The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
FDIC POLICY TOWARD B K FAILURES Walter A. Yarvel The marked increase in the number and size of banks that have failed in recent years has focused attention on the problems connected with bank failures and the appropriate aid bank regulatory agencies should provide to banks in distress. Since this aid is designed to maintain public confidence in the banking system, there is a need for greater public understanding of policies toward banks with serious problems. Special attention is given in this article to the activities of the Federal Deposit Insurance Corporation in providing assistance to insolvent banks. The different forms this assistance has taken over the years, as well as current FDIC policy as revealed in two recent large bank failures, are examined. Special problems relating to large bank failures raise questions concerning the adequacy of the size of the deposit insurance fund, the constraint this fund places on FDJC decisions, and the coordination and cooperation among bank regulatory agencies necessary to minimize the impact of bank failures. These issues will also be discussed. * The mandate given the FDIC by Congress in 1933 was quite cIear. Its purpose was to ‘lpurchase, hold, and liquidate . . . the assets of banks which have been closed; and to insure the deposits of all banks.” This prescribed order of duties supports the proposition that “the primary function of deposit insurance is, and always has been, protection of the circulating medium from the consequences of bank iailures. That insurance also serves the purpose of guarding the small depositor against loss from bank failures cannot be denied, but this function is of secondary importance” [3, p. 1911. Deposit insurance provides a safety mechanism against a sudden decline in the money supply through bank fail- The prevention of wholesale bank failures has been the expressed intent of Congress since the establishment of the Reconstruction Finance Corporation and the enactment of the Glass-SteagaIl bill in 1932. These emergency measures were followed by the passage of the Banking Act oi 1933, which was intended to be a permanent answer to the problem of widespread bank failures (over 11,000 banks failed between 1921 and 1933, nearly half of these after 1930). The Act established a national deposit insurance system under the FDIC. This was an important element in the fight to restore confidence in the commercial banking system and resulted in a precipitous decline in the number of failures after 1933, as shown in Chart 1. Over the years deposit insurance has helped to strengthen the banking system and has served as a stabilizing influence on the economy. FEDERAL RESERVE BANK OF RICHMOND 3 Rather than simply reures (see box insert). placing deposits of failed banks, deposit insurance liabilities of a failing bank protects depositors in full. It has been recognized, however, that the reduces the incidence of failure by assuring public that bank deposits are safe-thereby deposit assumption method has additional benefits not available through deposit payoffs. In some cases, the continuation of banking services to the community and minimizing the impact of the failure may be vital considerations. the pre- venting runs that can topple even sound banks. The Federal Deposit Insurance Corporation was established to assist in the protection of the nation’s money supply. Though the Corporation’s raison d’etre is widely accepted, its operating methods have been a controversial issue. The regulatory agencies in general and the FDIC in particular, charged with carrying out Congressional statutes, must decide which of their powers gives the greatest support to the banking system. The choice of methods employed by the FDIC has resulted from consideration of the financial status of the banks in question, different interpretations of Congressional intent, and Congressional inquiry itself. This choice has been a difficult one -one that has caused debate in the past and will undoubtedly continue to do so in the future. The FDIC has four alternative procedures that may be followed in assisting a failed or failing bank. These alternatives are : 1. DIRECT PAYMENT OF INSURED DEPOSITS: Acting as receiver of the bank’s assets and making direct payments to insured depositors; 2. DEPOSIT ASSUMPTION: merger with a healthy institution Facilitating a or replacement. by a new organization with new ownership and management, through loans and/or purchase of assets, thereby protecting 3. DIRECT 4. DEPOSIT LOANS: all deposits; Supplying direct financial aid in an effort to correct. deficiencies bank to continue in operation ; INSURANCE to allow the NATIONAL BANK: Operating a Deposit Insurance National Bank for a maximum of two years prior to a deposit payoff or deposit assumption. The first two methods of operation have been authorized since the establishment of the Corporation and have been, by far, the most commonly used. While the maximum deposit insurance protection has been increased from time to time (from $2,500 in 1933 to the present $40,000), legislators have taken the position that “it should never be the policy of Congress to guarantee the safety of all deposits in all banks” [9, p. 21. Under the direct payment to depositors method this mandate is maintained. If a bank is closed by the appropriate state or Federal authority and placed into receivership for liquidation of assets, the insured deposits are paid up to the maximum allowed by law. On the other hand, extension of advances to other banks to assume the deposit 4 ECONOMIC REVIEW, FDIC Activity and Congressional Supervision When distress situations occur, the FDIC has attempted to safeguard the public’s trust in banking largely through a varying policy of direct deposit payoffs and deposit assumptions via mergers. Table I outlines FDIC assistance to failed banks since 1946. It shows that FDIC officials avoided the direct payment of insured deposits between 1946 and 1954. Corporation officials felt that such procedures, with the loss of some depositors’ funds and an interruption of banking services, did not provide the support needed to maintain confidence in the banking system. Instead, mergers (usually consummated with financial aid from the Corporation) were the exclusive method used over this period. Congress challenged the Corporation’s avoidance of the direct payment method through receivership in 1951 [lo]. The contention was that the FDIC had insufficient evidence in some cases to base a decision on whether or not the assumption of assets by a healthy bank would reduce the risk or avert a loss to the Corporation’s insuranc:e fund as required by law. Some legislators argued that the FDIC did not know the full extent of its liability in all such cases, and therefore it may ble preferable for banks to be placed into receivership and direct payments on all insured deposits The Corporation, nevertheless, conbe made. tinued its established policies until 1955, when four deposit payoffs were experienced. The methods used to assist distressed banks have undergone Congressional scrutiny periodically since 1956, reaching full force in 1965. Seve:n banks failed in 1964-all being placed into retceivership and only the insured deposits paid. The direct concern of legislators this time, however, was not 100 percent insurance protection versus limited protection. Instead, the adequacy and quality of Federal banking supervision, examination, and interagency cooperation became the main subjects of Congressional inquiry. Of immediate concern to the Senate Committee on Government Operations [ 1 l] was the growing number of abuses by bank management that had been prime factors in the increasing incidence of SEPTEMBER/OCTOBER 1976 THE IMPACT OF BANK FAILURES To understand is first “high-powered and (3) to ratio as the deposits serves. bank of (H), three currency Federal of ratio by a fractional reserve result in an equal percentage The ratios of interest and the sion hold to money imposed and banking, is than The unity, public supply, of cannot rate of interest minants of rate minus net of service rate does of the to wish of are to deposits market to the currency the creation the confidence quences ‘Much of and this net of bank re- dollars high-powered things equal failures, banking to nation’s discussion is from bank unless of bank factors with these D/R Since deposits or impact is public’s is significantly currency. it the have two total essential. that can, the between on found capita in the the H constant). as convertibility per ratio is the In the been D/C net or expected ratio when paid money a study expected major net deter- may varies paid may with on to dump high-powered from for so bank (the the de- public If the withdrawal many money in the declines, negative normal the services changes deposits become it is only with and free deposits of money-currency. be forced ratio through on losses conditions, form this positively rate deposits proportionally between implicitly expected on such increases relation explicitly additional neutralized, problems Federal paid expensive) the income present Under (less form the to deposits. periods interest in shift Cagan unless of is its assets acquired on the to meet deposits and public. unless liquidity real which, lower connecting magnitude the reserve The in hri (assuming Phillip The the funds). needs the at During scale, its [l], alters fractional of a downward multiplicative to A deci- deposits. money either desire deposits +- D/C]. of payment ratio. reserves, the desira- the public’s in and of respec- relative D/R Under bank as long a less total formula deposit/currency (interest as desirable level currency the the assets impact the addition, on and in a decline States the of The determining losses). rate stock. deposit/reserve decisions decisions, and while high-powered can have interest his a large liquidity of that expected all more on its bank the of other by in composition. reduces of relative deposits the demands,of in for total is held several liquidity impact of earning expected deposits-i.e., lose the of (D/R), reserves member and banks ratic,? an -j- D/C)]/[D/R to United and ratio. high, responsibility Some the it (1) money plus up money, bank currency its system result variables greatest as aggregate currency on on occurs meet supply, money? reserves support determined by D/C consequent will ratio the show of paid even hold in results charges may in decisions. have in well money the in this for deposit/currency suspensions positor demand interest the H [ D/R(l the the fraction the on deposits interest the ratio of deposits of income, the M = currency Cagan’s real cash can of and held may contraction in D/C a shift paid the While expected net for as from determine the examination demand vault making are these balances viewing ratio: Since an the bank bank deposit/currency ratios balances affect cash smaller in a reduction determine is maintained. to stock the reserves held supply smaller the deposit/currency greater banks deposits useful cash affect a multiple the in influencing of money of desired law balances forces therefore, stock however, of High-powered change the money stock deposits reserves A in of these their by cash the ratio, and, mdney by proportion of deposit/currency of offered a withdrawal reserves system. the (D,‘C). plus bank in in the components as changes of a larger replaced, of deposits-thereby services amount otherwise two change magnitudes composition composition aggregate stock The the currency change the the Requirements bility of on interdependent. concerning tively. banking bank public public held decline changes D/C). effects are public and for the latter money under D/R the Reserve-the of deposits (namely by held high-powered in a sudden commercial held money will result accounting of to currency of the dollar can factors (2) the deposits amount with One failures recognize money” the defined as how necessary ON THE MONEY STOCK for banking prevent money other agencies can lead to massive banks-regardless to indiscriminant