rule, such as stability in the growth rate for the money stock, is best. When we have the knowledge, then we make the departure. The proponents of discretion, on the other hand, have seemed to assume that the knowledge exists or, at least, that judgment about the need for a given policy was superior to any simple rule. Even so, we simply note that central banking matters have often relied on a mystique. Those invoking the mystique as a substitute for knowledge have often seemed to present as their best defense, (1) an acquaintance with a frustrating array of facts and details, and (2) the attitude that study and research would confirm the validity of their new. This additional research is always in a promising future, despite all that has historically been completed. An example of the first characteristic is Mitchell's statement that "excessive concentration of our attention on any single variable, or even on any single group of related variables, would likely result in a potentially serious misreading of the course and intensity of monetary policy." An example of the second characteristic is the promise of the Fed-MIT, special-purpose, policy model. The challenge to the mystique has come from an outpouring of results from statistical analyses as well as from other research and 27 For an outline of the background of the committee's rules proposal, see Report of the Joint. Economic Committee (28, pp. 16-18). 21-570—68 14 202 writing, all from a variety of sources. Interestingly—at least with respect to the characteristic defenses of discretionary monetary policy—discussion in the forum of the Joint Economic Committee lias led to the position that the rule can be abandoned when an explanation can be given for doing so. Likely traditional defenses will suffice as explanations for some time. Even so, a course of developments with respect to the rules-discretion controversy seems to be indicated, notably: justify deviations from the simple rule with empirically verifiable explanations. SELECTED REFEKENCES 1. Barger, Harold, The Management of Money: A Survey of American Expert ence (Chicago: Rand McNally & Co., 1964). 2. Brimmer, Andrew, "Tradition and Innovation in Monentary Management," in Alan D. Entine (ed.), Monetary Economics: Readings (Belmont, Calif.: Wadsworth Publishing Co., 1968). 3. Bronfenbrenner, M., "Monetary Rules: A New Look," Journal of Law and Economics, October 1965. 4. Burns, Arthur F., and Paul A. Samuelson, Full Employment, Guideposts and Economic Stability (Washington, D.C.: American Enterprise Institute for Public Policy Research, 1967). 5. Clifford, Jerome A., The Independence of the Federal Reserve System (Philadelphia : University of Pennsylvania Press, 1965). 6. Chandler, Lester V., Benjamin Strong (Washington, D.C.: The Brookings Institution, 1958). 7. Christian, James W., "A Further Analysis of the Objectives of American Monetary Policy," Journal of Finance, June 1968. 8. Cloos, George W., "Pushing on a String: Monetary Conditions from the 193738 Recession to Pearl Harbor," Financial Analysts Journal, January-Februairy 1966. 9. Commission on Money and Credit, Money and Credit: Their Influence on Jobs, Prices, and Growth (Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1961). 10. de Leeuw, Frank, and Edward Gramlich, "The Federal Reserve-MIT Econometric Model," Federal Reserve Bulletin, January 1968. 11. Dewald, William G., and William E. Gibson, "Sources of Variation in Member Bank Reserves," Review of Economics and Statistics, May 1967. 12. Federal Reserve System Committee, "Reappraisal of the Federal Reserve Discount Mechanism," Federal Reserve Bulletin, July 1968. 13. Fels, Rendigs, and C. Elton Hinshaw, Forecasting and Recognizing Business Cycle Turning Points (New York: Columbia University Press for the National Bureau of Economic Research, 1968). 14. Frazer, William J., Jr., "Monetary Policy, Monetary Operations and National Economic Goals," Schweizerische Zeitsohrift fur Volkswirtschaft und Statistik, March 1968. 15. , "Monetary Policy, Business Loans by Banks and Capital Outlays by Manufacturing Corporations," Southern Journal of Business, April 1968. 16. , The Liquidity Structure of Firms and Monetary Economics ( dissertation approved at Columbia University, Department of Economics, February 1968). 17. , The Demand for Money (Cleveland: The World Publishing Co., 1967). 18. , and William P. Yohe, Introduction to and Analytics and Institutions of Money and Banking (Princeton, N.J.: D Van Nostrand Co., Inc., 1966). 19. Friedman, Milton, "Factors Affecting the Level of the Rate of Interest," Conference on Savings and Residential Financing: 1968 Proceedings (Chicago: U.S. Savings & Loan League, forthcoming). 20. , "The Role of Monetary Policy," American Economic Review, March 1968. 21. , "The Monetary Theory and Policy of Henry Simons," Journal of Law and Economics, October 1967. 22. Galbraith, John Kenneth, The New Industrial State (Boston: Houghton Mifflin Co., 1967). 23. Gibson, William E., "The Effects of Money on Interest Rates," Staff Economic Studies, No. 43, Board of Governors of the Federal Reserve System, March 1968. 203 24. Heller, Walter W., New Dimensions of Political Economy (New York: W. W. Norton & Co., Inc., 1967). 25. Haywood, Charles F., and Charles F. Linke, The Regulation of Deposit Interest Rates (Chicago: Association of Reserve City Bankers, 1968). 26. Hendershott, Patric H., The Neutralized Money Stock (Homewood, 111.: Richard D. Irwin, Inc., 1968). 27. Joint Economic Committee, Congress of the United States, Hearings, Standards for Guiding Monetary Actions (Washington, D.C.: U.S. Government Printing Office, 1968). 28. Joint Economic Committee, Congress of the United States, Report, Standards for Guiding Monetary Actions (Washington, D.C.: U.S. Government Printing Office, 1968). 29. Kane, Edward J., "Is There a Predicted Lock-In Effect?: A Look at Cross Section Experience in 1966." Technical Supplement to "Commercial Bank Tax Swaps," New England Business Review, March 1968. 30. , and Burton G. Malkiel, "The Term Structure of Interest Rates: An Analysis of A Survey of Interest-Rate Expectations," Review of Economics and Statistics, August 1967. 31. Malkiel, Burton G., The Term Structure of Interest Rates: Expectations and Behavior Patterns (Princeton, N.J.: Princeton University Press, 1966). 32. Meiselman, David, "The Policy Implications of Current Research in the Term Structure of Interest Rates," Conference on Savings and Residential Financing: 1968 Proceedings (Chicago: U.S. Savings & Loan League, forthcoming). 33. , "The New Economics and Monetary Policy," Financial Analysts Journal, November-December 1967. 34. Morrison, George R., Liquidity Preference of Commercial Banks (Chicago : University of Chicago Press, 1966). 35. Patman, Wright, "What Improvements are Needed in the Money Systems?" in Richard A. Ward (ed.), Monetary Theory and Policy (Scranton, Pa.: International Textbook Co., 1966). 36 . , "Supplementary Views," Report of the Joint Economic Committee, Congress of the United States, on the January 1965 Economic Report of the President (Washington, D.C.: U.S. Government Printing Office, 1965). 37. Roosa, Robert V., Federal Reserve Operations in the Money and Government Securities Market (New York: Federal Reserve Bank of New York, 1956). 38. Subcommittee on Domestic Finance, House Committee on Banking and Currency, Staff Report on Hearings, The Federal Reserve After 50 Years (Washington, D.C.: U.S. Government Printing Office, 1964). 39. Wallieh. Henry C., "Monetary versus Fiscal Policy," Conference on Sowings and Residential Financing: 1966 Proceedings (Chicago: U.S. Savings & Loan League, 1966). STATEMENT 0E MILTON FRIEDMAN, UNIVERSITY OF CHICAGO Unfortunately I cannot reply in full to your interesting question to economists because other commitments make it impossible for me to take the time that would be required. As a poor substitute, may I simply record my own views briefly on the appropriate guidelines for monetary policy. In my opinion, two points are critical—one with respect to the quantity of money, the other with respect to interest rates. 1. In the present state of our knowledge, I believe that the best— or least bad—guideline for monetary policy is steady growth of the quantity of money at a rate that on the average will mean stable prices of final products. The precise growth rate required for this purpose depends on the specific definition of money adopted. For a definition corresponding to currency plus all commercial bank deposits adjusted— demand and time—the appropriate rate is around 5 percent per year. For a definition limited to currency plus adjusted demand deposits only, the appropriate rate is a trifle lower. In my opinion, it would be desirable for Congress to instruct the Federal Eeserve to adopt this 204 policy. That would assure that the Federal Reserve System would provide a steady and stable background for private and public economic policy, instead of being itself a source of instability as it so often has been in the past and as it is currently being at this very moment. If Congress does not legislate this rule for the Federal Reserve System, the next best alternative is the one suggested by the Joint Economic Committee in its recent report; namely, that the Federal Reserve at least be required to report to Congress when it permits or forces the money supply to grow at rates outside a range such as the 2 to 6 percent per year specified by the Joint Economic Committee. 2. On interest rates, the desirable policy is to permit interest rates to be determined completely by free markets. The present prohibition on the payment of interest on demand deposits should be removed. The present authority given to the Federal Reserve System and to other governmental financial agencies to set the ceilings on the rates of interest that commercial banks and other financial institutions can pay on time deposits should be eliminated. Discounting should either be abolished or the discount rate should be linked to market rates and set higher than market rates so that it will be a penalty rate. The present limitations on the interest rates that the Government may set on its debt obligations should be removed. A free competitive market in loanable funds is no less desirable than a free and competitive market in other goods and services. The problems of the organization of the Federal Reserve System are extremely important so long as the two guidelines just outlined are not in effect. If these guidelines were put into effect, the problem of organization would become of little importance because the present power of the Federal Reserve System to introduce instability into the economy would be eliminated. As a matter of principle, I am opposed to independence for the Federal Reserve System. Monetary policy should be administered by a governmental agency in the regular administrative hierarchy ultimately responsible to Congress. I have discussed this issue at length in a paper entitled "Should There Be an Independent Monetary Authority ?" (in Leland B. Yeager, ed., In Search of a Monetary Constitution, Harvard University Press, 1962, pp. 219-243) of which I am enclosing a reprint. A full discussion of my views on monetary policy is presented in my book entitled A Program for Monetary Stability. Experience since this book was written has given me no reason to alter the major recommendations it contains. More recently, I have discussed the general problem of monetary policy in my presidential address to the American Economic Association in December 1967 which was published in the March issue of the American Economic Review under the title of "The Role of Monetary Policy." I have discussed current monetary policy in a number of columns in Newsweek. I am enclosing reprints of these items. In closing, may I express my appreciation and the appreciation of other academic students of monetary problems for the important role which you have played in keeping these significant issues in the forefront of public discussion, for assembling relevant evidence, and stimulating the Federal Reserve System to examine its owrn policies and procedures critically from time to time, as well as explaining them 205 to the public at large. My best wishes to you in the continuation of your good work. (The supplementary materials follow:) [From Leland B. Yeager (ed.), In Search of a Monetary Constitution, Harvard University Press, 1962] Cambridge, Mass.: SHOULD THERE B E AN INDEPENDENT MONETARY AUTHORITY? (By Milton Friedman) The text for this paper, to paraphrase the famous remark attributed to Poincare, is, "Money is too important to be left to the central bankers." The problem that suggests this text is the one of what kind of arrangements to set up in a free society for the control of monetary policy. The believer in a free society—a "liberal" in the original meaning of the word, but unfortunately not in the meaning that is now current in this country—is fundamentally fearful of concentrated power. His objective is to preserve the maximum degree of freedom for each individual separately that is compatible with one man's freedom not interfering with other men's freedom. He believes that this objective requires power to be dispersed, that it be prevented from accumulating in any one person or group of people. The need for dispersal of power raises an especially difficult problem in the field of money. There is widespread agreement that government must have some responsibility for monetary matters. There is also widespread recognition that control over money can be a potent tool for controlling and shaping the economy. Its potency is dramatized in Lenin's famous dictum that the most effective way to destroy a society is to destroy its money. It is exemplified in more pedestrian fashion by the extent to which control over money has always been a potent means of exacting taxes from the populace at large, very often without the explicit agreement of the legislature. This has been true from early times, when monarchs clipped coins and adopted similar expedients, to the present, with our more subtle and sophisticated modern techniques for turning the printing press or simply altering book entries. The problem is to establish institutional arrangements that will enable government to exercise responsibility for money, yet will at the same time limit the power thereby given to government and prevent the power from being used in ways that will tend to weaken rather than strengthen a free society. Three kinds of solutions have developed or have been suggested. One is an automatic commodity standard, a monetary standard which in principle requires no governmental control. A second is the control of monetary policies by an "independent" central bank. A third is the control of monetary policies by rules that are legislated in advance by the legislature, are binding upon the monetary authority, and greatly limit its initiative. This paper discusses these three alternatives with rather more attention to the solution through a central bank. A COMMODITY STANDARD Historically, the device that has evolved most frequently in many different places and over the course of centuries is a commodity standard; that is, the use as money of some physical commodity such as gold, silver, brass, or tin, or cigarettes, cognac, or various other commodities. If money consisted wholly of a physical commodity of this type, in principle there would be no need for control by the government at all. The amount of money in society would depend on the cost of producing the monetary commodity rather than on other things. Changes in the amount of money would depend on changes in the technical conditions of producing the monetary commodity and on changes in the demand for money. This is an ideal that animates many believers in an automatic gold standard. In point of fact, however, as the system developed it deviated very far from this simple pattern, which required no governmental intervention. Historically, a commodity standard—such as a gold standard or a silver standard—was accompanied by the development of alternative forms of money as well, of fiduciary money of one kind or another, ostensibly convertible into the montary commodity on fixed terms. There was a very good reason for this development. The fundamental defect of a commodity standard, from the point of view of the society as a whole, is that it requires the use of real resources to add to the stock of money. People must work hard to dig something out of the ground in one place—to dig gold out 206 of the ground in South Africa—in order to rebury it in Fort Knox, or some similar place. The necessity of using real resources for the operation of a commodity standard establishes a strong incentive for people to find ways to achieve the same result without employing these resources. If people will accept as money pieces of paper on which is printed "I promise to pay so much of the standard commodity," these pieces of paper can perform the same functions as the physical pieces of gold or silver, and they require very much less in resources to produce. This point, which I have discussed at somewhat greater length elsewhere 1 seems to me the fundamental difficulty of a commodity standard. If an automatic commodity standard were feasible, it would provide an excellent solution to the liberal dilemma of how to get a stable monetary framework without the danger of irresponsible exercise of monetary powers. A full commodity standard, for example, an honest-to-goodness gold standard in which 100 percent of the money consisted literally of gold, widely supported by a public imbued with the mythology of a gold standard and the belief that it is immoral and improper for government to interfere with its operation, would provide an effective control against governmental tinkering with the currency and against irresponsible monetary action. Under such a standard, any monetary powers of government would be very minor in scope. But such an automatic system has historically never proved feasible. It has always tended to develop in the direction of a mixed system containing fiduciary elements such as banknotes, bank deposits, or government notes in addition to the monetary commodity. And once fiduciary elements have been introduced, it has proved difficult to avoid government control over them, even when they were initially issued by private individuals. The reason is basically the difficulty of preventing counterfeiting or its economic equivalent. Fiducary money consists of a contract to pay standard money. It so happens that there tends to be a long interval between the making of such a contract and its realization, which enhances the difficulty of enforcing the contract to pay the standard money and hence also the temptation to issue fraudulent contracts. In addition, once fiduciary elements have been introduced, the temptation for government itself to issue fiduciary money is almost irresistible. As a result of these forces, commodity standards have tended in practice to become mixed standards involving extensive intervention by the state, which leaves the problem of how intervention is to be controlled. Despite the great amount of talk by many people in favor of the gold standard, almost no one today literally desires to see an honest-to-goodness full gold standard in operation. People who say they want a gold standard are almost invariably talking about the present kind of standard, or the kind of standard that was maintained in the 1930's, in which there is a small amount of gold in existence, held by the central monetary, authority as "backing"—to use that very misleading term—for fiduciary money, and with the same authority, a central bank or other government bureau, managing the gold standard. Even during the so-called "great days" of the gold standard of the 19th century, when the Bank of England was supposedly running the gold standard skillfully, the monetary system was far from a fully automatic gold Standard. It was even then a highly managed standard. And certainly the situation is now more extreme. Country after country has adopted the view that government has responsibility for internal stability. This development, plus the invention by Schacht of the widespread direct control of foreign exchange transactions, has meant that few if any countries are willing today to let the gold standard operate even as quasi-automatically as it did in the 19th century. Most countries in the world currently behave asymmetrically with respect to the gold standard. They are willing to allow gold to flow in and even to inflate somewhat in response, but almost none is willing either to let gold flow out to any large extent or to adjust to the outflow by allowing or forcing internal prices to decline. Instead, they are very likely to take measures such as exchange controls, import restrictions, and the like. My conclusion is that an automatic commodity standard is neither a feasible nor a desirable solution to the problem of establishing monetary arrangements for a free society. It is not desirable because it would involve a large cost in the form of resources used to produce the monetary commodity. It is not feasible because the mythology and beliefs required to make it effective do not exist. 1A Program for Monetary Stability pp. 4 - 8 . (New York: Fordham University Press, 1959), 207 AN INDEPENDENT CENTRAL BANK A second device that has evolved and for which there is considerable support is a so-called independent monetary authority—a central bank—to control monetary policy and to keep it from being the football of political manipulation. The widespread belief in an independent central bank clearly rests on the acceptance— in some cases the highly reluctant acceptance—of the view I have just been expressing about a commodity standard, namely, that a fully automatic commodity standard is not a feasible way to achieve the objective of a monetary structure that is both stable and free from irresponsible governmental tinkering. The device of an independent central bank embodies the very appealing idea that it is essential to prevent monetary policy from being a day-to-day plaything at the mercy of every whim of the current political authorities. The device is rationalized by assimilating it to a species of constitutionalism. The argument that is implicit in the views of proponents of an independent central bank—so far as I know, these views have never been fully spelled out—is that control over money is an essential function of a government comparable to the exercise of legislative or judicial or administrative powers. In all of these, it is important to distinguish between the basic structure and day-to-day operation wTithin that structure. In our form of government, this distinction is made between the constitutional rules which set down a series of basic prescriptions and proscriptions for the legislative, judicial, and executive authorities and the detailed operation of the several authorities under these general rules. Similarly, the argument implicit in the defense of an independent central bank is that the monetary structure needs a kind of a monetary constitution, which takes the form of rules establishing and limiting the central bank as to the powers that it is given, its reserve requirements, and so on. Beyond this, the argument goes, it is desirable to let the central bank have authority largely coordinate with that of the legislature, the executive, and the judiciary to carry out the general constitutional mandate on a day-to-day basis. In recent times, the threat of extension of government control into widening areas of economic activity has often come through proposals involving monetary expansion. Central bankers have generally been "sound money men," at least verbally, wThich is to say, they have tended to attach great importance to stability of the exchange rate, maintenance of convertibility of the nation's currency into other currencies and into gold, and prevention of inflation. They have therefore tended to oppose many of the proposals for extending the scope of government. This coincidence of their views in these respects with those of people like myself, who regard narrowly limited government as a requisite for a free society, is the source of much of the sympathy on the part of this group, whom I shall call "new liberals," for the notion of an independent central bank. As a practical matter, the central bankers seem more likely to impose restrictions on irresponsible monetary power than the legislative authority itself. A first step in discussing this notion critically is to examine the meaning of the "independence" of a central bank. There is a trivial meaning that cannot be the source of dispute about the desirability of independence. In any kind of a bureaucracy, it is desirable to delegate particular functions to particular agencies. The Bureau of Internal Revenue can be described as an independent bureau within the Treasury Department. Outside the regular government departments, there are separate administrative organizations, such as the Bureau of the Budget. This kind of independence of monetary policy would exist if, within the central administrative hierarchy, there were a separate organization charged with monetary policy which was subordinate to the chief executive or officer, though it might be more or less independent in routine decisions. For our purposes, this seems to me a trivial meaning of independence, and not the meaning fundamentally involved in the argument for or against an independent central bank. This is simply a question of expediency and of the best way to organize an administrative hierarchy. A more basic meaning is the one suggested above—that a central bank should be an independent branch of government coordinate with the legislative, executive, and judicial branches, and with its actions subject to interpretation by the judiciary. Perhaps the most extreme form of this kind of independence in practice, and the form that comes closest to the ideal type envisaged by proponents of an independent central bank, has been achieved in those historical instances where an organization that was initially entirely private and not formally part of the government at all has served as a central bank. The leading example, of course, is the Bank of England, which developed out of a strictly private bank 208 and was not owned by or formally a part of the Government until after World War II. If such a private organization strictly outside the regular political channels could not function as a central monetary authority, this form of independence would call for the establishment of a central bank through a constitutional provision which would be subject to change only by constitutional amendment. The bank would accordingly not be subject to direct control by the legislature. This is the meaning I shall assign to independence in discussing further whether an independent central bank is a desirable resolution of the problem of achieving responsible control over monetary policy. It seems to me highly dubious that the United States, or for that matter any other country, has in practice ever had an independent central bank in this fullest sense of the term. Even when central banks have supposedly been fully independent, they have exercised their independence only so long as there has been no real conflict between them and the rest of the government. Whenever there has been a serious conflict, as in time of war, between the interests of the fiscal authorities in raising funds and of the monetary authorities in maintaining convertibility into specie, the bank has almost invariably given way, rather than the fiscal authority. To judge by experience, even those central banks that have been nominally independent in the fullest sense of the term have in fact been closely linked to the executive authority. But of course this does not dispose of the matter. The ideal is seldom fully realized. Suppose we could have an independent central bank in the sense of a coordinate constitutionally established, separate organization. Would it be desirable to do so? I think not, for both political and economic reasons. The political objections are perhaps more obvious than the economic ones. Is it really tolerable in a democracy to have so much power concentrated in a body free from any kind of direct, effective political control? What I have called the new liberal often characterizes his position as involving belief in the rule of law rather than of men. It is hard to reconcile such a view wTith the approval of an independent central bank in any meaningful way. True, it is impossible to dispense fully with the rule of men. No law can be specified so precisely as to avoid problems of interpretation or to cover explicitly every possible case. But the kind of limited discretion left by even the best of laws in the hands of those administering them is a far cry indeed from the kind of far-reaching powers that the laws establishing central banks generally place in the hands of a small number of men. I was myself most fully persuaded that it would be politically intolerable to have an independent central bank by the memoirs of Emile Moreau. the Governor of the Bank of France during the period from about 1926 to 1928, the period when France established a new parity for the franc and returned to gold. Moreau was appointed Governor of the Bank of France in 1926, not long before Poincare became Premier after violent fluctuations in the exchange value of the franc and serious accompanying internal disturbances and governmental financial difficulties.. Moreau's memoirs were edited and brought out in book form some years ago by Jacques Rueff, who was the leading figure in the recent French monetary reform.2 The book is fascinating on many counts. The particular respect that is most relevant for our present purpose is the picture that Moreau paints of Montagu Norman, Governor of the Bank of England, on the one hand, and of Hjalmar Schacht, at that time Governor of the Bank of Germany, on the other; they were unquestionably two of the three outstanding central bankers of the modern era, Benjamin Strong of the United States being the third. Moreau describes the views that these two European central bankers had of their functions and their roles, and implies their attitude toward other groups. The impression left with me— though it is by no means clear that Moreau drew the same conclusions from what he wrote, and it is certain that he would have expressed himself more temperately—is that Norman and Schacht were contemptuous both of the masses— of "vulgar" democracy—and of the classes—of the, to them, equally vulgar plutocracy. They viewed themselves as exercising control in the interests of both groups but free from the pressures of either. In Norman's view, if the major central bankers of the wrorld would only co-operate with one another—and he had in mind not only himself and Schacht but also Moreau and Benjamin Strong— they could jointly wield enough power to control the basic economic destinies of the Western World in accordance with rational ends and objectives rather than 2 Emile, Moreau, [1954]). Souvenirs d'un gouvernenr <le la Banque de France (Paris: Genin, 209 with the irrational processes of either parliamentary democracy or laissez-faire capitalism. Though of course stated in obviously benevolent terms of doing the "right thing" and avoiding distrust and uncertainty, the implicit doctrine is clearly thoroughly dictatorial and totalitarian. It is not hard to see how Schacht could later be one of the major creators of the kind of far-reaching economic planning and control that developed in Germany. Schacht's creation of extensive direct control of foreign exchange transactions is one of the few really new economic inventions of modern times. In the older literature, when people spoke of a currency as being inconvertible, they meant that it was not convertible into gold or silver or some other money at a fixed rate. To the best of my knowledge, it is only after 1934 that inconvertibility came to mean what we currently (take it to mean : that it is illegal for one man to convert paper money of one country into paper money of another country at any terms he can arrange with another person.3 I turn now to the economic or technical aspects of an independent central bank. Clearly there are political obpections to giving the group in charge of a central bank so much power independent of direct political controls, but it has been argued, there are economic or technical grounds wrhy it is nevertheless essential to do so. In judging this statement, much depends on the amount of leeway that the general rule governing the central bank gives to it. I have been describing an independent central bank as if it could or would be given a good deal of separate power, as clearly is currently the case. Of course, the whole notion of independence could be rendered merely a matter of words if in fact the constitutional provision setting up the bank established the limits of its authority very narrowly and controlled very closely the polices that it could follow. In the 19th century, when wide support for independent central banks developed, the governing objective of the central bank was the maintenance of exchange stability. Central banks tended to develop in countries that professed to have commodity currencies, which is to say had a fixed price for the commodity serving as the monetary standard in terms of the nominal money of the country. For two countries on the same standard, this meant a fixed rate of exchange between the corresponding national currencies. In consequence, the maintenance of such fixer! rates had to be the proximate aim of the central bank if it was to achieve its major aim of keeping its currency convertible into standard money. The Bank of England, for example, was narrowly limited in what it could do by the necessity of keeping England on gold. In the same wray, in the United States when the Federal Reserve System was established in 1913, it never entered into the minds of the people who were establishing it that the System wTould really have much effective control internally in ordinary times. The Reserve System was established when the gold standard ruled supreme, and when it wras taken for granted that the major factor determining the policy of the System, and hence the behavior of the stock of money in this country, would be the necessity of maintaining external equilibrium with the currencies of other countries. So long as the maintenance of a fixed exchange rate between one country's currency and the currencies of other countries was the overriding objective of policy, the amount of leeway available to the central bank was narrowly limited. It had some leeway with respect to minor movements of a short-term character, but it ultimately had to respond to the balance of payments. The situation has changed drastically in this respect in the course of the past few decades. In the United States, which is of most immediate concern to us, the Reserve System had hardly started operations before the fundamental con3 Another feature of Moreau's book that is most fascinating but rather off the main track of the present discussion is the story it tells of the changing relations between the French and British central banks. At the beginning, with France in desperate straits seeking to stabilize its currency. Norman was contemptuous of France and regarded it as very much of a junior partner. Through the accident that the French currency was revalued at a level that stimulated gold imports, France started to accumulate gold reserves and sterling reserves and gradually came into the position where at any time Moreau could have forced the British off gold b^ withdrawing the funds he had on deposit at the Bank of England. The result wras that Norman changed from being a pr ucl boss and very much the senior partner to being almost a suppliant at the mercy of Moreau. Aside from the human drama, it emphasizes how important it is whether the rate of exchange is fixed 5 percent too low or 5 percent too high. Britain went back on gold in 1925 at a price of gold in terms < f thp pound that was probably something like 5 or 10 percent too low. and France went back <le facto at the end of 1926 and de jure in mid-1928 at a price of gold in terms of francs that was 5 or 10 percent to^ high. This difference meant the difference between the French being at the mercy of the British and the British being at the mercy of the French. 210 ditions taken for granted when it was established had changed radically. During World War I, most of the countries of the world went off gold. The United States technically remained on gold, but the gold standard on which it remained was very different from the one that had prevailed earlier. After the end of World War I, although other countries of the world gradually reestablished something they called the gold standard, the gold standard never again played the role which it had before. Prior to World War I, the United States was effectively a minor factor in the total world economy, and the necessity of maintaining external stability dominated our behavior. After the war, we had become a major factor to which other countries had to adjust. We held a very large fraction of the world's gold. Many countries never went back on gold, and those that did went back in a much diluted form. So never again has there been anything like the close domination of day-to-day policy by the gold standard that prevailed prior to 1914. Under these circumstances, "independence" of the central bank has become something meaningful, and not merely a technicality. One defect of an independent central bank in such a situation is that it almost inevitably involves dispersal of responsibility. If we examine the monetary system in terms not of nominal institutional organization but of the economic functions performed, we find that the central bank is hardly ever the only authority in the Government that has essential monetary powers. Before the Federal Reserve System was established, the Treasury exercised essential montary powers. It operated like a central bank, and at times a very effective central bank. More recently, from 1933 to 1941, the Federal Reserve System was almost entirely passive. Such monetary actions as were taken predominantly by the Treasury. The Treasury engaged in open-market operations in its debt-management operations of buying and selling securities. It created and destroyed money in its gold and silver purchases and sales. The Exchange Stabilization Fund was established and gave the Treasury yet another device for engaging in openmarket operations. When the Treasury sterlized and desterilized gold, it was engaging in monetary actions. In practice, therefore, even if something called an independent central bank is established and given exclusive power over a limited range of monetary matters, in particular over the printing of pieces of pape or the making of book entries called money (Federal Reserve notes and Federal Reserve deposits), there remain other governmental authorities, particularly the fiscal authority collecting taxes and dispersing funds and managing the debt, which also have a good deal of monetary power. If one wanted to have the substance and not merely the form of an independent monetary authority, it would be necessary to concentrate all debt-management powers as well as all powers to create and destroy governmentally issued money in the central bank. As a matter of technical efficiency, this might well be desirable. Our present division of responsibility for debt management between the Federal Reserve and the Treasury is very inefficient. It would be much more efficient if the Federal Reserve did all of the borrowing and all of the managing of the debt, and the Treasury, when it had a deficit, financed it by getting money from the Federal Reserve System, and when it had a surplus, handed the excess over to the Federal Reserve System. But while such an arrangement might be tolerable if the Federal Reserve System were part of the same administrative hierarchy as the Treasury, it is almost inconceivable that it would be if the central bank were thoroughly independent. Certainly no government to date has been willing to put that much power in the hands of a central bank even when the bank has been only partly independent. But so long as these powers are separated, there is dispersal of responsibility, with each group separately regarding the other group as responsible for what is happening and with no one willing to accept responsibility. In the past few years, I have read through the annual reports of the Federal Reserve System from 1913 to date, seriatim. One of the few amusing dividends from that ordeal was seeing the cyclical pattern that shows up in the potency that the authorities attribute to monetary policy. In years when things are going well, the reports emphasize that monetary policy is an exceedingly potent weapon and that the favorable course of events is largely a result of the skillful handling of this delicate instrument by the monetary authority. In years of depression, on the other hand, the reports emphasize that monetary policy is but one of many tools of economic policy, that its power is highly limited, and that it was only the skillful handling of such limited powers as were available that averted disaster. This is an example of the effect of the dispersal of responsibility among different authorities, with the likely result that no one assumes or is assigned the final responsibility. 