The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
EMPLOYMENT, GROWTH, AND PRICE LEVELS HEARINGS BEFORE THE JOINT ECONOMIC COMMITTEE CONGRESS OE THE UNITED STATES EIG H TY -SIX TH CONGRESS FIRST SESSION PURSUANT TO S. Con. Res. 13 SEPTEMBER. 28, 29, 30, OCTOBER 1 AND 2, 1959 P A R T 8— T H E E F F E C T OF IN C R E A S E S IN W A G E S , S A L A R IE S , A N D T H E P R IC E S OF P E R S O N A L S E R V IC E S , TOGETHER S IO N A L W IT H P R A C T IC E S U N IO N UPON AND P R IC E S , PROFES P R O F IT S , P R O D U C T IO N , A N D E M P L O Y M E N T Printed for the use of the Joint Economic Committee UN ITED STATES GOVERNMENT PRINTING OFFICE WASHINGTON : 1959 38563 For sale by the SliDGrintondent ol Documents, u.S. Government Printing Office Washington 26> D.O, Price 75 cents J O IN T E C O N O M IC C O M M IT T E E (Created pursuant to sec. 5(a) of Public Law 304, 79th Cong.) PAUL H. DOUGLAS, Illinois, Chairman WRIGHT PATM AN, Texas, Vice Chairman SENATE HOUSE OF REPRESENTATIVES RICHARD BOLLING, Missouri JOHN SPARKMAN, Alabama HALE BOGGS, Louisiana J. WILLIAM FULBRIGHT, Arkansas HENRY S. REUSS, Wisconsin JOSEPH C. O’MAHONEY, Wyoming FRANK M. COFFIN, Maine JOHN F. KENNEDY, Massachusetts THOMAS B. CURTIS, Missouri PRESCOTT BUSH, Connecticut CLARENCE E. KILBURN, New York JOHN MARSHALL BUTLER, Maryland WILLIAM B. W IDNALL, New Jersey JACOB K. JAVITS, New York Stu d y of E m ploym ent, Grow th, and P r ic e L e v e l s (Pursuant to S. Con. Res’ 13, 86th Cong., 1st sess.) O t t o E c k s t e in , Jam es II Technical Director W. L e h m a n , Administrative Officer W. K n o w l e s , Special Economic Counsel John CONTENTS W ITNESSES IN ORDER OP APPEARANCE Colin C. Clark, director o f research, the Econom etric Institute, In c.; ac companied by Peter W. Hoguet, president, and Mrs. Ruth Gomulicka, researchist_______________________________________________________________ Carl F. Christ, University of Chicago____________________________ _________ John R. Meyer, Harvard U niversity_______________________________________ Daniel B. Suits, University of C hicago____________________________________ Lloyd Ulman, professor of econom ics and industrial relations, University of California, Berkeley, Calif_______________________________________ 2515, Edward C. Budd, assistant professor of economics, Yale University______ George H. Hildebrand, professor of economics and director, Institute of Industrial Relations, University of California, Los Angeles, C alif______ Oscar A. Qrnati, associate professor of economics, and director of labor education center, New School for Social Research, New York, N .Y ____ Mark L. Kahn, W ayne State University___________________________________ George W. Taylor, professor of industry, W harton School, University of Pennsylvania____________________________________________________________ Jack Stieber, Labor and Industrial Relations Center, Michigan State University_______________________________________________________________ Charles C. fcillingsworth, Michigan State University_____________________ Frank C. Pierson, professor of economics, Swarthmore College___________ Martin Segal, D artm outh College_________________________________________ R obert Ferber, Bureau of E conom ic and Business Research, University of Illinois___________________________________________________________________ Eli Ginsberg, Colum bia U niversity________________________________________ Werner Z. Hirsch, Washington U niversity_________________________________ Neil W. Chamberlain, professor of economics, Yale U niversity_______ ___ Philip Taft and M erton P. Stoltz, Department o f Economics, Brown Uni versity___________________________________________________________________ John T. Dunlop, Harvard U niversity______________________________________ Page 2445 2483 2485 2492 2556 2516 2528 2543 2583 2588 2597 2628 2632 2635 2655 2661 2671 2703 2707 2733 S T A T E M E N T S A N D E X H IB IT S Budd, Edward C., assistant professor of economics, Yale U niversity_____ Exhibits: Average annual earnings and relative em ploym ent in strongly and weakly unionized sectors of the private econom y_________ Average annual earnings (including supplements) of, and propor tion of those engaged accounted for by, strongly and weakly unionized sectors of the private domestic econom y, selected years___________________________________________________________ Share of employee compensation in gross and net income origi nating in corporations, selected periods________________________ Share of employee compensation in gross and net national income (at factor cost) in strongly and weakly unionized sectors of the private econom y___ ____________________________________________ Share of em ployee compensation in net and gross income origi nating in strongly and weakly unionized sectors of the private domestic econom y, selected years______________________________ , Share of employee compensation in private domestic income for selected peacetime, nondepression years_______________________ Chamberlain, Neil W., professor of economics, Yale University___________ Christ, Carl F., University of Chicago_____________________________________ in 2516 2526 2522 2523 2527 2525 2525 2703 2483 Clark, Colin C., director of research, the Econom etric Institute, I n c .; accom panied by Peter W. Hoguet, president, and Mrs. Ruth Gomulicka, I’age researchist_______________________________________________________________ 2445 Exhibits: Comparable levels of taxation and the rate of price increase in various countries_______________________________________________ 2481 Econom etric Institute model of U.S. business fluctuations_______ 24-51 Extracts from Econom etric Institute forecasts___________________ 2449 General summary— U.S.S.R. net national product revalued in 2478 billions of dollars of 1950 purchasing pow er_________________ _ Prices, productivity and taxation________________________________ 2480 Productivity per man-hour in Soviet Union______________________ 2482 Statistics on Russian man-hours worked and real net product per