of neutralize runs on banks supply. . FEDERAL RESERVE BANK OF bank RICHMOND withdrawals their failures that of financial may and have position. It is maintain public serious conse- The FDIC, actions by bank management. charged with supervising the bank in receivership and making payments to depositors, and the Federal Reserve, responsible for making funds available to member banks in times of financial stress (over $9 million to San Francisco National), were precluded from entering a joint effort to rehabilitate the bank prior to its closing. The Investigation of several banks bank failures. involving dishonesty on the part of bank officials revealed a lack of interagency cooperation. The cause of the failure of the San Francisco National Bank in January 1965, for example, was not revealed to the FDIC or the Federal Reserve System until the bank was placed in receivershipover seven months following the finding of illegal Table INSURED BANK FAILURES, DEPOSIT I PROTECTION, FDIC DISBURSEMENTS AND 1946-1975 Banks in (FDIC YeCir Placed Banks’ Receivership Banks Deposits Receivership)* Direct Assumed 1 - 1947 From in FDIC FDIC Disbursements Deposit ($ in Payoffs Assumptions ($ Thousands) Thousands) - Disbursements Deposit - - 5 1946 FDIC Receiving Loans 265 1,724 - i948 - 3 1948 - 4 1950 - 4 - 2 - 1,884 1,340 1951 1952 - 3 - 2 - 1954 - 2 3,182 - 1953 269 2,552 - 5,039 902 1955 4(4) 1 - 4,459 2,343 1956 10) 1 - 2,981 463 - 1,056 1957 2(l) 1958 30) 1959 1961 1962 5(5) - 1963 m 1,856 - 4,799 - 10) - - 30) 1960 1 - - 6,191 - - - 1964 1965 - - 19,247 - 7(5) 231 2,801 12,471 - 3(3) 2 - 11,383 456 1966 l(l) - 732 14,339 1967 - 1968 4(4) - 6 - 7,864 - 5,053 1969 4(4) 5 - 7,652 i a,552 1970 4(4) 3 1971 5(3) 1972 l(l) 1 - 1973 3 3(3) - 1975 3(3)** ** *** One Deposit FDIC Note: Source: 6 bank placed insurance disbursements Deposit Annual poyoff Report into 10 national in state banks were in connection 1975 figures of of are the for Federal Dec. bank formed with 31 Deposit of authorities by the its Insurance in 1957; receiver insurance respective ECONOMIC 109,245 16,105 year of two 173,201 N.A.*** N.A.*** each closed responsibilities plus Corporation, REVIEW, two 167,748 16,781 - 1 - 4 receivership 19,696 53,739 1 - 3 1974 * 26,691 1 estimated in 1958, 1959, and 1971. banks. totalled additional $305.6 million. disbursements annually. SEPTEMBER/OCTOBER 1964, 1976 for the respective banks. members of the Senate committee concluded that cooperation and liaison among the Federal banking agencies were absolutely vital to the public This authority has been used only three times -July 1971, January 1972, and August 1976and then only with rigid constraints. In the first two interest. cases, the Corporation required that existing shareholders, not the FDIC, bear the existing loss potential on the bank’s assets. The FDIC also prohibited dividends from being paid, required new officers and directors to be subject to FDIC approval, and further restricted each bank’s activities. In the most recent case, direct assistance was granted to keep the bank going for three weeks until a deposit assumption could be arranged. Legislation designed to aid the regulatory agencies in protecting banks from criminal acts and gross mismanagement followed with the passage in 1966 of the Financial Institutions Supervisory and Insurance Act, which provided cease and desist powers and provisions for removal of officers and directors. An increase in the number of problem banks, plus the limited use of the newly provided supervisory powers spurred another Banking and Currency Committee investigation in 1971. The fourth alternative method for protecting depositors, the organization of a deposit insurance national bank, was utilized twice during 1975. Section 11 of the Federal Deposit Insurance Act authorizes the FDIC to transfer all the insured and fully secured deposits in the closed bank to the new bank. Those funds are then available to their owners to the same extent as they were in the closed bank. Deposit insurance national banks can remain in existence a maximum of two years, during which time the FDIC can make a public offering of stock in the new bank. Through this procedure, the Corporation hopes to encourage local communities to consider the establishment and capitalization of a new bank before a final disposition oi assets and transfer oi deposits from the insolvent bank. In his statement before the Committee [S, pp. 10-111, Frank Wille, Chairman of the FDIC, outlined the Corporation’s procedures and priorities concerning failing banks. Mr. Wille emphasized that the Corporation had no say in the closing of insured banks-this was the responsibility of its chartering authority: the Comptroller of the Currency in the case of national banks or the appropriate state authority in the case of state banks. It is mandatory, however, that the Corporation serve as the receiver of all national banks and serve as receiver of state banks when appointed. When this happens, the FDIC Board of Directors generally determines whether the deposit payoff or deposit assumption procedure should be foIlowed. The second method is utilized, however, only when the prospective cost to the Corporation is less than the cost through the deposit payoff alternative. A prerequisite to a deposit assumption, of course, would be an existing or newly organized bank that is willing to enter into such a transaction and that is acceptable to the appropriate chartering authority as well as to the FDIC. Congressional interest in the FDIC’s role in recent years, however, has shifted away from the metl~od that the Corporation uses to handle the protection of depositors’ funds to the question of the Agency’s role in the prevenfion of bank failures. The Corporation has not escaped criticism on its depositor insurance methods during this period: though. The Hunt Commission Report expressed the view that the dominant criterion used by Federal insurance agencies in meeting claims should be the needs and welfare of the community involved, not the minimization of payouts from the insurance fund [G, p. 731. The Commission’s report suggested the need for a reevaluation of deposit insurance legislation. This important issue had clearly been subjugated in legislative priorities, however, to the prevention of bank failures. Increasing emphasis has been placed on the Federal regulatory agencies’ responsibilities in preventing bank failures. These agencies have long sought to promote sound banking through examinations wherein management and financial conditions are evaluated. In the course of these The Corporation added new scope to its operations in 1971 when it used, for the first time, the direct loan authority granted in 1950. At that time, the Corporation was authorized to provide direct financial assistance to an insured operating bank in danger of closing whenever, in the opinion of the FDIC Board of Directors, the continued operation of such a bank was essential in providing adequate banking service in the community. Even in this case, assistance is withheld if individuals responsible for the bank’s poor condition will benefit financially or if it appears that assistance may be required over a prolonged period. FEDERAL RESERVE BANK OF RICHMOND 7 examinations, attempts are made to discover and correct unsafe or unsound practices or violations of law and regulations before such practices prove fatal to the bank. Congressional and regulatory attention has shifted to detection of bank problems at an early Congress and enough date to prevent failures. the financial community have come to expect bank regulators to step in and salvage a bank in trouble either as a corporate entity or as a party to a merger. The number of bank failures is not The Federal the Corporation’s concern, however. Deposit Insurance Corporation’s primary function has been the protection of the banking system from the consequences of bank failures-i-e., the creation of problems for otherwise healthy banks and destabilizing influences on the nation’s money Former FDIC Chairman Frank Wille suPPlY* interprets the Corporation’s mission to be one of minimizing the impact of a bank failure. When an insured bank, despite efforts at correction, progresses to the point where actual failure appears likely, FDIC . . conceives its mission to be not the prevention of failure at whatever cost but the protection of depositors and the maintenance of public confidence in the banking system as a whole despite the failure. We seek, in other words, a ‘soft landing’ which minimizes the impact of a bank failure in a community . . . [123. But how does the Corporation presently feel this responsibility is best carried out? For this answer it is best to look to recent experience. Two Recent Failures Examination of recent FDIC policy and procedure in handling bank failures is quite revealing. The largest failures in U. S. history, as well as the most publicized in recent years, have been those experienced by U. S. National Bank of San Diego (USNB) and Franklin National Bank, New York. Criminal charges have been filed in both instances alleging improper or illegal actions by top management. Each case reveals that conscious efforts were made to misrepresent the true financial conditions of the banks and to deceive regulatory authorities. The failure of U. S. National Bank of San Diego on October 18, 1973, at the time the largest bank in U. S. history to collapse ($934 million in deposits), was the subject of a hearing before the Bank Supervision and Insurance Subcommittee of the House Banking and Currency Committee. At that time, Mr. Wille pinpointed the steps taken by the Corporation incident to the transfer of certain assets and liabilities to Cracker National Bank, San Fran,cisco. Of particular inter8 ECONOMIC REVIEW, est to the Subcommittee were the FDIC’s involvement with USNB since its identification .as a problem bank and the Corporation’s consideration of the alternative methods available to it to protect the bank’s depositors and other creditors. In the last few weeks before USNB was closed, during which time the FDIC began preparations in the event the bank did fail, Corporation personnel went to San Diego for the purpose of obtaining specific and detailed financial information to be utilized in discussions with banks i:nterested in acquirin, v USNB’s offices and banking business. Concurrently, reviews of the Com:ptroller’s examination reports, provided to the Corporation, were started in order to measu’re the FDIC’s insurance risk. Estimates were th.at an insurance payoff in this case would necessitate an initial FDIC outlay of approximately $700 million and would result in the immediate loss Iof the use of nearly $230 million to the approximately 3,300 depositors whose deposits exceeded the $20,000 insurance limit in effect at that time. In the judgment of the FDIC Board of Directors and outside bankers involved in consultation, such action would have shaken public confidence in the nation’s entire banking system, with especially severe repercussions in California. considering such a payoff to be the last resort, the Corporation also rejected direct assistance to USNB because the statutory requirement that the continued operation of the bank was essential to provide adequate banking service in the cornmunity could not be substantiated. It was also felt that USNB’s controlling stockholder, responsible for many of the bank’s difficulties, would benefit financially from the assistance. The Corporation began to formulate a tran:saction proposal that it hoped would transfer substantially all the banking business of USNB at a sufficiently high price to satisfy the requirement, as interpreted by Congress and the FDIC, that the merger would minimize the loss to the Co:rporation’s insurance fund. It was recognized that if this could not be arranged the payoff method would be implemented. Serious discussions were begun with three major California banks that expressed an interest in acquiring USNB. In order to insure competitive bidding, the remaining four banks in the state capable of assuming nearly $1 billion in liabilities were also contacted. Two of these decided not to participate in thle bidding for internal reasons, while the other two confronted serious antitrust problems. After consultations with the Antitrust Division of the DeSEPTEMBER/OCTOBER 1976 partment of Justice, it was decided these last two banks would be contacted only if no acceptable bids were obtained from the other three banks. bids from se\-era1 Kew York banks and named European-American Bank and Trust Co. as the winner in the bidding to assume all oi the deposit liabilities and certain assets of Franklin with FDIC assistance. The next morning Franklin’s 104 branches in the Sew York area opened for business as usual as branches of European-American. The apparent ease with which the deposit assumption \=:as completed was, in fact, the end result of five difficult months of contractual negotiations wish potential buyers. During this period, the FDIC attempted to insure competitive bidding by more than one bank on a contractual basis acceptable to all parties.l The restriction placed on the use of the deposit assumption Once it became obvious that the failure was imminent, the FDIC took steps necessary to guarantee an efficient, expedient solution. Segotiations on a purchase and assumption proposal among the three banks and the Corporation were agreed upon. In case the bidding did not realize a premium that would conform to statutory requirements, a contingency plan for a payoff was Mr. Wille’s statement before the Subdrawn. committee was, therefore, carefu1 to emphasize that the Corporation, after careful consideration of all available alternatives, chose to meet its obligations through the method greatest benefit to the public within straints. method was :he decision assistance exceeding the necessary to pay off all terms of sale resulting in the deposit insurance fund the deposit payoff method lowed. that was of statutory con- Within three hours of the closing of USNB by the Comptroller and the FDIC being named as receiver, bids were accepted and analyzed as to their sufficiency, and court approval to the proposed acquisition was granted. The next morning all of USSB’s offices reopened at their usual business hours as branches of Cracker National Bank. The threat of destabilizing and disruptive influences on the American banking system was thus averted. not to contribute cash $750 million estimated insured deposits. If a smaller payout from could not be arranged, would have been fol- During the time of negotiations, it became apparent that rile assisted sale of Franklin would not be possible without a coordinated effort among the banking agencies. The Comptroller of the Current:. constantly monitored Franklin’s financial condition whiIe the Federal Reserve advanced the bank nearly $1.75 billion through its When Franklin Sational Bank ($1.7 billion in deposits) failed in October 1971, it captured the distinction of becoming the biggest bank failure in U. S. history. Once the twentieth largest bank in the country, Franklin’s failure resulted from a series of poor management decisions. Banking analysts generally agree that the bank’s lack of earning power, combined with relatively high loan losses, large losses in foreign exchange transactions, and heavy reliance on the use of short-term borrowings in the money market to back relatively long-term loans, made its failure a foregone conclusion. Of the 65 banks in its size category ($1 billion to $5 billion in deposits), Franklin ranked last in earnings power with a return on assets of only .23 percent. Massive withdrawals of deposits (53 percent of total deposits) followed the announcement of large foreign exchange losses in May 1971. Only heavy borrowings from the Federal Reserve System kept the bank afloat until Comptroller James Smith determined the bank to be insolvent and appointed the FDIC as receiver. : For a detailed disckxure of the FDIC’s participation in the solution to the Franklin pz~biem, see [IS]. Following earlier, the ?Federal Reserve s+.-antes were subsequently assumed by the FDIC and will be repaid hr~ely through liquidation of Franklin assets held by <be FDIC. the USXB Corporation discount RESERVE in an effort to seek an efficient Where widespread public reaction to a precipitous bank fanure is possible, and time is needed to work out a more orderly solution, either the Fed- eral Reserve or the FDIC may be willing to advance funds to the bank on a short-term, secured basis . FDIC concern for the level of uninsured deposits and the interruption of banking services within a community has clearly made the direct pavoff of insr?red deposits an undesirable alterd native in the case of large banks. Consideration of the impact a bank failure has on the financial community (in Franklin’s case both national and international in scope) has become of major im- precedent of a year immediateI>accepted FEDERAL window solution to the crisis.” Interagency cooperation may: in fact? have ad\-anced to the stage where the System was “buying time” for the best solution possible, as Mr. Wille implied. BANK OF RICHMOND 9 It is portance to the regulating authorities.3 entirely conceivable that a policy of minimizing If this view is accepted, there need for regulators or legislators the shock waves of a bank failure in the economy may eventually come into direct conflict with the potential disaster. requirement that a deposit assumption be shown to minimize a threatened loss to the Corporation insurance fund. This potential conflict certainly calls into question sole reliance on the comparison between direct liabilities of the FDIC under the deposit payoff and deposit assumption techniques as the basis for choosing between these methods. This comparison has become necessary due to great concern in the past with the absolute size of the insurance fund and its ability to cover excessive bank failures. Since the impact on the insurance fund has served as a constraint on the Corporation’s attempts to give maximum support to the nation’s money stock, examination of this restriction seems in order. fund were altered to exclude the contingency for failures resulting from the conduct of stabiliz.ation policy, assessment rates could be lowered to correspond with the experience of failures resulting from bank practices. A major practical problem of implementing such a program, howeve:r, would be in distinguishing bank failures attri’b- Adequacy of the Insurance Fund Kenneth Scott and Thomas Mayer, in an article  based upon research undertaken for the Hunt Commission, argue that insurance assessment rates have forced banks to bear substantially more of the costs of bank failures than they have generated. Acknowledging that banks should be expected to cover losses attributable to fraud, misconduct, and “normal” managerial failure, they present evidence supporting their contention that assessment rates have been sufficiently high to generate a large surplus over what is needed to cover these losses. The only justification for such rates is the contingency for failures due to gross perturbations in the economy attributable to the conduct of national fiscal and monetary policy. Deposit insurance for this fourth category of failures seems fully warranted on macroeconomic grounds as a safeguard against sharp and unplanned contractions in the money supply. The cost of this category of coverage, however, should be borne directly by the federal government as the party responsible-and not placed on banks . . . and their customers [7, p. 9001. 3If. in fact. we have 100 nercent den&t insurance for large banks. the-question arises whether the same protection should be-afg$ei small banks on equitable as well as competitive grounds. discussion of the need for review of present deposit insurance legislation, particularly concerning large bank failures, see  and . The latter argues that 100 percent deposit insurance would eliminate the conflict in social goals that arises when considering whether a large bank should be allowed to fail. Optimal resource allocation suggests that inefficient firms, regardless of size, must be allowed The stabilization goal. on the other hand, suggests that to fail. large bank failures should be prevented lest they lead to runs on Complete other banks and to a reduction in the money supply. protection for depositors (but not stockholders) would retain the disciplinary impact potential failure has on bank management but. at the same time, would serve to insulate the money stock from the hazards of large bank failures. Since the FDIC usually protects all deposits. eliminating the insurance ceiling de ju, as well as & facto would remove the uncertainty that large depositors now face. Such a policy would also eliminate the potential conflict between the objectives of minimizing the destabilizing impact of a bank failure and minimizing the cost to the deposit insurance fund. 10 ECONOMIC REVIEW, would be little to look upon a exhaustion of the insurance fund as a If the concept of the “adequacy” of the utable to stabilization policy from other causes. Past losses and disbursements have largely been attributable to the first three causes of failures. From this experience the accumulation of funds for insurance purposes may have been excessive. The argument for increased Government support of the insurance fund is not needed, howeve:r, to draw attention to the facts that the present fund is substantial, has never been threatened by depletion, and presently has a potentially urnlimited source of additional funds. The U. S. Treasury stands behind the FDIC in case the insurance fund is threatened. With a present reserve of approximately $6.7 billion, the Corporation also has what amounts to a blank check It can draw another $3 billion on the Treasury. immediately and after a short delay can obtain any additional amount if needed. Although 527 insured banks have failed since the Agency was established, additional Treasury funds have nev’er been used. Through 42 years of operation, the FDIC has incurred losses of $247 million, including estimated losses on active cases-approximately 3.7 percent of the present fund.4 This loss experience suggests the Corporation has prcotected the insurance fund in an extremely capable manner. Minimization of the loss to the insurance fund may interfere, however, with the pri- mary function zation shares of deposit insurance-the of the money supply-a with the Federal Reserve stabili- responsibility System. it FDIC and the Fed: A Common Bond The Federal Reserve, through the conduct of monetary policy, attempts to maintain the domestic money supply at levels consistent with the financial health of the nation’s economy. Through its dis‘The trend toward large bank failures may have further implications for the adequacy of the insurance fund. however. If one of the largest banks in the country were to fail, initial FDIC cash outlays would likely exceed the present level of the fund. This would be the case even if liquidation of the bank’s assets held by the Corporation resulted in a zero loss to the fund. Under such circumstances. Treasury assistance would presumably be required in the interim. SEPTEMBER/OCTOBER 1976 count mechanism and as supervisor of a large number of commercial banks, the Fed has acknowledged responsibility to provide funds on a secured basis to solvent but temporarily illiquid banks. The purpose of this “lender of last resort” function is to insure the viability of banks experiencing short-term liquidity problems-thereby protecting the public’s confidence in banking, thus preventing runs on bank deposits and destabilizing impacts on the money supply. Deposit insurance has a similar rationale. By minimizing the risk of deposit loss from bank failure, deposit insurance limits the potential cost of holding money in the form of deposits. This discourages the withdrawal of deposits that, if widespread, can cause a sharp reduction in the money supply. The distinction between a temporarily illiquid bank and an insolvent one provides the Federal Reserve a benchmark with regard to which the decision