211 Another defect of the conduct of monetary policy through an independent central bank that has a good deal of leeway and power is the extent to which the policy is thereby made highly dependent on personalities. In studying the history of American monetary policy, I have been struck by the extraordinary importance of accidents of personality. At the end of World War I, the Governor of the Federal Reserve System was W. P. G. Harding. Governor Harding was, I am sure, a thoroughly reputable and competent citizen, but he had a very limited understanding of monetary affairs, and even less backbone. Almost every student of the period is agreed that the great mistake of the Reserve System in postwar monetary policy was to permit the money stock to expand very rapidly in 1919 and then to step very hard on the brakes in 1920. This policy was almost surely responsible for both the sharp postwar rise in prices and the sharp subsequent decline. It is amusing to read Harding's answer in his memoirs to criticism that was later made of the policies followed. He does not question that alternative policies might well have been preferable for the economy as a whole, but emphasizes the Treasury's desire to float securities at a reasonable rate of interest, and calls attention to a thenexisting law under which the Treasury could replace the head of the Reserve System. Essentially he was saying the same thing that I heard another member of the Reserve Roard say shortly after World War II when the bond-supported program was in question. In response to the view expressed by some of my colleagues and myself that the bond-support program should be dropped, he largely agreed but said, "Do you want us to lose our jobs?" The importance of personality is strikingly revealed by the contract between Harding's behavior and that of Emile Moreau in France under much more difficult circumstances. Moreau formally had no independence whatsoever from the central government. He was named by the Premier, and could be discharged at any time by the Premier. But when he was asked by the Premier to provide the Treasury with funds in a manner that he considered inappropriate and undesirable, he flatly refused to do so. Of course, what happened was that Moreau was not discharged, that he did not do what the Premier had asked him to, and that stabilization was rather more successful. I cite this contrast neither to praise Moreau nor to blame Harding, but simply to illustrate my main point; namely, the extent to which a system of this kind is really a system of rule by men and not by law and is extraordinarily dependent on the particular personalities involved. Another occasion in U.S. history which strikingly illustrates this point is our experience from 1929 to 1933. Without doubt, the most serious mistake in the history of the Reserve System was its mismanagement of monetary matters during those years. And this mismanagement, like that after World War I, can very largely attributed to accidents of personality. Benjamin Strong, Governor of the Federal Reserve Bank of New York from its inception, was the dominant figure in the Reserve System until his death at a rather early age in 1928. His death was followed by a shift of power in the System from New York to Washington. The people in Washington at the time happened to be fairly mediocre. Moreover, they had always played a secondary role, were not in intimate touch with the financial world, and had no background of long experience in meeting day-to-day emergencies. Further, the chairmanship changed hands just prior to the shift of power and again in mid-1931. Consequently, in the emergencies that came in 1929, 1930, and 1931, particularly in the fall of 1930, when the Bank of United States failed in New York as part of a dramatic series of bank failures, the Federal Reserve System acted timorously and passively. There is little doubt that Strong would have acted very differently. If he had still been Governor, the result would almost surely have been to nip the wave of bank failures in the bud and to prevent the drastic monetary deflation that followed. A similar situation prevails today. The actions of the Reserve System depend on whether there are a few persons in the System who exert intellectual leadership, and on who these people are; its actions depend not only on the people who are nominally the heads of the System but also on such matters as the fate of particular economic advisers. So far, I have listed two main technical defects of an independent central bank from an economic point of view: first, dispersal of responsibility, which promotes shirking responsibility in times of uncertainty and difficulty, and second, an extraordinary dependence on personalities, which fosters instability arising from accidental shifts in the particular people and the character of the people who are in charge of the system. 212 A third technical defect is that an independent central bank will almost inevitably give undue emphasis to the point of view of bankers. It is exceedingly important to distinguish two quite different problems that tend to be confused: the problem of credit policy and the problem of monetary policy. In our kind of monetary or banking system, money tends to be created as an incident in the extension of credit, yet conceptually the creation of money and the extension of credit are quite distinct. A monetary system could be utterly unrelated to any credit instruments whatsoever; for example, this would be true of a completely automatic commodity standard, using only the monetary commodity itself or warehouse receipts for the commodity as money. Historically, the connection between money and credit has varied widely from time to time and from place to place. It is therefore essential to distinguish policy issues connected with interest rates and conditions on the credit market from policy issues connected with changes in the aggregate stock of money, while recognizing, of course, that measures taken to affect the one set of variables may also affect the other, and that monetary measures may have credit effects as well as monetary effects proper. It so happens that central-bank action is but one of many forces affecting the credit market. As we and other countries have seen time and again, a central bank may be able to determine the rate of interest on a narrow range of securities, such as the rate of interest on a particular category of Government bonds, though even that only within limits and only at the expense of completely giving up control over the total stock of money. A central bank has never been able to determine, at all closely, rates of interest in any broader or more fundamental sense. Postwar experience in country after country that has embarked on a cheapmoney policy has strikingly demonstrated that the forces which determine rates of interest broadly conceived—rates of return on equities, on real property, on corporate securities—are far too strong and widespread for the central bank to dominate. It must sooner or later yield to them, and generally rather soon. The central bank is in a very different position in determining the quantity of money. Under systems such as that in the United States today, the central bank can make the amount of money anything it wishes. It may, of course, choose to accept some other objective and give up its power over the money supply in order to try to keep "the" or "a" rate of interest fixed, to keep "free reserves" at a particular level, or to achieve some other objective. But if it wishes, it can exercise complete control over the stock of money. This difference between the position of the central bank in the credit markets and in determining the money supply tends to be obfuscated by the close connection between the central bank and the banking community. In the United States, for example, the Reserve banks technically are owned by their member banks. One result is that the general views of the banking community exercise a strong influence on the central bank and, since the banking community is concerned primarily with the credit market, central banks are led to put altogether too much emphasis on the credit effects of their policies and too little emphasis on the monetary effects of their policies. In recent times, this emphasis has been attributed to the effects of the Keynesian Revolution and its treatment of changes in the stock of money as operating primarily through the liquidity preference function on the interest rate. But this is only a particular form of a more general and ancient tendency. The real-bills doctrine, which dates back a century and more, exemplifies the same kind of confusion between the credit and the monetary effects of monetary policy. The banking and currency controversy in Britain in the early 19th century is a related example. The central bank emphasized its concern with conditions in the credit market It denied that the quantity of money it was creating was in any way an important consideration in determining price levels or the like, or that it had any discretion about how much money to create. Much the same arguments are heard today. The three defects I have outlined constitute a strong technical argument against an independent central bank. Combined with the political argument, the case against a fully independent central bank is strong indeed. LEGISLATED RULES If this conclusion is valid, if we cannot achieve our objectives by giving wide discretion to independent experts, how else can we establish a monetary system that is stable, free from irresponsible governmental tinkering, and incapable of being used as a source of power to threaten economic and political freedom? A 213 third possibility is to try to achieve a government of law instead of men literally by legislating rules for the conduct of monetary policy. The enactment of such rules would enable the public to exercise control over monetary policy through its political authorities, while at the same time preventing monetary policy from being subject to the day-to-day whim of political authorities. The argument for legislating rules for monetary policy has much in common with a topic that seems at first altogether different; namely, the Bill of Rights to the Constitution. Whenever anyone suggests the desirability of a legislative rule for control over money, the stereotyped answer is that it makes little sense to tie the monetary authority's hands in this way because the authority, if it wants to, can always do of its own volition what the rule would require it to do, and, in addition, has other alternatives; hence "surely," it is said, it can do better than the rule. An alternative version of the same argument applies to the legislature. If the legislature is willing to adopt the rule, it is said, surely it will also be willing to legislate the "right" policy in each specific case. How then, it is said, does the adoption of the rule provide any protection against irresponsible political action ? The same argument could apply with only minor verbal changes to the first amendment to the Constitution and, equally, to the entire Bill of Rights. Is it not absurd, one might say, to have a general proscription of interference with free speech? Why not take up each case separately and treat it on its own merits? Is this not the counterpart to the usual argument in monetary policy that it is undesirable to tie the hands of the monetary authority in advance; that it should be left free to treat each case on its merits as it comes up? Why is not the argument equally valid for speech? One man wants to stand up on a street corner and advocate birth control; another, communism; a third, vegetarianism; and so on, ad infinitum. Why not enact a law affirming or denying each the right to spread his particular views? Or, alternatively, why not give the power to decide the issue to an administrative agency? It is immediately clear that if we were to take up each case separately, a majority would almost surely vote to deny free speech in most cases and perhaps even in every case. A vote on whether Mr. X should spread birth control propaganda would almost surely yield a majority saying "no"; and so would one on communism. The vegetarian might perhaps get by, although even that is by no means a foregone conclusion. But now suppose all these cases were grouped together in one bundle, and the populace at large was asked to vote for them as a whole: to vote whether free speech should be denied in all cases or permitted in all alike. It is perfectly conceivable, if not highly probable, that an overwhelming majority would vote for free speech; that, acting on the bundle as a whole, the people would vote exactly the opposite to the way they would have voted on each case separately. Why? One reason is that each person feels much more strongly about being deprived of his right to free speech wrhen he is in a minority than he feels about depriving somebody else of the right to free speech when he is in the majority. In consequence, when he votes on the bundle as a whole, he gives much more weight to the infrequent denial of free speech to himself when he is in the minority than to the frequent denial of free speech to others. Another reason, and one that, is more directly relevant to monetary policy, is that if the bundle is viewed as a whole, it becomes clear that the policy followed has cumulative effects that tend neither to be recognized nor taken into account when each case is voted on separately. When a vote is taken on whether Mr. Jones may speak on the corner, it is not clearly affected by favorable effects of an announced general policy of free speech, and an affirmative vote will not produce these effects. In voting on the specific case, it is only peripherally relevant that a society in which people are not free to speak on the corner without special legislation is a society in which the development of new ideas, experimentation, change, and the like are all hampered in a great variety of ways. That these ways are obvious to all is due to our good fortune of having lived in a society that did adopt the self-denying ordinance of not considering each case of speech separately. Exactly the same considerations apply in the monetary area. If each case is considered on its merits, the wrong decision is likely to be made in a large fraction of cases because the decisionmakers are examining only a limited area and are not taking into account the cumulative consequences of the policy as a whole. On the other hand, if a general rule is adopted for a group of cases as a bundle, the existence of that rule has favorable effects on people's attitudes and beliefs and expectations that would not follow even from the discretionary adoption of precisely the same policy on a series of separate occasions. 214 Of course, the general rule need not be explicitly written down or legislated. Unwritten constitutional limitations supported unthinkingly by the bulk of the people may be as effective in determining decisions in individual cases as a written constitution. The analogy in monetary affairs is the mythology of gold, referred to earlier as a necessary ingredient of a gold standard if it is to serve as an effective bulwark against discretionary authority. If a rule is to be legislated, what rule should it be? The rule that has most frequently been suggested by people of a generally "new liberal" persuasion is a price-level rule; namely, a legislative direction to the monetary authorities that they maintain a stable price level. I think this is the wrong kind of rule. It is the wrong kind of rule because the objectives it specifies are ones that the monetary authorities do not have the clear and direct power to achieve by their own actions. It consequently raises the earlier problem of dispersing responsibilities and leaving the authorities too much leeway. There is unquestionably a close connection between monetary actions and the price level. But the connection is not so close, so invariable, or so direct that the objective of achieving a stable price level is an appropriate guide to the day-to-day activities of the authorities. The issue of what rule to adopt is one that I have considered at some length elsewhere.4 Accordingly, I will limit myself here to stating my conclusion. In the present state of our knowledge, it seems to me desirable to state the rule in terms of the behavior of the stock of money. My choice at the moment would be a legislated rule instructing the monetary authority to achieve a specified rate of growth in the stock of money. For this purpose, I would define the stock of money as including currency ouside commercial banks plus all deposits of commercial banks. I would specify that the Reserve System should see to it that the total stock of money so defined rises month by month, and indeed, so far as possible, day by day, at an annual rate of x percent, where x is some number between 3 and 5. The precise definition of money adopted and the precise rate of growth chosen make far less difference than the definite choice of a particular definition and a particular rate of growth. I should like to emphasize that I do not regard this proposal as a be-all and end-all of monetary management, as a rule which is somehow to be written in tablets of gold and enshrined for all future time. It seems to me to be the rule that offers the greatest promise of achieving a reasonable degree of monetary stability in the light of our present knowledge. I would hope that as we operated with it, as we learned more about monetary matters, we might be able to devise still better rules which would achieve still better results. However, the main point of this paper is not so much to discuss the content of these or alternative rules as to suggest that the device of legislating a rule about the stock of money can effectively achieve what an independent central bank is designed to achieve but cannot. Such a rule seems to me the only feasible device currently available for converting monetary policy into a pillar of a free society rather than a threat to its foundation. [From the American Economic Review, March 1968] T H E ROLE OF MONETARY POLICY* (By Milton Friedman**) There is wide agreement about the major goals of economic policy: high employment, /stable prices, and rapid growth. There is less agreement that these goals are mutually compatible or, among those who regard them as incompatible about the terms at which they can and should be substituted for one another. There is least agreement about the role that various instruments of policy can and should play in achieving the several goals. My topic for tonight is the role of one such instrument—monetary policy. What can it contribute? And how should it be conducted to contribute the most? Opinion on these questions has fluctuated widely. In the first flush of enthusiasm about the newly created Federal Reserve System, many observers attributed the relaA Program for Monetary Stability, pp. 7 7 - 9 9 . •Presidential address delivered at the 80th annual meeting of the American Economic Association, Washington, D.C., Dec. 29, 1967. * * I am indebted for helpful criticisms of earlier drafts to Armen Alchian, Gary Becker, Martin Bronfenbrenner, Arthur F. Burns, Phillip Cagan, David D. Friedman, Lawrence Harris*, Harvey G. Johnson, Homer Jones, Jerry Jordan, David Meiselman, Allan H Meltzer Theodore W . Schultz, Anna J. Schwartz, Herbert Stein, George J. Stigler, and James Tobin! 4 215 tive stability of the 1920's to the System's capacity for fine tuning—to apply an apt modern term. It came to be widely believed that a new era had arrived in which business cycles had been rendered obsolete by advances in monetary technology. This opinion was shared by economist and layman alike, though, of course, there were some dissonant voices. The great contraction destroyed this naive attitude. Opinion swung to the other extreme. Monetary policy was a string. You could pull on it to stop inflation but you could not push on it to halt recession. You could lead1 a horse to water but you could not make him drink. Such theory by aphorism was soon replaced by Keynes' rigorous and sophisticated analysis. Keynes offered simultaneously an explanation for the presumed impotence of monetary policy to stem the depression, a nonmonetary interpretation of the depression, and an alternative to monetary policy for meeting the depression and his offering was avidly accepted. If liquidity preference is absolute or nearly so—as Keynes believed likely in times of heavy unemployment—interest rates cannot be lowered by monetary measures. If investment and consumption are little affected by interest rates—as Hansen and many of Keynes' other American disciples came to believe—lower interest rates, even if they could be achieved, would do little good. Monetary policy is twice damned. The contraction, set in train, on this view, by a collapse of investment or by a shortage of investment opportunities or by stubborn thriftiness, could not, it was argued, have been stopped by monetary measures. But there was available an alternative—fiscal policy. Government spending could make up for insufficient private investment. Tax reductions could undermine stubborn thriftiness. The wide acceptance of these views in the economics profession meant that for some two decades monetary policy was believed by all but a few reactionary souls to have been rendered obsolete by new economic knowledge. Money did not matter. Its only role was the minor one of keeping interest rates low, in order to hold down interest payments in1 the Government budget, contribute to the "euthanasia of the rentier," and maybe, stimulate investment a bit to assist Government spending in maintaining a high level of aggregate demand. These views produced a widespread adoption of cheap money policies after the war. And they received a rude shock when these policies failed in country after country, when central bank after central bank was forced to give up the pretense that it could indefinitely keep the rate of interest at a low level. In this country, the public denouement came with the Federal Reserve-Treasury Accord in 1951, although the policy of pegging Government bond prices was not formally abandoned until 1953. Inflation, stimulated by cheap money policies, not the widely heralded postwar depression, turned out to be the order of the day. The result was the beginning of a revival of belief in the potency of monetary policyThis revival was strongly fostered among economists by the theoretical developments initiated by Haberler but named for Pigou that pointed out a channel—namely, changes in wealth—whereby changes in the real quantity of money can affect aggregate demand even if they do not alter interest rates. These theoretical developments did not undermine Keynes' argument against the potency of orthodox monetary measures when liquidity preference is absolute since under such circumstances the usual monetary operations involve simply substituting money for other assets without changing total wealth. But they did show how changes in the quantity of money produced in other ways could affect total spending even under such circumstances. And, more fundamentally, they did undermine Keynes' key theoretical proposition, namely, that even in a world of flexible prices, a position of equilibrium at full employment might not exist. Henceforth, unemployment had again to be explained by rigidities or imperfections, not as the natural outcome of a fully operative market process. The revival of belief in the potency of monetary policy was fostered also by a revaluation of the role money played from 1929 to 1933. Keynes and most other economists of the time believed that the great contraction in the United States occurred despite aggressive expansionary policies by the monetary authorities—that they did their best but their best was not good enough.1 Recent studies have demonstrated that the facts are precisely the reverse: the U.S. monetary authorities followed highly deflationary policies. The quantity of money in the United States fell by one-third in the course of the contraction. And it fell not because (there were no willing borrowers—not because the horse would not drink. It fell because the Federal Reserve System forced or permitted a sharp reduction J In [2], I have argued that Henry Simons shared this view with Keynes, and that it accounts for the policy changes that he recommended. 216 in the monetary base, because it failed to exercise the responsibilities assigned to it in the Federal Reserve Act to provide liquidity to the banking system. The great contraction is tragic testimony to the power of monetary policy—not, as Keynes and so many of his contemporaries believed, evidence of its impotence. In the United States the revival of belief in the potency of monetary policy was strengthened also by increasing disillusionment with fiscal policy, not so much with its potential to affect aggregate demand as with the practical and political feasibility of so using it. Expenditures turned out to respond sluggishly and with long lags to attempt to adjust them to the course of economic activity, so emphasis shifted to taxes. But here political factors entered with a vengeance to prevent prompt adjustment to presumed need, as has been so graphically illustrated in the months since I wrote the first draft of this talk. "Fine tuning" is a marvelously evocative phrase in this electronic age, but it has little resemblance to what is possible in practice—not, I might add, an unmixed evil. It is hard to realize how radical has been the change in professional opinion on the role of money. Hardly an economist today accepts views that were the common coin some itwo decades ago. Let me cite a few examples. In a talk published in 1945, E. A. Goldenweiser, then Director of the Research Division of the Federal Reserve Board, described the primary objective of monetary policy as being to "maintain the value of Government bonds. * * * This country" he wrote, "will have to adjust to a 2%-percent interest rate as the return on safe, long-time money, because the time has come when returns on pioneering capital can no longer be unlimited as they were in the past" [4, p. 117]. In a book on Financing American Prosperity, edited by Paul Homan and Fritz Machlup and published in 1945, Alvin Hansen devotes nine pages of text to the "savings-investment problem" without finding any need to use the words "interest rate" or any close facsimile thereto [5, pp. 218-27]. In his contribution to this volume, Fritz Machlup wrote, "Questions regarding the rate of interest, in particular regarding its variation or its stability, may not be among the most vital problems of the postwar economy, but they are certainly among the perplexing ones" [5, p. 466]. In his contribution, John H. Williams—<not only professor at Harvard but also a longtime adviser to the New York Federal Reserve Bank—wrote, "I can see no prospect of revival of a general monetary control in the postwar period" [5, p. 383]. Another of the volumes dealing with postwar policy that appeared at this time, Planning and Paying for Full Employment, was edited by Abba P. Lerner and Frank D. Graham [6] and had contributors of all shades of professional opinion— from Henry Simons and Frank Graham to Abba Lerner and Hans Neisser. Yet Albert Halasi, in his excellent summary of the papers, was able to say, "Our contributors do not discuss the question of money supply. * * * The contributors make no special mention of credit policy to remedy actual depressions. * * * Inflation * * * might be fought more effectively by raising interest rates. * * * But * * * other anti-inflationary measures * * * are preferable" [6, pp. 23-24]. A Survey of Contemporary Economics, edited by Howard Ellis and published in 1948, was an "official" attempt to codify the state of economic thought of the time. In his contribution, Arthur Smithies wrote, "In the field of compensatory action, I believe fiscal policy must shoulder most of the load. Its chief rival, monetary policy, seems to be disqualified on institutional grounds. This country appears to be committed to something like the present low level of interest rates on a long-term basis" [1, p. 208]. These quotations suggest the flavor of professional thought some two decades ago. If you wish to go further in this humbling inquiry, I recommend that you compare the sections on money—when you can find them—in the principles texts of the early postwar years with the lengthy sections in the current crop even, or especially, when the early and recent principles are different editions of the same work. The pendulum has swung far since then, if not all the way to the position of the late 1920s, at least much closer to that position that to the position of 1945. There are of course many differences between then and now, less in the potenev attributed to monetary policy than in the roles assigned to it and the criteria bv which the profession believes monetary policy should be guided. Then, the chief roles assigned monetary policy were to promote price stability and to preserve the gold standard; the chief criteria of monetary policy were the state of the money market, the extent of speculation and the movement of gold. Today, primacy is assigned to the promotion of full employment, with the prevention of inflation a continuing but definitely secondary objective. And there is major 217 disagreement about criteria of policy, varying from emphasis on money market conditions, interest rates, and the quantity of money to the belief that the state of employment itself should be the proximate criterion of policy. I stress nonetheless the similarity between the views that prevailed in the late 'twenties and those that prevail today because I fear that, now as then, the pendulum may well have swung too far, that now as then, we are in danger of assigning to monetary policy a larger role than it can perform, in danger of asking it to accomplish tasks that it cannot achieve, and, as a result, in danger of preventing it from making the contribution that it is capable of making. Unaccustomed as I am to denigrating the importance of money, I therefore shall, as my first task, stress what monetary policy cannot do. I shall then try to outline what it can do and how it can best make its contribution, in the present state of our knowledge—or ignorance. I. WHAT MONETARY POLICY CANNOT DO From the infinite world of negation, I have selected two limitations of monetary policy to discuss: (1) It cannot peg interest rates for more than very limited periods, and (2) It cannot peg the rate of unemployment for more than very limited periods. I select these because the contrary has been or is widely believed, because they correspond to the two main unattainable tasks that are at all likely to be assigned to monetary policy, and because essentially the same theoretical analysis covers both. Pegging of interest rates History has already persuaded many of you about the first limitation. As noted earlier, the failure of cheap money policies was a major source of the reaction against simple-minded Keynesianism. In the United States, this reaction involved widespread recognition that the wartime and postwar pegging of bond prices was a mistake, that the abandonment of this policy was a desirable and inevitable step, and that it had none of the disturbing and disastrous consequences that were so freely predicted at the time. The limitation derives from a much misunderstood feature of the relation between money and interest rates. Let the Fed set out to keep interest rates down. How will it try to do so? By buying securities. This raises their prices and lowers their yields. In the process, it also increases the quantity of reserves available to banks, hence the amount of bank credit, and, ultimately the total quantity of money. That is why central bankers in particular, and the financial community more broadly, generally believe that an increase in the quantity of money tends to lower interest rates. Academic economists accept the same conclusion, but for different reasons. They see, in their mind's eye, a negatively sloping liquidity preference schedule. How can people be induced to hold a larger quantity of money ? Only by bidding down interest rates. Both are right, up to a point. The initial impact of increasing the quantity of money at a faster rate than it has been increasing is to make interest rates lower for a time than they would otherwise have been. But this is only the beginning of the process not the end. The more rapid rate of monetary growth will stimulate spending, both through the impact on investment of lower market interest rates and through the impact on other spending and thereby relative prices of higher cash balances than are desired. But one man's spending is another man's income. Rising income wTill raise the liquidity preference schedule and the demand for loans; it may also raise prices, which would reduce the real quantity of money. These three effects will reverse the initial downward pressure on interest rates fairly promptly, say, in something less than a year. Together they will tend, after a somewhat longer interval, say, a year or two, to return interest rates to the level they would otherwise have had. Indeed, given the tendency for the economy to overreact, they are highly likely to raise interest rates temporarily beyond that level, setting in motion a cyclical adjustment process. A fourth effect, when and if it becomes operative, will go even further, and definitely mean that a higher rate of monetary expansion will correspond to a higher, not lower, level of interest rates than would otherwise have prevailed. Let the higher rate of monetary growth produce rising prices, and let the public come to expect that prices will continue to rise. Borrowers will then be willing to pay and lenders will then demand higher interest rates—as Irving Fisher pointed out decades ago. This price expectation effect is slow to develop and also slow to disappear. Fisher estimated that it took several decades for a full adjustment and more recent work is consistent with his estimates. 21-570—68 15 218 These subsequent effects explain why every attempt to keep interest rates at a low level has forced the monetary authority to engage in successively larger and larger open market purchases. They explain why, historically, high and rising nominal interest rates have been associated with rapid growth in the quantity of money, as in Brazil or Chile or in the United States in recent years, and why low and falling interest rates have been associated with slow growth in the quantity of money, as in Switzerland now or in the United States from 1929 to 1933. As an empirical matter, low-interest rates are a sign that monetary policy has been tight—in the sense that the quantity of money has grown slowly; high-interest rates are a sign that monetary policy has been easy—in the sense that the quantity of money has grown rapidly. The broadest facts of experience run in precisely the opposite direction from that which the financial community and academic economists have all generally taken for granted. Paradoxically, the monetary authority could assure low nominal rates of interest—but to do so it would have to start out in what seems like the opposite direction, by engaging in a deflationary monetary policy. Similarly, it could assure high nominal interest rates by engaging in an inflationary policy and accepting a temporary movement in interest rates in the opposite direction. These considerations not only explain why monetary policy cannot peg interest rates; they also explain why interest rates are such a misleading indicator of whether monetary policy is "tight" or "easy." For that, it is far better to look at the rate of change of the quantity of money.8 Employment as a criterion of policy The second limitation I wish to discuss goes more against the grain of current thinking. Monetary growth, it is widely held, will tend to stimulate employment; monetary contraction, to retard employment. Why, then, cannot the monetary authority adopt a target for employment or unemployment—say, 3 percent unemployment ; be tight when unemployment is less than the target; be easy when unemployment is higher than the target; and in this way peg unemployment at, say, 3 percent? The reason it cannot is precisely the same as for interest rates— the difference between the immediate and the delayed consequences of such a policy. Thanks to Wicksell, we are all acquainted with the concept of a "natural" rate of interest and the possibility of a discrepancy between the "natural" and the "market" rate. The preceding analysis of interest rates can be translated fairly directly into Wicksellian terms. The monetary authority can make the market rate less than the natural rate only by inflation. It can make the market rate higher than the natural rate only by deflation. We have added only one wrinkle to Wicksell—the Irving Fisher distinction between the nominal and the real rate of interest. Let the monetary authority keep the nominal market rate for a time below the natural rate by inflation. That in turn will raise the nominal natural rate itself, once anticipations of inflation become widespread, thus requiring still more rapid inflation to hold down the market rate. Similarly, because of the Fisher effect, it will require not merely deflation but more and more rapid deflation to hold the market rate above the initial "natural" rate. This analysis has its close counterpart in the employment market. At any moment of time, there is some level of unemployment which has the property that it is consistent with equilibrium in the structure of real wage rates. At that level of unemployment, real wage rates are tending on the average to rise at a "normal" secular rate, i.e., at a rate that can be indefinitely maintained so long as capital formation, technological improvements, etc., remain on their longrun trends. A lower level of unemployment is an indication that there is an excess demand for labor that will produce upward pressure on real wage rates. A higher level of unemployment is an indication that there is an excess supply of labor that will produce downward pressure on real wage rates. The "natural rate of unemployment," in other words, is the level that would be ground out by the Walrasian system of general equilibrium equations, provided there is imbedded' in them the actual structural characteristics of the labor and commodity markets, including market imperfections, stochastic variability in demands and sup2 This is partly an empirical not theoretical judgment. In principle, "tightness" or "ease" depends on the rate of change of the quantity of money supplied compared to the rate of change of the quantity demanded excluding effects on demand from monetary policy itself. However, empirically demand is highly stable, if we exclude the effect of monetary policy, so it is generally sufficient to look at supply alone. 219 plies, the cost of gathering information about job vacancies and labor availabilities, the costs of mobility, and so on.3 You will recognize the close similarity between this statement and the celebrated Phillips Curve. The similarity is not coincidental. Phillips' analysis of the relation between unemployment and wage change is deservedly celebrated as an important and original contribution. But, unfortunately, it contains a basic defect—the failure to distinguish between nominal wages and real wages—just as Wicksell's analysis failed to distinguish between nominal interest rates and real interest rates. Implicitly, Phillips wrote his article for a world in which everyone anticipated that nominal prices would be stable and in which that anticipation remained unshaken and immutable whatever happened to actual prices and wages. Suppose, by contrast, that everyone anticipates that prices will rise at a rate of more than 75 percent a year—as, for example, Brazilians did a few years ago. Then wages must rise at that rate simply to keep real wages unchanged. An excess supply of labor will be reflected in a less rapid rise i*n nominal wages than in anticipated prices,4 not in an absolute decline in wages. When Brazil embarked on a policy to bring down the rate of price rise, and succeeded in bringing the prise rise down to about 45 percent a year, there was a sharp initial riae in unemployment because under the influence of earlier anticipations, wages kept rising at a pace that was higher than the new rate of price rise, though lower than earlier. This is the result experienced, and to be expected, of all attempts to reduce the rate of inflation below that widely anticipated.5 To avoid misunderstanding, let me emphasize that by using the term ''natural" rate of unemployment, I do not mean to suggest that it is immutable and unchangeable. On the contrary, many of the market characteristics that determine its level are man-made and policy-made. In the United States, for example, legal minimum wage rates, the Walsh-Healy and Davis-Bacin Acts, and the strength of labor unions all make the natural rate of unemployment higher than it would otherwise be. Improvements in employment exchanges, in availability of information about job vacancies and labor supply, and so on, would tend to lower the natural rate of unemployment. I use the term "natural" for the same reason Wicksell did—to try to separate the real forces from monetary forces. Let us assume that the monetary authority tries to peg the "market" rate of unemployment at a level below the "natural" rate. For deflniteness, suppose that it takes 3 percent as the target rate and that the "natural" rate is higher than 3 percent. Suppose also that we start out at a time when prices have been stable and when unemployment is higher than 3 percent. Accordingly, the authority increases the rate of monetary growth. This will be expansionary. By making nominal cash balances higher than people desire, it will tend initially to lower interest rates and in this and other ways to stimulate spending. Income and spending will start to rise. To begin with, much or most of the rise in income will take the form of an increase in output and employment rather than in prices. People have been expecting prices to be stable, and prices and wages have been set for some time in the future on that basis. It takes time for people to adjust to a new state of demand. Producers will tend to react to the initial expansion in aggregate demand by increasing output, employees by working longer hours, and the unemployed, by taking jobs now offered at former nominal wages. This much is pretty standard doctrine. 3 It is perhaps worth noting that this "natural" rate need not correspond to equality between the number unemployed and the number of job vacancies. For any given structure of the labor market, there will be some equilibrium relation between these two magnitudes, but there is no reason why it should be one of equality. * Strictly speaking, the rise in nominal wages will be less rapid than the rise in anticipated nominal wages to make allowance for any secular changes in real wages. 5 Stated in terms of the rate of change of nominal wages, the Phillips Curve can be expected to be reasonably stable and well defined for any period for which the average rate of change of prices, and hence the anticipated rate, has been relatively stable. For such periods, nominal wages and "real" wages move together. Curves computed for different periods or different countries for each of which this condition has been satisfied will differ in level, the level of the curve depending on what the average rate of price change was. The higher the average rate of price change, the higher will tend to be the level of the curve. For periods or countries for which the rate of change of price varies considerably, the Phillips Curve will not be well definod. My impression is that these statements accord reasonably well with the experience of the economists who have explored empirical Phillips Curve. Restate Phillips' analysis in terms of the rate of change of real wages—and even more precisely, anticipated real wages—and it all falls into place. That is why students of empirical Phillips Curves have found that it helps to include the rate of change of the price level as an independent variable. 