man-hour_______________________________________________________ 2476 U.S. statistics on labor’s share of net national product (nonfarm) _ 2479 D unlop, John T., Harvard University_________________1 ________________ _ 2733 Ferber, Robert, Bureau of Econom ic and Business Research, University of Illinois----------------------------------------------------------------------------------------------- 2655 E xhibits: Change in deflated personal service expenditures, by category, 1036-58__________________________________________________ 2661 Estimated change in price of consumer services, bv category, 1936-58____________________________________________ _________ 2660 Influence of price in increase in personal service expenditures, by category, 1936-58_________________________________________ J 2661 Ginsberg, Eli, Columbia University________________________________________ 2661 Hildebrand, George H., professor of economics and director, Institute of 2528 Industrial Relations, University of California, Los Angeles, C alif______ Hirsch, Werner Z., Washington University________________________________ 2671 E xhibits: Total current expenditure (plus debt service) for public primary and secondary education, selected years, 1900-58________ 2672, 2673 T otal current expenditure (plus debt service) for public primary and secondary education minus expenditures for auxiliary services, selected years, 1900-1958____________________________*_ 2678 Kahn, M ark L., Wayne State U niversity__________________________________ 2583 Killingsworth, Charles C., Michigan State U niversity_____________________ 2628 M eyer, John R ., Harvard University______________________________________ 2485 Ornati, Oscar A., associate professor of econom ics, and director of labor education center, New School for Social Research, New York, N .Y ____ 2543 E xhibits: 2550 Em ploym ent, prices and wages in 1949, 1954, and 1958_________ Indexes of hourly money earnings in manufacturing in selected countries, 1946-57______________________________________________ 2547 Rise in the consumer price index in 16 industrial countries in Europe and North America; 1948-53, 1953-57_________________________ 2545 Pierson, Frank C., professor of economics, Swarthmore College___________ 2632 Segal, M artin, D artm outh College_________________________________________ 2635 Exhibit: Percentage increases in average hourly earnings, 1953-58__ 2639 Stieber, Jack, Labor and Industrial Relations Center, Michigan State University_______________________________________________________________ 2597 Suits, Daniel B., University of Chicago____________________________________ 2492 E xhibits: The econom ic outlook for 1960 as forecast by an econom etric model of the United States____________________________________ 2496 Review of forecasts_______________________________________________ 2493 T aft, Philip, and M erton P. Stoltz, D epartm ent of Econom ics, Brown U niversity________________________________________________________________ 2707 T aylor, George W., professor of industry, Wharton School, University of Pennsylvania____________________________________________________________ 2588 Ulman, Lloyd, professor of econom ic and industrial relations, University of California, Berkeley, Calif_______________________________________ 2515, 2556 A D D IT IO N A L IN F O R M A T IO N [Summary of the September 1955 and September 1959 classifications for labor market areas_______________________________________________________ 2717 EMPLOYMENT, GROWTH, AND PRICE LEVELS MONDAY, SEPTEMBER 28, 1959 C o n g r e s s of t h e U n i t e d S t a t e s , J o in t E c o n o m ic C o m m it t e e , Washington, D.C. The committee met, pursuant to recess, at 10 a.m., in room P-68, the Capitol, Hon. Paul H. Douglas (chairman) presiding. Present: Senator Douglas and Representative Widnall. The C h a i r m a n . I am very happy this morning to welcome an old personal friend, a very distinguished economist, Dr. Colin Clark of Oxford University. You started in life, I believe, as a physical chemist and have had a distinguished career on at least two continents, and it is always certain that you will provide a number of provocative ideas and predictions. Dr. Clark now has a connection with the Econometric Institute. I understand that Mr. Hoguet of that institute has a statement which I think it would be well if we printed in the record without your deliv ering it. (The statement referred to is as follows:) Statem ent of P eter W . H oguet , P r e sid e n t of t h e E conom etric I n s t it u t e , I n c . My name is Peter W. Hoguet. I am an attorney and an economist and I am president of the Econometric Institute, Inc., a New York business forecasting and advisory service. I am a graduate of Harvard University and New York University Law School and have divided my career between law and economics, business and government. In the early part of the war I was an assistant to the U.S. Minister in charge of Economic Warfare Division, U.S. Embassy, London, and later served with the Office of Strategic Services and Joint Chiefs of Staff, Washington and Paris. After the war I served successively as Chief of the United Kingdom, Eire, Ire land program for the Marshall plan and Government attorney in the Office of Price Stabilization, Economic Stabilization Agency, Washington. On leaving Government service I joined the Planning and Development Section of General Motors at Sao Paulo, Brazil. I joined the Econometric Institute early in 1957 and was elected its president in May of 1958. It is my great privilege today to introduce to the chairman and this committee, our distinguished director of research, Dr. Colin G. Clark. Dr. Clark is an inter nationally known econometrician. He was a Frances Wood prizeman of the Royal Statistical Society and he is a frequent visitor to Washington and various U.S. companies. He is well known in the United States. He was an associate of the late Allyn Young of Harvard University and has taught at the University of Chicago. Also, he is director of Institute for Research in Agricultural Eco nomics, Oxford University, and a fellow