to employ its lending function may be made. Our system of bank supervision and review usually provides the regulatory authorities with the information necessary to pass on the financial conditions of individual banks. Once it is determined that a bank cannot remain viable, the problem of how its operations and liabilities should best be handled arises. The FDIC disposes of those necessary failures in a manner that, while in the public’s best interest, gives masimum support to the circulating medium. regulatory process in banking has been initiated by Congress and will, undoubtedly, receive further attention in future years. Public confidence-the very foundation of the banking system’s existence-is based, fortunately, on more than just the banking agencies’ capacity to “bail out” banks in trouble. For in some cases, whether because of fraudulent actions by bank officials or the inability of regulators to correct management deficiencies, banks ynust be allowed to fail. Public confidence in the banking industry is based on the belief that banking authorities can assure stability through a coordinated program of regulation and supervision designed to limit bank failures only to unavoidable cases and to efficient disposition of fhose banks that do faiZ. Summary Recent experience has revealed extensive coordination and cooperation among Federal banking authorities in the handling of failing banks. This is both encouraging and crucial to the effort to support the banking system. This interagency cooperation has made it possible for banking authorities to lend maximum support to the nation’s money supply in those cases where a bank failure cannot be avoided. Adhering to a policy of minimizing the shock waves to the rest of the financial community, the FDIC has recently shown a decided preference for the deposit assumption method where statutory requirements can be met. But what will happen if the method that is in the public’s best interest comes into conflict with the constraint that the assumption route may only be used if it minimizes the loss to the Corporation’s insurance fund? What would have been the impact on the economy had US;?;B or Franklin Xational been placed in receivership and only the insured deposits paid off? It is doubtful that the degree of confidence in the banking system would have remained as high as it did had thousands of depositors lost millions of dollars in uninsured deposits. Yet it is clear what action the FDIC is required to take if such a conflict occurs. The concern for the effect individual failures have on the insurance fund could, under current legal requirements, eventually force the Corporation to resort to a large deposit payoff that may damage the public’s trust in banking and the regulatory authorities’ ability to support the nation’s money supply. If the latter continues as the objective of deposit insurance, a reevaluation of insurance legislation appears necessary to resolve the problems raised by large bank failures. Regulatory Review There is a recognized need for bank examiners and analysts to keep up with trends and innovations within the banking industry. The banking agencies’ capabilities in meeting their examining responsibilities are dependent on obtaining enough information to reveal the true condition of each bank. This places great importance on the supervisors’ investigative skills. Regulators have expressed a need for greater attention to the safeguards to bank soundness and stability. This concern joins the continuing goals of promotion of competition in banking and adaptation of the banking system to meet changing needs for credit as the focus of regulation. The Federal banking agencies, charged with supervising the country’s commercial banks, have acknowledged that current esamination procedures may be inadequate to the task of dealing with the sophisticated policies of today’s banks. A move toward continuous monitoring rather than single examinations is, therefore, underway. In addition, an estensive review of the entire FEDERAL RESERVE BANK OF RICHMOND 11 References 1. Cagan, Phillip. The Demand for Currency Relative to the Total Money Supply. Occasional Paper 62, National Bureau of Economic Research, 1958. Milton and Anna J. Schwartz. 2. Friedman, Monetary History of the United States, 1867-196$ Princeton : National Bureau of Economic Research, 1963. “The Deposit Insurance Legis3. Golembe, Carter. lation of 1933.” Political Science Quarterly, Vol. LXXXV No. 2 (June 1960), 194. “Failures of Large Banks: Impli4. Horvitz, Paul. cations for Banking Supervision and Deposit Insurance.” Journal of Financial and Quantitative Analysis, Vol. X No. 4 (November 1975), 589-601. 5. Mayer, Thomas. “Should Large Banks Be Allowed to Fail?” Journal of Financial and Quantitative Analysis, Vol. X No. 4 (November 1975), 603-10. 6. Report of the President’s Commission on Financial Structure and Regulation, Washington, D. C.: Government Printing Office, 1971. “Risk and 7. Scott, Kenneth and Thomas Mayer. Regulation in Banking: Some Proposals for Federal Deposit Insurance Reform.” Stanford Law Review, 23 (May 1971), 857-902. 12 ECONOMIC REVIEW, 8. U. S. Congress. House. Committee on Banking and Currency. Recent Bank Closings. Hearings before the Committee on Banking and Currency, 92d Cong., 1st sess., 1971. Senate. Committee on Banking 9. U. S. Congress. and Currency. S. Rept. 1269 To Accompany S. 2822, Slst Cong., 2d sess., 1950. 10. U. S. Congress. Senate. Committee on Banking and Curreicy. Nominations. Hearings before i subcommittee of the Committee on Banking and Currency, 82d Cong., 1st sess., 1951. 11. U. S. Congress. Senate. Committee on Government Operations. Investigations Znto Federallly Hearings before the Permanent Insured Banks. Subcommittee on Investigations of the Committee on Government Operations, 89th Cong., 1st sess., 1965. 12. Informal talk before the Arizona Wille, Frank. Bankers Convention, Phoenix, Arizona, October 11, 1974. 13. “The FDIC and Franklin National Wille, Frank. Bank : a Report to the Congress and All FDIC Insured Banks.” Delivered at the 81st annual convention of the Savings Banks Association of New York State. Reprinted in the American Banker, December 11, 1974. SEPTEMBER/OCTOBER 1976 ER DEFICITS, INFLATIO D MONETARY GRO CAN THEY PREDICT INTEREST RATES? William D. Jackson cial markets-to be determined mand and supply curves.3 This article traces the short-run impact of fiscal policy, inflation, monetary growth, and economic activity on interest rates. Its theoretical framework is a loanable funds theory of interest rate determination, which incorporates both neoclassical and Keynesian elements. This framework is useful for analyzing the crowding out effect, real versus nominal interest rates, the relative importance of Ml, X2, and M3, and the inflationsavings relationship in a financial markets setting. The implications of this theory are tested against interest rate movements during recent years. The resulting equations may be useful to investors in predicting the impact of fundamental economic changes on interest rates, an impact that may not be evident in term structure yield curves.’ Investment demand also responds to changes in output. If output rises, firms find it profitable to invest in plant, equipment, and inventories. As output rises, the demand for residential housing eventually increases. The investment schedule in Figure ! would thus shift to the right when LYield curves that relate short rates to long rates can shift dramatically over time. For comparisons of the predictive ability of economic and term structure interest rate mode%. see Michael E. Echols and Jan W. Elliott, “Rational Expectations in a Disequilibrium Node1 of the Term Structure,” Am&can Econonic Review (March 19X), pp. 28-44: Martin Feldstein and Gary Chamberlain, “Multimarket Expectations and the Rate of Interest,” downal of Monez/, Credit a%d Banking (November 1973). pp. S73-902: and Lacy H. Hunt, “Alternative Econometric Models for the Yield on Sgug?gn Corporate Bonds,” Butiness Economics (September 1973), . -. (Homewood: Prices (New FEDERAL 2 Several versions of loanable funds theories are described in Joseph W. Canard. An Inzrodaction to the T~GO~J of lntcrcst (Berkeley: Universiv of CalZornia Press, 1959); Frederich A. Lutz, The Theow oi Interest (Chicago: Aldine. 1969): and S. C. Tsiane. “Liquid%- Preference and Loanable Funds Theories, Multiplier and Velocity Analyses: 2 Synthesis,” Anzerican Economic Review fSeptember 1956). pp. 539-64. Less technical treatments appear in: John A. Cochran, Xoxw, Banking and the Economy (New York: Macmillan, 1967): Charles X. Henning et. al. Financial Markets and the Econowzy (Enzlewood Cliffs: Prentice-Hall, 19’76); SIurras E. P&&off et. (II’. Fincmcial Institutions and Marketa (Boston: Mifflin, 1970): and John G. Ranlett, Monet and Banking (New York: Wiley. 1969). Irwin, York: RESERVE by de- The RCSUZX~‘ jor Loanable Funds In the traditional theory, the demand for loanable funds was for the purpose of financing investment in real sector assets, such as commercial and residential construction, inventories, and plant and equipment. The demand for such investment depends upon the cost of capital, for which interest rates of investment serve as a proxy. The productivity -its rate of return-is determined by income, technolo,gy, and the existing stock of capital. The lower the cost of capital, the higher the net return from investment: its productivity less its interest cost. The same sort of relationship applies to household investment in residential housing, which is largely financed by mortgage borrotving. The investment schedule, the I line in Figure 1, shows that more investment is planned at lower interest rates. Loanable Funds Theory Current loanable funds theory builds on the foundation of an eighteenth century doctrine that was concerned with savings and investment in a barter economy with no governmental sector.” The modern inclusion of government finance, money, and inflation in the analysis allows “the .interest rate”-a composite of the spectrum of interest rates in related finan- ?&lark Blaw. Economic Tkeow in Retrospect 1968); Don Patinkin, Monczr, Interest, and Harper, 1965 ) . directly BANK OF RICHMOND 13 essentially on current irivestment demand, despite borrowing account by certain governments.4 The demand for loanable funds (LFD) thus consists of the sum of FD and I, as shown in Figure 1. (Consumer credit other than mortgages is treated as a deduction from savings.) The Supply of Loanable Funds The supply of loanable funds is a rather complex sum of savings by individuals and businesses, changes in the flow of credit extended by financial institutions, and variations in the public’s desire to hold money. real output rises, through relationship. the so-called accelerator The modern loanable funds theory recognizes that government deficit financing also creates a demand for loanable funds. Ma&ive government spending in recent years could not have been funded entirely by taxes without creating social unrest and reducing real output. Governments borrowed in private credit markets to fill the resulting gap. Federal Government demand for funds is insensitive to changes in interest rates. This interest-inelastic demand for funds is shown as the line FD in Figure 1. In a large-scale model of the economy, the FD demand for funds could determined by income be endogenous, i.e., through income taxes and by politically determined Government spending. In practice, Federal planners specify a given deficit as a measure of fiscal stimulus, making the deficit a largely predetermined (exogenous) policy tool. In contrast, funds raised by state and local governments largely represent capital expenditures on education, highways, housing, and public utility projects. These long-term projects resemble business capital expenditures in their sensitivity to interest rates. For example, state and local interest rate laws may prohibit new debt issues by these governments at rates exceeding specified ceilings. Their demand for funds is 14 ECONOMIC REVIEW, Savings by individuals respond positively to the reward for thrift at a given level of income:. The higher the interest rate, the greater the amount of future consumption that can be obtained by refraining from present consumption. Hence, the savings schedule S slopes upward in Figure 2. The supply of savings schedule also responds to changes in income, shifting to the right as higher income allows consumers and businesses to save more. This income effect may be more important that the interest effect on savings. The traditional theory of the supply of loan.able funds incorporates changes in the flow of bank credit, which result from changes in the supply of money. Newly created reserves (highpowered money) flow through the banking system when the central bank engages in open market purchases of Government securities. This causes banks to possess more nonearning reserves than they wish to retain and to use this liquidity to purchase financial claims until their cash is again in balance with their other desired portfolio holdings. The resulting increase in the supply of loanable funds is represented by the horizontal. distance Am in Figure 2. Commercial banks tend to increase their credit output derived from the new reserves more when interest rates are high than they do when rates are low. Banks decrease their excess reserves This when the reward for lending increases.5 ‘State and local governments as a grc.up generated a surplus of $51.7 billion from 1969 through 1975. mainly through their pension funds. Over half the new municipal security issues from 1964 through 1974 funded the four types of capital expenditures cited. (All statistics in this article are taken from Federal Reserve sources such as Flow of Funds accounts and Federal Reserve Bulletins.) 