220 But it describes only the initial effects. Because selling prices of products typically respond to an unanticipated rise in nominal demand faster than prices of factors of production, real wages received have gone down—though real wages anticipated by employees went up, since employees implicitly evaluated the wages offered at the earlier price level. Indeed, the simultaneous fall ex post in real wages to employers and rise ex ante in real wages to employees is what enabled employment to increase. But the decline ex post in real wages will soon come to affect anticipations. Employees will start to reckon on rising prices of the things they buy and to demand higher nominal wages for the future. "Market" unemployment is below the "natural" level. There is an excess demand for labor so real wages will tend to rise toward their initial level. Even though the higher rate of monetary growth continues, the rise in real wrages will reverse the decline in unemployment, and then lead to a rise, which will tend to return unemployment to its former level. In order to keep unemployment at its target level of 3 percent, the monetary authority would have to raise monetary growth still more. As in the interest rate case, the "market" rate can be kept below the "natural" rate only by inflation. And, as in the interest rate case, too, only by accelerating inflation. Conversely, let the monetary authority choose a target rate of unemployment that is above the natural rate, and they will be led to produce a deflation, and an accelerating deflation at that. What if the monetry authority chose the "natural" rate—either of interest or unemployment—as its target? One problem is that it cannot know what the "natural" rate is. Unfortunately, we have as yet devised no method to estimate accurately and readily the natural rate of either interest or unemployment. And the "natural" rate will itself change from time to time. But the basic problem is that even if the monetary authority knew the "natural" rate, and attempted to peg the market rate at that level, it would not be led to a determinate policy. The "market" rate will vary from the natural rate for all sorts of reasons other than monetary policy. If the monetary authority responds to these variations, it will set in train longer term effects that will make any monetary growth path it follows ultimately consistent with the rule of policy. The actual course of monetary growth will be analogous to a random walk, buffeted this way and that by the forces that produce temporary departures of the market rate from the natural rate. To state this conclusion differently, there is always a temporary trade-off between inflation and unemployment; there is no permanent trade-off. The temporary trade-off comes not from inflation per se, but from unaniticipated inflation, which generally means, from a rising rate of inflation. The widespread belief that there is a permanent trade-off is a sophisticated version of the confusion between "high" and "rising" that we all recognize in simpler forms. A rising rate of inflation may reduce unemployment, a high rate will not. But how long, you will say, is "temporary"? For interest rates, we have some systematic evidence on how long each of the several effects takes to work itself out. For unemployment, we do not. I can at most venture a personal judgment, based on some examination of the historical evidence, that the initial effects of a higher and unanticipated rate of inflation last for something like 2 to 5 years; that this initial effect then begins to be reversed; and that a full adjustment to the new rate of inflation takes about as long for employment as for interest rates, say, a couple of decades. For both interest rates and employment, let me add a qualification. These estimates are for changes in the rate of inflation of the order of magnitude that has been experienced in the United States. For much more sizable changes, such as those experienced in South American countries, the whole adjustment process is greatly speeded up. To state the general conclusion still differently, the monetary authority controls nominal quantities—directly, the quantity of its own liabilities. In principle, it can use this control to peg a nominal quantity—an exchange rate, the price level, the nominal level of national income, the quantity of money by one or another definition—or to peg the rate of change in a nominal quantity—the rate of inflation or deflation, the rate of growth or decline in nominal national income, the rate of growth of the quantity of money. It cannot use its control over nominal quantities to peg a real quantity—the real rate of interest, the rate of unemployment, the level of real national income, the real quantity of money, the rate of growth of real national income, or the rate of growth of the real quantity of money. 221 II. WHAT MONETARY POLICY CAN D O Monetary policy cannot peg these real magnitudes at predetermined levels. But monetary policy can and does have important effects on these real magnitudes. The one is in no way inconsistent with the other. My own studies of monetary history have made me extremely sympathetic to the oft-quoted, much-reviled, and as widely misunderstood, comment by John Stuart Mill. "There cannot * * *" he wrote, "be intrinsically a more insignificant thing, in the economy of society, than money; except in the character of a contrivance for sparing time and labor. It is a machine for doing quickly and commodiously, what would be done, though less quickly and eommodiously, without it: and like many other kinds of machinery, it only exerts a distinct and independent influence of its own when it gets out of order" (7, p. 488). True, money is only a machine, but it is an extraordinarily efficient machine. Without it, we could not have begun to attain the astounding growth in output and level of living we have experienced in the past two centuries—any more than we could have done so without those other marvelous machines that dot our countryside and enable us, for the most part, simply to do more efficiently what could be done without them at much greater cost in labor. But money has one feature that these other machines do not share. Because it is so pervasive, when it gets out of order, it throws a monkey wrench into the operation of all the other machines. The great contraction is the most dramatic example but not the only one. Every other major contraction in this country has been either produced by monetary disorder or greatly exacerbated by monetary disorder. Every major inflation has been produced by monetary expansion—mostly to meet the overriding demands of war which have forced the creation of money to supplement explicit taxation. The first and most important lesson that history teaches about what monetary policy can do—and it is a lesson of the most profound importance—is that monetary policy can prevent money itself from being a major source of economic disturbance. This sounds like a negative proposition: avoid major mistakes. In part it is. The great contraction might not have occurred at all, and if it had, it would have been far less severe, if the monetary authority had avoided, mistakes, or if the monetary arrangements had been those of an earlier time when there was no central authority with the power to make the kinds of mistakes that the Federal Reserve System made. The past few years, to come closer to home, would have been steadier and more productive of economic well-being if the Federal Reserve had avoided drastic and erratic changes of direction, first expanding the money supply at an unduly rapid pace, then, in early 1966, stepping on the brake too hard, then, at the end of 1966, reversing itself and resuming expansion until at least November 1967 at a more rapid pace than can long be maintained without appreciable inflation. Even if the proposition that monetary policy can prevent money itself from being a major source of economic disturbance were a wholly negative proposition, it would be none the less important for that. As it happens, however, it is not a wholly negative proposition. The monetary machine has gotten out of order even when there has been no central authority with anything like the power now possessed by the Fed. In the United States, the 1907 episode and earlier banking panics are examples of how the monetary machine can get out of order largely on its own. There is therefore a positive and important task for the monetary authority—to suggest improvements in the machine that will reduce the chances that it will get out of order, and to use its own powers so as to keep the machine in good working order. A second thing monetary policy can do is provide a stable background for the economy—keep the machine well oiled, to continue Mill's analogy. Accomplishing the first task will contribute to this objective, but there is more to it than that. Our economic system will work best when producers and consumers, employers and employees, can proceed with full confidence that the average level of prices will behave in a known way in the future—preferably that it will be highly stable. Under any conceivable institutional arrangements, and certainly under those that now prevail in the United States, there is only a limited amount of flexibility in prices and wages. We need to conserve this flexibility to achieve changes in relative prices and wages that are required to adjust to dynamic changes in tastes and technology. We should not dissipate it simply to achieve changes in the absolute level of prices that serve no economic function. In an earlier era, the gold standard was relied on to provide confidence in future monetary stability. In its heyday it served that function reasonably well. It clearly 222 no longer does, since there is scarce a country in the world that is prepared to let the gold standard reign unchecked—and there are persuasive reasons why countries should not do so. The monetary authority could operate as a surrogate for the gold standard, if it pegged exchange rates and did so exclusively by altering the quantity of money in response to balance-of-payment flows without "sterilizing" surpluses or deficits and without resorting to open or concealed exchange control or to changes in tariffs and quotas. But again, though many central bankers talk this way, few are in fact willing to follow this course—and again there are persuasive reasons why they should not do so. Such a policy would submit each country to the vagaries not of an impersonal and automatic gold standard but of the policies—deliberate or accidental—of other monetary authorities. In today's world, if monetary policy is to provide a stable background for the economy it must do so by deliberately employing its powers to that end, I shall come later to how it can do so. Finally, monetary policy can contribute to offsetting major disturbances in the economic system arising from other sources. If there is an independent secular exhilaration—as the postwar expansion was described by the proponents of secular stagnation—monetary policy can in principle help to hold it in check by a slower rate of monetary growth than would otherwise be desirable. If, as now, an explosive Federal budget threatens unprecedented deficits, monetary policy can hold any inflationary dangers in check by a slower rate of monetary growth than wrould otherwise be desirable. This will temporarily mean higher interest rates than would otherwise prevail—to enable the Government to borrow the sums needed to finance the deficit—but by preventing the speeding up of inflation, it may well mean both lower prices and lower nominal interest rates for the long pull. If the end of a substantial war offers the country an opportunity to shift resources from wartime to peacetime production, monetary policy can ease the transition by a higher rate of monetary growth than would otherwise be desirable—though experience is not very encouraging that it can do so without going too far. I have put this point last, and stated it in qualified terms—as referring to major disturbances—because I believe that the potentially of monetary policy in offsetting other forces making for instability is far more limited than is commonly believed. We simply do not know enough to be able to recognize minor disturbances when they occur or to be able to predict either what their effects will be with any precision or what monetary policy is required to offset their effects. We do not know enough to be able to achieve stated objectives by delicate, or even fairly coarse, changes in the mix of monetary and fiscal policy. In this area particularly the best is likely to be the enemy of the good. Experience suggests that the path of wisdom is to use monetary policy exi>licitly to offset other disturbances only when they offer a "clear and present danger." III. HOW SHOULD MONETARY POLICY BE CONDUCTED? How should monetary policy be conducted to make the contribution to our goals that it is capable of making? This is clearly not the occasion for presenting a detailed "Program for Monetary Stability"—to use the title of a book in which I tried to do so (3). I shall restrict myself here to two major requirements for monetary policy that follow fairly directly from the preceding discussion. The first requirement is that the monetary authority should guide itself by magnitudes that it can control, not by ones that it cannot control. If, as the authority has often done, it takes interest rates or the current unemployment percentage as the immediate criterion of policy, it will be like a space vehicle that has taken a fix on the wrong star. No matter how sensitive and sophisticated its guiding apparatus, the space vehicle will go astray. And so will the monetary authority. Of the various alternative magnitudes that it can control, the most appealing guides for policy are exchange rates, the price level as defined by some index, and the quantity of a monetary total—currency plus adjusted demand deposits, or this total plus commercial bank time deposits, or a still broader total. For the United States in particular, exchange rates are ail undesirable guide. It might be worth requiring the bulk of the economy to adjust to the tiny percentage consisting of foreign trade if that would guarantee freedom from monetary irresponsibility—as it might under a real gold standard. But it is hardly worth doing so simply to adapt to the average of whatever policies monetary authorities in the rest of the world adopt. Far better to let the market, through floating exchange rates, adjust to world conditions the 5 percent or so of our resources 223 devoted to international trade while reserving monetary policy to promote the effective use of the 95 percent. Of the three guides listed, the price level is clearly the most important in its own right. Other things the same, it would be much the best of the alternatives— as so many distinguished economists have urged in the past. But other things are not the same. The link between the policy actions of the monetary authority and the price level, while unquestionably present, is more indirect than the link between the policy actions of the authority and any of the several monetary totals. Moreover, monetary action takes a longer time to affect the price level than to affect the monetary totals and both the time lag and the magnitude of effect vary with circumstances. As a result, we cannot predict at all accurately just what effect a particular monetary action will have on the price level and, equally important, just when it will have that effect. Attempting to control directly the price level is therefore likely to make monetary policy itself a source of economic disturbance because of false stops and starts. Perhaps, as our understanding of monetary phenomena advances, the situation will change. But at the present stage of our understanding, the long way around seems the surer way to our objective. Accordingly, I believe that a monetary total is the best currently available immediate guide or criterion for monetary policy—and I believe that it matters much less which particular total is chosen than that one be chosen. A second requirement for monetary policy is that the monetary authority avoid sharp swings in policy. In the past, monetary authorities have on occasion moved in the wrong direction—as in the episode of the great contraction that I have stressed. More frequently, they have moved in the right direction, albeit often too late, but have erred by moving too far. Too late and too much has been the general practice. For example, in early 1966, it was the right policy for the Federal Reserve to move in a less expansionary direction—though it should have done so at least a year earlier. But when it moved, it went too far, producing the sharpest change in the rate of monetary growth of the post-war era. Again, having gone too far, it was the right policy for the Fed to reverse course at the end of 1966. But again it went too far, not only restoring but exceeding the earlier excessive rate of monetary growth. And this episode is no exception. Time and again this has been the course followed—as in 1919 and 1920, in 1937 and 1938, in 1953 and 1954, in 1959 and 1960. The reason for the propensity to overreact seems clear: the failure of monetary authorities to allow for the delay between their actions and the subsequent effects on the economy. They tend to determine their actions by today's conditions—but their actions will affect the economy only 6 or 9 or 12 or 15 months later. Hence they feel impelled to step on the brake, or the accelerator, as the case may be, too hard. My own prescription is still that the monetary authority go all the way in avoiding such swings by adopting publicly the policy of achieving a steady rate of growth in a specified monetary total. The precise rate of growth, like the precise monetary total, is less important than the adoption of some stated and known rate. I myself have argued for a rate that would on the average achieve rough stability in the level of prices of final products, which I have estimated would call for something like a 3 to 5 percent per year rate of growth in currency plus all commercial bank deposits or a slightly lower rate of growth in currency plus demand deposits only.6 But it would be better to have a fixed rate that would on the average produce moderate inflation or moderate deflation, provided it was steady, than to suffer the wide and erratic perturbations we have experienced. Short of the adoption of such a publicly stated policy of a steady rate of monetary growth, it would constitute a major improvement if the monetary authority followed the self-denying ordinance of avoiding wide swings. It is a matter of record that periods of relative stability in the rate of monetary growth have also been periods of relative stability in economic activity, both in the United States and other countries. Periods of wide swings in the rate of monetary growth have also been periods of wide swings in economic activity. By setting itself a steady course and keeping to it, the monetary authority could make a major contribution to promoting economic stability. By making that course one of steady but moderate growth in the quantity of money, it would make a major contribution to avoidance of either inflation or deflation of prices. 6 In an as yet unpublished article on "The Optimum Quantity of Money." I conclude that a still lower rate of growth, something like 2 percent for the broader definition, might be better yet in order to eliminate or reduce the difference between private and total costs of adding to real balances. 224 Other forces would still affect the economy, require change and adjustment, and disturb the even tenor of our ways. But steady monetary growth would provide a monetary climate favorable to the effective operation of those basic forces of enterprise, ingenuity, invention, hard work, and thrift that are the true springs of economic growth. That is the most that we can ask from monetary policy at our present stage of knowledge. But that much—and it is a great deal—is clearly within our reach. REFERENCES 1. H. S. ELLIS, ed., A Survey of Contemporary Economics. Philadelphia 1948. FRIEDMAN, "The Monetary Theory and Policy of Henry Simons," Jour. Law and Econ., Oct. 1967, 10, 1-13. 3. 9 ± Program for Monetary Stability. New York 1959. 4. E. A. GOLDENWEISER, "Postwar Problems and Policies," Fed. Res. Bull., Feb. 1945, 81, 112-21. 2. MILTON 5. P. T. HOMAN AND FRITZ MACHLUP, ed., Financing American Prosperity. New York 1945. 6. A. P. LERNER AND F. D. GRAHAM, ed,, Planning and Paying for Full Employment. Princeton 1946. 7. J. S. MILL, Principles of Political Economy, Bk. I l l , Ashley ed. New York 1929. MILTON FRIEDMAN ON CURRENT MONETARY POLICY I have been watching with increasing apprehension, concern and incredulity the behavior of the quantity of money over the past 8 months. The Federal Reserve System clearly dioes not intend to produce a serious recession in 1967. Yet continuation of their present policy will make such an outcome all but inevitable. The accompanying chart shows the reason for concern. It plots two monetary magnitudes: MI, the total usually designated "the money supply" by the Fed; and M2, a broader total that includes also time deposits at commercial banks. The striking feature of the chart is the sharp reversal in both totals in April 1966. Before then, both totals were growing rapidly. Since April, MI has actually declined—something it has rarely done except before and during severe recessions—and M2 has grown at a sharply reduced rate. Since September, both totals have been declining. This is the sharpest turnaround since the end of the war. Slower monetary growth was badly needed in order to stem inflation—but a good thing was carried too far. Do changes in the quantity of money matter? There is massive historical evidence that they do. Every economic recession but one in the United States in the past century has been preceded by a decline in the rate of growth of the quantity of money. And the sharper the decline, the more serious the subsequent recession—though this tendency is far from uniform. Changes in monetary growth affect the economy only slowly—it may be 6 or 12 or 18 months or even more before their effects are manifest. That is a major reason why the connection is easily overlooked. Recent experience conforms to the historical record. Acceleration of monetary growth in 1962 was followed by economic expansion. The monetary growth rate was too high—but it took until 1965 for its cumulative effects to produce rising prices. The price rise started) the Fed talking about the need for tighter money, but it acted in the opposite direction; monetary growth accelerated still more, intensifying inflationary pressure and producing the rapid price rises of recent months. The sharp braking of monetary growth in April of 1966, has in turn only recently been showing up in spreading signs of pending recession. Why has the Fed permitted the quantity of money to behave so eratically? Primarily, I believe, because it has used misleading criteria of policy—it is inconceivable that the quantity of money as measured by MI could decline for 8 months if the Fed had been determined to have it grow. It is as if a space vehicle took a fix on the wrong star. No matter how sensitive and sophisticated its guiding apparatus, it would go astray. Similarly, the men who guide the Fed have been going astray because they have been looking at interest rates and other measures of credit conditions rather than the quantity of money. 225 N'owswcek—Van Dyke } Interest rates began rising in 1965 because of the sharp rise in the demand for credit that accompanied the onset of inflation. The Fed slowed the rise by accelerating monetary growth—but the rates continued rising, and the Fed interpreted this as a sign that it had tightened, whereas in fact it had eased. Similarly, interest rates are currently showing some weakness—because the demand for credit has been declining in response to monetary restriction since April 1966. Yet the Fed interprets the weakness as a sign that it has eased— whereas only the Fed's continued tightness prevents interest rates from falling more rapidly. The Fed's erratic policy reflects also its failure to allow for the delay between its actions and their effects on the economy. Said Governor Robertson of the Board in a recent speech: "Monetary policy will be formulated by the Federal Reserve, day by day, in the light of economic conditions as they emerge." This is a formula guaranteed to produce bad policy. If it is followed, the Fed will continue to step too hard on the brake until the recessionary effects are clear and unmistakable, and then will step too hard on the accelerator. Like a good duck hunter, the Fed should lead its target, not shoot where it now is. What policy should the Fed now adopt? It is almost surely too late to prevent a recession—that damage has already been done. It is not too late to prevent the recession from turning into a severe downturn. To that end, the Fed should at once act to increase the quantity of money at a rate of about 5 percent per year for M2. If the Fed adopted and persisted in such a policy, it could moderate the coming recession without paving the way for a new burst of inflation. [From Newsweek, June 3, 1968] MONETARY POLICY (By Milton Friedman) In two earlier columns on monetary policy, I was highly critical of the Federal Reserve System for acting too late and then, when it did act, for overreacting. This time, I come to praise, not to criticize. Since November 1967, the Fed has moved not only in the right direction but also by about the right amount. 226 The recent record is summarized in the accompanying table, which gives the annual rate of growth for two monetary totals, for industrial production, which is a sensitive index of changes in economic activity, and for consumer prices. Because it takes time for monetary changes to exert their influence, the rates of growth of production and prices are given for periods that begin 6 months later than the corresponding periods for money. Rate of change (percent per year) Period for money April 1965 to April 1966 April 1966 to January 1967 January 1967 to November 1967. November 1967 to April 1968 Money1 Mi M2 6.0 -.2 7.7 4.7 9.6 3.7 11.9 5.3 Industrial production 9.9 -2.3 5.2 Consumer prices 3.7 2.3 4.0 Period for production and prices October 1965 to October 1966. October 1966 to July 1967. July 1967 to April 1968. * Mi = currency plus adjusted demand deposits. M 2 = M i plus time deposits in commercial banks. As an aid in interpreting these numbers, let me note that a long-term rate of growth in M2 of about 5 percent per year would be consistent with roughly stable prices. The 5 percent would match the growth in output and leave a little over to satisfy the desire of people to hold somewhat more money relative totheir income as they become richer. From April 1965 to April 1966 the Fed permitted the money supply to grow rapidly despite signs that inflation was accelerating. At long last, in April 1966, it stepped on the brake—abruptly and, as the table shows, too hard. The result was the so-called money crunch in the fall of 1966, the slowdown in the economy recorded in the decline in industrial production, and a cut in price inflation from a rate of 3.7 percent per year to a rate of 2.3 percent per year. The sharper response of production than of prices is typical. An inflationary process, once underway, develops an inertia of its own. It takes an economic slowdown to stop the acceleration of prices and, even then, it takes a long time to restore price stability. That is why it is so important to prevent inflation from gaining momentum. Concerned by the signs of emerging slack in the economy, the Fed reversed policy in January 1967. This time, to its credit, it acted more promptly than usual. But, as usual, it reacted too sharply, not only restoring, but exceeding the earlier excessive rate of monetary growth. As a result, I wrote last October, "it is almost surely too late to prevent an appreciable price rise—that damage has already been done. It is not too late to prevent the price rise from turning into a severe inflation. To that end, the Fed should at once act to limit the increase in the quantity of money to a rate of about 5 percent per year for M2. If the Fed adopted and persisted in such a policy, it could moderate the coming inflation without paving the way for a new recession." It was too late to prevent an appreciable price rise. Prices have recently been rising at 4 percent per year. But in November 1967 the Fed did reverse its policy and M2 has been growing since at only slightly more than 5 percent per year. There has not yet been time for this moderate policy to have much effect—as is reflected in the absence of any entries for production and prices in our table matching the final period for money. But if the Fed persists in its present policy, the exuberant expansion in the economy will taper off later this year and so will the rate of price rise—whether or not there is a tax increase. There may also be some rise in unemployment before the price inflation is brought under control, though any rise is likely to be small. But will the Fed persist? Will it keep its cool? Or will continuing inflation lead it, as in April 1966 to step still harder on the brake in the hope of getting quicker results? Alternatively, will the first signs of reduced expansion and increased unemployment lead it, as in January 1967, to start the printing presses whirring again and set off a new burst of inflation? The Fed's steadiness in the past 6 months—despite the gold crisis, high, and rising interest rates and the controversy over Government expenditures and taxes—is a hopeful augury. STATEMENT OF GARY FROMM, THE BROOKINGS INSTITUTION Thank you for the opportunity for submitting a statement in regard toH.R. 11. While I find much in the act that I would approve, I also see certain dangers. In particular, allowing the President directly to control monetary policy makes that stabilization and growth instrument subject to the viscissitudes of shortrun fluctuations in political opinions. Moreover, mistakes in forecasting would tend to be magnified if all countercyclical policies were closely coordinated. At this juncture, I would recommend that the Federal Reserve be given clearer congressional guidelines as to its economic objectives and be required more fully to disclose its anticipated stabilization program and report on the results. Formalization of the coordinating relationships vis-a-vis the Council of Economic Advisers, the Bureau of the Budget, and the Treasury Department might be helpful, too. As to administrative reforms, certainly some are in order. However, at present, greatly weakening the Board's autonomy might be deferred until the response to improved guidelines, reporting, and coordination can be assessed. Should you desire, I would be happy to comment on these matters at greater length. STATEMENT OF TILFORD C. GAINES, MANUFACTURERS HANOVER TRUST CO. H.R. 11 1. QUESTIONS ON MONETARY POLICT GUIDELINES AND OPENMARKET OPERATIONS I. Do you believe that a program coordinating fiscal, debt managemerit, and monetary policies should be set forth at the beginning of each year for the purpose of achieving the goals of the Employment Act, or alternatively should we treat monetary and fiscal policies as independent mutually exclusive stabilization policies? It w^ould be helpful to have a program set forth at the beginning of each year aimed at broad coordination of public economic policies. Such a program would have to be flexible to permit adaption to changing circumstances and, in most cases, would deal only with objectives rather than with techniques for achieving those objectives. 2. If you believe a program should be specified, do you believe that the President should be responsible for drawing up this program, or alternatively should such responsibility be dispersed between the Federal Reserve System and agencies responsible to the President? (Please note that informal consulting arrangements can be made as desired whether responsibility is assigned to the President or divided between the President and the Federal Reserve. The concern here is with the assignment of formal responsibility for drawing up the economic program). The responsibility for drawing up the program should be dispersed between the Federal Reserve System and the agencies responsible to the President, with the end product a single document signed both by the Federal Reserve and the Council of Economic Advisers (as agent for the President). Differences between the agencies on the economic 228 outlook and on broad policy objectives would seldom arise, and if they should arise there would be advantages in having them discussed in open debate rather than suppressed. Only two cases come to mind of such differences in recent years: in 1950-51, when the Federal Reserve abandoned support of a pegged interest rate structure and in late 1965 when the Federal Reserve raised the discount rate contrary to the wish of the administration. In both cases, subsequent events vindicated the Federal Reserve. This does not argue that the Federal Reserve possesses infallible wisdom, but it does suggest the danger of imputing such wisdom to the executive branch. There is the further point that the tradition of central bank independence grew out of a long history of currency debasement by monarchs and elected officials. Perhaps this is no longer an important consideration, but it does suggest that the central bank might best be permitted its present independence unless there are clear benefits to be derived from restricting that independence. There has been close cooperation between the agencies of the executive branch and the Federal Reserve in developing economic policies. The only benefit that might be realized by changing the present arrangements would be the concentration of policy power in the executive branch. This would seem to be a dubious benefit. 3. Concerning monetary policy guidelines: A. Should monetary "policy be used to try to achieve the goals of the Employment Act via intervention of money supply {defined as desired) as provided in H.R. 11, or alternatively should H.R. 11 ~be amended to make some other variable or variables the immediate target of monetary policy; for example, interest rates, bank credit, liquidity, high powered or base-money, total bank reserves, excess reserves and free reserves? Please define the target variable or combination of variables recommended and state the reasons for your choice. (If desired, recommend a target variable or variables not listed here). It wouid be most helpful if, in providing the reasons for your choice, you list the actions the Federal Reserve should take to control the target variable (or variables) and also explain the link between your recommended target of monetary policy and the goals of the economy as defined by the Employment Act. The immediate target variable of Federal Reserve operations in trying to achieve the goals of the Employment Act should meet two requirements: First, it should be related causally to economic activity; second, it should be a variable that the Federal Reserve System is able to regulate within rather broad limits. Based on my observation of the American financial system, money supply—however defined—does not meet either of these requirements. The rate of change in money supply is primarily a result rather than a cause of the rate of change in current dollar economic activity and, in any case, there is reason to doubt that the Federal Reserve is able to regulate rate of change in money supply. The second quarter of this year provides an illustration of the impotence of the Federal Reserve in regulating money supply growth. Demand deposts in the second quarter grew at an 8.5-percent annual rate, in spite of severely restrictive Federal Reserve policy. For the Federal Reserve to have attempted to reduce that rate of growth to fall within any reasonable target range would have required policy actions that would have been devastating to the orderly functioning of the national and international money market. The popular monetary theories are rooted in experience in compara- 229 tively primitive 19th century economies, experience that is not wholly applicable to our highly complex financial system. The simple fact is that our financial mechanism offers a vast array of competitive financial assets among which the holder of financial claims may select, and there is no realistic way that the Federal Reserve may prevent the conversion process among these claims. None of this is to say that money supply would not grow at a fairly orderly rate if the economy itself were growing at an orderly rate, but the cause-effect relationship runs from orderly economic growth to orderly monetary growth rather than the reverse. A preferable target variable would be the total and composition of credit flows within the flow of funds accounting system. The broad economic projection prepared by the Council of Economic Advisers each year is translatable into rough approximations of the credit required, by sectors if the growth targets in the Council's projections are to be achieved. Through its operations in the Government securities market, and perhaps in Government agency securities, the Federal Reserve System is in a position to have a marginal influence upon the overall availability of credit and the availability of credit at different maturity sectors. B. Should the guidelines of monetary policy be specified in terms of some index of past, present, or future economic activity, or alternatively in terms of the target variable''s value or growth? For example, should, the President's 1969 program for achieving the goals of the Employment Act be formulated to require consistency with some set of overall indicators of economic activity, or alternatively so that your target variable attains a certain value or growth regardless of the economic winds? Please indicate the reasons for your preference. Policy guidelines should be derived consistent with the growth target established in the report of the Council of Economic Advisers. The relationship between the financial variables which the Federal Reserve might use as an immediate target, including the proposed flow of funds target, is not invariable with respect to real economic growth. Therefore, to establish guidelines in terms of some fixed rate of growth in the immediate target variable would not guarantee optimum results for the ultimate variable, real economic growth. G. For only those persons toho recommend that some index of economic activity be used to guide the monetary authorities in controlli/ng the target variable: Should we use a leading (forward looking), lagging (backward looking) or coincident indicator of economic activity? It would be most helpful also if you would identify the index you would like to see used and specify how the target variable should be related to this index. There is no single index of economic activity that is wholly adequate for guiding the monetary authorities in controlling the target variable. All business condition analysis is an effort to predict the future from a wide array of data measuring the immediate past and present. Mechanically, perhaps the most useful approach would be continuous measurement and prediction of the economy through the medium of a comprehensive predictive model, such as the MIT-Federal Reserve financial flows model. The month-by-month output of such a model would measure the extent to which the economy was achieving and was expected to achieve the targets for economic growth established by the Council of Economic Advisers in collaboration with the Federal Re- 230 serve System. The model would simultaneously test the extent to which the financial flow targets were being achieved and the extent to which failure to achieve the immediate financial flow targets might be impeding achievement of the economic growth target. Concerning debt management policy: Given the goals of the Employment Act, what can debt management do to help their implementation? (If you believe that debt management has no role to play in this matter, please explain why.) The principal contribution of debt management should be to avoid financing operations or practices that would needlessly interfere with the olderly flow of funds to other users of credit. At times of huge Treasury deficits, as in the past 2 years, there is little that debt management can do to avoid being a disruptive influence, since it is inescapable that some part of the money raised by the Treasury will be preempted from other uses, with resulting distortions in sectorial economic and financial balance. During what one may hope w ill be more typical periods of only moderate deficit or surplus, the principal area of interest in Treasury financing is the refunding of existing debt rather than the financing of new debt. At such times, three considerations might guide the Treasury. First, every effort should be made to maintain an orderly maturity distribution of the debt in order to avoid a piling up of short debt that would require more frequent and larger Treasury financing operations. This guideline does not imply that the Treasury must attempt to sell large amounts of long-term bonds; it does imply that the Treasury design its financing so as regularly to place new securities in the intermediate maturity range. Second, the Treasury should avoid massive shifts in the debt between maturity areas. For example, the refunding of several billions of dollars of maturing debt into the long-term area might be feasible at a time of easy bond market conditions, but the resulting preemptng of long-term funds and freeing of short-term funds would have selective effects upon the availability of credit to different types of borrowers. In particular, such financing could have serious backlash effects upon the mortgage market. Third, Treasury financing should be devised with an eye to the international money market and the U.S. balance of payments. The effect of Treasury financing decisions upon the term structure of interest rates might, at certain times, sharply influence the movement of short-term funds into or out of the United States. This should not be a determining consideration, but it should have an influence upon Treasury planning. 