of the Econometric Society. With the committee’s permission, I would like to file a statement on the de tails of Dr. Clark’s background and the past history of the Econometric Institute. I would also like to mention that Dr. Clark is the author of a number of well known books on applied economics including “Conditions of Economic Progress,” “Economics of 1960,” and a “Critique of Russian Statistics.” He first predicted that countries with high tax levels would be faced with persistently rising prices in an article in the Economic Journal in 1945. In this 2445 2446 EMPLOYMENT, GROWTH, AND PRICE LEVELS article he stated that the safe limit of taxation of any country was 25 percent of net national income. The following is a statement of the details of the development of the econo metric model and the past history of the Econometric Institute plus the back ground of Dr. Colin G. Clark. THE ECONOMETRIC MODEL OF THE U.S. ECONOMY The model is a framework of equations that encompasses the entire economy and I would like to point out that the model has taken years to perfect and has been programed for processing through an IBM 650 computer on a quarterly basis, covering a period of a year and a half. This is the first time that such quarterly forecasts have been computed from an integrated econometric model of the U.S. economy. The IBM 650 computer makes possible intricate calculations in a short period of time, which formerly required weeks. This means that expertly processed data and forecasts are now available to business and finance such sooner than heretofore. As I have stated, the institute prepares econometric studies from current, pri vate and public economic and financial data and publishes weekly and monthly forecasts of several key industries for which the model is used as a forecasting tool. The model contains some 80 equations, 29 of which are behavioral. These equations are the result of lengthy research and study and are based on data determined to be the prime factors in influencing the fluctuations of the various components in our economy. The model has previously been formally tested by feeding it with information available for the fourth quarter of 1956. Based on this data the model accurately “forecast” the sudden and sharp business reces sion beginning in the middle of 1957 and the start of the recovery in 1958. THE HISTORY OF THE INSTITUTE (1938) The Econometric Institute, Inc., was organized on April 1, 1938, to meet the needs of business for economic research and to interpret the effects on individual businesses of changing economic conditions. Econometrics, as developed by the Econometric Institute, Inc., is a statistical science of economic forecasting. The purpose is to enable businessmen to make successful plans for the short-, the intermediate- and the long-term conduct of their activities. On the basis of a comprehensive analysis of current facts con cerning the flowing streams of money and goods, the institute forecasts the future national income and its relation to production and distribution in the various fields of industry and agriculture. In this way, wise corporate and indi vidual planning is made possible by reliable forecasting which substantially nar rows the field of guesswork. In its forecasting work the institute uses scientific techniques. Its officers and staffs have for years been engaged in developing these techniques and are well prepared to guide businesses in the setting up and staffing of statistical con trol departments. Although the nature of economic forecasting is such that 100 percent accuracy can scarcely be expected, surprisingly good results can be obtained in business and economic forecasting if one is willing to make a substantial investment in econometric studies. All aspects of business activity are, of course, interrelated. For example, in dustrial production adjusted for technological changes yields man-hours worked in industry. Average hourly earnings times these man-hours worked is highly correlated with industrial income. Other income, such as agricultural, salary, dividends, etc., is, in turn, related to this industrial income. The total of busi ness sales is correlated with production times the average of price of industrial products. This price, in turn, depends upon the cost of production, that is, es sentially man-hours times hourly earnings, and, in the case of consumers’ goods upon the consumers’ ability to buy, that is, upon his disposable income (per sonal income after taxes). In the case of producers’ goods, the current demand and price depend upon previous corporate purchasing power and interest rates. The institute’s three weekly services, “Economic Measures,” “Trade Measures,” and “Financial Measures,” provide its clients with short- and long-term forecasts of production of and demand for consumers’ perishable, semidurable and durable goods, construction materials, capital goods, fuels, materials and supplies, shortand long-term forecasts of personal and disposable income retail trade, textiles EMPLOYMENT, GROWTH, AND PRICE LEVELS 2447 and apparel, electrical appliances, homefurnishings, commitments, business loans, credit conditions, interest rates, earnings o f industries, com m odity price indexes, labor supply and demand, and foreign exchange rates. Clients o f the institute include such companies as G ulf Oil, Armco Steel, Bank o f Am erica in California, Central Trust Co., Cincinnati Gas & Electric, Firestone Tire & Rubber, Ford M otor Co., General Electric, General Motors, and GeorgiaPacific. Each year the institute has enlarged its activities and today serves as a con sultant on interpretation o f economic trends and business planning to many man agements. Its consultants are in daily contact with the officers and technical staffs o f leading companies. These consultants are supplied w ith research mate rial by the permanent technical staff o f the institute. The Econom etric Institute has also made exhaustive studies o f economic forces that affect growth or decay of economic regions. It has made comprehensive studies o f the long-term