5 When earning asset returns are high enough, banks not only practice this form of asset management but also increase the size of their portfolios by borrowing nondeposit funds: certificates of deposit, discounts and advances from the Federal Reserve, etc. Funds borrowed at the discount window increase the money supply, as well as bank credit. SEPTEMBER/OCTOBER 1976 behavior increases Am) curve relative interest elasticity to the S curve. of the (S + ing markedly portant form when of the these deposits public’s wealth are an imholding.? The increased supply of loanable funds Am may be derived in practice from changes in Ml, M2, or M3. Ml, the sum of currency and demand The supply of loanable funds also varies with the public’s demand for money. For example, financial innovations such as credit cards lower deposits, is directly responsive to changes in monetary policy. M2, defined as Ml plus noncertificate time and savings deposits at banks, and M3, defined as M2 plus similar deposits at nonbank financial institutions, are more inclusive measures of liquidity in the economy.6 the public’s demand for cash and demand deposits. The supply of loanable funds increases when the public desires to exchange Ml balances for financial claims. Such an exchange of Ml for financial claims is known as dishoarding and results in a higher ratio of income to money, or higher velocity, for the economy. This increase in the supply of credit is represented by an increase in the horizontal distance between S and the total supply of loanable funds in Figure 2the distance DH. (Below some low interest rate level, such as R, in Figure 2, the public will prefer additional liquidity rather than the inconvenience of low-yielding financial claims and will hoard.) * The sum of savings S, changes in These monetary aggregates are important determinants of the supply of credit funds to mortgage and other longer-term borrowers by financial institutions. Increased savings and/or shifts from the public’s desired Ml balances into insured earning assets result in increases in consumer time and savings deposits (part of M2 or M3), which are quickly supplied to credit markets by financial institutions after provision for (rather low) required reserves. credit flows Am, and net dishoarding DH defines the total supply of loanable funds-the LFS curve in Figure 2. The increased supply of loanable funds Am is a multiple of any increase in reserves through the well-known credit multiplier. The size of this multiplier is sensitive to changes in the public’s desire to hold time and savings deposits, increas- Interest Rate Determination As in other markets, the price of loanable funds is determined by the intersection of supply and demand. With income held constant, the market for loanable funds may be represented in Figure 3 by LFD, the demand ; LFS, the supply ; and Rf, the equilibrium price or interest rate. The quantity of loanable funds offered and accepted is Qf. e See Alfred Broaddus, “Aggregating the Monetary Aggregates: Concepts and Issues,” Economic Review. Federal Reserve Bank of Richmond, (November/December 1975), pp. 3-12. Y Changes in credit can be several times the amount of the change in high-powered money. One version of the potential credit expansion multiplier is defined “if the public holds demand deposits, currency. and [time and savings deposits] in the proportions 1:c:t . . . the combined acquisition of credit instruments by banks and intermediaries” would be: 1+c+t rd + c + (rt + rdrs)t X where X is excess reserves available to support credit expansion, rd is the reserve requirement for demand deposits. rt is the reserve requirement for time and savings deposits held at the central bank. and r is the subjective “reserve requirement” for intermediary DH deposiis held in demand deposits of commercial banks. The larger the proportion of time and savings deposits, particularly those of nonbank intermediaries, that the public desires to hold, the larger the potential multiplier. Warren L. Smith, “Financial Intermediaries and Monetary Controls,” Qwwterly Journal of Economics (November 1959). pp. 533-53. I > ’ Loonable FEDERAL BThe treatment of net dishoarding as an addition to the supply of loanable funds is based on the increase in the velocity of Ml shown later. Dennis H. Robertson, “Mr. Keynes and the Rate of Interest,” in Readings in the Theory of Income Distribution, ed. by William Fellner and Bernard F. Haley (Philadelphia: Blakiston, 1946). High velocity, one consequence of high interest rates, dampens them in the next time period. See John Kraft and Arthur Kraft, “Income Velocity and Interest Rates.” Journal of Money, Credit and Banking (February 1976). pp. 123-5. Funds RESERVE BANK OF RICHMOND 15 I The Crowding framework Out Effect The loanable is well suited to the analysis funds of crowd- ing out. This concept refers to the displacement of private borrowings by Federal deficit financing. Repeating the previous schedules in Figure 4, at the rate Rf Federal deficit financing at the level FD and private investment financing at the Suppose that the Federal deficit level I1 occur. increases to FD’. The demand for loanable funds shifts rightward to LFD’ by the increase in the deficit. If the supply of loanable funds schedule remains constant, the interest rate increases from Rt to Rz. The Federal sector borrows FD’ despite creased to 02; the larger deficit then displaced (92 - Q) -(FD’ - FD) = (I1 - 1~) of private sector funds. In any case, the rise in interest rates is one indicator of the resulting pressures on private capital expenditures. If the deficit is successful in raising income during a depression, investment spending may not be excessively depressed. But when income rises, this rightward shift in LFD reinforces the Investment will be damprise in interest rates. ened over time. An additional effect of deficit financing on the state of investor confidence that influences the the higher interest rate structure. But the higher rate Rz depresses business investment. If income position of the I curve has been hypothesized. For example, in one Keynesian model, and the state of investor confidence remain unchanged, investment capital funds decline from 11 to 12. under conditions of a budget deficit there exists ;A’ inverse relationship between investment and The fall in investment will not usually equal the rise in Federal borrowings. The extent of the crowding out depends on the elasticity and position of the I curve. If investment is highly interest elastic, capital espenditures will decline markedly. If investment is fairly insensitive to interest rates, most planned capital expenditures will continue to be made. In the example of Figure 4, private capital funds declined by less than the increase in deficit financing. At the higher rate Rz the total supply of loanable funds in- [the change in Government bonds]. , . . [the] appearance of public hostility and fear of deficit spending (adverse expectations) can, in theory, profoundly interfere with the stimulative eaiy;. of the fiscal action causing the deficit. extreme, a perverse result, i.e., a negative spendings multiplier . . . might even be obtained.9 Inflation While the above analysis assumed a noninfIationary economy, the loanable funds framework is well suited to the analysis of inflation and financial markets. Inflation erodes th.e purchasing power of Ioanable funds. ?Vhen th:is loss of purchasing power is subtracted from th.e nominal rate, the real rate of interest is obtained. This real rate equals the nominal rate only when prices remain constant. If, for esample, the interest rate is 7% w-hen the price level is rising steadily at 4%, the real rate is 3%. Most loanable funds theorists, following Irving Fisher, assume that borrowers and lenders reac:t Borrowsymmetrically to anticipated inflation. ers recognize that they will repay their debts i,n The productivity of investmen.t cheaper dollars. in nominal terms rises by exactly the anticipate’d rate of inflation. Similarly, lenders recognize that they will receive debt repayments in less valuable dollars. Their real reward for saving declines by the anticipated rate of inflation. Under these assumptions, the demand for funds would shift upward to the right by the expected rate of inflation, while the supply of funds would shift upward to the left by the same amount. The nominal rate of interest would rise by exactly the Neither the real amount of expected inflation. rate nor with the quantity inflation. This of credit flows hypothetical would var:? situation 2 Richard J. Cebula, “Deficit Spending, Expectations, and Fiscd Policy Effectiveness,” Public Finance (19X3), pp. 365-6. 16 ECONOMIC REVIEW, SEPTEMBER/OCTOBER 1976 i.s repayment will be in cheaper dollars, but also because the productivity of new capital rises.l’ The total demand for loanable funds may not increase by the full extent of the anticipated inflation, however. Some users of capital find that their borrowing capacity cannot keep pace with the total cost of capital investment. These users, such as price-regulated utilities, many potential home buyers, and some state and local governments, may find that they are priced out of the They are very sensitive to the capital market. nominal rate of interest, as well as to the nonMoreover, interest cost of capital investment. Federal deficit financing should not be stimulated by inflation in the short run. LFD thus shifts upward by an amount less than the inflation. In Figure 5, the demand for loanable funds will shift to a position such as LFD” if a rate of inflation 7~is anticipated based on actual inflation. Borrowers as a group would pay Ri to obtain the pre-inflation quantity of funds Qf. Inflation also affects the supply of loanable funds, but not in the manner prescribed by Fisherian loanable funds theory. As discussed earlier, illustrated in Figure 3.