5. Concerning open market operations: H.R. 11 requires that the FOMC conduct open market transactions "in accordance with the programs and policies of the President pursuant to the Employment Act of And in this connection, H.R. 11 provides that "The Federal Reserve Board shall submit a quarterly report to the Congress stating, in comprehensive detail, its past and prospective actions and policies under this section and otherwise with respect to monetary affairs, and indicating specifically how such actions and policies facilitate the economic program of the Presidents A. H.R. 11 makes no provision whatever for conducting open market operations for so-called "defensive" or uroad-clearing" purposes, that is to counteract seasoned and other transient factors affecting money market and credit conditions. Do you see any merit in using open market operations for defensive purposes or should they be used 231 only to facilitate achievement of the Presidents economic program and the goals of the Employment Act? What risks and costs, if any, must be faced and paid if open market transactions are used to counteract transient influences? There is no realistic way that the Federal Eeserve can avoid some amount of "defensive" open market operations. At different seasons of the year, over specific holidays, etc., there are very large changes in available reserves. For example: deposits increase seasonally each fall and winter, causing required reserves to increase; deposits are pulled down sharply over the Labor Day weekend and the Fourth of July as consumers withdraw currency for long weekends or vacations. It is difficult to see what point would be served by permitting these predictable influences upon bank reserves alternatively to ease or tighten the money market. To the extent that the Federal Reserve goes to a longer reserve averaging period, liberalizes use of the discount window, or permits reserve excesses as well as shortages to be carried into the next averaging period, the need for day-to-day defensive open market operations will, of course, be reduced. The various amendments to Federal Reserve regulations currently being proposed will move in this direction, but some intervention to deal with seasonal variations in reserve needs will still be required. It would seem that the heart of this question is not whether all defensive operations should be avoided, reducing open market operations to a mechanical provision of a predetermined amount of new reserves each week, but whether all of the daily in-and-out operations now undertaken to steady the "tone of the market" are necessary. Expressed in this way, and in consideration of the changes in the regulations now pending, it does seem the money market could be given greater latitude for making its own day-to-day adjustments without the continuous mothering of the open market desk. B. Do you believe that monetary policy can be effectively and efficiently implemented by open market operations? C. For what purpose, if any, shoidd (a) rediscounting, (b) changes in reserve requirements, and (c) regulation Q be used? TIo%o might H.R. 11 be amended to implement your recommendations? Open market operations are the most powerful of the instruments available to the Federal Reserve System, but probably should not be relied upon exclusively for achieving immediate Federal Reserve targets. For example, if open market operations are to avoid the dayto-day defensive actions aimed at limiting fluctuations in the reserve base or in the "tone of the market," it is essential that liberal discount facilities be available in order that banks may make day-to-day or week-to-week adjustments through the discount window for shortterm fluctuations in total reserves. Also, as distasteful as any form of price control may be, the record suggests that regulation Q may be used in certain circumstances to achieve restraining effects on bank credit, or to assist in directing savings flow through intermediaries, that cannot be as effectvely achieved through open market operations. Ideally, one might like to see all types of interest rate regulation abandoned, so as to permit the private credit markets to allocate credit competitively. Realistically, however, traditional attitudes toward interest rates and their reflection in political concern suggests that the objective of a truly competitive financial system is not a likelihood for some while. One in- 232 strument of Federal Reserve policy that might be foregone is changes in reserve requirements. In fact, reserve requirements themselves (particularly on time deposits) are in the nature of a discriminatory tax on commercial banks, and the very principle of reserve requirements should be carefully examined. Given their existence, however, there can be little justification for the periodic increases in requirements that have been made in recent years when open market operations could as easily and more equitably have accomplished the same thing. I would have no specific recommendations for amendments to H.R. 11 that might incorporate these comments on the use of instruments other than open market operations in the executive of Federal Reserve policy. D. Do you see any merit in requiring the Federal Reserve Board to make detailed quarterly reports to the Congress on past and prospective actions and policies? Are there any risks and costs in this procedure? In what ways, if any, would you modify the reporting provision? What information do you believe should be included in such reports as you recommend the Federal Reserve submit to the Congress? The principal advantage that might be obtained from requiring the Federal Reserve Board to make detailed quarterly reports to the Congress would be a better informed Congress, better able to exercise its responsibilities for the management of the Nation's money. There would be obvious costs in the preparation and hearings on such reports, but such costs would be nominal when viewed against the benefits that might be obtained. There should be no risks in this requirement unless the hearings were used as the forum for a vendetta against the Federal Reserve or unless the hearings led to instructions from the Congress that the Federal Reserve achieve highly precise targets, an outcome that could be realized only at the cost of Federal Reserve flexibility and financial market stability. In its reports to the Congress, the Federal Reserve should review the broad economic targets toward which policies have been aimed, the immediate targets expressed in terms of credit flows, the actions that have been taken to achieve those targets, and the degree of success obtained on both the immediate targets and the ultimate economic targets. E. What costs and benefits would accrue if representatives of the Congress, the Treasury, and the CEA were observers at Open Market Committee meetings? The suggestion that representatives of the Congress, the Treasury, and the CEA attend Open Market Committee meetings as observers probably will encounter serious opposition because of the high degree of confidentiality essential to these meetings. In the abstract, it is by no means obvious that this opposition is well grounded. After all, carefully selected Members of the Congress and of other Government agencies participate in the most highly confidential discussions within such critical functions as the Defense Department. There is no reason for presupposing that representatives from the Congress, the Treasury, and the CEA would be less responsible in guarding the confidentiality of Open Market Committee meetings than are the 40 or 50 Governors, bank presidents, and senior staff people who now attend these meetings. It was my privilege to attend meetings of the Open Market Committee for a number of years while I was associated with the Federal Reserve Bank of New York. It occurs to me that an important benefit from having outsiders, particularly Members of the Congress attend these meetings would be 233 the impression that the observer would form of the intelligence and dedication to responsible money and credit management that characterizes the meetings. The observers would be useful as liaison between the Federal Reserve, the Congress, and the other departments of Government. In particular, the Members of the Congress attending these meetings would be much better able to understand the objectives that the Federal Reserve had been pursuing and the often intricate reasons why those objectives were not fully realized. Unfortunately, all of this is in the abstract. In the actual case, if the observers were people prejudiced against the Federal Reserve System and committed to its destruction, the result of their attendance at meetings of the Open Market Committee would be disruptive to Federal Reserve policy and injurious to the welfare of the United States. It follows that if such observers are to be invited to FOMC meetings, the Federal Reserve should have some voice in their selection. II. APPRAISAL OF THE STRUCTURE OF THE FEDERAL RESERVE * II.R. 11 provides for the following structural changes in the Federal Reserve System: 1. Retiring Federal Reserve bank stock; 2. Reducing the number of members of the Federal Reserve Board to five and their terms of office to no longer than 5 years; 3. Making the term of the Chairman of the Board coterminus with that of the President of the United States; Ip. An audit for each fiscal year of the Federal Reserve Board and the Federal Reserve Banks and their branches by the Comptroller General of the United States; and 5. Funds to operate the Federal Reserve System to be appropriated by the Congress of the United States. Please comment freely on these several provisions. In particular, it ivould be most helpful if you would indicate any risks involved in adopting these provisions and discuss whether their adoption would facilitate the grand aim of H.R. 11, which is to provide for coordination by the President of monetary and fiscal policies. Some of the structural changes in the Federal Reserve System provided in H.R. 11 are rather unimportant while appearing to be dramatic while others of great substance appear to be innocent. Specifically, retiring the Federal Reserve bank stock now owned by the member banks appears to be a fundamental alteration in the character of the Federal Reserve System, shifting it from private to public ownership, but logically this should not be a significant change. The Federal Reserve System has been wholly independent of the member banks from its founding, and the ownership of the Reserve Banks by the member banks has not given the member banks any control or influence upon Federal Reserve policy. The present form of corporate structure has given each Reserve bank a board of directors, providing the mechanism for an important flow of information from industry and finance into the Federal Reserve. But these boards could be redesignated advisory committees and serve the same function. The matter of providing that the Comptroller General of the United States audit the Federal Reserve Board and the Federal Reserve banks each year has been strongly resisted by the Federal Reserve as a 21-570—68 16 234 foot in the door of their independence that might lead to ultimate, complete loss of independence. To a reasoning person it might seem that this is carrying the principle of the camel's nose somewhat further than could be supported. There is a clear dividing line between the right of the Government to audit the Federal Reserve System and a decision to bring Federal Reserve policy under control and determination by the executive branch of the Government. Similarly, there should be no objection to making the term of the Chairman of the Federal Reserve Board coterminous with that of the President of the United States. In fact, the incumbent Chairman on numerous occasions has proposed that this be done. The practical issue involved here is that a harmonious relationship between the Executive and the Federal Reserve is much more likely to be achieved if the President is able to select his own Chairman from among the members of the Board at the beginning of his term of office. In turn, the improved harmony between the Executive and the Federal Reserve should tend to promote a better meshing of fiscal and monetary policies toward common economic ends. Even in the case of these logically innocuous changes, however, the question arises as to whether under the actual circumstances the changes would be desirable. What is important in addition to the logic of the case is the motivation behind the proposed changes. If the motivation is no more than a tidying up of unimportant anomalies in the law, then it would be difficult to argue against the changes. But if the motivation is a desire to weaken or substantially alter the Federal Reserve System through a process of gradually nibbling away at the arrangements that safeguard Federal Reserve independence, then the changes should be strongly resisted. The minor benefits to be obtained from the proposals would not justify the dangers to the economic welfare of the country of a vulnerable, weakened central bank. The other structural changes proposed in H.R. 11 could be of major significance for the effective functioning of the Federal Reserve System. The proposal to reduce the number of Governors to five and to reduce their terms from 14 to no longer than 5 years is such a change. There certainly is nothing sacred about the present sevenman board and the present 14-year term of office, but a board consisting of seven men has functioned well and has brought in a broad range of interests and points of view. Also, the 14-year term has enabled the incumbent Governors to learn in depth the often complex theoretical and practical issues with which monetary policy must deal. In actual fact, a number of Governors have not served out their full terms, so that each President in recent years has had an opportunity to appoint more new Governors to the Federal Reserve Board than would be indicated by the 14 years provided in the Federal Reserve Act. At the same time, the long years served by any number of the Governors has made possible a continuity of policy and a depth of wisdom and understanding that shorter terms and steady turnover would not. As in all proposed changes of existing law or custom, the first question that must be anwered is whether or not change serves a useful purpose of sufficient importance to justify its enactment. In this case, the system of seven Governors 235 serving 14-year terms has worked well and there is no apparent overriding reason why this arrangement should be changed. Similarly, H.R. 11 would eliminate the Open Market Committee and concentrate responsibility for open market operations in the Board of Governors. Anyone familiar at firsthand with the work of the Open Market Committee is aware that the inclusion of five of the presidents on the committee has contributed importantly to the debt of discussion. Of course, Reserve bank presidents could continue to advise on developments in their respective districts, so that the regional contribution now provided for through meetings of the Open Market Committee would not be wholly lost. But once again, the question should be answered as to precisely what the advantages are that would be derived from this change. The Congress established the Federal Reserve System in a form consistent with our Federal Republic—a central bank with the various regions of the country represented in the policymaking organization. The system has worked well and there is no apparent reason why it should be changed. Finally, the proposal that funds to operate the Federal Reserve System be appropriated by the Congress is an apparently innocent proposal but one that is potentially dangerous. The long history of money management has repeatedly emphasized the need for the central bank to be as independent as possible from the political process. There is no more certain way to get the central bank involved in the political process than to make the appropriation of funds necessary for its existence subject to action by the Congress. By comparison with other Government agencies, the total staff and total expenses of the Federal Reserve System are nominal. There certainly can be no suspicion that the Federal Reserve is spending lavishly and that such expenditures might be curtailed if subjected to congressional scrutiny. It is not possible to foresee economies from this proposal; its only apparent purpose would seem to be to bring Federal Reserve policy directly under congressional control, and the only reason for this objective would be to enable the Congress to direct the Federal Reserve as to the kind of policies that should be followed if its appropriations are to be approved. The System has worked well as now constituted and almost surely over the history of the Federal Reserve has worked better than would have been possible if Congress had been calling the tune. STATEMENT OF WILLIAM I. GREENWALD, CCNY Replies to questions on monetary policy guidelines and open market operations, invited by Congressman Wright Patman, chairman, House of Representatives, Subcommittee on Domestic Finance of the Committee on Banking and Currency, on September 18, 1968. answers—I (1) I endorse a program coordinating fiscal, debt management, and monetary policies being set forth at the beginning of each year for the purpose of achieving the goals of the Employment Act. (2) The target program, for administrative simplicity, might be 236 best vested in the Office of the President, assuming formal consulting arrangements are mandated. (3) (A) Monetary policy to achieve targets and goals via intervention of the money supply is supported, again, in order to minimize the number of interventions. (3) (B) The guidelines for monetary policy need specification in terms of future economic activity (real terms), sustained by a certain value of growth (money terms). The reason for my preference is the unsettled and unresolved scientific theory and evidence regarding monetary theory. (3)( C) No comment. (3) (D) No comment. (3) (E) No comment. (3) (F) No comment. (4) Debt management policy has to be coordinated (subordinated) with monetary-fiscal guidelines set for the target program. (5) (A) Very short term factors—that is, seasonal and transient elements—should be neglected and minimally influenced by monetary intervention. (5) (B) No. (5) (C) The power to control and change margin requirements were given to the Federal Reserve System by section 7 of the SEA of 1934. The power over this selective and particularistic control might better be transferred to the Securities and Exchange Commission. (5) (E) In my opinion, the costs and benefits, aside from those relating to communication of economic information, primarily are administrative matters. II (1) No comment. (2) No comment. (3) No comment. (4) I endorse an annual audit of the type indicated. (5) I support a continued financial independence of the Federal Reserve System from either the executive or legislative branch, in order to minimize the political factors in the System. No comment. i n S T A T E M E N T OP H E R S C H E L I. GROSSMAN, B R O W N UNIVERSITY H.R. 11 raises three basic and distinct questions regarding the formulation and execution of monetary policy. I. What variable or variables should be used as proximate indicators of the consistency of current monetary policy with its goals and as proximate targets for monetary policy? 2. What should be the relative importance of fixed rules as contrasted with current discretion in selecting and determining the appropriate value of these proximate targets? 3. To the extent that current discretion is to be exercised, ivho should have the authority to exercise it? These comments begin with a discussion of the goals of monetary policy and then consider each of these questions in turn. In summary, the conclusions reached are as follows: 237 1. Interest rates and the money supply should be used as complementary indicators and proximate targets for monetary policy. 2. Discretion should be permitted in the execution of policy, so long as it is agreed to give priority to macroeconomic goals and to make explicit© the rationale for policy actions. 3. The formal authority to formulate and execute monetary policy probably should not be transferred from the Federal Reserve System to the administration, although to do so would not be likely to have significant consequences. I. GOALS OF MONETARY POLICY All three questions posed above deal directly with some operational aspect of monetary policy. However, in each case the answer will vary depending upon the ends or goals of monetary policy. First, proximate indicators or targets, which are not themselves goals of policy, may be useful because the effects of policy upon the goals are either not predictable with certainty or not immediately manifested (or at least immediately observable). A good proximate indicator and target would be both a good predictor of the goal variables and itself more immediately observable than the goal variables. Second, rules may be preferable to discretion because they may be an effective way to prevent the pursuit of undesirable goals. Third, the operationally most significant consequence of assigning the authority to carry out policy may well be that certain goals will be favored. Consequently, the first step in discussing these questions must be to specify the goals which monetary policy is to serve. Monetary policy traditionally has been concerned with a variety of goals, wThich are often incompatible. Incompatibility implies that tradeoffs must be made among the goals. These historical goals may usefully be classified according to whether they involve the value of money (the price level) or the allocation of resources. The classic goal of price level stability immediately presents a conflict with another macroeconomic goal of low-aggregate unemployment of scarce resources. Unfortunately, policymakers have frequently obscured the need for establishing priorities in this area by advancing meaningless arguments such as "an unacceptable rate of price increase, if not halted by accepting modest unemployment now, inevitably will lead to severe unemployment at some future time." However, although the trade-off between price level stability and low unemployment appears rarely to have been explicitly recognized in the public statements of monetary policy formulators, it has undoubtedly been important in practice. We may suppose that policy typically aims for the minimum aggregate unemployment consistent with a maximum tolerable proportionate rate of price level increase (denoted by P*). If the higher the acutal rate of price level increase (denoted by P) the lower aggregate unemployment, P* becomes a target as well as a maximum rate. Examples of resource allocation goals of monetary policy have been facilitation of (cheap) financing of the public debt, stability (profitability) of financial institutions, high level of activity in the construction industry, and stability of exchange rates and the price of gold. There seems to be a general consensus among economists that monetary policy should be principally, if not exclusively, concerned with 238 macroeconomic goals. Allocation goals seem inappropriate for two principal reasons: First, whereas monetary policy is just about unique in being capable of direction toward macroeconomic goals (fiscal policy seems too inflexible to be generally a practical alternative), other policies may be used to achieve allocation goals. Consequently, both sets of goals could in principle be achieved if monetary policy gives precedence to the macroeconomic goals in any case of conflict. For example, construction activity could be insulated from the effects of high-interest rates by the removal of legal constraints and by direct subsidies. Second, and perhaps more basic, these particular allocation goals seem to be in themselves generally unworthy because they imply a less economically efficient allocation of resources. For example, why should a ceiling be put on interest rate increases in order to protect financial intermediaries—whose essential activity is speculation on the future level of interest rates through borrowing short and lending long—from their own mistakes? Why should the relative riskiness of this activity be artificially reduced ? Similarly, why should foreign exchange dealings be made relatively less risky by exchange rate stabilization % II. PROXIMATE INDICATORS OF MONETARY POLICY The allocation goals of monetary policy all have the common characteristic that they more or less directly imply limitations on the behavior of market interest rates. The effects of monetary policy upon interest rates are both very predictable and quickly observable, at least relative to the effects upon the price level and aggregate unemployment. Consequently, if and when any of these allocation goals are paramount, the need for and interest in proximate indicators is minimal. However, regardless of the relative importance of the various allocation and macroeconomic goals, so long as monetary policy is at all seriously concerned with either of the macroeconomic goals, the question of choice of proximate indicators becomes interesting. Suppose, for simplicity, that the conflict between goals is resolved by saying that monetary policy should ignore allocation considerations, and should aim for constant equality between P* and P. The question is how can one tell whether current monetary policy is consistent with this equality. One cannot simply look at current P. Because of lags, currently observed P is probably completely independent of current monetary policy. Three sorts of lags should be distinguished, each of which contribute additively to the total lag: First, the latest available data on P is not current P. The process of collecting and reporting information takes time. (This observation lag might be termed technical, whereas the other two lags relate to economic behavior). Second, the reaction time of aggregate excess demand to monetary policy is likely to be finite. Third, the reaction time of P to changes in aggregate excess demand is also likely to be finite. In addition, even once these reactions begin they are likely to be distributed over time, with relatively small initial weights. Consequently, even though monetary policy could in principle bring about any desired change in P with only these reaction times separating cause and effect, in reality such relatively large doses 239 of monetary policy as to be impractical are likely to be necessary to effect large changes in P even that quickly. The consequence of any of these lags is that evaluation of current monetary policy requires prediction of the effects of current monetary policy upon future P. Such forecasting might be simplified by finding an endogenous variable whose reactions to monetary policy are more quickly observable than those of P, but whose value gives a reliable indication of the future course of P. Such a variable could serve as a proximate indicator and target for monetary policy. Two prominent and conceivably qualified classes of candidates for this role are indexes of the level of interest rates and measures of the supply of money. Both of these classes have much shorter observation lags than does P. Also, both these classes have the potential for indicating the level of aggregate excess demand in the near future and thus for predicting P in the more distant future. In addition, both these classes contain certain variables—for example, money market rates and high-powered or base money—which appear to be immediately, or almost immediately responsive, to monetary policy. In order to infer from the observed level of interest rates (denoted by r) information about the future relationship between P and P*, one must know how r compares with that level of interest rates (denoted by r*) which would generate (with a reaction lag) a level of excess aggregate demand which in turn would generate (with a reaction lag P equivalent to P*. Of course, r* is not observable. Nor is it easy to infer, because not only does it vary with exogenous shifts in the excess demand function, but it is also dependent, through the distributed lag in the response of aggregate demand to the level of interest rates, upon past value of r. The obvious alternative to r as a proximate indicator is the rate of growth of some concept of the money supply (denoted by M). We may suppose that the excess flow supply of money balances will correspond fairly closely to some subsequent level of excess aggregate demand. In order to use M as an indicator, we must know the rate of desired additions to money balances (denoted by M*) which would be associated with r*. Of course, like r*, M* is not observable, and the inferences about it involve analogous difficulties. Ideally, both the correct interest rate and money supply indicators should imply the same evaluation of current monetary policy. However, given the uncertainties involved in ascertaining both r* and M*, it would seem foolish to rely exclusively on either one. At any moment, policymakers may be more confident about their estimates of either r* or M*, and should draw their conclusions accordingly. As forecasting techniques improve, confidence about the reliability of either of these indicators wTill increase. At the same time, better forecasting, perhaps paradoxically, will lessen the value of proximate indicators by improving the reliability of direct inferences about the effects of monetae policy upon P. However, for the foreseeable future, the element of uncertainty in forecasting is likely to remain substantial. In this situation, presumably all available knowledge should be utilized. Interest rate and money supply indicators should be regarded as complements, not as substitutes. 240 Policymakers often refer to and apparently utilize other variables as proximate indicators of monetary policy. Of these, the two which seem most prominent are various concepts of the supply of bank credit and the quantity of free reserves. Regarding bank credit, it is hard to see how this could even potentially be a relatively useful indicator since its value reflects both the total supply of credit as well as the desired degree of intermediation between ultimate borrowers and lenders. Regarding free reserves, as frequently noted, the actual quantity is only meaningful in relation to the unobservable desired quantity. Consequently, the use of free reserves as an indicator is roughly equivalent to the use of high-powered or base money. III. R U L E S VS. DISCRETION The preceding discussion suggests that the selection of good proximate indicators and targets for monetary policy and the determining of their appropriate values is a relatively sophisticated task. The question then arises as to whether this process should be left to current discretion or guided by simple rules, to be reviewed only periodically. The advocacy of rules is, prima facie, a defeatist attitude; for it almost certainly excludes the attainment of the degree of perfection which in principle would be possible under discretion. Nevertheless, rules have been advocated because it is felt that in practice they would enhance the achievement of the chosen goals of monetary policy. This improvement might result either from better execution of the means of monetary policy or from better formulation of the goals themselves, or from both. As regards means, advocacy of rules implies a low opinion of either the degree of understanding or of the forecasting ability possessed by any possible discretionary formulators of monetary policy, or of both. As regards goals, advocacy of rules which are incompatible with the pursuance of certain goals implies disagreement with these goals and belief that a discretionary authority might pursue them. The question of rules vs. discretion, of course, cannot be answered in general, but only in reference to specific rules and to a specific monetary authority. The only rule which seems to be currently receiving serious consideration would involve a constant M, for some concept of the money supply. Viewed as a method for achieving equality between P and P*, the performance of this rule would depend upon the constancy of M* for the selected money supply concept. Advocates of this rule concede that it probably would not smooth out minor cyclical divergences, but argue very convincingly that it would exclude the possibility of major inflations or contractions. However, given the current state of knowledge and understanding, even on the part of those currently responsible for formulating monetary policy, such a rule does not appear attractive. Currently, the chances of discretion resulting in a major mistake are probably nil. Moreover, discretion, now, probably w^ould be capable of near perfection in achieving equality P and P* even in the very short run, if it were directed exclusively toward that goal. The record of monetary policy during the 1960's seems not to be inconsistent with this view. 241 Ail entirely independent motivation for advocating the constancy of M is that such a rule would generally be inconsistent with pursuance of the allocation goals of monetary policjr. Consequently, to the extent that the authorities are likely to give priority to allocation goals in times of goal conflicts, such a rule could enhance attainment of the macroeconomic goals. However, if the essence of the dispute does relate to goals, clarity would certainly be enhanced by confrontation directly with this issue, rather than through the subterfuge of operational rules. Moreover, if agreement were achieved regarding the priority of macroeconomic goals, the full potential of discretionary policy formulation could then be realized in achieving them. The distinction between rules and discretion becomes in principle purely formal if the rules are reviewed very frequently. In practice, however, even in these circumstances an operationally significant difference may exist. For example, it has been suggested that rules or guidelines be formulated which the policymakers could violate so long as they stated their reasons. In effect, this arrangement would permit complete discretion, but, at the same time, would cause the rationale for policy actions to be made explicit. Such explicitness would prevent sloppy or casual formulations and better and more widespread understanding of policy. The danger of explicitness generating unfavorable expectations is probably minimal. IV. THE DISTRIBUTION OF AUTHORITY Given that discretion is to be exercised, the question then must be answered as to how to distribute the authority to exercise it. In the present context, this issue is equivalent to the question of how independent the Federal Reserve System should be of the administration and the Congress. Again, in regard to this issue, both the means and ends of monetary policy may be considered, although the latter are probably more relevant. With respect to means, there would seem to be no a priori reason to expect that either the Fed or administration officials would generally have greater understanding or knowledge of effects of monetary policy. Only the Congress, because of the sluggishness with which it operates, is obviously not qualified to execute monetary policy. The argument about the distribution of authority arises primarily because the exercise of discretion in the execution of monetary policy carries with it the ability to exercise discretion in establishing priorities among the possible policy goals. The possibility of separating these two aspects of the policymaking process is limited by the fact that what are for the policy executor ends in themselves can usually be represented to be means to achieve more generally agreed upon goals. For example, the allocation goals of monetary policy are often argued quite plausibly, if fallaciously, to be prerequisites to the attainment of the macroeconomic goals. One argument which could resolve this issue as a matter of principle is that the independence in goal choice of the monetary authority from the immediate will of the elected representatives of the people is undemocratic. However, historical precedent suggests that the contrary principle of effective checks and balances is just as relevant. In any event this issue is more likely to be decided as a matter of practice than of principle. 242 At present the Fed has, on paper at least, a significant degree of independence in determining the course of monetary policy. Moreover, because the political process makes fiscal policy very inflexible, authority over monetary policy implies ultimate authority over stabilization policy. However, in exercising this authority the Fed is subject to pressure from the administration. The force of this pressure is not directly observable by outsiders. However, it would appear likely that it is generally great enough to make the administration the ultimate arbitor of monetary policy. Consequently, so long as the administration still received the advice and counsel of Fed officials, very little of consequence would likely be changed by a formal transfer of authority from the Fed to the administration. Moreover, the present formal separation of power does force the Fed and the administration to explain, at least to each other, the arguments for the policies which they are advocating. In addition, the present arrangements offer an amount of political convenience to the administration in that they allow for blaming the Fed for ultimately beneficial policies which may be unpopular in the short run, for example, "tight money." S T A T E M E N T OF S E Y M O U R E. H A R R I S , E M E R I T U S , H A R V A R D VERSITY AND UNIVERSITY OF C A L I F O R N I A , SAN UNI- DIEGO INTRODUCTION The integration of monetary policy with fiscal and debt policy could be greatly improved. It is simply upside-down economics for the Fed to go one way and the President and his advisers another. We cannot, as cannot other nations, afford the luxury of independence for the central bank, nor control of monetary policy for 15 years by one official who, however devoted, nevertheless is ruled by an excessive fear of inflation. H.R. 11 is a long step forward toward integration and reduced powers for the Fed. I - L . THERE SHOULD BE A COORDINATION OF MONETARY A N D FISCAL POLICY An appropriate mix of monetary and fiscal policy is needed. In the years 1961-63, for example, the Government relied on both expansionist fiscal and monetary policy in order to get out of a moderately stagnate economy. In 1965-66 recourse was had to monetary restriction because the Fed feared an inflation. Inadequate recourse was had to fiscal policy, with the result that excessive dependence on monetary restriction induced the dangerous crunch of 1966. In 1967 and 1968, inadequate use was made of restrictive fiscal policy and the Fed, though fearful of a return to the crunch of 1966, returned to monetary restriction. 1 - 2 . W H O IS TO BE RESPONSIBLE? The current attack is to discuss the issues of economic policies among the Treasury, the Budget Bureau, and the Council of Economic Advisers. They tend to determine fiscal policy. But they also consult with the Fed. In my opinion the three agencies of the executive should recommend policies to the President. Since monetary policy is a necessary weapon, some negotiation with the Fed is necessary. 243 But the role of the Fed should be limited. The executive should have increased control over monetary policy. The Fed through most of its history has been excessively fearful of inflation, and therefore has generally provided less than the optimum monetary resources. Independence for the Fed is not supportable. It is the responsibility of the executive to determine the supply of money and its price. It is unwise for the Fed to operate in one direction and the President in another—as in 1966. This does not mean that the Fed operates merely to further the Treasury's objective of low money rates. In the early post-World War I I period the Government could be criticized for imposing a cheap-money policy in order to reduce the cost of money to the Treasury. Under Eisenhower, there was much talk of the independent monetary policy. Actually the Fed, generally interested in restrictive policies, gave Eisenhower what he wanted. The President feared inflation and hence cooperated with the Fed in the pursuant of high money rates. Under Kennedy, the President paid lipservice to the independence of the Fed. But he insisted upon supplies of money adequate enough to assure a return to prosperity. The supply of money increased substantially and yet long-term rates were stabilized. Interest rate on Aaa bonds, for example, were as follows: 1952 = 2.96; 1960=4.41; 1965=4.49. In the 8 Eisenhower years, the average rise of real GNP was 3 percent ; in the first 5 years of Kennedy and Johnson, under expansionist monetary policies, was 5 percent. Despite the expansionist monetary policies from 1961 through 1965, the rise of prices was but 1.3 percent a year as compared with 2 percent from 1952 to 1960. In 1966,1967, and 1968, the results of policy were not as good as in earlier years. Prices rose by 1.7, 2.9, and 2.9 percent in the years 1965, 1966, and 1967, and may well rise by 4 percent in 1968; and despite restrictive monetary policies in 1966 and 1968. Moreover, in 1967 and 1968, the rate of rise of real GNP tended to decline. One explanation of this fact is that wages' rates were beginning to rise at a more rapid rate and also excess capacity had been greatly reduced. Additional resources were not provided in proportion as demand rose; and hence the rise of demand brought some additional inflation. I - 3 - A . W H A T GUIDELINES FOR MONETARY POLICY? Monetary policy should be enlisted for achieving maximum growth and employment, relative stability of prices, the latter both to stimulate output and improve the competitive position of the United States; and as a means of attaining money rates low enough in relation to expected profits to stimulate investment and cut down unemployment. Creation of money as a means of reducing unemployment should be used with caution. By 1965-66, with unemployment falling below 4 percent, there was evidence of increased pressure on prices. Money rates, indebtedness of member banks, reserves of member banks, free reserves are not guidelines of monetary policy. They are tools for achieving an expansive or restrictive monetary policy. 