prospects o f Buffalo, Chicago, Cleveland, Cincinnati, Dayton, Detroit, Indianapolis, Los Angeles, Miami, Milwaukee, Minneapolis, Pittsburgh, San Francisco, Washington, and a number o f smaller cities. It has compared working and housing conditions in these and other cities and areas and has suggested plant locations. Each solution o f special problems fo r business provides the raw’ material for additional econometric research. Every new research accumulation aids in the solution o f a new problem. F or instance, knowledge of the demand for automo biles, railroad equipment, building and various consumers’ durable goods sheds penetrating light on the demand for steel, copper, zinc, lead, and other metals. Studies o f the basic industries make possible the forecasting o f railroad carloadings and revenue ton-miles by commodity groups and these, in turn, indicate the demand fo r certain kinds o f rail equipment. These different demands are the starting points fo r calculations o f labor and material requirements. Historical and current figures on income by geographic areas facilitate the forecasting of sales for department stores and the setting o f sales quotas fo r companies selling to or through retail outlets. In short the institute’s past w ork has included special studies and forecasts o f : sales, costs, prices, demand, supply, profits, capacity expansion, financing methods, inventories, financial requirements, population and its shifts, regional and national income, employment, production and wage rates. All economic series and phenomena are interrelated, but in general there are differences in phase or timing. These differences make it possible to build up a system o f scientific forecasting based upon a series of econom etric relationships differing in phase relationship. This econom etric method, developed by the Econom etric Institute, Inc., is unique and no other organization has its compre hensive set of tested relationships. Time and again these various econom etric studies have enabled the Economet ric Institute to part company w ith other forecasters and to achieve correct forecasts. For example, the Econometric Institute in 1945 forecast postw ar boom conditions ; in September 1948 it forecast recession for 1949; in June 1949 it forecast r e cov ery ; in April 1953 it forecast recession ; in May 1954 it forecast r ecov ery ; and in August 1956 it forecast recession in industrial production but steady to rising personal income after the yearend. Some of these have been noted in various articles in the press especially recently when the institute published the results o f its new integrated model of the U.S. economy. DR. COLIN G. CLARK Dr. Colin G. Clark is now the director o f research fo r the Econometric In stitute, a well-known U.S. business advisory service located at 230 Park Avenue, New York, N.Y. His previous background is given on page 566 o f “ W ho’s Who, 1958” (B r itis h ), the Macmillan Co., as follow s : CLARK, Colin Grant, M.A. (O xon ) ; M.A. (Cantab.) ; D irector o f Institute for Research in Agricultural Economics, Oxford, since 1953; Fellow of the Econometric Society; b. 2 Nov. 1905; s. o f James Clark, merchant and manu facturer, Townsville and P lym ou th : m. 1935, M ajorie T a ttersa ll; eight s. one d. E d u c.: Dragon S ch ool; W in chester; Brasenose Coll., Oxford. Took degree in chem istry; Frances W ood Prizeman of the R oyal Statistical Society, 1928; Assistant to the late Prof. Allyn Young of H a rv a rd ; worked on the New Survey of London Life and Labour, 1928-29, and Social Survey o f Merseyside, 1929-30 ; on Staff of Economic Advisory Council, Cabinet Offices, 1930-31; University 2448 EMPLOYMENT, GROWTH, AND PRICE LEVELS Lecturer in Statistics, Cambridge, 1931-37; Contested (L ab.) North Dorset, 1929, and W avertree (L iverp ool), 1931; South Norfolk, 1935. Visiting Lecturer at Universities of Melbourne, Sydney, and Western Australia, 1937-38. UnderSec. o f State for Labour and Industry, Dir. o f Bureau o f Industry, and Financial Adviser to the Treasury, Queensland, 1938-52. P u blication s: The National In come, 1924-31, 1932; Economic Position of Great Britain (w ith Prof. A. C. P ig ou ), 1936; National Income and Outlay, 1937; National Income o f Australia (w ith J. G. C raw ford), 1938; Critique of Russian Statistics, 1939; The Condi tions of Economic Progress, 1940 (revised edns. 1951 and 1957) ; The Economics of 1960, 1942; W elfare and Taxation, 1954; Australian Hopes and Fears, 1957; other pamphlets and numerous articles in Econom ic periodicals. R ecreations: camping, housework. T . : 57775. Club: Johnsonian (B risban e). Dr. Clark’s recent publications include the “ Economics o f 1960” and a ‘ ‘Critique of Russian Statistics” published in 1939 and “ The Real Productivity of Russia,” to be published December 1959. H is present address is The Econometric Institute, Inc., 230 Park Avenue, New York, New York. The C h a i r m a n . Mr. Clark, we will be pleased if you would proceed in your own way. Do you wish to introduce your colleagues ? STATEMENT OF COLIN G. CLARK, DIRECTOR OF RESEARCH, THE ECONOMETRIC INSTITUTE, INC. (ACCOMPANIED BY PETER W. HOGUET, PRESIDENT, AND MRS. RUTH GOMULICKA, RESEARCH OFFICER OF THE INSTITUTE) Mr. C l a r k . Mr. Hoguet is the president of the Econometric In stitute, and Mrs. Gomulicka is the research officer engaged in this work. Having been engaged in preparing econometric predictions for the next 18 months by the use of an interconnected system of equations, I think I had better begin by giving the conclusions which we have reached, that although many businessmen are now quite anxiously thinking about the prospect of another recession, we do not see any such recession taking place, at any rate, between now and the middle of 1961. Extracts from the institute’s forecasts are on this table wThich I shall submit for printing in the record. EMPLOYMENT, GROWTH, AND PRICE LEVELS 2449 Extracts from Econometric Institute forecasts 1958: 1..................... 2___ ________ 3.................. . 4..................... 