*O The LFD” and LFS” curves fully embody the rate of inflation rr (R, Th e quantity of loanable funds flowminus R,). ing through credit markets remains Qi. u The demand for external finance will increase even when persistent inflation lowers the return on existing capital investment. John Lintner, “Inflation and Common Stock Prices in a Cyclical Context.” in Anmud Report. (New York: National Bureau of Economic Research, 1973). pp. 23-36: and Lintner. “Inflation and Security Returns,” Jownd of Finance (May 1975). pp. 25940. The true relation between inflation, the nominal rate, and the real rate is, however, more complex than in the above scenario. Both nominal and real rates are affected by asymmetrical inflation-induced shifts in LFD and LFS. Inflation stimulates LFD, as is well known. It enhances the nominal dollar returns available from current investment. Future output can be sold at higher dollar prices. Moreover, physical investments made today should be less costly than those made in the future, when their prices are expected to be higher. The probability of capital gains from selling capital assets then rises. Inflation also raises expected wages. Employees demand protection of their standard of living through higher nominal wages. Minimum wage levels are raised in response to the inflation, reinforcing the rise in labor costs by setting everhigher floors underneath wages. Employers then attempt to substitute capital for labor. The investment demand curve increases under inflationary conditions, not only because expected debt lo Donald J. Mullineaux, “Inflation Insurance: An Escalator Clause for Securities,” Business Review, Federal Reserve Bank of Philadelphia, (October 19’72). pp. 11-12. FEDERAL RESERVE BANK OF RICHMOND 17 that theory would have LFS shift upward to the left in response to inflation. As will be shown however, the supply of Ioanable funds actually shifts to the right in response to inflation. While this reaction may not occur in a hyperinflationary it has occurred in recent American economy, esperience. Clearly, inflation reduces the expected future value of present cash holdings. W~ealth holders attempt to reduce their Ml balances when inflation “taxes” the value of their money holdings. This dishoarding increases the supply of funds available to purchase interest-bearing financial assets that are partially protected against inflation by nominal interest payments.12 LFS shifts to the right by the distance DH” in Figure 5. The partial supply of loanable funds curve (LFS + DH”) increases more rapidly as higher inflation is expected to deplete the value of Ml. Moreover, of expected inflation increases future real income the uncertainty Most streams. people feel that a high rate of actual inflation, particularly if it exceeds a “normal” rate of inflation, indicates that their future expenses will increase more rapidly than their future incomes. This feeling is particularly rational when (1) is imported from abroad cost-push inflation through cartelized commodities or devaluations and (2) inflation shifts individuals into higher income tax brackets and raises other taxes. Most individuals feel that they cannot raise their income to match these uncontrollable increases in Furthermore, the probability the cost of living. of complete income compensation for inflation decreases as the rate of inflation increases. Even ;j the prospect of higher real income appears as likely as the prospect of lower real income during inflations, the resulting increased variance of expectations of real earnings decreases the confidence with which most people view the future. To hedge against. this uncertain future, lessconfident consumers increase their rate of current saving.13 Contrarv d to the conventional wisdom, r”Dean S. Dutton, “The Demand for Money and the Expected Rate of Xmwy, Credit and Banking (Noof Price Change,” Jounal vember 19X), pp. 861-V: Robert A. Mundell, “A Fallacy in the Interpretation of Macroeconomic Equilibrium,” .Jounal of Politicd Economy (February 19651, pp. 61-6: Mundell, “Inflation and Real Interest,” Journal oj Political Econom?~ (June 1963), pp. ‘230-3: Lester D. Taylor, “Price Expectations and Households’ Demand for Financial Assets.” Ezpbmtions in Economic Research (Fall 1974). pp. 268399. I3 F. Thomas Juster and Paul Wachtel, “Inflation and the ConBrookinos Paper.s on Ecmwmic Activity (No. 1. 1972). pp. 71-121; Hayne E. Leland, “Saving and Uncertainty: The PreJovmal oj Economics cautionary Demand for Saving.” Quarterly (August 1968). pp. 466-73: Agnar Sandmo, “The Effect of Uncertainty on Saving Decisions.” Review of Economic Studies (July 19’iO). pp. 353-60. sumer.” 18 ECONOMIC REVIEW, consumers then save by reducing their spending on purchases of durable goods-automobiles and household furniture and fixtures.lJ If the inflation is unanticipated, consumers may even reduce their expenditures on nondurable goods and services to increase their savings. In addition, the desire of most individuals to protect the capitalized value of their earning asset holdings stimulates saving behavior when interest rates rise during inflationary periods. The real value of portfolio earning assets declines in inflationary periods, not only because the earnings expected from capital are received in depreciated dollars, but also because the rate of discounting of this earnings stream-the “pure” rate of interest plus a premium for assuminlg financial risk-also rises.l” This wealth effect, which dampens consumption and stimulates saving, is not balanced out by net debtors feeling wealthier in real terms during an inflation. Mos#t debt is owed by businesses and governments, whose real wealth position does not directly enter into most individuals’ evaluation of their personal portfolio positions. Finally, inflation does not directly diminis‘h the very large supply of funds that institutiona. The purinvestors provide to credit markets. chasing power of money is not an important factor in the investment decisions of bank and nonbank institutions whose liabilities are mea.sured in dollars. They seek the highest “prudent” nominal rate of return from their financial assets once the size of their portfolios is determined.16 A large this form body of empirical of saving behavior evidence confirms in the -4merican I4 The large expenditures on consumer durable goods in 1972-73 stemmed partly from the artificial restraint on their prices dictated by price controls. These prices were expected to rise rapidly when controls would be removed. ‘j Financial wealth can be defined as: WA+?+? r P where W is wealth, M is the quantity of money, B is the quantity of bonds expressed in terms eq;livalent to perpetual bonds with a 31 coupon, r is the current market interest rate, E is the expected earninns stream from real capital, and P is the market-determined rate of discount for profits. Deflating all terms by the price level defines “real” financial wealth. Joseph R. Bisiqnano, “The Effect of Inflation on Savings Behavior,” Economic Review, Federal Reserve Bank of San Francisco, (December 1976), p. 21. It can be shown that when inflation raises the nominal rate of discount r for riskless bonds, it increases the nominal rate of discount P for risky financial investments to an even greater extent. The prices of equities fall with the resulting increase in perceived financial risk, as well as with the increase in required return du,@ to higher interest rates. lsLintner, Thomas Piper, and Peter Fortune, “Investment Policies of Major Financial Institutions Under Inflationary Conditions,” in National Bureau of Economic Research, op. tit., p. 98. SEPTEMBER/OCTOBER 1976 economy during recent inflations.17 Inflation shifts the savings schedule (given income) by a distance S” in Figure 5 ; inflation does not decrease it. The suppIy of IoanabIe funds schedule increases from LFS to LFS” (LFS + DH” + S”) in an inflationary climate typical of recent experience. positively with income Y and anticipated inflation rr.ig The Federal deficit FD is assumed to be exogenous. The supply of loanable funds is: Under these conditions the demand for funds exceeds the supply of funds at the no-inflation interest rate Ri. With this excess demand for credit, the nominal rate of interest rises to R,. where savings vary positively with the real rate, income, and anticipated inflation. Changes in credit Am based on changes in money are treated as exogenous in the short run. The inclusion of (2) the dishoarding I + where the investment inversely with interest FD = I(r, Y, rr) + = S(r,Y,n) term is discussed interest later (p. 21). rate, subtract (1) and collect terms. shows the determi- A number of previous studies of the determinants of rates were reviewed before completely specifying the equations to test the loanable funds theory. 2o The results of these studies are generally consistent with the loanable funds framework, but they contain enough contradictory findings to warrant a new investigation. FD 1s Smithy ‘Monetary Theories of the Rate of Interest: A Dynamic Synthesis,” Review of Economics and Statistics (February 1958). pp. 15-21; Tsiang, Zoc. cit. 20Leonall C. Andersen and Keith M. Carlson, “An Econometric Analysis of the Relation of Monetary Variables to the Behavior of of Price DeterPrices and Unemployment,” in The Econometrics mination, ed. by Otto Eckstein (Washington: Board of Governors of the Federal Reserve System, 1972). PP. 166-83; J. A. Cacy. “Budget Review, Federal Reserve Deficits and the Money Supply,” Monthly Bank of Kansas City, (June 19751, PP. 3-9; G. Marc Choate and Stephen II. Archer, “Irving Fisher, Inflation. and the Nominal Rate of Interest,” Journal of Financicrl and Quantitative Analysis (November 1975). pp. 675-85; Donald M. DePamphilis. “Long-term Interest Rates and the Anticipated Rate of Inflation,” Business Economics (May 1975), pp. 11-18; Echols and Elliott, ZOO. cit.; Fddstein and Chamberlain, ZOC. cit.: Feldstein and Eckstein. “The Fundamental Determinants of the Interest Rate,” Review of Economics and Statistics (November 1970). pp. 363-75: William E. EcoGibson, “Interest Rates and Monetary Policy“ in Monetary no&es, ed. by Gibson and George G. Kaufman (New York: McGrawHi& 1971). pp. 311-29; Gibson. “Price-Expectations Effects on Interest Rates,” in Gibson and Kaufman, Ibid., pp. 339-51; Gibson and Kaufman. “The Sensitivity of Interest Rates to Changes in Money and Income,” Journal of Political Economy (June 1968), pp. 472-8; Stephen M. GoIdfeId, Commercial Bank Behavior and Economic Activity (Amsterdam: North-Holland, 1966) : Michael 3. Hamburger and William L. Silber, “An Empirical Study of Interest Rate Determination,” Review of Economics and Statistics (August 1969), PP. 369-81; Hunt, Zoc. cit.; Thomas J. Sargent, “Commodity Price Expectations and the Interest Rate.” in Gibson and Kaufman, op. cit.. pp. 330-S; Robert H. Scott, “Liquidity and the Term Structure of Interest Rates.” Quarterly Journal of Economics (February 1965). pp. 135-45: Silber. Portfolio Behavio+ of Financial Institutions (New York: Holt, 1972): and Yohe and Karnosky. Zoc. tit. Ii The saving rate is significantly related to measured uncertainty in the economy. For example. from 1962 I through 1975 II, personal savings/disposable personal income was correlated -0.68 with the Survey Research Center Index of Consumer Sentiment. This Index was correlated -0.79 with the rate of inflation. Correspondingly, the personal saving rate was correlated 0.54 with the annualiaed rate of change in the Consumer Price Index over this period. More extensive confirmation of these relationships is provided by: Susan W. Burch and Diane Werneke, “The Stock of Consumer Durables, Inflation and Personal Saving Decisions,” Review of Economics and Statistics (May 1975), PP. 141-54: Saul II. Humans. “Consumer Durable Spending: Explanation and Prediction,” Bsookings Papers on Economic Activity (No. 2! 1970). pp. 173-99; Juster and Taylor, “Towards a Theory of Savings Behavior,” American fk’n;? Revrew (May 1975). PP. .203-24: Juster and Wachtel, ’ . George Katona. Psychologzcal Economics (New York: Elseviery 1975) : William Poole, “The Role of Interest Rates and Inflation in the Consumption Function,” Bmokings Papers on Economic Activity (No. 1, 1972), PP. 211-20; Burkhard Strumpel et. al.. teds.), Human Behavior in Economic Affairs (San Francisco : Jossey-Bass, 1972 ) ; Taylor, “Price Expectations:” and Taylor, “Saving Out of Different Types of Income,” Brookings Papers on Economic Activity (No. 2, 1971). pp. 383-415. 