244 I - 3 - B . GUIDELINES OF MONETARY POLICY? The guidelines should be especially given by movements in GNP, employment, prices, rate of interest. When prices are rising too much in relation to output, that guide should suggest some moderation in the expansion of money. In the first 5 years of the Kennedy-Johnson administration, for example, GNP (real) rose about five times as much as prices. This was an index of successful general policy, and particularly of monetary policy. But in 1967, GNP rose by 2y2 percent and prices by 3 percent. Even a rise of GNP 5/6 as much as prices may not be a bad record. A rise of GNP is a plus item; but the increase of prices is a minus item. But the issues are not merely economic. Ideological issues are also relevant. Rising prices hurt those whose incomes do not respond to the inflation, and notably the old and savers. In general, the rise of prices is likely to be accompanied by higher levels of employment and less unemployment. Hence a case may be made for a policy of some inflation since it brings less unemployment, and hence a lesser degree of concentration of ill fortune on the unemployed. I—3—C. GUIDES TO MONETARY AUTHORITIES The indicators of the NBER are useful and especially the leading indicators (LI) which anticipate change. But individual Li's often vary in the direction of movement. They raise many other problems also. Basing policy on past behavior of Li's may be misleading, and in part because in more recent years Government intervention has become a more important factor. Hence the past movements of (say) stock market prices or rate of interest may not be a very good guide for the future. In fact, one can raise some general issues of projections along these lines. Thus a projection of GNP for 1969 may prove to be off by a good margin, the explanation being a Government policy not anticipated. A prognosticator may also prove to be right only because unanticipated Government policy brought about the growth (say) anticipated. The economist who guesses right because he also guessed right on Government's contribution is to be applauded. But if his projections are confirmed by history only because of policies he did not anticipate, his credit should be limited. At any rate, the use of Li's should give better results if recourse is had to a dozen (say) that in the past proved reliable (and not all 36), and if the trends for several months, rather than one or two, are considered. I—3-D. VARYING GUIDELINES Indeed, guidelines should vary from year to year. A strong case can be made out, for example, for seeking a 5-6 percent rise of output per year in the years 1961-64, years of large excess capacity and output substantially below potential; and at the same time should seek a rise of output restricted to 4 percent in 1967 and 1968. Why the difference ? In the earlier period, a substantial rise of output is possible without inflation because part of the rising demand can be absorbed out of excess capacity. But in 1968 the country can profit primarily from an increase of numbers on the labor market and a 2-3 percent increase of productivity. Insofar as demand rises above 4 percent, which measures the gains from these channels, then the excess of demand 245 would be reflected primarily in a rise of prices rather than one of output. The policy of tax increases and expenditure reduction of 1968 has been supported in order to contain growth to a level of around 4 percent. Of course the task of the Government depends on the contribution of the private economy. Should consumption, private investment, anticipated Government expenditures, and the excise of exports yield an unacceptable level of output and unemployment, then the Government has to increase its contribution by raising expenditures and (or) reducing taxes. Moreover, the Government has to consider the strain on the economy associated with the automatic rise of taxes that accompanies increasing income. In planning for the end of the Vietnam War, the Government takes into account the reduction of military expenditures and the rising income and taxes without change of tax structure. Should military expenditures decline by $15 billion, and tax receipts automatically rise by $10 billion with a $40 billion gain in income, then the Government will have to provide $25 billions through increased spending on nonmilitary items and cuts in taxes. (I assume that the military will not make up for the lag in new weapons of recent years.) I - 3 - E . VARIATION I N TARGETS GROWTH? The variations should be related to the potential of the economy. Labor supply, hours of work, productivity gain, (related) the contribution of capital and technology will suggest potential growth. Over long periods of time, the annual gain of man-hour output—say 3 percent—can, through the operation of the compound interest law, greatly increase output. In 25 years output would double; in 50 years increase by 3% times. I-3-F. MONETARY AUTHORITY ALLOWED TO ADJUST TARGETS VARIABLES DURING T H E YEAR ? I would give the monetary authority full authority to adjust to changing prospects. Projections for 3 or even 6 months yield reasonably good results. But beyond 6 months, the actual change may diverge greatly from the projected one. If by June, the economy seemed to be on the downgrade surely the Fed should be tempted to introduce expansionist policies. In early 1968, the tax increase and spending cut seemed to support an anti-inflationary policy. But even in the first half of 1968, there were signs of trouble: high and rising interest rates, excessive inventory building, unsatisfactory housing, reduced Government contributions to spending, greater doubts about the economy—all of these raised questions. By the middle of 1968, a growing consensus feared an economic decline. Surely the Fed restrictive monetary policy should be abandoned. (I write in August 1968). All planned economics operate on the principle that when developments deviate from the expected, freedom to change policies should prevail. 1 - 4 . T H E ROLE OF DEBT M A N A G E M E N T Debt management can contribute much. For example, when the economy is just beginning to recover, it is mistaken policy for the Government to issue long-term securities in large amounts. The effect 246 is then to raise interest rates just when the economy needs the lift of a decline in rate. The Eisenhower administration made such mistakes. Again, when the economy needs high short-term money rates in order to increase imports of capital and discourage outward movements, then the appropriate policy is to issue large amounts of short-term paper and thus get the short-term rate up. Hence the balance of payments would improve. Moreover, the long-term market could then be protected to some extent against rises, with favorable effects on the economy. In the great crisis of 1966, the country, and especially the housing industry, were starved for money. The Executive helped prevent a disaster by withdrawing large issues that were to be sold and in this way reduced competition in a badly provisioned market. I—5—A. USE OF OPEN MARKET OPERATIONS I am not sure I understand the thrust of this question. In general, I would use open market operations to increase or reduce the reserves of member banks. The size of reserves largely determines the amount of money that banks create. This is of course not the only determinant. There is something to the idea that when reserves are maintained or built up through a rise of indebtedness to the Fed, then the expansive effects of a rise of reserves is less than when the additional reserves stem from purchases of securities to raise the reserves of member banks. A comparison of Federal Reserve bank credit and member bank reserves from 1960 to 1967 will suggest the contribution of open market operation. [In billions of dollars] December 1960 Federal Reserve credit outstanding P.S. held Discounts and advances Gold stock Currency in circulation Member bank reserves 27.2 27.3 .1 18.0 33.0 19.3 June 1968 51.3 51.3 .7 10.4 47.5 25.7 These figures reveal: 1. That the major origin of open market operation was the financing of additional money in circulation and the loss of gold. Without Fed intervention, reserves of member banks would drop with large losses of gold or increased money in circulation. Rise of Federal Reserve credit outstanding Financing: Rise of currency in circulation and loss of gold Providing additional M.B. reserves Billion $24.1 17.1 6. 4 It is clear that the largest source of Federal Reserve credit originates in offsetting rises of money in circulation and loss of gold, not purchases to expand the total reserves of member banks. Offsetting gain of Federal Reserve credit Expansion due to open market operations Billion $17.1 6. 4 I see no alternative but to use open market operations for "defensive" reasons (i.e. $17.1 B in 7% years), and also to satisfy needs for expansion. In the 7 years, 1960-67, the rise of money was as follows : 247 Rise—December 1960-Deceryiber 1967 Billion $40.4 11.5 29.0 110.9 I. Money supply (a) Currency (&) Demand deposits II. Time deposits What is especially striking is the average rise of money and money plus time deposits, December 1952-December 1960 and December 1960-Deoember 1967. [In billions of dollars] 1952-60 Money.. Money plus time d e p o s i t s . . . 1.7 5.8 1960-67 5.8 21.6 In the Eisenhower years, money increased only about one-third as much per year as in the 7 Kennedy-Johnson years, and about onequarter as much for money plus time deposits. Despite the large demands put on the system by large losses of gold and rises of money in circulation, the Fed, under pressure from the two Presidents in 196167 expanded monetary supplies at a rate of about three times that of the Eisenhower period. I - 5 - B . OPEN-MARKET OPERATIONS A N ADEQUATE TOOL? I am not sure I understand this question: "Do you believe that monetary policy can be effectively and efficiently implemented solely by open-market operations?" Open-market operations are probably the most potent weapon. To some extent the Fed, by raising rates or reducing them, can directly influence the price of money. But ever since the midtwenties open-market operations have been the major weapon of monetary policy. Through influencing the volume of indebtedness and through changes of reserve requirements, it is possible also to influence the supply of money. I - 5 - C . USE OF REDISCOUNTING, CHANGES I N RESERVE REQUIREMENTS, AND REGULATION Q Kediscounting is a tool for increasing the effectiveness of Fed policy. But the total impact is not ordinarily great. By changing reserve requirements the Fed can induce a rise or fall in required member bank reserves. But this is an overall weapon and may upset some parts of the money market. But similar unwanted effects may follow open-market operation. Kediscounting to this extent has an advantage over open-market operations, for rediscounting relates to the amount of cash needed by individual banks. One troublesome aspect of changes in reserve requirements is that they result in windfalls or penalties to banks accordingly as requirements are reduced or increased. Thus, allowance of currency and coins as reserves in 1959-60 greatly increased reserves of banks, and thus substantially improved the profit position of banks. Recourse of changes in reserve requirements results in varying impact on the price (yield) of different assets. Dependence on open-market operations improves the prices and reduces the yield on Government securities. By the middle of 1968, the Federal Reserve banks held more than $51 248 billion in Government securities as compared with but $150 million in June 1929. Even by 1939, the portfolio amounted only to $ 2 ^ billion. Clearly, open-market operations greatly favor the Treasury as sellers of securities, whereas a reduction of reserve requirements and accompanying expansion of money provides additional demand and higher prices for all kinds of assets. Under regulation Q, the Fed sets maximum rates on savings and time deposits held by commercial banks. By allowing higher rates under regulation Q, the Fed has, in fact, favored commercial banks against savings banks and S. & L.'s. In the last few years, the result has been the capture of a much larger part of the savings and time deposits by the commercial banks, with unpleasant consequences for S. & L.'s. The housing market in California was especially hit. Another effect has been a large relative increase in time deposits for financial institutions and hence greater interest of financial institutions in long-term assets. I 5 - D . I SEE MERITS I N FEDERAL RESERVE REPORTING TO CONGRESS EVERY 3 M O N T H S " O N PAST A N D PRESENT PROSPECTIVE ACTIONS A N D POLICIES" If the Fed reveals its hand to Congress, then policies favorable to the economy are more likely to prevail. In most major countries, the central banks have to reconcile differences with the executive, with the Government the ultimate source of responsibility and authority. I - 5 - E . REPRESENTATIVES OF CONGRESS, T H E TREASURY, AND COUNCIL OF ECONOMIC ADVISERS AS OBSERVERS A T OPEN-MARKET COMMITTEE MEETINGS At present the Secretary of the Treasury, the Council of Economic Advisers, and the Director of the Budget meet once a week for a discussion of issues. As senior consultant to the Secretary of the Treasury, I am invited to these meetings and often attend them. I think the meetings are very helpful, and are attended by about 12 high officials. Ordinarily the Chairman of the Federal Reserve Board is not present though there are some meetings with Mr. Martin. I see no reason why issues of open-market policy should not be discussed with the three relevant departments, both with the Chairman of the Federal Reserve Bank and other members. This might be better than the "observer" policy. The Fed has some special responsibilities in helping to stabilize the economy, and hence they should be allowed to discuss their responsibilities freely and without other agencies and Congressmen breathing down their necks. I have generally held the view that the Fed is too independent of both the Executive and the Congress. Yet I would not go as far as introducing these observers. But there should be much more free discussion between the Fed and the three other relevant agencies, and even with responsible Members of Congress who, by virtue of their knowledge and position, are strong candidates for keeping themselves informed. Surely if the Fed and the others are to work out a proposal to the President, the three departments, and to some extent the Congress should discuss open-market policies fully and frankly. I I - l . Retiring Federal Reserve stock.—I have no strong views on this issue. 249 II—2. REDUCING T H E NUMBER OF MEMBERS OF T H E FEDERAL RESERVE BOARD TO FIVE AND TERMS TO 5 YEARS In general, I approve this recommendation. The Board would be more efficient with five members, and in this way the next administration will be able to tie Federal policy more to the views of the Executive. It may be well to get rid of the less able members of the Board now. I say that though I am aware with the appointments in particular of Messrs. Mitchell, Maisel, and Brimmer, the average quality has greatly improved. The staff also is much improved over the staff of earlier years. It is certainly a mistake for one member of the Board to control monetary policy over a period of more than 15 years. A limit of 10 or even 6 years should be placed upon the Chairman. It is particularly unfortunate that the Chairman, though a devoted public servant with high moral standards, is obsessed by the fears of inflation when justified and not so justified. Fortunately^ the new members have pruned the authority of the Chairman and it must be said that in the 1960's—except for 1966 and 1968—the Chairman has abandoned to some extent the obsession with inflation. But it is risky to allow one member of the Board to control monetary policy and to that extent the economy. II—3. M A K I N G T H E TERMS OF T H E BOARD AND T H E PRESIDENT COTERMINOUS This is an excellent idea. The possibilities of a well-integrated policy would be greatly increased. It would be helpful if the Chairman's term might be extended by 1 or 2 years additional. II—4, 5. A N AUDIT OF THE FEDERAL RESERVE BOARD AND FEDERAL RESERVE B A N K S BY T H E COMPTROLLER AND APPROPRIATIONS B Y CONGRESS I see no serious objection. I do not see why the Fed should profit from special financial arrangements not open to other parts of the Government. III. COMMENTS ON RECENT MONETARY POLICY I have commented elsewhere in this paper. In general, I would say that monetary policy in the ^ears 1961-68 overall were better than under Eisenhower. Martin yielded to the pressures of Presidents Kennedy and Johnson, and he has to share his authority increasingly since 1961. Of course there are exceptions. The policy in 1965 beginning with the December 1965 rise of rates was almost disastrous. Even before December 1964, there were numerous statements bv the Chairman of the Federal Reserve Board aibout the dangers of inflation and the need of restraint Yet from 1964 to the middle of 1968, GNP (in 1958 prices) rose from $582 to $702 billion, a rise of 21 percent in 3y2 years, or about 6 percent a year. Moreover, in this period, prices rose only by 11 percent, or 3 percent a year. This is not a perfect record; but note that real GNP rose twice as much as prices. Thus, policy would have been much more sterile had the authorities listened to Martin's concentration on fears of inflation. 21-570—68 17 250 The hard-money policy of 1966 could scarcely be justified. First, because it was the first real defiance of executive control over economic policy. Second, because the rate of interest rose spectacularly and the supply of money responded most inadequately. This is evident in the following changes in interest rates: 1964 Taxable bonds. Corporate bonds, Aaa FHA new home mortgage bonds 3-month Treasury bill 4.15 4.40 5.45 3.55 1966 1967 4.65 5.13 6.29 4.88 July 1 9 , 1 9 6 8 4.85 5. 51 6. 55 4.32 5.14 6.26 5.467 These are unprecedented increases in rates, and especially long-term rates. Only the passage of the tax increase stopped the rise. But the decline of rates has been of most modest proportions. The Fed still hesitated to give the economy a lift even though rates continued to rise through most of 1958, until a passage of the tax bill, and there were increasing signs of an economic decline. I might comment on one omission in the questionnarie. The cost-push aspect of inflation requires consideration. With wage rates rising 6 percent (prices 4 percent), the burden put upon both fiscal and monetary policy is greatly increased. STATEMENT OF L O W E L L H A R R I S S , COLUMBIA UNIVERSITY M A K I N G M O N E T A R Y A N D FISCAL POLICIES W O R K : C O M M E N T S ON H . R . 11 The important topics covered by H.R. 11 deserve exhaustive analysis on a broad scale. The committee deserves high commendation for its undertaking. The press of many demands prevents me at this time from giving the points the direct attention I should like. Within recent years, however, my writings have touched on several of them. Selection and recasting of materials from several sources permits me to respond, incompletely and inadequately but most respectfully, to some of your inquiries.1 T H E AVERAGE W A G E R A T E : A BASIC FACTOR OFTEN OVERLOOKED 1. The magnitudes appropriate for policy decisions depend upon a factor too often overlooked—the average wage rate. Total aggregate demand of $900 million (GNP) would buy 150 billion hours of labor at $5 an hour (total cost to the employer), leaving $150 billion for other factors. If aggregate demand rises by $10 billion (through fiscal or monetary policy) and if this increase all goes to labor, an additional 2 billion hours of employment are created—providing wages remain at $5 an hour, average or, more accurately, at the margin). If average wages were to rise to $5.20 (4 percent), however, the $760 billion now available to labor would not even pay for 150 billion hours of employ1 Adapted from the following by C. Lowell Harriss: The American Economy, 6th ed. (Homewood, 111.; R. D. Irwin, Inc., 1968) - with (W. J. Shultz) American Public Finance, 8th ed. (Englewood Cliffs, N.J., Prentice-Hall, Inc., 1965) ; Money and Banking, 2d ed. (Boston: Allyn & Bacon, 1965) ; "Fiscal Action To Influence Employment and the Price Level: Some Criteria," Government Finance Brief, No. 4, 1966, Tax Foundation (from testimony for the Joint Economic Committee) ; "Inflation's Hidden Effects," "The National Debt: Looking Ahead, Tax Review, July 1967 and March 1968 (Tax Foundation) ; "Making Tax Policy Work," Manufacturers Hanover Trust Co., October 1967; "Objective® of Fiscal Policies: Looking at Goals," Business Papers, Oklahoma State University, No. 5, 1967. 251 ment (144 billion). Depending upon the cost of labor, as well as of other factors of production, dollar increases in aggregate demand do not necessarily insure more employment. 2. More to the point, the volume of real employment from any given monetary demand relates to the price of labor. The figures above are too rounded to be entirely realistic but will serve to illustrate. A 4percent rise in the average price of labor will "cause" a drop of about 4 percent in man-hours paid for from any given dollar amount of total demand. 1. The effects of fiscal policy depend upon monetary actions.—Shifts in Government deficit and surplus can tip the balance and keep the economy from going too far in either direction, and certainly prevent expansion or contraction from cumulating dangerously. It is vital, to distinguish between two points (a) the amount of Government spending, and (b) the source of the funds. If the Government cuts or increases its spending while private spenders make equal, but opposite, changes, there will be no net alteration in aggregate demand. 2. If inflation threatens, how can fiscal action cut overall spending ? The Government can take more money (in taxes) from private spenders than it spends itself. By reducing disposable income in this way, it cuts business and family buying. The Treasury can use the surplus to retire securities held by commercial banks. The effect will be greater if debt changes are those of the Federal Reserve banks or if currency is retired. The monetary authorities must, of course, prevent increases in bank loans to business that offset declines in Government debt. The imperative necessity of the coordination of monetary with fiscal policy must never be forgotten. The total volume of M (money) will fall. Or the Treasury, by merely holding the surplus as idle bank deposits can give them a V (velocity) of zero. 3. If recession seems in the offing, Government can lower taxes to leave families and businesses with more disposable income. Government can also increase total demand by itself spending more without taking more from the public. Where can it get the money to make up the difference ? By increasing M. How ? It can borrow from commercial banks. (If it has been holding deposits idle, it can use them, raising the V from zero; currency issue is also possible.) 4. The effective force changing national income (in the short run) comes not from the amount of Government spending, nor even solely from changes in such spending. What counts more will be the way the changes are financed. Some economists question the effectiveness of monetary policy in a recession because businesses cannot be counted upon to react promptly to easy money by borrowing and spending enough more to give full employment. By fiscal action, however, Government can do enough, by creating purchasing power and injecting it into the stream of national income, directly and quickly. 5. The results of deficit financing will depend, not only upon the amounts involved, but also upon the kinds of tax and spending changes. More important is the way the money to cover the deficit is raised. If the borrowing is accomplished by selling savings bonds in a wTay to absorb purchasing power that otherwise would be spent on consumption or investment, the ultimate net effect of the whole process may be little change in national income. But if the debt is sold to commercial banks, the economy can get a true stimulus as net new 252 purchasing power is injected. There can be such net injection only if bank lending that would otherwise occur is not curtailed. 6. A shift from deficit or balanced budget financing to surplus financing operates, in essence, as the reverse of a shift to deficit financing. In general, a move to surplus financing has a contractive effect on the economy as more money is taken from the taxpaying public than is paid back in expenditure. The crux of the matter is the use of the funds received by the former holders of the debt. If banks are repaid and the money stock thereby reduced, the downward pressure on business activity can be substantial. If the funds made available to the private economy by debt retirement go into financing an increase in private investment, or consumption, the contractive result will not be significant. 7. Changes in the total flow of income resulting from fiscal operations depend significantly upon increases or decreases in the country's stock of money. Changes in bank lending are the chief agency through which fiscal operations introduce purchasing power into the income flow or withdraw purchasing power from it. Changes in the volume of deposits in commercial banks are monetary developments. BANKS AND T H E HATE OF INTEREST 1. Borrowers want money. They have no reason to care whether the money is newly created or whether it is the result of saving out of income. The significance for the economy, however, is tremendous. 2. The demand for loans, often expressed as a demand for credit, can be met out of the supply of savings. There is a price, an interest rate, which will balance the quantities of credit demanded and supplied. The "thing" which changes hands is money. For money as such, there is a demand and also a supply. The supply-demand of credit and the supply-demand of money are essentially different. Much misunderstanding results from the fact that commercial banks both grant credit (make loans) and create money (demand deposits). 3. Some of the supply of loanable funds, and especially some of the change in the amount supplied in the short run, results from commercial bank lending in the form of money creation. Funds for lending of this sort can involve virtually no real cost in the sense of labor and materials. Under some circumstances the creation of demand deposits for some borrowers may not require any sacrifice of desirable alternatives. Government can influence the amount of such lending by the rules which regulate banks in their creation of money. 4. Is it not possible, then, to use the banking system to control the rate of interest? More tempting, perhaps, is the possibility that by enabling banks to "manufacture" money more easily, man can lower the cost of borrowing. At any given time, society can permit banks to expand their loans—extend more credit—even though there has not been a penny more of voluntary savings. This fact is undeniable. Another fact is also undeniable. Such loan expansion increases the stock of money. 5. An increase in the volume of funds available for lending will tend to reduce interest rates. But this is not the only result. As the new money is spent, the volume of economic activity in dollar terms will go up. So may the price level. 253 6. Will a rise in the price level affect interest rates? It ought to induce some lenders to try to get a higher yield as compensation for the loss in value of money. Some owners of bonds will sell, perhaps to buy stock or real assets. Some holders of money will shift to goods. Borrowers will tend to seek larger loans to benefit from rising prices. Thus, both demand and supply changes (after the initial injection) move to raise interest rates. 7. Unless banks add another injection of money, interest rates tend to rise. Except temporarily, banking policy cannot lower interest rates, unless injections of new money into the economy are continued. In a world of essentially full employment, except as the quantity of new money matches growth of productive capacity, such a practice must be self-defeating. As the growing stock of money circulates, the upward pressure on prices will continue. 8. When there is substantial unemployment, of course, the effects of the increase in bank-created funds for lending—money—may appear primarily as greater output, with little rise in prices. Higher incomes will both stimulate saving and raise the demand for borrowed funds. The relative magnitudes will depend upon a host of conditions whose net effect will not be clear before the process has gotten underway. But I find difficulty in believing that the interest rate will not go up. If so, the actions to lower interest rates are successful only temporarily, successful, that is, in terms of the narrow goal of influencing the one price, the rate of interest. 9. What about the power of the monetary authorities and the banking system to raise interest rates ? Again, the basic forces of demand and supply dominate. Banks have influence in the short run. By reducing reserves or raising reserve requirements, the monetary authorities could force banks to reduce loans outstanding. Interest rates would tend to rise. But for the longer pull they would not rise a great deal unless the process were repeated and repeated, something which is improbable, partly because the downward movement of national income would reduce demand for borrowed funds. 10. Briefly, and explicitly excluding inflationary possibilities, bank actions over any extended period will not have much effect on the level of interest rates. The influence in the short run is quite another matter. Changes in short-term rates that are large in percentage terms can result from alterations in bank lending. Such changes in lending may result from forces apparently outside the banks—a decline or a rise in business demand for loans—or from official actions which alter the capacity of banks to make loans. OUR MONETARY POLICY AND ITS USE Ambitions are more easily raised than satisfied. Some of us have come to expect a great deal from fiscal and monetary action, to seek both big and varied objectives. Yet what can we really expect to accomplish? What is possible? What are attainable objectives? Several factors affect the attainments which are possible. 1. The monetary authorities cannot directly change the income stream. What they do will not in itself constitute an increase or decrease in the flow of expenditures through the economy. The Federal Reserve changes the availability of money, or of close money substitutes. 254 2. The effects of monetary change cannot be pinpointed. Even if an initial impact were to be focused by some sort of selective action, such as short-term business (self-liquidating commercial) loans or special treatment of mortgage lending, the results will gradually spread over the economy; other sectors feel the results, some promptly, others only after delay. 3. Monetary action will not build houses or raise the productivity of manufacturing or make people friendlier. It will not in itself modify the underlying real elements of the economy. Monetary conditions, however, are an essential element of the framework within which we produce and consume. 4. The tools or implements of monetary policy, of course, help determine the "possible." The chief tools influence the stock of money and its close substitutes, (a) The monetary authorities cannot control the willingness of banks to lend or of customers to borrow. (&) Even if the quantity of money were controlled rigidly, no central authority would be able to control velocity or the changing desires for liquidity. (c) The authorities do not know by how much either changes in interest rates or in the prices of securities will influence the volume of investment. Apparently, however, demand schedules for many types of investment goods are highly inelastic with respect to interest. We certainly do not know how big the marginal responses will be. 5. The choice of what we ought to try to do will depend upon the extent of confidence in our ability to predict the effect of different kinds of action. Prediction remains an uncertain art. 6. Another factor affecting the possible is the quality both of the personnel who will make and administer policy and of their advisers. Will the people who must make the decisions be well qualified ? What assurance can the public have that those responsible for policy are competent to act as well as is required for success ? A person or group not qualified to do the job well (no matter how competent to act on other problems) may in fact have decisive influence. For example, a banker who is well qualified to judge the quality of business loans may have little competence for deciding how much change in the quantity of bank lending will serve the public best. Serious errors may produce mostly mistakes. The existence of this risk underlies the argument for trying to rely upon carefully devised rules in preference to frequent exercise of judgment. 7. What might seem possible in theory may in fact be impossible because of practical realities. For example, two such different things as the temporary strength of a few leaders in Congress or delays m getting a half dozen pieces of key information may modify significantly the hopes of success in dealing with a particular situation. The freedom to use monetary policy also depends in part upon what monetary authorities in other lands are doing. The "possible" in one country rests somewhat upon the "actual" in others. 8. The intermixture of monetary and credit policy seems certain to continue to create confusion. Society's need for money—as a medium of exchange, a store of value, a unit of account, a standard of deferred payment—is different from its need for credit. The attempt to meet desires for loans by creating money—or the destruction of money when the demand for credit slackens—is not likely to produce the best monetary policy. Yet the two are grouped together, with officials in and out of Congress often more concerned about credit than about 255 the stock of money. The desire to help one section of the economy (farming) or restrict another (the stock market) can be satisfied, perhaps, by credit policy. Monetary policy, however, is not an efficient instrument for such purposes. 9. Some goals are simple, some complex. Some are clearly the means toward other ends, while others are more nearly ends in themselves. Shall we, for example, settle upon a rather specific goal for the monetary authorities, such as a stable or slowly rising quantity of money? Or should we seek a goal embracing much more of economic life, such as maximum employment or the rapid growth of real income ? Monetary goals, like money, are means to other ends. It is hopeless to expect monetary action in itself to bring us to our real objectives—the goods and services we would get from the full employment operation of a free economy. Yet it is also hopeless to expect to achieve the real objectives if monetary currents are running strongly against their realization. 10. The issue of rules versus authorities is this: (a) Monetary policy and directives for action can be stated in terms that are clear, definite, and precise. (The statements would be embodied in laws passed by Congress.) Human beings in carrying out the policy would be left little (or no) room for judgment or direction, (b) In contrast, the monetary system can be framed to give power to be used by authorities (an individual or a group) as they think best in the light of the conditions which actually develop. The first kind of policy is one of rules, the second one of authorities. Between the extremes lies possibilities of compromise. 11. Arguments in favor of rules.— (a) One argument for fixed rules derives from the fact that the monetary system makes up such a vital part of the economic framework. If this part is certain, then individuals and businesses can arrange their affairs more efficiently than if they are uncertain about what may develop in this important part of economic life. Expectations will be more certain. There is less room for surprise. Society can largely eliminate one source of risk—and risk is a cost, (b) If rules are fixed, changes in monetary policy will not make things worse. Adhering to a rule assures protection against some human errors. Even if the rule may not be best for every situation, there is no danger of bad selection of alternative actions or bad timing as authorities try to meet changing conditions. Thus, the public avoids not only the cost of uncertainty but some of the risks of poor policy. At the worst, we may sacrifice little to obtain this gain because the potential superiority of flexible over fixed policy will not be large, whereas the losses from shifting to an inappropriate policy can be substantial. (c) The record of discretionary management has by no means been brilliant. Evaluating the record is difficult; one cannot know what would have been produced by different actions. Nevertheless, the accomplishment does in itself provide a convincing testimonial to the superiority of authority over rule. 12. Arguments against rules.—The major argument against reliance upon fixed rules is that discretion may be used wisely to meet needs as they develop. No two sets of economic conditions, after all, are identical ; the future is unknown. How can men, with all their limitations as human beings, set a general rule for the future which will serve as well as the best that men can devise as conditions actually develop ? Can we not get the best results if the monetary system is adaptable ? 256 A fixed rule can hardly serve all desirable goals equally effectively, and a policy well suited to achieving one goal (such as price level stability) may be poorly adapted for another (economic growth), which may increase m relative importance. Insulation from troubles coming from 'other countries may require flexibility in monetary policy. Not enough is known about the ability of officials to implement a rule to be confident of success. (Two more or less fixed rules—the gold standard and the balanced Government budget—had considerable weight for decades. Yet they did not control in the sense of displacing other guides of policy.) 13. Perhaps the clearest-cut monetary goal would be to regulate the quantity of money itself. The objective is less ambitious than others to be discussed later. And it could be achieved. The control envisaged would be more rigorous than is possible now, when private decisions also influence the expansion and contraction of bank loans—and therefore the stock of money. The quantity of money to be outstanding would be determined. Anyone seeking to borrow money would have to get it from the existing stock. Commercial bank creation of deposits wTould be impossible. The requirement that demand deposits be backed by 100 percent reserves offers a possibility worth more attention than can be discussed here. The monetary authorities controlling reserves would then have one-for-one control of the quantity of money. If such a plan or, more realistically, one less rigid were adopted, what might be the guides or standards of the amount of money ? 13a. One possibility would be to fix the quantity of money once and for all. Clearly, then, no economic disturbance could arise out of changes in the quantity of money. A single objection, however, dooms this suggestion: In a growing economy, the price level, including wage rates, would decline as total output increased. Price declines on the scale that would be involved would create intolerable strain. Rigidities are too numerous and too severe. 13b. A more realistic possibility would be to increase the stock of money at some steady and definite rate. No one, then, need have any doubt about the amount of money in the economy, next month or next year. Once the policy had been determined and necessary changes made in financial institutions to assure practical implementation, the monetary authorities would have no discretion. Every week or month the Federal Reserve would buy enough assets (Government debt) in the open market to build the money stock as agreed upon, possibly with seasonal adjustment. Or the Treasury would adjust its financing —debt refunding, new borrowing, and currency issue—to provide the expansion. The injections of new money might be set to equal as nearly as possible the expected growth in the economy so that the price level would be essentially stable. A proposal so rigid may seem unduly restrictive. Would the gain— the removal of what has been a source of economic instability, changes in the rate of money creation and destruction—be worth the loss of freedom to change the quantity of money? History records many occasions on which discretionary monetary action rather than strict adherence to a rule could have served the public interest. Nevertheless, an advocate of the strict control of the stock of money must not necessarily abandon his case when presented with such evidence. He may argue that the trouble which seems to call for remedial action would 257 not have developed if his plan had been in effect earlier. The strict rule does not conform to our tradition. More important, however, is the belief that in a world of freedom and uncertainty men should try to adapt monetary policy to the conditions which do actually arise. And one weakness is clearer today than it was even a few years ago. "Money" is not easily defined. Close, and not-so-close, substitutes exist. Changes in the totals of substitutes can upset the results expected from a policy of fixed growth of the money stock. 14. Any policy based on strict control of the quantity of money has an inherent weakness. It ignores velocity. Is there, then, a policy which deliberately takes account of velocity? One approach would seek to regulate either the total of money payments (MY), or the total of income payments (MVy). The goal would probably be some stated, regular increase in the aggregate. Regulation of the total of money payments would be a more powerful instrument of economic control than one limited to the quantity of money. Achievement, however, would be very much more difficult. Velocity is the result of the way millions of economic units act. Though a central authority can control M within narrow limits, it cannot control the use of M. (Among the problems for which there is yet no good solution is the changing importance of "nonincome" uses of money, such as the purchase and sale of securities.) The use of money is the most decentralized, the most dispersed, of economic realities. A proposal which includes allowances for changes in V is quite a different thing from one which involves M only. In fact, the working out of the policy would have to rely upon those changes which the monetary authorities are actually able to make—changes in M. These would be larger or smaller, depending upon the expected movements in V. Is a goal which can be achieved only by correct forecasts of V overly ambitious ? So it would seem. Yet the monetary policies most widely accepted today do, at least indirectly, involve this more ambitious objective. This fact is true of the pursuit of price level goals. 