1959: 1____ _______ 2____ _______ 3___________ 4___________ 1960: 1___________ 2....................... 3___ ____ 4....................... Prices base 1947-49 Man-hours bill annual rate 127.3 127.6 127.9 128.6 129.1 129.6 131.4 131. 0 131.5 132.6 133.6 117.2 117.3 117.5 118.8 119.7 122.4 121.7 121.2 121.4 121.9 122.1 Disposable income dollar bill annual rate 1958: 1.................... 2...................... 3........................ 4..................... 1959: 1....................... 2....................... 3....... ................ 4....................__ 1960: 1........................ 2....................... 3 ..................... 4....................... 310.3 312.9 320.4 322.9 327.4 335.3 340.0 341.4 345.4 350.8 356.4 1958: 1.................................................. 2................................................. 3___________ _____________ 4_________________ _______ 1959: 1_________________________ 2.......................... ...................... 3_____________ ___________ 4_________________ _______ 1960: 1.......... ...................................... 2............................................... 3_________ _______________ 4......................................... Earnings cents per hour 208.7 209.1 213.0 214.2 217.6 220.0 225.0 227.4 230.1 233.1 236.5 Gross national products dollar bill annual rate 431.0 434.5 444.0 457.1 470.2 484.5 491.5 492.5 498.8 508.4 516.0 Corporate profits dollar bill annual rate 31.5 33.8 38.0 43.5 45.5 48.9 49.3 47.6 48.2 50.2 51.9 Dividends dollar bill annual rate 12.7 12.6 12.6 12.0 12.8 13.0 13.7 13.7 13.9 14.3 14.8 Savings Consumption Consumer Nondurables House con struction dollar bill dollar bill durables dollar bill annual annual dollar bill annual dollar bill annual annual rate rate rate rate rate 22.9 22.0 26.0 23.7 23.5 24.1 21.4 21.2 22.1 22.6 23.2 278.3 290.9 294.4 299.1 303.9 311.2 318.6 320.2 323.3 328.2 333.2 36.9 36.7 37.1 39.8 41.3 44.1 46.5 45.6 45.7 46.6 47.6 139.5 141.5 143.1 143.6 145.3 147.7 148.5 148.6 149.4 151.0 152.8 Other con struction dollar bill annual rate Producer durables dollar bill annual rate Inventory change dollar bill annual rate Unfilled orders nondurable dollar bill 18.4 17.7 17.4 17.4 17.8 17.9 18.9 19.0 18.8 18.9 19.0 23.8 22.6 22.2 23.2 23.9 26.0 31.2 30.7 30.1 30.1 29.7 -6 .9 -5 .8 -3 .4 .8 6.1 10.4 2.9 1.8 3.4 5.4 5.5 2.5 2.7 2.6 3.0 3.4 3.6 3.4 3.2 3.1 3.0 2.9 17.1 16.9 18.0 19.9 21.9 23.1 20.8 20.5 21.3 22.1 22.7 Unfilled orders durable dollar bill 44.6 43.8 43.9 44.9 47.2 47.7 52.4 52.8 53.7 57.5 63.1 2450 EMPLOYMENT, GROWTH, AND PRICE LEVELS We do see—and these extracts from our machine calculations have been circulated—rather the opposite, that the rise in prices, which will be temporarily checked in the first half of 1960, will again be resumed in the latter part of 1960. You will notice, also, that we see wage earnings going up. We see corporate profits going up, a lesser rise in dividends, and we see savings failing to rise. We see also from this table a rise in the sales of producer durables, but not a very adequate one. We attach particular importance to the figures of unfilled orders, because they exert a powerful influence on other elements in the economic system. And you will notice our figures for unfilled orders for nondurable goods are declining. We do not see any great pres sure of orders in the nondurable sector, but among all types of durable goods we see a progressively increasing amount of unfilled orders. We see 1960 a boom year for the capital goods and durable goods industries. You will also notice that in the column for inventory change the second quarter for 1959 was altogether exceptional because of antici pation of the steel strike, but you will see now that while we predict some slowing off of inventory building in the next 9 months, we show it again accelerating in the latter part of 1960. I do not want to try to predict further ahead than the time for which we have mathematical information, but I should add that the feature of all past recessions has been a period of excessive inven tory building, followed by an unpleasant period of falling pro duction and inventory liquidation. And while w^e cannot see any recession at any definite date, we can, as it were, see some of the ingredients for recession being brewed, in the form of rapid buildup of inventories. May I draw attention to this picture of the model of the system of equations on which wre are working. I submit a copy of the dia gram for printing in the record. EMPLOYMENT, GROWTH, AND PRICE LEVELS STO C KS M E A S U R E D IN CONSTANT $ LA G GED OR PARTLY LA G GED 2452 EMPLOYMENT, GROWTH, AND PRICE LEVELS The model is extremely intricate and has been the fruit of a great many years of research. Since I have been connected with the Econometric Institute, I have been able to amalgamate the equations and relationships of Dr. Charles F. Roos, the founder of the institute, with others which I developed myself. The red lines are what might be called mechanical relationships, such as the statement that if production is $11 billion and con sumption is $10 billion, the addition to inventory must be $1 bil lion. There are a great many of these. But the green and black lines show how one variable influences businessmen’s behavior in re spect to another. They are behavioral or psychological equations. They can be fitted approximately from past experience, but we know that they are liable to change. The most intricate part of the whole model is the interrelationship between inventories, orders, and un filled orders, and it is to that problem that we have devoted a great many years of research. We reach a conclusion which is both interesting and alarming. The buildup of inventories and the buildup of unfilled orders is a process which develops a considerable momentum of its own, and the same applies to the converse process, when businessmen are liquid ating inventory and reducing their new orders. It is a process which builds up a considerable momentum of its own, and it is very hard for any action of the Government to check these upward and down