1sWilliam P. Yohe and Dennis S. Karnosky, “Interest Rates and Price Level Changes, 1952-69.” Review, Federal Reserve Bank of St. Louis, (December 1969), pp. 18-38; A. John Steigmann, “On Inflation and Interest Rates,” Business Economics (May 1975). pp. 72-3. RESERVE DH Nominal and real rates increase when the Federal Government runs a deficit and when the Nominal and real interest money supply falls. rates rise when real output increases if the incomeinduced investment exceeds the income-induced Nominal interest rates rise during inflasaving. tionary periods if investment demand rises more than the supply of savings plus dishoarded money. Finally, the theory developed above postulates that real rates fall during inflations. demand for funds varies rate r-a real rate-and FEDERAL Am + equation (2) from equation The resulting relationship nants of interest rates. Loanable Funds Theory and Predicting Interest Rates The loanable funds theory can be stated in equation form. The demand for loanable funds is : = S + To solve for the nominal Inflation stimulates financial flows : loanable funds flowing through financial markets rise from Qf to Qg in Figure 5. The greater flows of funds are associated with an incomplete adjustment of the nominal interest rate to inflation. The dishoarding and saving adjustments to inflation, increases in the supply of credit by financial institutions, and the inability of some borrowers to adapt to inflation prevent the full adjustment of LFD and LFS to experienced infIation in a period less than the very long run. Only then could all desired income and portfolio adjustments to presumably fully anticipated inflation be made. LFD = +=+DH But Rs is less than Rf plus the inflation rate V; the real rate of interest clearly declines. This lower real rate increases desired investment along LFD”.18 (1) LFS BANK OF RICHMOND 19 Interest Rate Equations The empirical findings of previous studies and the loanable funds theory Which Monetary Aggregate Influences Interest Rates? There has been much discussion in recent outlined nominal years concerning the proper definition of money. Of the various aggregates suggested, the riskless and highly liquid Ml, M2, or 313 seems appropri- (3) above suggest equations for estimating interest rates of the form: RATEt I, = The time + 11 2 ciFDt-i i=O where the following pated : COY cos > + + bYt-1 n $ di6t-i i=O coefficient 0, a<O, subscript a& b>O, t refers values Sci>O, are antici- Sdi>O. to monthly observa- tions. RATE is the nominal rate. The constant term COS captures the effects of any influences that are not explicitly considered, such as a tendent>- for rates to assume some “normal” level. gron-th of money ,ir is The annualized rate of the foundation upon which resulting larger credit changes Am are based. The lagged unemployment rate serves as an inverse proxy for the level of real output Y. This closely watched coincident indicator reflects excess demand in the labor and It reflects the difference beproduct markets. tween actual output and capacity output.“l It is also associated with the state of investor confidence in the economy.” Moreover, since it is not defined in monetary units, it should not be subject to inflationary distortions of measurement. Unlike personal income, which includes transfer payments and which tends to increase despite industrial fluctuations, the unemployment rate should reflect variations in real GYP, which is not available on a monthly basis. The esogenous Federal deficit FD should affect the economy the annualized rate of I\-ith a I;L~. Similarly. markets price change I? should affect financial These lags arc based on over a long period. investor reactions to trends in these volatile series, reflecting delayed incorporation of information into expectations. The necessity of incorporating a dishoarding term into equation (3) requires a slight digression on the definition of money. ate in the loanable funds model. Broader aggregates incorporate credit instruments themselves, which are subject to risk of default if less than AAA quality and which are subject to capital loss of varying extent if interest rates rise. These securities are generally either illiquid (U. S. savings bonds) or beyond the reach of most ind.ividuals (commercial paper, Treasury bills). Any of these three behaviorally appropriate aggregates could be used as the money variable in this model. The question is, which one of these measures influences interest rates most strongly. One answer to this question emerges from the relationship between changes in these aggregates and credit flows. X’ew Ml, flowing through the banking system, was 8.2 percent of total funds advanced in credit markets from January 1967 through December 1975. The more rapidly growing new M2 was 23.6 percent of these funds. And explosively growing new 313, flowing through nonbank depository institutions as well as through banks, accounted for 40.7 percent of the credit market funds advanced in this period. This evidence suggests that growth in M3 is more closely related to the change in the supply of credit than growth in Ml or M2.‘” A second answer emerges from the velocity of these monetary aggregates. Dishoarding of Ml has occurred in recent years. The income velocity of Ml increased secularly from 4.3 in the fourth quarter of 1965 to 5.3 in the fourth quarter of 1975. The income velocity of 312, however, re:mained remarkably constant during this period. It was 2.4 in the fourth quarter of 1966 and 2.4 in the fourth quarter of 1975. The income velocity of X3 \-aried slightly around its beginning and ending value of 1.5 during this period. 27Throufih Okun’s Law. “the unemployment rate can be viewed as a pro.=- variable for all the ways in which output is affected by idle resources.” Arthur M. Okun, “Potential GSP: Its Measurement and Significance,” Procacdin.os of the Business and Economic Stetistics Section. American Statistical Association (1962). p. 99. Andersen and Carlson. Zoc. cit.; Gary Smith, “Okun’s Law Revisited,” Qwzrterly Revieau oi Economics and Business (Winter 19X), pp. 37-S. Inflation, institutional factors such as changes in the payments mechanism, and increasing activity by nonbank financial institutions have evidently lessened the traditional role of Ml. This shift away from desired holdings of Ml, particularly from currency, into interest-bearing dep0sit.s stimulates the supply of loanable funcls through reduced reserve ratios and the correspondingly higher potential loan/deposit ratios. Many sav- ZIt is highly related to the Index of Consumer Sentiment, for example. See the references in footnote 17, and Dwight M. Jaffee, “Cyclical Variations in the Risk Structure of Interest Rates,” Journal of Monetar?J Economics (1975), pp. 309-25. ‘3The calculations in this and the following paraaraph are based on Flow oi Funds data. See footnote 4 for references. 20 ECONOMIC REVIEW, SEPTEMBER/OCTOBER 1976 ings and loan associations have loan/deposit greater than unity, for example.24 ratios The considerations that money should behave as a medium of exchange for goods and services with a fairly constant velocity and that it should serve as a store of real purchasing power (at least partly protected against inflation by interest payments), suggest that the growth of M2 and M3 may serve as better indicators of liquidity than the growth of Ml. Essentially zero dishoarding of M2 and M3, as indicated by their stable velocity in recent years, correspondingly suggests that the DH term is not required in empirical interest rate equations. Methodology The extent to which the basic economic influences of income, inflation, deficit spending, and changing credit flows influence interest rates may vary with the quality and term to maturity of various securities. To what extent do the short- and long-term, new issue or seasoned, taxable and tax-free, and risky and riskless characteristics of securities alter the response of their interest rates to fundamental economic factors? To study these questions, equations of the form (3) were estimated for the following rates: the 3-month new issue Treasury bill rate, Moody’s 3-5 year U. S. Government securities rate, Moody’s Industrial A seasoned long-term bond rate, Moody’s new issue Municipal A rate, and the long-term -Government bond rate reported by the Federal Reserve. The equations are estimated on a monthly basis from December 1966 through December 1975. Since the analysis is concerned with short-run interest rate responses to economic factors, the maximum time lag is limited to twelve months. Economic Interactions: The Fed’s Dilemma Interactions among fiscal policy, inflation, money, and unemployment over longer periods reduce the ability of single-equation models to identify causality. In particular, financing the Federal deficit involves the indirect purchase (“monetization”) of part of the resulting Federal debt by the central bank. This causes the money stock to rise. The resulting excess supply of money may create later excess demand in the commodity market, as well as current excess demand in the credit market, and Iead to subsequent inflation. The monex See footnote ‘7 and the other loanable funds credit multipliers shown in Smith, “Financial Intermediaries.” A shift from currency into nonbank deposits could increase loanable funds bu almost four times the amount of the shift in Smith’s analysis. tary authority thus faces a cruel dilemma when extensive deficit financing occurs. Should the money decline supply increase enough to cushion the in investment in the current period, it may generate inflation later. If monetary growth is large enough to hold down current nominal interest rates despite the deficit financing, it may raise inflationary expectations and interest rates in the future. If money does not increase enough to allow most planned investment to be made, future productive capacity will be markedly lower than it would have been without the deficit. This condition of lower-than-otherwise output may result in shortages and future inflation. Interest rates may then rise to high levels unless the demand for goods and services falls. Interest Rate Equations The estimated relationships of interest rates to Federal deficits, inflation, monetary growth, and unemployment are reported in Appendix Tables I and II. Appendix A discusses their technical aspects in detail. For the general reader, the empirical results may be summarized briefly. While the equations estimate nominal rates, realized real rates may be implied from the Iabb =ed coefficients on the inflation rate. If yearly inflation terms are less than unity, ex post real rates tend to decline. Chart 1 illustrates interest rate equations in the sample period. the effectiveness of the in matching actual events In the chart actual rates appear as solid lines, and rates predicted ex post appear as dashed lines. These equations explain 92 to 99 percent of the variation in interest rates over the period. (The predicted rates tend to lag very slightly behind actual rates, as would be expected from their use of lagged predictors.) The predicted rates exhibit no secular tendency to over or underpredict actual rates. In general, Federal deficit spending increases interest rates with a four- to six-month lag. These deficits generally continue to drive up both Federal and private borrowing costs throughout the remainder of a twelve-month period. The resulting interest rate pressure is larger, more significant, and more prolonged for the Industrial A and Municipal A rates than it is for the similar maturity long-term Government bond rate. Risk-averse investors in the long-term bond market evidently require a larger “risk premium” on medium-grade private securities when deficit spending reduces their state of confidence. This rise in interest rates restricts the effectiveness of FEDERAL RESERVE BANK OF RICHMOND 21 the deficit in raising income.