15. Fiscal and monetary policies are by no means perfect substitutes for each other. They are not fully interchangeable. Nevertheless, the effects of any fiscal policy must work out in an environment which depends significantly upon monetary policy. Both public debate and advanced professional analysis often benefit from assuming "other things being the same." Real world processes, however, do not permit the simplification which involves a fiscal policy change which has no monetary effect. Economists disagree in their weighting of the relative importance of monetary and fiscal actions under different combi« nations of conditions. Such differences of view, however, do not justify what sometimes seems to be the denial, by implication, that monetary policy will significantly influence the outcome of fiscal action. 16. Who can possibly judge the effects of different possible fiscal actions next month or next recession without making assumptions about monetary conditions? The leaders of our Government have the potential power to assure themselves of a much higher degree of certainty about monetary policy than has been the case to date. True, velocity of circulation will remain beyond direct control of official agencies. But changes in the stock of mone^—defined as currency plus demand deposits—can be controlled within a moderate range, not 258 necessarily from week to week but for short periods relative to phases of a business cycle. Changes in the amount of money added to the economy do more than influence interest rates when newly created deposits add to the supply of loanable funds as the money is injected into the economy. The money continues to exist, to pass from hand to hand, to "be used in transactions. 17. Other policies, notably those affecting wage rates, also influence employment and price levels. The higher the level of average wage rates, the greater the dollar total of demand needed for any total of employment. Raising the minimum wage and extending coverage would aggravate the problems of achieving full employment with price level stability. The resulting wage rate structure would obstruct the absorption into the employed labor force of young people and others whose productivity has not yet reached the legal minimum. ON T H E USE OF FISCAL POLICY 1. Fiscal policy results from (1) the recommendations of numerous elements of the executive branch, (2) the actions of revenue-raising committees and the appropriations committees (and their subcommittees) in both House and Senate, and (3) The Houses of Congress themselves. Monetary policy is made by the Federal Reserve subject to an undeterminable influence from the executive branch, Congress, foreign central banks, and other sources. 2. An outsider cannot evaluate the "real-life" working of these arrangements. But I have read much of what has appeared in print. It leaves me uneasy. The men who have made the decisions do not seem always to have understood the issues, processes, mechanisms—including the ties between monetary and fiscal policies—as well as we should like. Perhaps, however, the past is a poor guide to the future. Will not everyone have learned ? Unfortunately, some of us are slow learners. Even more to be regretted, the "truth" is not always crystal clear. 3. The validity of one point, however, seems beyond question: The public may justifiably expect that the two groups of decisionmakers coordinate policies. Where arrangements fail to assure coordination,, what needs to be done? My few suggestions assume no major change in relations among the branches of our Government. 4. More remains to be done in providing evidence about what has (just) happened and in analyzing the probable results of alternative courses of action. Much merit supports the recent efforts to get the Federal Reserve to inform the Joint Economic Committee of the bases for monetary policy actions. Congressional hearings advance understanding. Nevertheless, they cannot do all that is reasonably possible in threshing out tough questions—and many are tough. The public forum has some disadvantages as a means of examining complex and controversial issues. "Second thoughts" cannot get into the discussion when there is no second round. How can any committee of Congress be certain that it is getting the full and complete thinking of Federal Reserve and executive agencies, with articulation of doubts and differences of view among the men and women with a rightful claim to competence ? 259 5. A summary which links the revenue changes to the general level of economic activity, is as follows: Within the limits of existing capacity, the amount the economy produces will depend upon the total dollar demand for goods and services—the buying of families, businesses, and governments. At the prevailing level of prices, including wage rates, some total dollar amount of demand will be needed to buy the output which would be produced at full utilization of capacity. If total demand falls short, unemployment will result because wage rates and many prices are inflexible downward. On the other hand, a total of demand greater than "needed" will bid up prices—will be inflationary. In either case, fortunately, action by the Federal Government may improve matters. A reduction in tax collections will leave taxpayers with more dollars and thus raise private demand. A rise in tax collections, however, will force the private sector to spend less and dampen forces of inflation. 6. The process by which revenue change influences national income will depend in part upon the responses of the financial system. A reduction in their tax bills will leave families and businesses with more money to spend, but the Treasury will get fewer dollars for Federal spending. Without a budget surplus, would not the Treasury then need to curtail its purchasing by about as much as consumers increase theirs ? If so, total buying by consumers, Government, and business can hardly rise to give the economy much of a boost. If Federal expenditure programs, those which are presumably justified on their own merits, are to be maintained while consumer and business buying expands, how can the Treasury get dollars to make up for those withheld from it by tax reduction ? By borrowing—but where ? What if all the funds being supplied for lending are being taken by businesses, home buyers, State-local governments, and others? Treasury borrowing which would absorb part of this limited supply would thereby force others to scale down their plans. A tax cut cannot lead to net stimulation of total demand when the resultant Federal borrowing deprives private boi^rowers of loan funds they icould have spent. 7. On the other hand, the Treasury may be able to borrow in an environment in which commercial banks create enough in demand deposits to permit the new Federal securities to be sold without depriving other borrowers of funds. Federal Reserve cooperation can assure such a result. The Fed has power to provide the banking system with enough lending capacity to enable the Treasury to be accommodated without requiring curtailment of business and consumer borrowing. 8. The process by lohich tax reduction leads to an increase in total purchasing power, therefore, requires monetary policy that permitsthe Treasury to borroio more without reducing the funds available toother borrowers. Directly or indirectly, the Government must get dollars which would otherwise not be used. Federal expenditure of newly created money will add to the total demand for business output. Furthermore, the effects do not end with the first use of new money. The added dollars will continue to circulate and therebyfinancesome transactions which would not otherwise be made. The cumulative stimulus from dollars created to finance tax reduction will exceed the original amount somewhat—but no one knows by how much. Failure of the Federal Reserve to expand bank lending capacity, or failure of the 260 financial system to perform "normally" may frustrate hopes. In such cases, tax reduction will not spur the economy through expansion of total purchasing power. 9. The recent U.S. record and the postwar experience of many other economies must confirm the fear that expansionary fiscal and monetary policies are likely to bring price-level increases before full utilization of capacity. During Mr. Kennedy's Presidency, the economy did not press on the limits of productive capacity; and the price level, according to many observers, was essentially stable. Even rapid increases in the money supply, however, need not have much influence on the price level, if the new dollars are used to buy goods and services that would not otherwise have been produced. Clearly, however, the nearer the economy comes to full utilization of productive capacity, the greater the likelihood that additions to demand will bid up prices. 10. A rise in the taxpayments of individuals and corporations will reduce the funds available to them for other purposes. To varying degrees, families, and businesses will curtail their buying, thereby reducing this element of total demand. More dollars, however, are pouring into the Treasury. Experience suggests that their availability will tempt lawmakers to raise Federal spending, perhaps with only a short lag. 11. If tax collections exceed Federal spending, the Treasury can use the "unspent" dollars to retire debt. Assume that the Treasury retires debt. The financial institutions which get cash as a result are in a position to increase their lending; upward pressure on interest rates may ease. The demand for loans tends to be high under conditions whicn lead to anti-inflationary action, and borrowing can be expected to rise. The kinds of things bought with the borrowed funds will differ considerably from the consumer items forgone because of the tax increase. Borrowed funds, for example, are used more for capital goods. But total dollar demands on the economy may be about the same. To the extent that this is so, an increase in tax revenue and reduction of Federal debt will have little or no restraining, anti-inflationary effect. 12. Is there some way to prevent the dollars which are used for debt retirement from reappearing as demand in the private sector ? Not as a practical matter. But roughly equivalent restraining effects can be obtained through Federal Reserve action. Many sources of savings feed into the stream of loanable funds to which the Treasury will be making additions. Another source of loanable funds consists of additions to demand deposits by the banking system. The Fed has power to prevent, or slow down, the growth of bank deposits. Bank lending, therefore, can be kept lower than if the Treasury's debt repayment were not adding to funds available for lending. Such restraint imposed by the Federal Eeserve can seek to keep total outlays on investment goods around the level expected if there had been no tax increase. The latter has forced down consumption, however, so that total demand will press less heavily on productive capacity. Another result is that the stock of money grows more slowly than otherwise. Therefore, fewer dollars will be created for future transactions. 13. The tie between a tax increase and monetary policy is often presented in slightly different terms. At any given time the Fed believes that there is a maximum by which it can permit bank lending to rise 261 without creating inflationary pressures. If tax collections were greater, a higher rate of monetary expansion would be consistent with price level stability. M A N A G E M E N T OF THE DEBT Even when the debt is not growing its existence continues to present problems. 1. Some aspects of debt management are technical, but others have significance for the economy generally. A borrowing of $10 billion to repay an equal amount will not have neutral effects on the economy; nor will all holders of old debt accept new debt in return. The Treasury may not borrow from the same groups as hold the maturing debt. It obtains funds in some areas of the market and then makes fund available elsewhere. Consequently, the management of the debt will change interest rates and the relative ease or tightness of credit in sectors of the market. Within limits, which shift from time to time, the choices made in refinancing outstanding Federal debt can serve a constructive purpose. 2. Yet the choices can also have unwelcomed effects, since distinguishing the desirable from the undesirable cannot always be done with absolute certainty. Nevertheless, the men close to the credit and money markets—the Treasury and its advisers—can generally make better policy decisions if the range of available alternatives is broad rather than narrow. Anything in law or tradition which limits the scope for choice in refunding also reduces the possibility of making debt management an effective instrument of policy and creates more than a slight danger of unwise actions. The economy as a whole will function at least a little differently, for better or worse, as a result of decisions about managing outstanding debt. 3. The existence of the debt also affects the economy in terms of liquidity. Monetizing the debt proceeds as each passing day reduces the remaining life of outstanding debt. An obligation due within a few days or weeks has very different liquidity characteristics from one due in 20 years. The shorter the debt, the more it resembles—and serves the liquidity purposes of—money. A Federal debt that does not grow, one that may remain in about the same hands, can gradually become more and more like money. The effects can be at least a little like the inflation-creating tendencies of additions to the money supply. One consideration for the Treasury in managing the debt, therefore, is to have a structure which conforms as well as possible with the economy's apparent needs for liquidity. 4. Some balance of long-, intermediate-, and short-term debt will be better than another. To achieve this balance the Treasury must be able to sell the kinds of issues required. But if new long-term debt cannot in fact be sold, the maturity structure will deteriorate somewhat. Even with ingenious mixes of new issues offered in advanced refundings, the Treasury cannot always achieve the results it would like when the law limits its f reedom of maneuver. 5. A ceiling of 4% percent on interest payable on debt with a maturity over 5 years was a legacy from World War I. Congress in 1967 met part of a Treasury request for authority to sell notes maturing up to 7 years at whatever interest rate the market requires. For longer debt, however, the 4^4-percent limit continues. Imagine trying to fi- 262 nance an industrial enterprise or a public utility or housing with such a restriction,—no borrowing for more than 7 years. For long the 4^4-percent limit had no operational significance. In the 1920's when interest rates were higher, the Treasury was retiring debt. After that, either the level of interest rates or the structure enabled the Treasury to manage a large and growing debt with little or no interference from the ceiling. But this is not the case today. Nor will it always be the case in the future. Long-term loans of top quality may often command more than percent. Why ? The productivity of capital—which underlies much of the demand for private borrowing—will often be appreciably higher. And may not inflation reduce willingness to lend for long periods at interest rates acceptable in the past ? Neither the Treasury nor the Federal Reserve makes interest rates; both are among many elements in the market. Interest rates result from operation of the forces of demand and supply. Part of the demand comes from the National Government. The Treasury in managing the outstanding debt must often borrow. Its demands can have significant effects for a time in a part of the market even though its total debt is not growing. If market forces lead to interest rates which are above 4*4 percent, what can the Treasury do as it faces the need to refund old debt ? Will generous citizens lend "below the market" as a favor? Not extensively. The law permits the Treasury to borrow on short term, now up to 7 years, and pay whatever rates the market demands. On such debt, the Treasury will pay much over 4*4 percent. The legal limit, therefore, does not determine the interest cost of the debt. 6. The ceiling does affect the structure of the debt—and in a way which can be unfortunate. Today, it forces more debt into relatively short>-term form. The shorter the debt, the greater the liquidity and "moneyness." Debt becomes more nearly monetized. When the Federal debt seems likely to endure for generations, a ceiling which prohibits borrowing for 30 or 20 or even 10 years prevents the best adjustment to reality. The ceiling forces issuance of relatively liquid debt forms. To call it an engine of inflation is to exaggerate, but some such result does occur. Today, and very probably in the years ahead, the most skillful management cannot keep the debt from becoming more liquid, from making the economy a little more inflation prone. 7. There is a widespread belief that Government borrowing to finance spending has an expansive effect on business, and debt retirement a contractive influence. This view is oversimplified, to say the least. The economic effects of Government borrowing differ according to whether such borrowing: (a) absorbs funds that would otherwise be spent for consumption; (b) absorbs savings that would otherwise finance investment (that is, creation of new capital properties); or (#) results in the creation of money. The effects are intimately related to what happens in the monetary and financial system. 8. Absorption of funds that would otherwise be spent.—When the Federal Treasury persuades workers to buy savings bonds through payroll-deduction plans, or other arrangements which involve commitments to buy regularly, the economic effect of the borrowing will be somewhat the same as that of a payroll tax—within limits. Some of the purchasing power they divert into bond purchases would otherwise have gone into consumption spending. 263 9. Absorption of funds being saved or already saved.—When individuals or institutions (other than commercial banks) buy newly issued Government bonds with funds they have saved or received from savers, the immediate effect is much the same as if the accumulated savings had been absorbed by a tax. The effect may be the same whether the bond purchaser uses funds on hand and awaiting investment or sells some other form of security to buy the Government bond. In the latter case, the securities or properties sold may eventually find their way to some holder of accumulated savings who used them in this way instead of financing new investment. However, if the eventual purchaser of the assets which are sold borrows from a commercial bank to pay for them, the situation falls into the category of Government borrowing in which bank loans and demand deposits are created. Businesses and individuals seeking funds to pay for new investment goods find conditions tighter. Borrowing costs will certainly rise, and the sale of newly issued shares of stock will be more difficult. Some planned investment will be forgone because of higher financing costs. Government absorption of private savings, therefore, tends to contract national income, more or less offsetting the expansive effects produced by spending the proceeds of the loan. 10. Borrowing that results in creation of money (deposits created by bank lending).—Government borrowing can result in the creation of money—new demand deposits at commercial banks. The borrowing government gets newly created purchasing power, upon which it draws to pay its expenses. The spending of the borrowed funds has an expansive effect upon the economy. The newly created deposits move into the country's stream of monetary payments. The public has more money to spend than otherwise. Total spending, and therefore national income (at least in a monetary sense), are above what they would be if the money had not been created. Money used is not used up. It continues to circulate. 11. However—and this point is often overlooked—the net effect of government borrowing depends upon what happens to private borrowing. If bank lending capacity is so limited that lending to government leads to less lending to business, the net expansive effect is correspondingly smaller than one might expect from looking at government actions alone. 12. Debt reduction.—When a government uses an excess of current revenue to retire a debt issue held by banks, the economic effect is the reverse of floating such an issue. The government repays the debt with bank deposits which it has accumulated, and these deposits go out of existence. Although, in a strict sense, the actual transaction of the debt retirement does not affect the level of business, the process of obtaining the money withdraws purchasing power from the public. And the destruction of deposits in a significant sense passes the contractive effect on into the future—unless private borrowing from banks rises to offset the drop in government borrowing. The raising of the funds to retire the loan would, of course, contract national income. 13. When a government retires a debt issue previously bought by individuals or institutions with saved funds, the effect of the retirement is the reverse of that of the borrowing. The funds received by the bondholders are now available for new investment or consumption spending. The added availability of funds, plus any increase in consumer spending, would stimulate investment. Some time 264 would be required, however, to offset the contractive effects of the taxation that raised the funds. 14. Debt and currency.—Currency is, to some extent, a substitute for Federal debt. If government were to issue currency to retire some debt, the stimulating effect could be great indeed. If the debt were that held by private savers—individuals or institutions—they would generally try to use the funds to buy new income-providing securities, real estate, or other property. The stimulus to investment could be strong. If the government, in contrast, were to borrow from savers to retire currency, the contractive effects would be substantial. In general, the higher the "moneyness" of the debt form, the closer it is to currency. 15. The Federal Government can influence the level of economic activity through management of its debt. For a large part of the outstanding debt, it ought to have the discretion of shifting between issues with a low degree of "moneyness"—long-term obligations of the kind purchased with real savings by individuals and financial institutions—and debt which is nearly money—the issues sold to commercial banks. (The 414-percent ceiling now limits the freedom.) To help check inflation, the Treasury could sell more long-term debt and use the proceeds to pay off debt held by commercial banks. To stimulate the economy, the Treasury could increase its borrowings from the banking system and pay institutional and private investors (assuming appropriate monetary policy). Such a policy would not be costless. It would require the Treasury to do more refunding into longterm issues than it might otherwise plan to do during a boom when interest rates were high, and to forgo opportunities to refund into long-term issues when interest rates were low during a slump. STATEMENT O F E . C. H A R W O O D , A M E R I C A N ECONOMIC INSTITUTE FOR RESEARCH This is in reply to your letters dated July 9, 1968, and September 18, 1968. After conducting extensive research during the past 45 years on money-credit problems, I have concluded: {a) That money-credit developments in recent decades constitutes a repetition of past major economic blunders on a larger scale than ever before; and, (b) That the increasingly complicated plans proposed for coping with the problems that have arisen cannot be useful in the long run; and, (c) That the only hope for avoidance of seriously adverse consequences lies in correction of past errors by restoring sound commercial banking. The results of research that support the findings and conclusions above are presented in the accompanying Economic Education Bulletin, "Why Gold ?" I hope that this reply to your letters, as well as the accompanying bulletin will be helpful to you and your committee. Space limitations prevented printing the above mentioned Bulletin in full. Reprinted below are sections IV and V of "Why Gold?" 265 I V . THE GOLD STANDARD AND THE GOLI> DOLLAR ELEMENTS OF THE GOLD STANDARD For a country to be on the full gold standard the following conditions must be met: 1. The standard monetary unit is a fixed amount of gold. 2. All domestic currency and coin are freely exchangeable at their face value for gold, and whoever obtains gold is free to use it in any way he chooses. 3. There is no limit on the amount of gold that may be brought to the mint for coinage. 4. Gold is full legal tender for payment of all obligations. 5. There is no restriction on the importation or exportation of gold. VARIANTS OF THE GOLD STANDARD A country is on a gold coin standard when the foregoing requirements are met and its currency and nongold coin are freely exchangeable for gold coins that are multiples of the amount of gold constituting the monetary unit. The dollar, for example, as legally defined, is 15.238 grains of gold, nine-tenths fine. If $10 gold pieces were being minted, each one would consist of 152.38 grains of gold, nine-tenths fine. In a country that is on a gold bullion standard, currency and nongold coin are redeemable in gold bullion at a fixed amount of gold for each monetary unit, the minimum obtainable being the amount in a gold bar of a certain specified weight. When the purpose of this kind of gold standard is to discourage the circulation of gold and so economize in its use, the gold bars in which the currency is redeemable are of considerable weight. For example, when England was on the gold bullion standard from 1925 to 1931, currency could be redeemed only in gold bars weighing 400 ounces. This variant of the gold standard has been objected to on the grounds that "A gold bullion standard is a rich man's standard, operating above, and out of reach of, the man of small means. It would be responsive only to the behavior of the rich man, the banks other business enterpries, and the government * * * [with a gold coin standard] there is no discrimination against any individual, particularly the man with small capital, and it is most important that he be able to exercise his preferences and thereby to register his doubts because these are part of the machinery of automatic braking which the gold coin standard provides." 1 In a country that is on the gold-exchange standard, currencies of foreign countries are used as reserves for the domestic currency, often in conjunction with some gold. The principal currencies thus used are U.S. dollars and British pounds. Although the country's monetary system is thus one stage removed from gold, the value of its currency is kept at the established gold parity because of the indirect convertibility of the currency into gold. The principal hazard inherent in this variant of the gold standard is the possibility, which can never be entirely absent, that at some time the currencies used as reserves will be devalued, as was the British pound in November 1967. A serious disadvantage of the widespread use of the gold-exchange standard is described elsewhere in this study. The most important features of the gold standard, whether redemption is in gold coin or gold bullion, are the fixed amount of gold in the monetary unit and the freedom with which gold and currency are interchangeable at the Treasury or central bank. Fixity in the amount of gold in the monetary unit makes gold and the currency representing it an efficient medium of exchange and a relatively stable measure and store of value. The free inter changeability of gold and currency by citizens and foreigners alike provides an automatic mechanism that tends to restrain unsound monetary and fiscal practices. Extreme currency and credit inflation is not possible while the rules of the gold standard prevail. The significance of these features of the gold standard will be considered at greater length in subsequent sections of this study. Here it is sufficient to note that it is the lack of full and free redeemability of U.S. currency for gold that keeps the United States from being on a full gold bullion standard. Although foreign governments and central banks can exchange U.S. currency for gold at the U.S. Treasury, individuals and firms, whether American or foreign, are not allowed to do so, except under license for certain purposes. Thus, the United States, while internationally on a restricted gold bullion standard, is on an in1 Walter E. Spahr, The Case for the Gold Standard, New York, Economists' National Committee on Monetary Policy, 1940, pp. 28-29. 21-570—68 IS 266 convertible paper currency standard domestically. This bifurcated departure from the full gold standard has been at the root of the Nation's money-credit problem for three decades. "PAPEB GOLD" AND THE SDE PROPOSAL Some authorities, including those in the United States, would supplement or replace the gold-exchange standard with some sort of international "paper gold'* for use as a monetary reserve asset by national governments. Advocates of such a "paper gold," standard hope that such "paper gold" will be printed and distributed in sufficient quantities eventually to replace gold as the basic monetary reserve of the world. During 1967 United States authorities made a concentrated effort to persuade the member countries of the International Monetary Fund to accept a plan for creating a paper asset to be used for settling payments imbalances and held as monetary reserves. Agreements made at meetings of the Fund membership in London and Rio de Janeiro during the summer of 1967 were heralded by U.S. officials as momentous steps toward revolutionary reform of the international monetary system. However, the substance of these agreements soon was seen to be little more than a face-saving concession to the United States. France and other European countries agreed to a plan for augmenting world monetary reserves if such reserves became inadequate, but only on the conditions that they collectively would have sufficient voting power to veto the plan as well as other important operations of the International Monetary Fund. Thus, only tentative endorsement was made of a plan to grant each member of the Fund "special drawing rights" (sometimes referred to as SDK's) in addition to its existing rights to borrow from that organization. Such drawing rights would be transferred within specified limits from the central bank of one country to that of another in settlement of international payments imbalances. If such a plan ever is implemented, it will represent merely an extension of the present inflationary gold-exchange standard rather than a radically new development. The possibility of the plan becoming operational seems remote. Continental European countries presumably will not permit its implementation as long as they continue to accumulate dollar claims as a result of the continuing imbalance of payments of the United States. Recent international monetary developments have made authorities of these countries even more reluctant than before to continue accumulating dollar claims and British pounds. Such developments have emphasized the superiority of gold to "paper gold" for use as monetary reserves. Current world sentiment, particularly in Europe, is for closer, rather than looser, monetary ties to gold. W H A T IS A DOLLAR? An article in the Financial Times (London), September 1, 1960, included the following statement: "The U.S. dollar made a further recovery, rising one-half cent to regain its peak level of $35.24-26." Americans who read this surely wondered what it could mean. It must have surprised them to be told that a .dollar was $35.24-26. Moreover, the terms "recovery," "rising," "regain," and "peak level" all misled because they implied that what occurred was favorable to "the U.S. dollar." The sentence quoted epitomizes the confusion of terms, unwarranted connotations, and disregard for accurate definition that are often apparent in discussion of monetary matters. Perhaps the most common semantic error in the field of money and credit is that due to mistaking claims on dollars for the dollars themselves. This error appears in various ways, one of which is the phrase, "the dollar price of gold." Those who use this expression are thinking perhaps of the dollar as a piece of paper currency or a credit to a checking account. However, the fact is that "dollar" is the statutory or legal name for a specified amount of gold. By authority of the Gold Reserve Act of 1934 the dollar was specified as 15%i grains of gold nine-tenths fine, which is the equivalent of one thirty-fifth of an ounce (troy weight) of pure gold. Paper currency and paper credits are not dollars. They represent and may serve as claims upon actual dollars, each of which is the specified amount of gold. To refer to the dollar price of an ounce of gold is like referring to the price of potatoes in terms of potatoes. It would be foolish to say that the price of a bushel of potatoes is 60 pounds of potatoes. If, however, there were certificates ,or book credits representing potatoes, one might have to offer in the marketplace 267 certificates or credits representing 61 or more pounds of potatoes in order to obtain a 60-pound bushel of actual potatoes. This might be the situation if there were some cost or other impediment to the conversion of the certificates or credits directly into the potatoes that they purported to represent. The situation in the London gold market and other gold markets abroad is similar to this whenever the amount of U.S. currency required to obtain an ounce of gold exceeds $35 plus the cost of obtaining gold from the U.S. Treasury and transporting it to the foreign market. In the London market British currency is exchanged for gold (subject to certain restrictions, among which is the prohibition of gold purchases by British subjects), but the amount of that currency required to obtain an ounce of gold can be translated easily inito an equivalent amount of U.S. currency which a foreign holder who wanted gold could use to buy the necessary British currency in the foreign exchange market. Thus, the financial reporter quoted above should have said, in order to be accurate, that ithe amount of U.S. currency required to purchase an ounce of gold in London represented $35.24 (or $35.26). He should have made clear that two different thirtgs were involved in the exchange. V . ECONOMIC BENEFITS OF THE GOLD STANDARD In discussing the economic benefits of the gold standard, we should face frankly the fact that it is not a panacea for all economic ills. Those who offer patent medicines alleged to be cures for all the physical ailments of mankind have long been regarded as quacks. So likewise should we regard individuals who offer any single simple remedy for all economic ills. Particularly important is it to realize that there are both short-term and longrun economic problems, the business cycles of boom and depression as well as the long-term trend. The major economic benefits of a return to the gold standard will not include solutions for all present or short-term economic problems but will be related more definitely to the industrial progress and even the survival of the United States in the longer run. INHIBITING UNWISE FISCAL AND BANKING POLICIES Readers who think that an inflationary trend may be quickly reversed by some happy coincidence of events may be underestimating the temptations involved. In France and most other nations the takings of government-planned embezzlement through depreciation of the currency have been relatively small, because the savings and life insurance of the citizens have been only a fraction of the amount per capita here in the United States. In the United States the present value of life insurance policy reserves, social security trusit funds, individual holdings of Government and other bonds, and the savings accounts of American citizens approximates $850 billion. This can readily be stolen by the subtle processes of inflating and repeated devaluations. If a redeemable currency were restored, the wiser and more farsighted of the Nation's citizens who saw the dangers in unsound fiscal or banking policies could demand gold, and Treasury and bank officials ordinarily would act with awareness of this fact. A possible resulting outflow of gold would force the Federal Reserve authorities and the Government to reconsider unwise policies. Instead of depending on the wisdom of a selected few who might err disastrously, the Nation would provide freedom of action for the many hundreds of thousands of its wisest citizens who presumably can best foresee the probable effects of unwise policies and so act as to counteract those policies. In the absence of the combined judgment of a multitude of keen and experienced observers, there may be no effective check on unwise policies until they have resulted in serious disaster. No wise monetary authority or fiscal policymaker should want to be without such an important guide to policy as a redeemable currency provides, and no foolish monetary authority should be permitted to disregard that guide. Although all money-credit systems require some control, if only to prevent abuses, the automatic features of the gold standard give early warning of credit abuses or unsound procedures and therefore facilitate corrections by a minimum of managements. Because it minimizes the excuse for controls, the gold standard is especially disliked by those who seek to enmesh the economy in a network ><of socialistic restrictions. 