ward movements. I am hopeful that this may be done at some future date, but it is going to be a very much harder task than is generally realized. I should add that the construction of this model has been extremely laborious. The Econometric Institute is a large research organization, and it is principally through the altogether superhuman labors of our research officer and programer, Mrs. Gomulicka, that we have been able to complete this work and have it put on an electronic machine. Our data processing had reached the stage where direct hand calcu lations were quite impracticable. I think you will see, gentlemen, that the number of interrelations to be dealt with are so many and so intricate that an electronic calculation is the only way of analyzing the problem and of taking account of several different assumptions. We have to make independent or exogenous assumptions about what Federal Reserve policy will be and what the Government will spend, and one or two minor components, such as exports. On all the other matters, prices and output and sales and inventories, we do not find it necessary to make any assumptions, because we find that they are all interrelated to each other and to the things which have gone before. Two other features of our current predictions may have impressed you. Our prediction of employment remains comparatively low. In fact, it will not be until the end of 1960 that the number of man-hours worked gets back to the rather high level of the second quarter of 1959. During 1960, the labor force will have risen by rather over 1 percent. On the other hand, there will have been some reduction in the average working week. But taking those two factors together, we reached the rather unhappy conclusion that unemployment in 1960 will be slightly higher than it has been in recent months in 1959. And also, while showing a rise in industrial construction and producers’ EMPLOYM ENT, GROW TH, AND PRICE LEVELS 2453 durables, we show a rise which is not entirely adequate, which does not properly meet the depreciation of old industrial assets. One of the worst consequences of the persistent rise in prices in recent years is that depreciation is hopelessly understated. The official estimate of capital consumption is at the rate of $40 billion per annum, whereas the true figure ought to be nearer $60 billion per annum. And new investment at present is not properly keeping pace with depreciation. It is true that the amount of capital required per unit of output is now falling. It is not necessary fully to replace the old capital and volume. On the other hand, each piece of capital now depreciates more rapidly than its predecessors did, and is also much higher priced. The price of capital goods in relation to the price of other products is rising. So we conclude that present investment is not entirely adequate. In making these predictions, we have assumed that interest rates will go on rising and reach a level of 5 percent on triple A bonds by the middle of 1960. We tried an alternative estimate, where we assumed that interest rates would remain at 4 ^ percent on triple A bonds, and corresponding levels for other investments. The results were slightly improved throughout. There was no serious acceleration of price increase. I am afraid prices will increase whatever the rate of interest is. But residential construction by the fourth quarter of 1960 wall rise from $22.7 billion to $23.3 billion if interest rates are low. Producers durables will rise from $29.7 billion to $30 billion. The rate of inven tory building will be accelerated. Employment will be slightly higher. National product will be higher. Profits will be higher. And Gov ernment net revenue will also be considerably higher. So on the face of it, while I know there are considerations which lead the Federal Reserve Board to maintain a hard money policy, nevertheless lower interest rates, if they could be obtained, would have generally desirable effects except on this one point of inventory building. They would accelerate the inventory building and so might make the next recession a bit worse when it came. Representative W i d n a l l . Could I interrupt at this point ? What did you say that would do to prices? Mr. C l a r k . An interest rate lower by half a percent would have no appreciable effect on prices. We have not calculated what would be the effect of a really violent reduction of interest rates, but a half percent reduction of interest rates would have no appreciable effect. We also made an estimate of what would happen if Government spending were higher. We have assumed for these purposes that Government spending will go rising at about the rate of recent years, and at present all Government spending, including State and local, is at the rate of $129 billion per annum. We estimated that it will be up to $137 billion by the end of 1960. We made an alternative esti mate, putting it $314 billion higher. The main effect is that that will accelerate the price rise by about half a percent. It would lead to a slight increase in employment, although the direct effect of increased employment on Government work would be about half offset by some decrease in employment in the private sector. There would be 2454 EM PLOYM ENT, GROW TH, AND PRICE LEVELS an increase in corporate profits, a decrease in producers’ durables, and a substantial decrease in the rate of inventory accumulation, if Gov ernment expenditure were about $3 billion per annum higher. We made one further alternative assumption, of what would happen if there were a 5 percent reduction in working hours or a reduction in the labor force through earlier retirement: which would have the same economic effect. The results there were rather unexpected. The first effect would be that both employment and profits would fall quite sharply. But that would not last very long. And within 18 months, we would have employment considerably higher than be fore and also profits getting about back to their normal level. We would have considerably higher prices. Producers’ durable in vestment would fall at first