“” This evidence ports the view that crowding out, measured directly through interest rates, has occurred some extent in our economy in recent years. su’pinto Inflation stimulates nominal rates very significantly, with both current period and lagged effects. The Treasury bill rate, for examplet reacts strongly to inflation : approximately half of the impact of a sustained rate of inflation pears in this rate over a ten-month period. cent inflation encourages inventory building, apRere- sulting in heavy demand for bank loans and commercial paper. This puts upward pressure on all short-term rates, including the Treasury bill Longer-term rates, however, adapt less rate. strongly to inflation. The 3-5 year Treasury note, Industrial A, and Municipal A rates embody less than one-quarter of realized inflation rates within When inflation occurs, the Industrial A a year. rate reacts very rapidly, while the I-7. S. 3-5 year security rate reacts more slowly, and the Municipal A rate generally takes still longer to respond. The long-term Government rate incorporates only about one-eighth of the actual inflation rate during a twelve-month period. These findings are consistent with the infiation-induced shifts in the supply and demand curves of the loanable funds theory above. Real rates fall when the price level increases rapidly, although to a different extent for each rate. Th.e length of the period of past inflation that realasset investors use to anticipate inflation over the period of their borrowing should be positively related to the length of the borrowing contract. Increasing the rate of monetary growth lowers interest rates. But the effects of varying growth rates of money are erratic or insignificant in equations that examine them for lagged time periods.“0 Growth in AI.3 lowers rates more than In turn, growth in Ml lowers growth in M2. rates to a still lesser extent, sometimes not significantly. Monetary growth is more imp0rtan.t for shorter rates than for longer ones. Appendix B examines these liquidity effects in more detail. Realized income has the influence on interes,t rates that theory suggests. High unemploymenr, typifying weak private sector excess demand (investment minus savings) for credit, lowers ail 2s Carlson and Roger W. Spencer, “Crowding Out and Its Critics.” Review, Federal Reserve Bank of St. Louis, (December 19%). PP. 2-11: Spencer and Yohe, “ The Crowding Out of Private Expenditures by Fiscal Policy Actions,” Rsthto, Federal Reserve Bank of St. Louis. (October 1970), pp. 12-24. 2s Similar results appear in Gibson, “Interest Rates and Xonetar~ Policy,” lot. cit. 22 ECONOMIC REVIEW, SEPTEM5ER/OCTOBER 1976 five interest rates. *’ affect shorter-period period Current rates exemption business conditions more than longer- determinant ones. Finally, for municipal of this difference The longer-term the constant terms incorporate the atory effects of other factors that are not explicitly considered. For example, the constant in the Municipal A rate equation is more than 100 basis points below the constant in the similar quality Industrial A rate equation. The income tax year Appendix Tables I and II present ing equations for the five interest equations ment, Bill Rote I I results, and a useful specifica- economic forces when informed framework supjudg- for pre- rates. OF INTEREST RATES value, while the more typical deficit is indicated by a positive figure. The inflation rate is defined as the annualized rate of change of the consumer price index. The distributed lags on Federal deficits and inflation employ the smoothing technique of third-degree Almon polynomial approximation without constraints on beginning or ending This technique finds a time response values. Appendix Equation Statistics interest Nonetheless, A EXPLANATION CHARACTERISTICS factors may provide dicting the estimatrates. The by other These of role in shorter- an operational of fundamental markets. plemented expectations markets. provide tion of the effects explan- bill and U. S. 3-5 play a larger longer-term intercepts. have better Near-term factors than on financial between equations equations. these rates are measured in basis points (100 basis points equal one percent). The growth rates of money are given as revised seasonally adjusted annual rates. The unemployment rate is expressed as a seasonally adjusted percentage. The Federal budget deficit is expressed in units of $trillions/lO. A surplus is indicated by a negative STATISTICAL note is an important than the Treasury term APPENDIX AN ECONOMETRIC ability institutional ?: It is not significant in the Industrial A equation. The simple correlation between lagged unemployment and the Industrial A rate is 0.71, suggesting that unemployment reduces investor confidence in these slightly risky securities. The confidence effect evidently aImost overcomes the income effect for this rate. See William D. Jackson, Deteminants of Long-Tern Bond Risk Premiums. Federal Reserve Bank of Richmond, (1976). bonds Table I OF ESTIMATED r Security Rote INTEREST Industrial A Bond Rote RATE EQUATIONS T Municipal A Bond Rate I - 1 long-term Government Bond Rate t t Coefficient Statistic Coefficient Statistic Coefficient15 StatisticCoefficient StatisticCoefficient Statistic Predictor M3 Growth Rote (1) Unemployment Rote (t-l) FederalDeficit (Sum of Coefficients t to t-11)’ Inflation Rote (Sum of Coefficients t to t-11)* Constant ii* Standard Error Durbin-Watson f’ peon of Dependent Variable -4.21 -54.2917 -4.7622 -3.71 -2.1123 -3.76 -3.62 - 25.0579 -1.77 -0.2153 -0.03 - 2.0967 -13.8815 -2.22 -1.1516 -1.29 - 11.4576 1470.5396 1.11 1653.1787 1.26 1168.5569 1.87 1956.4399 1.87 888.9832 1.15 46.7221 629.9043 5.91 8.96 23.6375 682.9395 2.49 7.32 12.4733 679.8755 4.69 12.82 18.8627 570.9873 2.34 6.41 12.7252 624.8027 2.15 9.52 0.9412 34.23 1.90 0.8493 0.9187 29.45 1.71 0.9244 0.9863 13.02 1.26 0.9609 0.9549 21.86 1.54 0.9660 0.9541 16.10 1.64 0.9573 575.7180 647.4338 758.1765 563.5535 603.5989 C * Individual distributed coefficients shown in Appendix Table II. log ore FEDERAL RESERVE BANK OF RICHMOND 23 Appendix DISTRISUTED Time Log TroClskUy Bill Rote lnflotion Rote Federal Deficit LAG COEFFICIENTS Table II FOR INTEREST u.s.3-5 Year Security Rote Federal Deficit 37.5810 (0.30) I RATE EQUATIONS Industrial A Bond Rate Municipal A Bond Rate Inflation Rote Federal Deficit Inflation Rote 1.2608 (1.19) 89.7999 (1.61) 1.4806 (3.17) 28.4906 (0.30) 1.1662 (1.49) 3.1411 0.5053 Federal Deficit t -57.4630 (-0.41)" 3.8679 (3.15) t-1 -3.1176 (-0.02) 4.6892 (3.64) -8.1634 (-0.06) 1.1932 (1.00) 47.5372 (0.79) 1.2841 (2.36) 5.3250 (3.67) -10.5670 (-0.07) 1.3638 (0.98) 28.6808 (0.43) 1.3210 (2.04) lpt;5 Inflation Rote I Long-term Government Bond Rate Federal Deficit -29.3344 (-0.42) Inflation Rote 1.1606 (2.01) 3.1280 (0.04) 0.7233 (1.07) ;;:go 37.4223 (0.46) 0.5914 (0.74) t-2 53.9294 (0.35) t-3 109.5951 (0.69) 5.7489 (4.08) 19.6125 (0.13) 1.6928 (1.20) 28.6467 (0.42) 1.5233 (2.28) 56.4860 (0.49) 0.4099 (0.37) 70.9156 (0.84 46850 (0.83) t-4 159.7966 (1.06) 5.9345 (4.86) 71.6716 (0.50) 2.1007 (1.61) 42.8509 (0.W 1.8230 (2.87) 114.5848 (1.02) 0.77B9 (0.73) 100.9691 (1.22) 0.9243 (1.18) t-5 200.4509 (1.41) 5.8552 (5.52) 134.8705 (0.97) 2.5078 (2.09) 66.7093 (1.02) 2.1520 (3.59) 179.9039 (1.65) 1.2945 (1.28) 124.9479 (1.55) 1.2295 (1.66) t-6 227.4750 (1.62) 5.4846 (4.93) 198.4789 (1.45) 2.8345 (2.32) 95.6379 (1.48) 2.4423 (4.03) 242.1455 (2.24) 1.8585 (1.83) 140.2160 (1.76) 1.5206 (2.03) t-7 236.7859 (1.62) 4.7962 (3.59) 251.7659 (1.80) 3.0011 (2.22) 125.0527 (1.91) 2.6258 (4.05) 291.0112 (2.64) 2.3725 (2.18) 144.1376 (1.77) 1.7179 (2.15) t-a 224.3008 (1.48) 3.7637 (2.46) 284.0007 (1.98) 2.9281 (2.00) 150.3696 (2.26) 2.6345 (3.87) 316.2039 (2.83) 2.7383 (2.40) 134.0768 (1.63) 1.7415 (2.07) t-9 185.9368 (1.23) 2.3605 (1.52) 284.4339 (2.03) 2.5357 c1.m 167.0049 (2.58) 2.4003 (3.47) 307.4258 (2.83) 2.8576 (2.60) 107.3978 (1.35) 1.5115 (1.87) t-10 117.6107 (0.83) 0.5602 (0.42) 242.3937 (1.86) 1.7444 Cl.44 170.3742 (2.86) 1.8552 (3.39) 254.3783 (2.54) 2.6320 (2.86) 61.4646 (0.83) 0.9481 (1.40) t-11 15.2397 (0.10) 147.0898 (1.12) 0.4746 (0.44) 155.8937 (2.64) 0.9311 (1.97) 146.7636 (1.48) 1.9632 (2.47) i -1.6638 (-1.33) -6.3586 (-0.09) -0.0286 (-0.04) . * The parentheses contain+ statistics the coefficients for immediately above. without constraining the adjustment path to a predetermined shape.’ The summed coefficients appear in Appendix Table I, while the individual tinlc. coefficients appear in Appendix Table II. 7‘11~ .+:niiicance of the coefficients is given by their t statistics. An absolute value of t of 1.29 or more indicates a statistically significant relationship. The fi” statistics have been corrected for degrees of freedom (98). The Cochrane-Orcutt correction for first-order verted to units of the original variables. This technique is largely effective in removing autocorrelation, as shown by the Durbin-Watson statistic, which is satisfactory for all except the Industrial A and Municipal A equations. Their high R-“s and ability to explain interest rates on a month-by-month basis during recent years suggest that the remaining positive autocorrelation is not a serious problem. Several variants of these equations were est:i- autocorrelation is used.' This technique corrects a common problem in time series analysis : “runs” of successive overprediction and underprediction. Its correction factor for autocorrelation is p. The values of p indicate that these equations are essentially first-difference transformations recon- mated. Substituting the index of industrial production and its changes for the unemployment rate produced insignificant t values for these prosies of income and its change. Anticipatoryexpectations proxies for future income, such as the new (deflated) index of leading indicators and stock prices, are so correlated with inflation, monetary growth, and unemployment that they added essentially no new information to the analysis. 1 Phoebes J. Dhrgmes, Distributed Lags: P-roblsms of Estimation and Fornwdatim (San Francisco: Holden-Day, 19il); James L. Murphy. Introductory Eeonmnetrics ( Homewood: Irwin. 1973 ) . ~MurphY. lot. cit. 24 ECONOMIC REVIEW, SEPTEMBER/OCTOBER 1976 APPENDIX B Ml, M2, M3, AND INTEREST Does the increasing use of interest-bearing time and savings accounts as stores of liquidity mean that the growth of M2 and M3 lowers interest rates more than the growth of Ml? Alternative versions of the interest rate equations test this hypothesis. The monetary growth coefficients appear in Appendix Table III. All of the other esplanatory variables possess the same sign and general significance, or M.3 represents Growth economy’s whether the k growth in Ml, M2, term. in M3 is a more valid indicator of the liquidity than is growth in Ml. Appendix COEFFICIENTS OF MONETARY Growth RATES in M3 indicates the liquidity of the econ- omy to a lesser extent than growth in M3. A traditional indicator of monetary policy, growth in Ml has a weak influence on interest rates in this specification. Its liquidity effect is less than one-quarter of the liquidity effect of M3, falling to insignificance in the Municipal A and longterm Government rate equations. These empiri- cal results suggest that consideration of broader monetary aggregates in the implementation of monetary policy is a proper the monetary authority. Table move on the part of 111 GROWTH IN INTEREST RATE EQUATIONS Long-term Rate Treasury of Ml U. S. 3-5 Year Bill Growth Security Rate Rate -1.1568 M2 -4.6281 - - * The parentheses contain 6.4509 t statistics (-2.03) 3.2594 - - 4.7622 -2.1123 (-3.71) for the coefficients 1.3622 (-3.41) (-3.59) (-4.21) Rate -0.5126 (-1.75) (-4.29) M3 Bond -1.1013 (-1.62)’ Industrial A (-3.76) immediately above. Municipal Bond Rote -0.4634 (-1.12) -1.6106 (-2.44) - 2.0967 (-2.22) A I-! Government Bond - Rote 0.2350 (-0.77) - 0.7274 (-1.48) - I.1516 (-1.66) The ECONOMIC REVIEWis produced by the Research Department of the Federal Reserve Bunk of Richmond. Subscriptions are available to the public without charge. Address inquiries to Bank and Public Relations, Federal Reserve Bank of Richmond, P. 0. Box 27622, Richmond, Virginia 23261. Articles may be reproduced if source is given. Please pro&de the Bank’s Research Department with a copy of any publication in which an article is used. FEDERAL RESERVE BANK OF RICHMOND 25