268 The Federal Reserve Board should be free to act when there are warning signs of unsound financial developments. An independent agency to provide the minimum degree of appropriate control has been proved by long experience to be the only effective means of managing a nation's money-credit system. Confidence in the future worth of the dollar is essential to long-term Government financing. Experience has shown that a nation that meets its promises to pay enjoys the best credit standing. When the United States resumed gold redemption of its currency in 1879, one immediate and striking result was the reestablishment of the Government's credit standing. Government bonds could be floated at substantially lower interest rates. Redeemable promises to pay presumably would not be issued as recklessly as irredeemable promises might be, in fact usually have been. Throughout the course of history, governments relieved of fulfilling their promises to redeem currency on demand sooner or later have taken advantage of such an invitation to reckless irresponsibility. The fixed amount of gold in each dollar has a relatively stable purchasing power in the long run. Specifically, when not disturbed by the inflating or deflating of other purchasing media, the exchange value of gold has remained remarkably stable for generations. Changes in prices based on gold usually have been gradual rather -than seriously disruptive. When prices continue rising in a country on the gold standard, gold tends to move out from the reserves securing currency and bank deposits, thereby limiting or preventing the further expansion of credit and a subsequent rise in prices. Gold is universally accepted as a medium of exchange. Even when practically all nations of the world have been "off the gold standard" as far as domestic redeemability was concerned, they have sought gold; and the people of the world, whenever there was widespread fear of monetary depreciation, have done likewise. Gold is universally recognized as a valuable substance that does not deteriorate in storage. The fact that a currency is convertible into gold should assure for that currency virtually the same value in exchange and acceptability as gold itself. Such was the Nation's experience for the several decades during which the United States adhered to the full gold standard. Moreover, history shows that experiments with managed paper-money standards, tried by many of the principal nations of the world in the past 250 years, inevitably failed. RESTORATION OF INVESTORS' CONFIDENCE Another important benefit of the full gold standard would be a restoration of confidence, among those who save, in the future value of their savings. Here in the United States, the small annual savings of individuals, largely accumulated by those who labor, reach an astronomical total each year. For example, during 1967 personal savings totaled about $38 billion. Since mid-1950 the confidence of small investors in the future value of U.S. savings bonds seems to have decreased. Redemptions have exceeded, sales with the result that the total held has decreased in the past decade despite increasing population. To what extent the cashing of savings (bonds has reflected distrust of the future value of the dollar, no one knows. Nevertheless, long experience in many countries of the world clearly indicates that such developments, on a much larger and far more devastating scale, are to be expected when public confidence in money diminishes. As our Nation figuratively walks a monetary tightrope above the abyss of national disaster, little imagination is required to visualize the torrent of demand that could flood the Nation's market places if fear finally impelled the multitude of small savers to buy goods, whether needed or not, as an alternative to seeing the value of their savings rapidly diminish. Also to be considered in this connection is the desirability of ending the search for "hedges" against inflation. Small investors are faced with the question, How much of my funds should I invest in such a way that I will be protected against continuing depreciation? Unfortunately it is those individuals who can least afford to take the risks that all equity investments involve wTho may not have enough income, even for the necessities of life, unless they risk such losses in the hope that an increasing income will offset any further rise in living costs. When assured of the future value of their savings, men have confidence and are willing to invest. Such confidence and the resulting long-term commitments facilitate orderly progress. When they have a fixed standard and a redeemable currency the future of which is not being questioned, men can recognize "bargains" and act accordingly. Probably this accounts in part for the fact that firm 269 adherence to the gold standard has invariably hastened recovery from business depressions. When there is no fixed standard and redeemable currency or its future is in question, men have an inadequate basis for judging "bargains" accurately. Comsequently, they hesitate to make commitments; and while potential employers hesitate, the unemployed wait in Government-induced idleness. Such was the Nation's experience during the money-juggling years after 1932. ENDING ADJUSTMENTS FOR A FLUCTUATING DOLLAR Another benefit that could result from return to the full gold standard is the ending of those statisticians' nightmares, the adjustments of indexes for fluctuations in the value of the dollar. Of course, the gold standard would not prevent in the future, any more than it did in the past, the serious distortions of economic values that are attributable to credit inflation and deflation, (the twin evidences of incompetent and unwise banking). Such distortions, however, are relatively limited in magnitude and duration as long as a fixed gold standard is maintained. Return to the gold standard would make unnecessary the elaborate efforts to adjust statistical value-series that have been necessary in recent years. Without adequate bases for economic comparisons even the simplest representations by labor seeking a higher wage or by capital seeking a greater reward become almost unintelligible. The complications resulting from adjustments for a rubber dollar are to many people incomprehensible. Thinking for the moment only of labor's aspect of the problem, we should ask how John Doe is to judge what a pension payable in 1980 dollars is worth to him today? Will the dollar he knows today shrink in value as rapidly as the dollar has during the past two decades, perhaps even more? If so, what is the use of social security and company pension benefits anyway? Is the whole game of trying to provide for the distant future to be a fruitless one that few laboring men can hope to understand? But labor is not the only economic factor cheated by the fluctuating dollar. When plant depreciation charges are based on values long outdated by a shrinking dollar, capital likewise loses. The ordinary books of account reveal profits that are illusory and encourage policies that ultimately can lead many business firms into bankruptcy. When both parties to a contract have a fixed standard unit as the measure of their respective obligations, they can judge the risks involved far more accurately than when their contract promises are stated in the necessarily vague and fluctuating terms of a managed currency. Although, under the gold standard, the buying power of the dollar may be distorted temporarily either by unsound credit expansion (inflation) or the collapse of such unsound credit expansion (deflation), in the long run the exchange value of gold varies relatively little; no other medium of exchange as yet has proved to be so stable. In the long run, with an irredeemable paper-currency system, the inevitable distortions of economic value judgments are reflected in the attitudes of individuals and business organizations. Expedient adaptation to the exigencies of the near future becomes the dominating policy; long-range considerations are forgotten or disregarded. When such views predominate, will the United States continue to be one of the leading nations of the world ? LIFTING THE VEIL THAT CONCEALS BASIC ECONOMIC PROBLEMS Another economic benefit that would result from returning ito the full gold standard would be removal of the "money veil" or "money illusion" that conceals from most people the Nation's basic economic problems. Removing the ''veil" would not guarantee that those now in a position to solve the problems would see them clearly and promptly take appropriate actions, but throughout the Nation many for whom the problems were clarified might press in various ways, including the political, for itheir solution. For example, the present monopoly power of organized labor in basic industries may not be understood by the general public as long as inflationary additions to the supply of purchasing media readily permit wage and price increases. This monopoly power has been made possible by discriminatory statutes and abused by some labor leaders with despotic control on an industrywide basis. While the inflationary process goes on, however, such monopoly becomes more firmly entrenched, more determined to get an increasing share of the wealth 278 produced. The longer the public remains blind to the issue, the more difficult and disruptive the final settlement probably will be. Insofar as the depreciation charges of business are based on prices lower than those currently prevailing and are therefore inadequate for the replacement of capital equipment, business profits lare in part illusory; they reflect a hidden consumption of capital. Taxes based on illusory gains are destructive of real wealth. They hamper the sound growth of the economy, the continued growth that appears to be essential to survival in the world as we know it today. An inflation-stimulated boom is not sound prosperity. There have been several such booms in the Nation's history, and all have been followed by severe depressions. Never have managers of a "managed irredeemable money" been able to create a sound and lasting prosperity. The creation of deposits and currency based on Government debt and noncommercial bank loans does not create real wealth. Such a procedure only deludes those who strive to measure and exchange wealth, and it invites overspeculaition; paper gains are lost in the inevitable depression aftermath. Men are free to the extent that the culture or society in which they live permits them to plan and choose their goals, provides equality of opportunity to act effectively in pursuit of those goals, and permits them to retain the fruits of their labors. There is much evidence to indicate that increasing departure from economic freedom is destroying Western civilization, including our own country; but the economic relationships involved are so obscured by the veil of manipulated money that all too few have any understanding of what the Nation's greatest economic problem is. "TO PROMOTE MAXIMUM EMPLOYMENT, PRODUCTION, AND PURCHASING POWER" 1 The experience of history shows that an irredeemable currency endangers the economic system that uses it. Innumerable instances testify to the truth of this assertion, and none refute i t A fixed monetary standard, on the other hand, facilitates the achievement of equilibrium among the economic factors of production, without which there can be neither full employment nor optimum output of products to be purchased. Only if labor has a fixed and simple standard of value free from the misconceptions attributable to a depreciating currency can it judge the real value of its present gains and possible future pension benefits. The efforts of those who labor to obtain social security benefits and company pensions will be fruitless if the depreciation of the dollar continues at the rate of the past two and onehalf decades. John Maynard (Lord) Keynes, who was a leader among the advocates of a socalled managed irredeemable currency, openly avowed that his scheme was a means of deceiving those who labor and who neither understand nor are in a. position to take advantage of the vagaries of such an irredeemable paper money system. (See Keynes The General Theory of Employment, Interest, and Money_ ch. 2, pt. II.) Only the shrewd speculator and the man of great wealth can expect to profit in the long run from a "managed" irredeemable currency. In the absence of a fixed monetary standard and a redeemable currency thepressure for continued inflation tends to rob those who have the least economic power, depleting the only resources they have. The widows and orphans, the elderly and the ill in health, are virtually defenseless against the ravages of a depreciating dollar that diminishes the buying power of their savings and depreciates the values of life insurance and annuities. Technological progress, given a fixed monetary unit and sound fiscal and banking policies, ordinarily would result in a gradual lowering of costs and prices that would benefit all consumers. Especially beneficial would this be for those whom most men strive hardest to protect, their potential widows and their children. FACILITATING LONG-TERM INDUSTRIAL PROGRESS But of all the benefits to be expected from a return to the full gold standard, perhaps the most important in the present and foreseeable future would be the achievement of the most rapid rate of long-term industrial growth that the economy can sustain. We do not imply that a return to the gold standard alone would insure this highly desirable outcome (future military strength will, even more1 The declared purpose of the Employment Act of 1946. 271 than in the past, depend on industrial progress) ; but failure to restore the gold standard almost certainly will prevent the optimum rate of economic growth. Restoration of the full gold standard would, in all probability, be followed by a long-term downward trend of prices that might continue for some years as the inflationary purchasing media now in circulation were slowly eliminated. That a long-term downward trend of prices could be a greater aid to industrial progress than continually rising prices may surprise many who have accepted the widely publicized notion that the reverse is true. However, the idea that perpetually rising prices are better than falling prices in the interests of an expanding economy in the long run seems to have no basis in recorded economic experience and lacks even theoretical justification. As far as can be discovered, the only seemingly valid argument in support of this notion is based on the fact that prices usually rise during the recovery phase of businesscycle changes. But business cycles are short-term changes; we must turn to other considerations to learn the truth about the relation between rising or falling prices in the long run and economic progress. First, what light can be obtained from a brief review of the relationships that appear to be involved? Industrial progress results from taking advantage of the scientific and technological advances that make possible more effective use of the three basic factors of production: land, labor, and capital. Now the scientific and technological advances are not distributed evenly over all industries or all types of industries, nor are they evenly distributed over all the companies in any one industry. Therefore, in order to derive the utmost industrial advantage from new developments, land, labor, and capital must be shifted from those companies, industries, and types of industries where the scientific and technological advance is slow or nonexistent and must be moved into the companies and industries where the technological advance is the most rapid and most fruitful at particular times. As it happens, only one effective way to judge the relative economic worth of various technological changes has been discovered, and that is the test of relatively free competitive markets. Continually rising prices induced by more and more inflating as a longrun policy have the effect of permanently distorting the markets as long as the policy is continued. Under such circumstances, company A, if it is leading the technological advance, will enjoy not only the profits attributable so that leadership but also the "windfall" profits attributable to inflation. Company B, if it is lagging in the technological advance, might be experiencing losses in the absence of inflation, but under the conditions assumed might be able to report profits in spite of its inability to keep up with the technological progression. Under the circumstances just described, what will happen? The management of company A will, of course, try to expand rapidly and will have profits available that can be used for that purpose. But A can expand its plant and labor force only by bidding a higher price for the three economic factors than B can. If B is encouraged by inflationary profits to continue its existing rate of operations or perhaps even expand somewhat, from what source can A get those factors of production? One would expect industrial growth to be hampered by such conditions, because resources cannot be shifted readily to more effective uses from inefficient companies that continue to operate. On the other hand, when technological progress is reflected in a price level not artificially supported by inflationary monetary manipulation and accompanying currency depreciation, one could expect the long-term trend of prices to be downward, gradually perhaps, but nevertheless downward. And the absence of inflation, accompanied by a downward trend of prices, would prevent windfall profits in the lagging companies and industries, which would experience losses. Such concerns would be forced to release land, labor, and capital to other uses. The shifts that must be made if industrial progress is to be at the optimum rate will then be readily effected. Surely there are few facts more obvious that that the price of progress is change. So much for theoretical aspects of the problem. Is there any proof that the reasoning offered is sound ? In 1879, 14 years after the end of the Civil War, the United States returned to the gold standard. For the decade of the 1870's, the average level of commodity prices measured by one comprehensive index was 117.5. For the three successive 5-year periods beginning with 1880-84, the average levels of commodity prices measured by the same index were 101, 84, and 78. The decline 272 was almost continuous, and by the end of the 15-year period following 1879, prices were down 33 percent from the average level for the 1870*8. During the same period industrial production increased at the most rapid rate for the most prolonged period in the Nation's history. Specifically, if the average physical volume of manufacturing production for the 1870's be considered the base or 100-percent level, the average index numbers for each of the next three successive 5-year periods were 158, 196, and 245. In 15 years the gain was 145 percent, more than 6 percent compounded annually. Moreover, for 11 of the 15 years, industrial production remained well above the estimated long-term trend. These developments showed how groundless were the widespread fears, preceding the return to a currency redeemable in gold, that resumption of the gold standard would be efialamitous and that a prolonged fall of prices must inevitably be accompanied by industrial stagnation. Lest the foregoing be misunderstood, we should make clear that the figures just presented are not offered as conclusive proof that the preceding theory is sound. But the facts of history seem to place the burden of proof on those who sponsor and defend the Nation's long-continued, inflationary monetary policies, who urge that prolonged inflating and continually rising prices will insure Tapid long-term industrial progress. We have a right, even a duty, to ask the proponents of inflation, Are you not •denying to your country the known and demonstrable benefits of a sound monetary system; are you not jeopardizing the industrial progress on which our survival in a possibly hostile world will have to depend? MAINTAINING THE FREEDOM OF AMERICAN CITIZENS Foreign governments and central banks can obtain gold on demand in exchange for the Government's promise on our paper currency to make such payments ; but American citizens cannot obtain from their Government fulfillment of its specific promise to pay on demand. All Federal Reserve notes (except the new ones being issued) carry the unequivocal pledge that the United States •"will pay to the bearer on demand" the number of dollars indicated. Instead of giving to the bearer on demand the dollars promised, the Treasury merely will give other paper promises to pay dollars. Such subterfuge, the substitution of promises for promises instead of the thing promised, is unworthy of a great Nation and an honest people. A fully redeemable currency would restore to the people some degree of control over unsound banking and spendthrift government. Since the departure from the gold standard in 1933, the people of the United States have lost, in large part, their control over the public purse. The full gold standard would restore that control and help prevent the large losses that continuing inflation causes. It would help to preserve the system of free enterprise and free markets that has made the Nation the leading industrial power of the world, and without which the people cannot remain free. It would provide the best assurance that this Nation would remain free and would continue to grow stronger than its enemies. STATEMENT OF GABRIEL HAUGE, TRUST MMSTUTACTTIKEES HANOVER CO. D E A R M R . P A T M A N : Upon returning to my office I have found your letter of July 9th inviting me to reply to questions pertaining to hearings to be held on H.R. 11. Upon inquiry here, I find that our economist, Dr. Tilford C. Gaines, received a similar letter and has been in the process of preparing his replies before your September 1, 1968, deadline. In view of the fact that whatever replies I might have made to your questionnaire would have been developed in consultation with him. I hope you will accept his submission on behalf of both of us. 273 STATEMENT OF T H O M A S M. H A V R I L E S K Y , UNIVERSITY OF MARYLAND ( 1 ) A COORDINATED MACROECONOMIC POLICY PROGRAM Part 1 elicits an opinion on the advisability of coordinating fiscal, monetary, and debt management policies at the beginning of each year under the aegis of the President. Part 2 asks whether this program should be developed solely by the President or jointly by all agencies. I shall briefly answer these questions and then enumerate additional suggestions. The obvious interrelationship between fiscal, monetary, and debtmanagement policies makes "independent" policy formulation an implausible alternative to a coordinated program. The office of the President by force of circumstance is the most feasible instrumentality for harmonizing monetary, fiscal, and debt-management policies. Therefore the executive must author the program. Coordination, however, should evolve from an exchange of opinion among the Federal Reserve, the Treasury, the Council of Economic Advisers, the Bureau of the Budget, and other agencies of macroeconomic policy. Some Additional Observations Polices cannot be skillfully coordinated unless all parties to the program have identical or compatible objectives. Any program of coordination should cite the desired values for the goal-variables 1 of economic policy. The goals of macroeconomic policy should be clearly stated each year; for example, a specific percentage of unemployment^ a specific acceptable rate of price level change, etc. If desired values of the goal-variables are incompatible, for instance, the economy may not be able to produce both 3 percent unemployment and a near zero rate of price inflation, priorities or weights would have to be assigned.2 The coordinated policy program should be premised upon an exchange of opinion among fiscal, monetary, and other agents of the Federal Government; the results of the interchange could be made more operational if all parties used similar hypotheses about the structure of the economy. I do not suggest that various conceptions about the state of macroeconomic nature need be identical or absolutely valid for all classes of phenomena, but hopefully the monetary and fiscal agents and advisers will not pour their respective opinions into alien coordinate systems. If the President, as an arbiter and the ultimately responsible author of the program, would resolve conflicts between the hypotheses of say the Federal Reserve System and the Treasury, the program might proceed more effectively. The coordinated stabilization policy program should also sift and consolidate predictions of the course of future economic activity and forecasts of noneconomic policy force impinging upon achievement of the goals of policy. (See (3), p. 278.) This would promote, at the outset of the year, compatible stabilization policy actions by monetary and fiscal authorities. The absence of such coordination has created 1 This hybrid term is used to designate the variables in the economic structure which monetary and fiscal authorities ultimately try to influence, such as the level of unemployment and the rate of change of the price level. 2 This might be achieved for some goals by setting a feasible desired rate of increase of aggregate demand as an intermediate target for all macroeconomic policy agencies. This suggestion is a logical extension of the target strategy discussed in the text of the statement. 274 some difficulty in the past. For instance, in recent years monetary policy actions may have been inappropriate because of inadequate forecasts of defense spending plans. (Pee (8) p. 282.) If these steps were taken, the President could consequently issue a general statement of fiscal and monetary policy recommendations as seen from a first-of-the-year vantage point. Because of the lack of perfect foresight of economic events, the Office of the President cannot effectively specify monetary and fiscal stabilization policy for the entire year; it cannot effectively prescribe the desired value of the monetary guideline-variable, because the appropriate value of this variable will change from time to time. (See (3) p. 278.) In summary, ideally the Office of the President should promote standardized, or at least consistent hypotheses about the structure of the economy among all parties to stabilization policy; it should cite the specific goals of policy and their relative importance; it should consolidate economic predictions and forecasts; it should set the initial tenor of stabilization policy for all agencies for the forthcoming year. (2) A GUIDELINE OH INTERMEDIATE TARGET STRATEGY Part 3A inquires as to the advisability of imposing an intermediate guideline- or target-variable upon the monetary authority. I first present a summary statement of my views on this issue and then develop these views at greater length. Summary The adoption of a guideline strategy, as discussed below, offers considerable promise. However, I am skeptical that we can yet specify a money supply guideline, or any guideline, for monetary policy. I do not favor the President selecting any guideline for the Federal Eeserve System at this time without thorough apprehension of the issues involved. Because of these issues, as discussed in sections (a) through (d) below, choice of a guideline-variable should be deliberated and prominence should be accorded the opinions of the monetary authority. To select a guideline-variable for the Federal Eeserve System or any agency without further study of the issues seems to invite more radical (I use the term in the philosophical rather than the political sense) change than many seem to realize. (See sec. (d) p. 278.)3 As a separate matter, I believe that no guideline-variable, when one is selected, should have a fixed desired value or a fixed range of desired values. The laudable design of macroeconomic policy coordination seems to have become mistakenly predicated upon the concept of a fixed monetary rule. If, after studied consideration of the issues, a uddeline-variable were adopted, I favor that the Federal Eeserve ystem be permitted to announce different desired values of that variable as frequently as necessary. (See (3) p. 278.) Because the Office of the President would issue a coordinated program but once a year, it could not continually prescribe the desired value of the guideline-variable. The Federal Eeserve System, on the § 3 To avoid misinterpreting the author's position the staff notes that in concluding this discussion the author states, "Yet I believe that enough is known that . . . I favor a total reserve target-variable or a monetary base target-variable or a little less conservatively a narrow money supply target-variable." 275 other hand, has available not only the earlier forecasts of the President's program, but also has current forecasts non-monetary-policy forces, such as defense spending, which affect the goal-variables, such as the level of unemployment, as well as current predictions of future economic activity. (See (3) p. 278.) Therefore, the monetary authority can better specify the desired level or rate of change of the guidelinevariable. It would, of course, be held responsible for these changes within the framework of H.E. ll's proposed "quarterly report to Congress, stating in comprehensive detail its past and prospective actions." The role of a monetary policy guideline-variable A target- or guideline-variable is a variable, such as the long-term interest rate or the money supply, which is affected by monetary policy with less lag than monetary policy affects ultimate goal-variables, such as the level of unemployment.4 The target-variable, in turn, affects the goal-variables after some lag. The advantage of a target-variable is that by adjusting policy to affect adroitly the target-variable the monetary authority can (to the extent that the target-variable is predictably related to the goal-variables and to .the extent that policy can predictably and/or swiftly affect the target-variable) bring about desired values of the goal-variable. For instance, if the long-term interest rate's effect upon the goal-variables of unemployment and the price level is well known and if the long-term interest rate can be affected by monetary policy with little lag, the Federal Reserve, by readily attaining desired values of the long-term interest rate, can easily achieve a desired level of unemployment and a desired (acceptable) rate of price inflation. Some issues in selection of a target-variable There are many problems in target-variable selection. One of these is finding a target-variable which is both affected with no lag and related with complete certainty to the goals of policy. Such an ideal target-variable probably does not exist. For instance, the Federal Reserve can control free reserves with very little lag but the relation of this very proximate potential target-variable to the goal-variables is known only with considerable uncertainty. Therefore, this readily influenced target-variable does not allow good control of the ultimate goal-variables. On the other hand, while the relation of a less proximate potential target-variable, such as aggregate demand, to the goalvariables is quite well understood,5 it is not easily controlled by monetary policy; therefore, it, too, does not allow good control of the ultimate goal-variables. Monetary research ought to suggest an optimal 4 1 believe the linkage between monetary policy action and the target-variables and goalvariables is as follows. Open market transactions in short-term securities affect money market conditions (marginal reserve measures, short-term interest rates, etc.), after a earning assets eventually respond, the money supply and short-term interest rates are affected ; effects are transmitted in the market after a lag to long-term interest rates and a range of credit conditions. Certain components of investment spending respond after a short lag total reserves and the monetary base respond; as bank purchases or sales of lag to the changed short- and long-term interest rates and credit terms via the cost effect as well as the effect on the price of real assets relative to their supply price. There may, in addition, be a wealth effect wrought by changes in the stock of financial assets. Eventually aggregate investment and aggregate consumption respond and finally the goalvariables of policy are affected. This final reaction overlooks the earlier response of certain measures of the balance-of-payments problem to the change in short-term interest rates. 5 See footnote 2. 276 target-variable, one which is influenced as readily as possible by monetary policy and, concurrently, relates as closely as possible to the goal-variable. Now let us systematically relate the degree of controllability of both the target-variable and the goal-variable to the foresight required of the monetary authority. As a general rule: (a) A preferable target-variable should reduce the foresight required of the monetary authority.—It is reasonable to assume that target-variables chronologically less proximate to monetary policy action (affected after a sizable lag) are chronologically more proximate to the goal-variables (affecting them after a short lag). It has been recognized, in addition, that the greater the lag between monetary policy action and the target-variable, the more distant and hence the less accurate the monetary authority's current forecasts of non-monetarypolicy forces affecting the target-variable. Also, after the target-variable is affected, the greater the lag between it and the goal-variables, the more distant and hence the less accurate the monetary authority's forecasts6 of the non-monetary-policy forces affecting the goal-variables. It then follows that less accuracy of forecasts of non-monetarypolicy forces affecting the target-variable must be sacrificed for more accuracy of forecasts of non-monetary-policy forces affecting the goalvariables. The optimal target-variables can neither be too proximate to policy action (because forecasts of the effects of non-monetary-policy forces on the goal-variables will be quite poor) nor too proximate to the goal-variables (because forecasts of the effects of non-monetarypolicy forces on the target-variable will be quite poor and undesired variation in the target-variable will cause undesired variation in the goal-variables). From this sketch it is not obvious that the money supply is an optimal target. Even if the money supply can be fairly accurately controlled, the long lag between the money supply and (unspecified) goalvariables and nence the inaccuracy of forecasts affecting the goalvariables, could make systematic countercyclical policy influence on the goal-variables highly unreliable. Other aspects of the issue of policymaker foresight and knowledge have already received some attention in discussion of the target problem and therefore require little elaboration here. Generally, they all recognize the need for accuracy of the monetary authority's knowledge of both the relation between policy action and the target-variable and the relation between the target-variable and the goal-variable.7 Most of the earlier discussion has centered on the latter linkage. For instance, it has been aptly contended that a money supply target-variable requires knowledge of the money demand function and that a longterm interest rate target-variable requires knowledge of the marginal real rate of return. To impart some generality to and to systematize this problem, my discussion has focused on the relation between the lags and the forecasts of exogenous disturbances affecting the transmission of monetary policy. 6 These latter forecasts are made at the time the target-variable is affected by policy action and influence the desired value of the target-variable. (See (3) p. 278.) This discussion is based on my paper "The Optimal Proximity of a Monetary Policy TargetVariable". 7 In addition, the issue of target-variable measurability has been mentioned. For instance, if data on an otherwise good target-variable are not readily available its usefulness is restricted. I feel, however, that the resources of the Federal Reserve System can be reallocated to make data on otherwise attractive target-variables more immediately available and meaningful. For instance, Federal Reserve reporting of the money supply series has improved immensely in recent years. 277 (b) The selection of a target-variable depends on specific weights attached to the goals of policy and the domino/at tool of monetary policy.—If the rate of income growth is an important goal-variable, some form of a quantity target-variable may be preferable to an interest rate target-variable. For example, a specified desired growth rate of the monetary base, the money supply, or bank credit may be a simpler concomitant of income growth than an interest rate level. On the other hand, if some measure of the balance-of-payments problem is an important goal-variable, an interest rate target-variable may be preferable; that is, to lessen certain aspects of the balance-of-payments problem all one need know would be certain international interest rate spreads. Target-variable choice depends also on the dominant tool of monetary policy; for example, a short-term interest rate target-variable might relate more closely to the discount rate tool than a total reserves target-variable. (c) The selection of a more useful target-variable requires that the Congress rescind its implicit mandate to the monetary authority for money market stabilization.—The Federal Reserve System has in the past seemed to derive satisfaction from achieving an "announced value"8 of a target-variable irrespective of its relation to the goalvariables. There are two rather popular reasons for this. First, there is the well-known Federal Keserve tradition of stabilizing money market conditions. This may result both from the well-publicized influence on the central bank of money market operators and bankers, who are notably averse to volatile short-term interest rates, as well as from the historical mandate from Congress for orderly money market conditions. Second, the Federal Reserve System and especially the manager of the open market account are probably wary of error (missing "announced" targets) in daily or weekly operations.9 If the monetary authority gives preference to achieving the desired targets irrespective of the closeness of the target-variable's relation to the goal-variable, more proximate, more easily attained target-variables, such as variables which measure the condition of the money market, will be selected and control over the goal-variables will be sacrificed. As outlined in section (b), the less the lag between monetary policy and the target-variable, the greater the lag between the target-variable and the goal-variables. This implies less accurate forecasts of non-monetary-policy forces affecting the goal-variables; consequently, control over the goal-variables will be sacrificed.10 8 Although currently these "announcements" have been somewhat impressionistic statements about the "tone" and "feel" of money market conditions, they may perhaps be translatable by some arcane process into a bounds for one or several of the variables which measure the condition of the money market. ®Two likely reasons for this come to mind. First the Federal Reserve System strongly prefers to hit an "announced" target because by avoiding errors it reduces the blame it receives for adverse changes in the goal-variables. This is not to be confused with the more reprehensible Federal Reserve practices of ambigously alluding to a bewildering assortment of unused longer run target-variables, before and after the fact. Second, as suggested by section (d) below, the Federal Reserve System and the open market account manager are probably not oriented to learning by error and receive less benefit from a missed announced target than would a more experimentally minded central bank. This is elaborated in my paper "Some Risks of Monetary Policy Innovation," forthcoming in Quarterly Review of Economics and Business. 10 Some economists have suggested that both money market stabilization and longer run targets could be pursued, others have suggested that they are incompatible in a world of uncertain knowledge. The point here is that if the Federal Reserve is concerned with achieving one or several targets independently of the effect on the goal-variable, a more proximate target-variable will be chosen, and by the assumptions of section (b) above, goal-variable control will be sacrificed. 278 The tradition of money market stabilization will disappear as the banker's influence on the Federal Reserve System continues to wane. This tradition would further fade if Congress would rescind its mandate to the Federal Reserve System to stabilize money market conditions. Given these two influences, the central bank would then become less fearful of error (missing "announced" targets). The confluence of all these tendencies ought to make the monetary authority more willing to adopt target-variables less proximate to policy action. Closer control of the goal-variables would then be more likely. This means that some "defensive" operations directed at money market stabilization must be sacrificed. There are probably numerous economists who now would not hesitate to impose an operationally "better" target-variable upon the Federal Reserve System to help dispense with its apparent aversion to missing announced targets. However, I believe that this would be a rather reckless obtrusion. (d) The selected target-variable should not force the Federal Reserve System into a contrived risk-assuming posture.—It seems to me that the monetary authority is a risk-averse technologist who performs his assigned task according to often obsolete theories (perhaps inapplicable to any phenomena such as the free reserves doctrine), or rules of thumb. (When confounded by theoretically oriented critics, the officials of the Federal Reserve often and perhaps aptly contend that monetary policy is an "art", that is, it is handled by rule of thumb.) In this case, exacting the adoption of an announced target which can be hit only after a tedious trial-and-error process would probably impose radical institutional change upon the central bank. Indeed, a space vehicle manufacturer would not be expected to "launch" new projects until all but the uninsurable risks have been reduced to tolerable limits. Is learning by trial-and-error experiment the proper domain of the central bank ? What effects would such radical change have upon economic behavior?11 While I believe that the central bank's aversion to error will abate both as the banker's influence wanes and if Congress rescinds its mandate for money market stabilization. These questions should be considered before a particular target-variable is chosen. Having little empirical evidence and being cognizant of the gravity of some of the above constraints, it is difficult to offer a specific targetvariable for adoption. Yet I do believe enough is known that the targetvariable of monetary policy may be extended beyond variables which measure money-market conditions such as free reserves and the shortterm interest rate. I favor a total reserve target-v airable or a monetary base target-variable or a little less conservatively a narrcno money supply target-variable. [Emphasis supplied.] ( 3 ) ALTERING T H E DESIRED VALUE OF T H E TARGET-VARIABLE Parts 3B and 3C seek an indication of the factors which influence the setting of the target-variable's desired value. After outlining this I 11 Society recognizes that risk-assuming, innovative behavior by many organizations could be disruptive and sets maximum acceptable limits to this activity by many means; including law, insurance underwriting, custom, etc. It is possible, for instance, that if the central bank's stabilization of money-market conditions were terminated, the financial asset demand functions of households and businesses would shift thus vitiating current estimates of these important relations, reducing the policymaker's knowledge even further and making stabilization policy less effective. 279 adduce several reasons why the target-variable should not have a fixed value. The role of leading indicators and forecasts The desired level or rate of change of the target-variable should be altered periodically depending on changes in leading indexes of future economic activity (which relate to the goal-variables) and changes in forecasts of non-monetary-policy forces affecting the goal-variables. Indexes must lead economic activity by the size of the lag between policy action and its effect on the goal-variables. The particular leading index chosen depends on the length of the lag which in turn depends on the specific goal-variables and their respective weights. For instance, if the price level responds to aggregate demand after a greater lag than unemployment, monetary policy for price stability must use an index which leads economic activity by a greater degree than the index used by monetary policy for low unemployment. The length of the forecast of the impact of non-monetary-policy forces depends on the length of the lag between the target-variable and the goal-variable. For instance, if residential construction is a targetvairable, forecosts of the impact of nonmonetary forces on the goalvariables need be less proximate than if the monetary base were the target-variable. The monetary authority must also use forecasts of non-monetarypolicy forces affecting the target-variable. These forecasts will effect the strength of policy action but will not affect the desired value of the target-variable unless they are correlated to non-monetary-policy forces affecting the goal-variable. Many economists believe that lags of monetary policy are variable. This would clearly weaken the usefulness of forecasts because the policymaker would know neither the necessary temporal length of his forecasts nor what leading indicator of economic activity to consult. It has been suggested that the presence of varying excess demands and excess capacities in the economy causes a change in the rate of change of the money supply to affect economic activity with a varying lag.12 If these excess demands and excess capacities were better understood, this Friedman lag might be more predictable and the monetary authority would know how far in the future he must forecast and what leading indicator of economic activity he must consult. On the other hand, the variability of the rate of change of the money supply has been suggested as a cause of the Friedman lag.13 Varying money supply impulses are believed to cause economic activity to be affected after a varying lag. For instance, if the monetary authority maintains a target value of, that is, pegs, the short-term interest rate, the money supply would show considerable variability, as it is used to moderate the effect of changing credit demands upon the short-term rate. However, if the money supply instead were actually employed as the target-variable and were made to react consistently to identical stimuli, ceteris paribus, its variability and the variability of the lag would be reduced. 12 Franco Modigliani, Statement before the Joint Economic Committee Congress of the United States, 90th Cong., second sess., May 8, 1968, p. 12. 13 Karl Brunner, "The Role of Money and Monetary Policy" Review. Federal Reserve Bank of St. Louis, July 1968, p. 20. 