but by the end of the year would be back to its normal level, and the rate of inventory building would be very considerably accelerated. I now come to what I believe is the central issue before this com mittee, namely, the long period rate of economic growth. And I give reasons in my statement for expecting a growth rate of about 3y2 percent per annum in the volume of real national product. The C h a i r m a n . Before you do that, may I ask you this question. As you know, I have a very high respect for you. This is the first set of quantitative forecasts that I have seen. The period of the 1920’s was when quantitative forecasts were made, sometimes of a rather sweeping nature. A friend of mine, Professor Cox of the University of Chicago, appraised the results as to the degree to which they had successfully forecasted what was going to happen, but on the whole it was found that the record was rather poor; that the group that made the best forecasts were the Babsons, the method that had been most criticized scientifically, and the group that made the worst record was Harvard, which was supposed to be the most scientific. The Harvard people, after the stock market break of 1929 occurred, issued a prediction saying there was going to be a recession of short duration and of only limited extent. And the results, of course, were the complete opposite of what the Harvard people predicted. And the Harvard people then went out of forecasting and ever since then have treated it as a matter of bad form to refer to this venture of theirs into the field of forecasting. Now I would like to ask you this question. These are very daring forecasts. Have you tested out your past predictions to see whether or not life has approximated them? That is, have you had a past rec ord o f historical experience ? Mr. C l a r k . What you are saying is, I am afraid, entirely true, Mr. Chairman. Harvard in the 1920’s tried a so-called diffusion method. They thought that they could measure which series moved before and which after the main run of business. I am sorry to say some economists are trying to apply this diffusion method still. I think it is a most dangerous one. It is also true that Harvard was entirely deceived in 1929. It is also an interesting sidelight on business history that Keynes was one of those who was equally mistaken. The C h a i r m a n . I was not trying to build up Keynes in comparison with Harvard. EM PLOYM ENT, GROW TH, AND PRICE LEVELS 2455 Mr. C l a r k . But it was the shock which Keynes then received which led him to rethink all his theories. No, these methods we are using have been tested. Roos’ system, which was a forerunner of this, was correct regarding the 1949, 1953, and 1957 recessions. This model which we have before us now was given a formal test quite recently when we supplied it with 1956 data only and asked it to predict 1957 and 1958, and it gave the recession occurring just at the right time. The C h a i r m a n . Y o u did this prior to the recession ? Mr. C l a r k . N o ; this was after the event. But let me now come to what was done prior to the event with my version of the model, which was somewhat different from Roos’. A t that time, between 1954 and 1957, I was working for two Chi cago businesses, the Union Stockyards and Central Manufacturing District, and a simpler version of this model was prepared in March 1957, and copies were distributed to these businesses and to one or two others, which showed a sharp recession beginning in July 1957. So we do come with considerable confidence. These methods have been tested in actuality. The 31/2 percent long period growth rate which we foresee-----The C h a i r m a n . I just want to say: Now we will wait with interest to see whether history in the next 18 months conforms to your pre dictions. Mr. C l a r k . I should also add, Mr. Chairman, that in March and April 1958, when a lot of business opinion was very pessimistic. Econ ometric Institute advised all its clients to look for a sharp recov ery in the autumn of 1958. The C h a i r m a n . As I understand it, you say there will not be a recession until sometime in the summer of 1961? Mr. C l a r k . I will put it the other way. I do not want to imply that there will be one even then. The C h a i r m a n . I understand. Mr. C l a r k . But we see a slight lull in 1960 and then a further up ward movement of prices and of orders in late 1960. The 31/2 percent per annum growth rate-----The C h a i r m a n . Is this an American growth rate ? Mr. C l a r k . This is American. Its principal component is the growth of product per man-hour, which is a growth at the rate of 2.3 percent per annum. This rate has prevailed with only minor inter ruptions since the 1890’s. This is my own figure, but it is entirely identical with that obtained by the National Bureau of Economic Research, examining the same data. The rate in most other countries is equal to or lower than the American. The British rate is outstandingly lower. The only coun tries which for any long period have had a growth rate above the American have been Sweden, Italy, and Japan. The C h a i r m a n . That is a long-term rate ? Mr. C l a r k . That is a long-term rate. The C h a i r m a n . In recent years the rate in Western Europe has had a higher growth rate than we? Western continental Europe? Mr. C l a r k . Yes. ^ But these figures are of their nature misleading, because you are taking countries which have been devastated by war, and you are measuring their rate of recovery. I think it is a fair 2456 EMPLOYMENT, GROW TH, AND PRICE LEVELS parallel to take a doctor who is treating a child recovering from a serious illness, plotting its weight on a diagram, seeing how rapidly it rises, and saying that within a year the child will weigh more than its father. I f a doctor did that, we would regard him as unfit to prac tice medicine. But economics is still a very unsophisticated science. And a lot of these claims which are made about the Russian growth rate are entirely wide of the mark. The idea seems to have got gen eral acceptance that the Russian