280 CRITICISM OF A FIXED M O N E T A R Y RULE I do not favor a fixed (or fixed range) desired value for the targetvariable for five general reasons that have been expertly discussed elsewhere and need only be reviewed here. First, countercyclical monetary policy should not be abrogated by resort to rule because of the imminent possibility of continued improvement of Federal Reserve knowledge and effectiveness. Second, as has been popularly contended, fixed rules are inconsistent with cyclically changing goal-variables because the relation between a target-variable such as the money supply and a goal-variable, such as the level of employment, is not constant over the business cycle. Third, a fixed rule would be inconsistent with observed variation in the secular growth of the economy. For instance, technological change would alter the rate of growth of the economy and make obsolete a predetermined growth rate of the money supply or some other financial quantity. Fourth, fixed rules may actually have a destabilizing effect upon the cyclical movements of the goal-variables. This is surely true of a pegged interest rate and may be true of a pegged rate of money supply growth.14 Fifth, by some occult process, the laudable purpose of macroeconomic policy coordination and congressional review of monetary policy has been predicated upon the monetary rule concejpt. A rule is not necessary for Congress to readily discover what the monetary authority is doing. The quarterly required reports featured by H.R. 11 would serve this purpose. Finally there is no reason to believe that "rule" is intrinsically superior to discretion. This belief is highly suspect because it is grounded in the philosophy that Central Government activism is innately evil and irrevocably undemocratic. Indeed, the absence of an effective activist Central Government would directly abandon more economic and political prerogative to the few inordinately powerful private interests. Popular freedom and welfare are already subjugated to such concentrated private political and economic power that the relinquishing of more power by the Central Government (where there is, at least, a hope of its being democratically influenced) should not be permitted. (4) DEBT M A N A G E M E N T POLICY A N D DEFENSIVE OPERATIONS Parts 4 and 5A inquire as to the role of debt management policy and money market stabilization through "defensive" open market operations. Debt management can best assist macroeconomic policy by continued implementation of the well-known techniques by which the debt may be more efficiently refunded in order to reduce instability arising from Treasury activity in the money market. This is, of course, a concomitant of the proposal that Congress relieve the monetary authority of its undue concern for attaining an intermediate target-variable irrespective of its effect upon the goal-variables of policy as discussed in (2), sections (c) and (d) p. 274. In other words, because of the undue weight the Federal Reserve attaches to attaining an announced target, they adopt target-variables which can easily be influenced 14 William Brainard and James Tobin, "Econometric Models: Their Problems and Usefulness," American Economic Review supplement, May 1968. 281 (variables which measure conditions) money market. Under conditions discussed in (2) page 274 this results in sacrifice of control of the goal-variables. To the extent that defensive open-market operations place undue emphasis on money market stabilization for its own sake, they would be superfluous if Congress explicitly relieved the Federal Reserve of this responsibility. However, to the extent that "defensive" operations are used to maintain a desired value for the target-variable which is consistent with desired values of the goal-variables, they are necessary. Short-term money market variability may very well be suppresed by improved debt-placement technques. (5) SECONDARY TOOLS OF MONETARY POLICY Parts 5B and 5C ask about the role of the other tools of monetary policy. These issues have been extensively discussed for some years and only opinion need be registered here. The discount rate has impeded the implementation of monetary policy because it is not changed frequently enough to discourage member bank borrowing for profit. If not altered more frequently, the discount rate should be tied to the short-term Treasury bill rate. Regulation Q has also impeded monetary policy and should be abolished. Reserve requirements, if raised as well as lowered more effectively could result in potentially less Government debt in the banks in the long run. This would benefit the taxpayer because more interest would be returned from the Federal Reserve banks to the Treasury. Open market policy should remain the principal tool of monetary policy. (6) FEDERAL RESERVE REPORTING TO CONGRESS AND OBSERVERS AT OPEN MARKET COMMITTEE MEETINGS Parts 5D and 5E ask about the costs and benefits of: (1) requiring quarterly Federal Reserve System reports to Congress and (2) the. presence of observers at Open Market Committee Meetings. Complete written record should be made at least as often as the desired level or rate of change of the target-variable is altered. These should be collected and sent to Congress quarterly. The rationale for desired changes in the target-variable would include alterations in leading indicators as they predict independent variations in the goalvariables and new forecasts of the impact of non-monetary-policy forces on the goal-variable. The risks of this procedure to the Federal Reserve would be minimal if the issues regarding target-variable selection were given their due consideration. (See (2), sections (c) and (d) P- 2 7 4 -) As part of the coordination program a fairly extensive and enriching dialogue between monetary and fiscal agents and advisors would include extra-agency observers at Federal Reserve as well as Treasury and CE A meetings. (7) ALTERING T H E STRUCTURE OF THE FEDERAL RESERVE SYSTEM Part I I concerns numerous proposals for altering the structure of the Federal Reserve System. 21-570—68 19 282 I interpret these provisions of H.E. 11 as helping to assure that the goals of macroeconomic policy, popularly mandated to the President,, are more directly effectuated by monetary policy. This is premised on the principle that the Federal Eeserve System should not be a money supply trustee and the public should ultimately have jurisdiction over monetary conditions. This principle can be effectuated in part by relieving the monetary authority of its short-run money market orientation. (See (2), sections (c) and (d) p. 274.) The most formidable constraint on this process should be the necessity of having the Federal Eeserve System remain a fairly viable source of economic opinion. Therefore I generally favor all provisions of this part of H.E. 11 except for having the Federal Eeserve seek congressional appropriation. This feature could inhibit the effective cultivation of the relatively independent economic opinion necessary for an effective program. Congressional or executive whim might temporarily vitiate the research-critical capacility of the CEA and the Treasury but could not immediately effect the Federal Eeserve System as long as it were fiscally independent. Other features of the bill might permit central government to reduce the valuable critical independence of the System but surely not as quickly as the fiscal provision. My support of the1 provisions, other than the appropriations feature, is based on the belief that Central Government will not long remain in a scientific dark age. (8) RECENT MONETARY POLICY Part I I I seeks an opinion on recent monetary policy. My opinions on monetary policy since 1964 shall be confined to a few remarks. In December 1965 the Federal Eeserve System now appears to have acted too late in raising the discount rate and launching tighter money. However, I recall the dominant opinion at the time among many economists (especially in government) was that the System's action was premature; this historical instance argues for preserving the independence of the Federal Eeserve's critical opinion within a coordinated policy program and suggests caution in reform of the Federal Eeserve System's structure. (See (7) p. 281.) Through December 1966 the reduced growth in the narrow money supply was appropriate, grounded as it was on forecasts of the effects of non-monetary-policy forces; namely, defense spending, upon the economy. The swing to buoyant growth in the money supply in 1967 was only roughly suitable given indicators of the expected decline of economic activity and the increased demand for liquidity. However, after mid1967 money supply growth was excessive, primarily because of inaccurate forecasts of defense spending and inaccurate assessment of the prospects for passage of the tax increase during that time. Again this instance testifies to the desirability of a coordinated program. (See (1) p. 273.) In retrospect the earlier 1968 increases in the money supply growth seem to have been anticipating the promised economic slowdown of late 1968 and early 1969. 283 S T A T E M E N T OF D O N A L D D. H E S T E R , U N I V E R S I T Y OF WISCONSIN H.E. 11 is a dangerous bill for, while it commendably introduces some long overdue reforms, it simultaneously endangers the existence of one of our most successful and effective public agencies. Specifically, the bill is desirable in striking out the myth that the Eeserve System is somehow owned and hence controlled by the banking community. The fiction of Federal Reserve bank stock is worth discarding. It is also on sound footing when it urges closer coordination between the executive branch of the Government and Federal Eeserve policy. Toward this end it is probably a valuable contribution to make the term of the Chairman of the Board coterminous with that of the President. However, it is not necessary for the President to have additional powers or controls over the Board for he and the Congress can always largely offset Federal Eeserve policy through Treasury actions. If a really intransigent Board failed to cooperate with the executive branch, appropriate legislation could then be drawn up in time to avert any serious disruption in the economy. Eeducing the number of and/or term of Federal Eeserve Board members, auditing the accounts of, and/or authorizing congressional appropriations for the Federal Eeserve System seem to be unnecessary and to be capable of annihilating this efficient and effective agency. In my judgment the recent performance of the Federal Eeserve System compares favorably to other Federal financial agencies both in its dayto-day supervision of private sector firms and its longer run planning of our financial system. It has made mistakes, of course, but its continuing research programs and its flexibility in coping with new situations have proven to be admirable safeguards. Both its long-term research and its flexibility are likely to be seriously compromised if H.E. 11 is enacted. The financial structure of the American economy is complex; only long experience and study of our System can guide effective policy formation. The present 14-year terms of Board members is an appropriate span because it insures both a mature understanding of the System and continuity in System research programs. Flexibility of policy and research programs are likely to suffer if the Board must seek congressional approval for its research budget. Other agencies which must receive congressional approval of expenditures, such as the SEC or the FHLBB, have suffered from inadequate staff and research funds; their effectiveness has been corresponding^ curtailed. New services, such as nationwide computerization of check clearing, are sure to suffer from budget cuts just when the American banking system is about to become overburdened with paperwork. If evidence of corruption or gross inefficiency at the Federal Eeserve were available, then congressional appropriation of System funds might be desirable. However, I knowT of no such evidence. Along the same lines, a regular audit of Federal Eeserve System accounts seems an unnecessary and expensive control procedure at this time. It obviously could allow confidential System plans and bank examination data to be leaked to the public in a damaging and costly fashion. More importantly it might endanger the integrity of the Federal Eeserve System. Congressmen, acting in behalf of lobbies, could exact substan- 284 tial concessions from the System in return for generous appropriation votes. The founders of the Federal Reserve System wisely recognized that money is power and that politics is the game of acquiring power. They were equally wise in insulating the prime supervisory agency from congressional supervision of budgets, from rapid turnover in Board membership, and from executive branch auditing of accounts. Until clear evidence of weakness in this arrangement is available, I urge rejection of H.R. 11. The list of questions concerning monetary guidelines flows from a philosophical approach to policy implementation which I oppose. In responding to these questions I shall attempt to suggest both my preferences for policy formation and the deficiencies of the "guideline," "indicator", or "target-variable" approach. Questions 1-1, 1-2.—Policy, monetary, fiscal, or debt management, should not be constrained by a program drawn up on January 1; all three policies should be coordinated so as to reach the goals of the President and, of course, those stated in the Employment Act of 1946. A principal advantage of both monetary and debt management policy is that the lag between a decision to act and the debt or monetary action is arbitrarily short. As the recent surtax experience amply testifies, such immediate implementation is not always possible for fiscal policy. It would be very unwise to impair our most flexible instruments by tying them to a program. It also serves no good purpose to tip off speculators about the direction in which interest rates are likely to move. This is not to say, however, that these policies should be applied independently of present and prospective fiscal policies. The Economic Report of the President is a valuable document in its own right; it does not need to be supplemented by statements about the likely direction of monetary policies. The President should continue to be responsible for producing the report. Question IS.—A target-variable, rather tautologically, is something one aims at. If many variables are "targets," as I believe should be the case, then many guns or "instruments" will be needed. I see no j ustification for shooting at the money supply or some interest rate; they are simply instruments. What matters are targets involving the level of employment, the rate of change of consumer prices, the rate of growth of GNP, the balance of payments, the distribution of income and wealth, and the equality of opportunity. The relations between these targets and the instruments of monetary policy are complex and can be represented only in the form of a rather detailed analytic model. A well-structured model is likely to suggest explicitly the relation between various policy instruments and the goals which, say, the Council of Economic Advisers may wish to achieve. The model will also suggest whether certain goals are within our reach. It will not yield exact prescriptions because most equations are not exact relationships. It nonetheless can be a very useful aid in developing policy and in setting values for different instrument variables. Existing econometric models are imperfect in various ways and are being improved as time passes. Each model specifies a link between various policy instruments and targets. This linkage varies from model to model because different model builders have different ways of repre- 285 senting the economy. It also differs because models have been estimated from different time periods. The legal environment, preferences, and technology of our great economy change as time passes; there is no reason to expect stationary relationships. Unfortunately, different estimation techniques are also a source for different linkages. I am not able to write down a multiequation model which is completely adequate for conducting the economic affairs of the Federal Government. Indeed, I do not think such models are likely to appear during the next decades. I do think that various interest rates, tax rates, Government expenditureflows,money supply measures, and legal restrictions such as usury laws, reserve requirements, interest rate ceilings, and minimum wages are important instruments. Because the economy constantly changes, I would always hope that policy implementors would use such models with a good bit of discretion and with a sense for how the system is changing. But, a few things are clear. First, if the economy cannot be adequately described with many equations, it certainly cannot be adequately described with few. Second, it is sometimes argued that because some variables are historically correlated, they may be expected to continue to be so in future years. Moreover, such a relationship is thought to be a valuable policy guide. Almost all sensible aggregative econometric models exhibit high correlations. High correlations are no substitute for clear thinking and they are a weak basis for policy formation. Third, except in very fortuitous circumstances if policy instruments are intended to hit a number of goals (targets), it will be necessary to have a number of instruments simultaneously applied at specific levels. When goals change, many of the instruments will have to change as well if policy is to succeed. Finally, to concede that our knowledge about the relation between policy instruments and goals (targets) is imperfect and sometimes misleading does not mean that models are evils. We should not opt for simplistic, ignorant policy rules simply because we lack omniscence. I believe that it is sheer insanity to look for some touchstone or crude univariate rule of thumb to keep our economy in order. The economy is complicated and our Government is capable of dealing with it more responsibly. Mathematical models applied with discretion are the promising path. Question —I am very uncertain about the importance of debt policy. Most recent empirical studies have uncovered little evidence of a relation between interest rates and change in the maturity structure of the Federal debt. Nevertheless, intuitively I believe that lengthening the debt maturity will tend to discourage real investment. The failure of this result to appear in empirical studies probably reflects our ignorance about the very complex simultaneous structure of asset markets. Debt management, in my unsubstantiated view, can be valuable in determining the level of employment and in fostering intermediate-term economic growth. Question 1-5-—I have previously commented upon the inadvisability of adhering strictly to some predrawn policy schedule for the year ahead. I can see no advantages which the economy might realize by having Congress regularly informed. 286 Seasonal and other transient factors should enter directly into the large models discussed above. If the solutions to those models suggest that defensive policies are desirable, then of course they should be pursued. There is no reason to consider such factors independently; reaching the goals in the President's economic program is the sole objective. I should stress that a defensive smoothing of seasonal fluctuations is not necessarily a desirable activity of the Federal Reserve System. In its absence, private sector firms would make allowance for the existence of such fluctuations when managing their cash positions. However, it is probably true that economies of scale exist in such smoothing operations and that the economy as a whole is best served by having the Federal Reserve offset transient money fluctuations. I think it would be a serious mistake to restrict monetary policy to a single vehicle, open-market operations. As suggested above there are considerable advantages in having a number of policy instruments available in order to reach the many objectives to which the country and its leadership aspire. Variations in reserve requirements, margin requirements, discount rates, and so forth, are all potentially useful. I do not think that monetary policy can be efficiently implemented solely by open-market operations, although they are a very effective vehicle. Changes in other instruments such as discount rates affect many variables which are importantly related to our economic objectives. In the absence of the aforementioned aggregative economic model, it is irresponsible for me to guess the intensity with which each targetvariable is affected by changes in instruments. Orders of magnitude can be inferred from some recently published econometric model simulation studies. Below I suggest that recent interest rate regulations may have worked rather severe hardships on the housing industry, but these regulations were not initially imposed by the Federal Reserve System. I have previously argued that confidentiality is essential in implementing monetary policy. Chance disclosures can result in immense profits. In addition, applying monetary policy is not an exact science. It is easy to imagine that individuals interested in discrediting monetary policy could irresponsibly publicize minor technical flaws. These critics might in turn cause monetary authorities to act with less speed than was in the national interest. I am not sympathetic to regular reports to Congress and I have no suggestions for what should be included in such reports. Similarly, I think that it is unnecessary for Congress, the Treasury, or the CEA to have observers at Open Market Committee meetings. Those organizations have ample channels to transmit information to the committee under the present arrangement. Question II.—At the beginning of this statement I indicated my opinion of H.R. 11. Items 2, 4, and 5 are potentially dangerous and will not help the President implement policy. As suggested above, these proposals are likely to compromise both the effectiveness and quality of our monetary system. A more appealing reform proposal would be for the Congress to reduce the current number of Federal agencies regulating financial markets and to put the survivors under an agency modeled after the Federal Reserve System. I have the general impression that these other agencies' policies are less coordinated with current fiscal policy than are actions of the Federal Reserve. It might also be a good idea to increase the control which the Board of Gover- 287 nors has over the 12 individual Federal Eeserve banks although this is a relatively minor reform. Question III.—The following summary of recent financial developments is an abridged version of a paper to be published elsewhere. It does not purport to be a complete evaluation of recent policy, but only an incomplete exposition of how monetary policy may have operated. The evaluation has two main messages. First, interest rates and interest rate restrictions are powerful instruments in allocating investment funds writhin the economy. Monetary policy works and is potent. Second, the disruptive allocative effects described below would have been unnecessary if the administration and Congress had followed the advice of most professional economists and promptly raised taxes in 1966 or 1967. It is no fault of the Federal Eeserve System that our Government irresponsibly failed to allow for the costs of our Vietnam war involvement. Open market operations in 1966 doubtlessly considerably retarded inflationary pressures which are now so ominous. If the Federal Eeserve had not acted, price increases and associated hardships would be far more troublesome today. Precisely this flexibility in policy formation must not be sacrificed with the passage of H.E. 11. Table 1 reports recent quarterly financial acquistions by the household sector as measured in the flow-of-funds accounts.1 In addition it reports how individuals distributed their disposable income between consumption and saving in the national income accounts. All flows are seasonally adjusted and reported at annual rates. The household sector is the largest sector reported in the flow-of-funds accounts. TABLE 1—RECENT AGGREGATIVE QUARTERLY STATISTICS, 1963-68 [In billions of dollars] ( N E T ) financial acquisitions by households ( S A ) 1963—1st quarter 2d quarter 3d quarter 4th quarter 1964—1st quarter 2d quarter 3d quarter 4th quarter 1965—1st quarter 2d quarter 3d quarter 4th quarter 1966—1st quarter 2d quarter 3d quarter 4th quarter 1967—1st quarter 2d quarter 3d quarter 4th q u a r t e r . . . . 1968—1st quarter Disposable personal income (SA) Personal saving (SA) 369.7 400.7 406.9 414.1 423.9 435.8 443.1 449.6 456.0 464.0 479.4 489.3 497.5 503.3 512.4 522.0 532.7 540.0 548.2 557.9 571.5 19.3 19.2 18.8 22. 5 22.0 27.7 25.6 29.5 24.5 24.0 30.9 29.3 26.6 28.7 29.2 34.6 38.8 36.0 38.5 41.6 38.0 Demand deposits and currency Commercial bank savings accounts Savings institution savings accounts Life insurance reserves Pension fund reserves 2.2 4.1 13.7 6.5 .9 5.9 11.7 -2.2 1.4 .5 8.1 13.6 14.2 7.3 14.5 8.3 8.5 11.2 13.5 9.8 15.8 13.8 11.6 14.3 13.0 8.5 18.0 17.9 18.7 8.5 15.8 17.1 15.2 13.3 12.3 13.2 12.5 9.8 4.9 5.3 9.3 16.9 20.6 18.0 10.7 4.4 4.2 4.2 4.6 4.8 4.9 4.8 4.7 4.7 4.6 4.6 5.3 3.9 4.6 5.3 11.7 10.8 12.0 10.4 11.6 11.2 11.7 13.0 11.5 13.5 15.1 12.4 14.7 14.9 15.2 Credit market instruments 7.6 1.8 .9 -2.9 11.7 3.7 -1.0 13.7 15.4 11.1 2.0 -12.9 -18.0 -1.1 16.8 Sources: ( a ) Survey of Current Business, June 1968; ( b ) Business Statistics, the 1967 biennial supplement to the Survey of Current Business; and ( c ) Federal Reserve Bulletin, October 1966 and May 1968. 1 Households Include persons as members of households and personal trusts and nonprofit organizations serving individuals such a foundations, private schools and hospitals, labor unions, churches, and charitable organizations. About 10 percent of household sector assets are believed to be controlled by personal trusts and nonprofit organizations. (Flowof-funds accounts 1945-67, p. I. 33.) 288 First, note the marked increase in the saving rate by individuals. Roughly speaking the average propensity to save was 5 percent in 1963, 6 percent m 1964-66, and 7 percent since then. Classical economists would have expected this pattern because interest rates were rising quite steadily until September 1966 and then reached a new high region beginning in late 1967.2 These saving data are consistent with an hypothesis which argues that monetary policy is effectively transmitted throughfluctuationsin bond interest rates. The remaining columns in the table show net acquisitions of financial assets by households. Quarterly acquisition flows into life insurance and pension fund reserves have been quite stable through time. Contractual agreements for these assets prevent wide quarter-toquarter fluctuations. Net acquisitions of demand deposits and currency by households fluctuate greatly. These data do not appear particularly illuminating when attempting to describe recent events. Net acquisitions of commercial bank savings account deposits by households also do not appear useful for describing recent events. A slight slowdown in deposit growth occurred in 1966 and a strong recovery began in the first quarter of 1967, but this pattern only weakly mirrored events at savings institutions. The remaining two series, savings institution deposits and credit market instruments, appear to have been very interactive. In the first three quarters of 1966, households greatly reduced their acquisitions of savings deposits and simultaneously shifted heavily into various credit market instruments. During the first two quarters of 1967, they disposed of many of these securities and rapidly reacquired savings institution deposits. During the last quarter of 1967 the process again seems to be reversing itself. In table 2 it can be seen that among credit instruments, most of the swing involved debt instruments of the U.S. Government. TABLE 2.—DETAILS OF RECENT SWINGS I N CREDIT MARKET INSTRUMENTS [Financial acquisitions by households, seasonally adjusted] U.S. Government securities 1965—4th quarter 1966—1st quarter 2d quarter 3d quarter 4th quarter 1967—1st quarter 2d quarter 3d quarter 4th quarter 1.7 8.7 11.0 7.0 4.9 -8.6 -9.9 15.1 State and local obligations 3.0 2.7 -3.1 4.4 4.8 -.5 -2.4 1.9 1.1 Corporate and foreign bonds -2.1 3.5 3.4 .8 -2.8 -.7 .9 3.6 2.6 Corporate stock -3.4 1.2 4.3 -1.6 -5.3 -2.6 -4.8 -6.2 -2.1 Mortgages -.3 -2.3 -.2 .6 .5 -.5 -1.7 -.4 .2 Source: Federal Reserve Bulletin, May 1968. It appears that households were responding very sensitively to the levels of Government security interest rates. During the latter part of 1965, 1966, and 1967, interest rates paid on deposits at savings and loan associations were relatively stationary when compared to bond market interest rates. Households bought U.S. Government securities heavily during the two periods of high interest rates, 1966 and late 1967 ; they sold them during the first half of 1967 when interest rates were lower. It is very difficult to escape the conclusion that Government 2 A brief rise In the saving rate in 1964 above 1965-66 levels can reasonably be attributed to the tax cut of 1964. 289 securities and savings institution deposits are close substitutes. Openmarket operations are likely to be especially effective in diminishing flows through savings institutions when interest rates on savings deposits fall behind rates of return on Government securities as they did during this period. The savings and loan industry is an extremely complex financial intermediary w^hich is regulated by the Federal Home Loan Bank Board. About two-thirds of savings institution assets are in savings and loan associations. For brevity, attention will be confined to this indusrty. Approximately 85 percent of the industry's assets are in first mortgage loans; savings and loan associations are the largest suppliers of 1-4 family conventional mortgages in the American economy by a wide margin. Except for New England and the Mid-Atlantic States, savings and loan associations dominate local mortgage markets across the country. The industry grew rapidly until the end of 1963 and then more slowly until 1966 when growth nearly ceased for a year. The slowdown in growth after 1963 partly reflected a profit squeeze which the industry experienced because of rising interest (or dividend) rates paid on deposits (or shares) and falling new mortgage loan interest rates. In March 1965 the Bank Board became alarmed that rising deposit interest rates might seriously endanger the stability of the industry Thus: * * * there is clear evidence that for some time escalation of dividends has accelerated the flow of savings to institutions or markets where performance did not fully justify further injections of mortgage money * * *. The Board found that institutions were raising dividend rates under conditions which did not appear to require more aggressive competition for funds * * * the Board determined that it would be unwise to extend advances to members increasing dividend rates until it had an opportunity to evaluate the effect of * * * [recent increases in dividend rates by some associations]. (Federal Home Loan Bank Board. Annual Report, 1965, p. 51.) These advances restrictions were modified several times in the subsequent months. During the month of June [1966], it became increasingly evident that the policy of restraining dividend increases by restricting access to Board credit was losing effectiveness * * *. In late 1966, the Board terminated all restrictions under this program. (Annual Report, 1966, p. 47.) In their place on September 21, 1966, Congress empowered the Federal Home Loan Bank Board, after consulting with the Board of Governors of the Federal Reserve System and the Board of Directors of the Federal Deposit Insurance Corporation, to put ceilings on rates paid on deposits; these powers continue to the present day. As one might have expected, deposit growth weakened very significantly in response to interest rate restrictions beginning in mid1965. This weakness continued until late 1966 when declines in bond rates occurred. During the first 6 months of 1968 the experience of 1966 appears to be recurring. Thus, during these months the net deposit inflow was $2 billion in 1966, $6 billion in 1967, and $3.6 billion in 1968.3 The principal difference between 1966 and 1968 is that interest rates of commercial banks and mutual savings banks were more effectively controlled in the latter year. 3 Sources are recent releases from the Data Management Division of the Federal Home Loan Bank Board. 290 The effect of these interest rate restrictions on mortgage flows appears to have been dramatic. Until mid-1964, mortgage acquisitions were steadily! rising, if allowance is made for seasonal variations. Acquisitions leveled off in 1965 and then collapsed in 1966 and early 1967. They partially recovered in 1967, but in recent months again seem to be falling. Mortgage acquisitions by insured savings and loan associations in 1967 were below those in 1962; 1968 is likely to be no better. Given the growth in the American population and the rising price of houses, it is apparent that the number of mortgage transactions executed per capita has fallen considerably in recent years. Eising mortgage interest rates have sharply curbed demand for new mortgage loans. The above picture is broadly consistent with the following interpretation although it is certainly not unique in that respect. Because the average maturity of their assets, essentially mortgage loans, is longer than the effective average maturity of their deposit liabilities, rises in interest rates will temporarily depress the net worth of savings and loan associations. In order to protect liquidity and, in the case of stock associations, the rate of return on invested capital, associations may initiate further rises in deposit interest rates to attract new funds. The Federal Home Loan Bank Board, in its role of protecting the solvency and stability of the industry, becomes alarmed at declining net worth and with rises in foreclosures and reacquired real estate which naturally occur with mortgage loan growth. It therefore curbs further deposit interest rate increases. By this action it makes deposit flows into associations very sensitive to changes in other interest rates in the money market. In effect, the Bank Board has greatly increased potential short-term interest rate fluctuations in mortgage markets by stabilizingfluctuationsin deposit interest rates. Table 3 reports seasonally adjusted quarterly investment series from the National Income Accounts. The series suggest that monetary policy discriminated heavily against investment in residential structures and that this burden was concentrated in late 1966 and early 1967. Accompanying this investment slump was a very pronounced acceleration of Federal defense spending associated with the Vietnam war buildup. Government spending tends to grow steadily in the United States, but a rapid acceleration of this magnitude must be offset elsewhere in a full-employment system if price stability is to be maintained. A relatively compelling interpretation of table 3 is that the housing industry bore much of the burden of the defense buildup. This need not have been the case if fiscal policy and especially tax increases had been introduced at an earlier date. Monetary policy did a big job as well as could be hoped for. Hopefully the lesson from this experience wil not go unheeded if further spending increases become necessary. Deflationary monetary policy works by inducing the private sector to reduce its spending and increase its saving. This may be achieved by driving up interest rates which discourages investment and increases the reward for saving. In recent years households have saved more and invested less in new residential construction. Monetary policy has worked to offset inflationary pressure. 291 TABLE 3.—QUARTERLY INVESTMENT FLOWS, 1963-68 ( S A ) Nonresidential structures 1963—1st quarter 2d quarter 3d quarter 4th quarter 1964—1st quarter 2d quarter 3d quarter 4th quarter 1965—1st quarter 2d quarter 3d quarter 4th quarter 1966—1st quarter 2d quarter 3d quarter 4th q u a r t e r . . . . 1967—1st quarter.. 2d quarter 3d quarter 4th quarter 1968—1st quarter 18.8 19.7 19.4 19.9 20.4 21.1 21.4 21.8 23.1 24.7 25.1 27.3 28.3 27.5 28.2 27.7 27.7 26.3 26.6 26.7 28.5 Producers' durable equipment Nonfarm residential structures 33.2 33.8 35.5 36.8 37.9 39.0 41.0 41.6 44.1 44.6 46.8 48.3 50.0 51.2 53.1 55.1 54.2 55.2 56.2 57.3 58.7 Change in business inventories 25.5 26.2 26.5 27.4 27.1 26.6 26.5 26.3 26.6 26.5 26.4 26.2 26.5 25.3 23.2 20.4 20.9 22.5 25.0 27.0 27.6 Federal defense expenditures 4.7 4.8 6.0 8.1 4.8 6.1 4.8 7.7 10.6 8.8 9.4 9.9 9.9 14.0 11.4 18.5 7.1 .5 3.8 9.2 2.7 51.2 50.5 51.0 50.3 50.5 50.7 49.8 48.9 48.4 49.2 50.3 52.4 55.1 58.4 63.0 65.6 70.2 72.5 73.3 74.2 76.7 Sources: ( a ) Survey of Current Business, June 1968. ( b ) Business Statistics, the 1967 biennial supplement to the Survey of Current Business. STATEMENT OF W A L T E R E. H O A D L E Y , BANK OF AMERICA I appreciate the opportunity to comment to your committee on the general question of the coordination of monetary and fiscal policy and onH.R.ll. Certainly most economists would agree that both monetary and fiscal policy have a major role in the achievement of the goals of "maximum employment, production, and purchasing power" as set out in the Employment Act of 1946. Moreover, monetary and fiscal policy have a crucial role in helping to alleviate the whole range of new problems caused by the substantial achievement of full employment itself. Maintaining full employment, once achieved, probably poses one of the most serious economic problems facing the Nation over the period ahead. In the past 3 fiscal years, increases in Federal revenues have fallen far short of increases in Federal expenditures and the result has been clearly inflationary fiscal policy. Monetary policy actions have tended to be very restrictive in an attempt to offset the inflationary impact of fiscal policy. ^ Although we can all agree as to the desirability of increased cordination of monetary and fiscal policy, care must be exercised that they are not tied together in such a way that inflationary fiscal policy is reinforced by inflationary monetary policy. Under present institutional arrangements, monetary policy is much more flexible than fiscal policy in practice. Thus, when evidence becomes available that economic projections or projected Federal receipts or expenditures are in error, monetary policy can be more quickly adapted to the changed circumstances. In this context, I fear that there would be little net gain and very possibly a net loss to the economy by setting out a program coordinating fiscal, debt management, and monetary policies at the beginning of each year. This does not imply that monetary and fiscal policies are independent and mutually exclusive, but only that the Federal Reserve 292 should maintain a degree of flexibility to adapt monetary policy to changing conditions. While I firmly believe that forecasting is an indispensable and inevitable element in all decisionmaking, I am concerned that specific public forecasts of monetary policy would become self-defeating. This is likely because of the vagaries of international monetary and political developments as well as unpredictable domestic events not to mention the present well-advanced degree of sophistication in United States and worldwidefinancialmarkets. As recent experience amply demonstrates, we need to improve our ability to forecast economic activity and find wrays to increase the flexibility of fiscal policy and the timeliness of its application. The tools of monetary policy must be improved if we are to avoid the selective impact of monetary restraint on certain sectors of our economy, especially residential construction. However, at this time, I know of no single economic indicator which could serve as an exclusive guide to the monetary authorities. Debt-management policy could be used as more of an adjunct to monetary and fiscal policy if the 4% percent ceiling on Treasury bonds were removed. At present, inflationary fiscal policy is reinforced by debt management policy as the deficit must be financed with shortterm, highly liquid securities. Removal of the interest rate ceiling would allow a more orderly and effective debt management policy. The modification in the discount window operation recently proposed by the Federal Reserve would largely eliminate "defensive" open market operations. Instead normal market operations could then be used to implement monetary policy. Regulation Q should not be used as a tool of monetary policy. The Federal Reserve now reports in great detail annually to the Congress on past actions. In addition, reports are made available on open market committee meetings after a reasonable delay. As suggested, it would be difficult and presumably unwise for the Federal Reserve to report on prospective actions because of the present margin of error in worldwide economic forecasts and the need for a flexible monetary policy response to changing economic conditions. I see little benefits to additional reports nor in the presence of Members of the Congress, the Treasury, or the CEA at Open Market Committee meetings. I'm still not convinced that the suggested changes in the structure of the Federal Reserve System as provided in H.R. 11 would be an improvement; there is real value, however, in reviewing the structure periodically to insure that it keeps abreast of national needs and changing money market and institutional conditions. In any event, the members of the Federal Reserve Board need to be as free as possible from short-term partisan political considerations. The period since 1964 has been a period in which fiscal policy was overly stimulative and contributed to inflationary pressures. The attempts of the Federal Reserve to control inflation without precipitating a recession have generated many stresses and strains in the financial markets and depressed residential construction activity. And, although I may not have always agreed completely with the actions taken by the Federal Reserve Board, I think the Federal Reserve System performed rather well during this difficult period. 293 In closing, let me make it clear that I strongly support the use of the tools of fiscal and monetary policy and debt management to achieve the goals of full employment and rapid economic growth within the context of price stability. Any action which the Congress could take to make these tools more effective in achieving these goals would certainly be welcome. However, on the whole, I believe that H.R. 11 would be counterproductive in this regard and t