growth rate is two or three times the American. In fact, over the long period—and I submit the detailed evidence— the Russian growth rate in product per man-hour is very consider ably below the American. The C h a i r m a n . Going back to 1913? The Census of Manufac turers and National Income? Mr. C l a r k . Yes. W e have a reasonable idea of what Russian pro ductivity was in 1913, and the per man-hour growth since has aver aged only 1.2 percent per year. I submit a diagram. The whole story is told in the diagram, which is on the last page of the tables. When farming was collectivized, in 1928, the result was a violent fall in productivity, and productivity settled, went to a new plateau, in the late 1930’s. Then in the 1950’s there was a short period of rapid rise, and later as might have been expected, as is happening in Germany and similar countries, a flattening out. And the long period growth rates are only 1.2 going back to 1913 or 1.6 if instead you take 1939 as your starting point. It is just the same with figures of Germany and Japan. I f Ger many and Japan had propagandists as bold and unscrupulous as the Soviet propagandists, they would set out to prove that in another 10 years Germany and Japan will surpass the U.S. production, which is purely absurd, and so is the Russian claim. The C h a i r m a n . Which raises this claim. The Russians are un doubtedly unscrupulous at handing figures. How can you be clever enough to detect the degree of their unscrupulousness and deflating figures properly ? Mr. C l a r k . It has taken more than 2 0 years of continuous study, Mr. Chairman. I first published a book on this subject in 1939, and I have relied a good deal on other critical work, particularly by Jasny, Nove, and Chapman. I should also add that the wrork done by Professor Nutter in the National Bureau of Economic Research comes to exactly the same conclusion using an entirely different method, because I have worked using figures of actual consumption and investment. He has worked by using manpower productivity in sectors of manufacture. And he has come to just the same conclusion. You can understand the zeal with which Soviet and fellow travel ing propagandists tried to make their case, because the Russian philosophy is a materialist one. The Russian people have been called upon to sacrifice their property and their freedom and their family life and their religious beliefs and everything else for the sake of productivity, and it is a bit hard to find, as they are finding, that all they have is a very second-class improvement in productivity in re turn for sacrificing everything else. EM PLOYM ENT, GROW TH, AND PRICE LEVELS 2457 Coming back to American prospects, I should say there is 110 seri ous prospect of the 2.3 per annum productivity growth being in creased. Or, if it is to be changed, it will only be changed very slowly. These productivity changes in nearly every country take place continuously and slowly. The rate of growth of the labor force at present is slightly down. That is the reason why some economists profess to find some slow ing down of the rate of economic growth in the last few years. But the labor force will again beign to grow much faster, about 1963, when the postwar babies begin to come on to the labor market. But I am afraid that this economic growth is accompanied by these persistently rising prices. And a persistent rise in prices in peace time is a new phenomenon. It is quite mistaken to believe— it is not good economic history to say that there is a perpetual upward trend turning of prices. What happens is that prices go up rather violently in periods of war and social disturbance, but it is certainly not a normal tendency to have them going up all the time. Some economists used to claim that there was an economic necessity for rising prices; that it provoked more investment and stable eco nomic growth. I have been studying all the available evidence on economic growth for nearly 25 years, and I would say that there is no justification for that claim at all. Economic growth goes on in periods of stable prices or even falling prices just as much as in periods of rising prices. And while the so-called economic case for rising prices is non existent, the case against them on grounds of social justice is over whelming. The principal sufferers from rising prices are the old peo ple and the savings of poor families, which have to be kept in insurance policies and savings deposits, and the principal gainers are wealthy speculators. And anybody who, knowing all the consequences of his action, deliberately advocates persistently rising prices must be ad judged guilty of the meanest kind of plundering of poor men’s savings. I think however that most of those who advocate rising prices can be excused on grounds of invincible stupidity and of not fully under standing the consequences of all they do. But I should add that many of the second-rate economists and publicists who are so persistently clamoring for rising prices are themselves speculators of a rather amateur nature. They do not have the skill and knowledge to gain by speculation on a stationary market, and they want a perpetual bull market. Also, these persistently rising prices have a very depressing effect on those who live on fixed salaries, on public officials, on civil servants, on teachers, soldiers, policemen, clergymen, and other im portant elements in society. I am not suggesting that many of these men will leave their pro fessions because of the falling rate of remuneration, but it will be difficult to recruit the right men to these fixed salary professions. And especially when you should bear in mind that soldiers and civil servants and others in the past have put up with low salaries in the expectation of a fixed and valuable pension, and if you create a world in which people expect prices to go on rising forever, such pensions lose their value, and the discipline of the public services 38563— '59— pt. 8-------2 2458 EM PLOYM ENT, GROW TH, AND PRICE LEVELS may begin to deteriorate rather alarmingly. This is a slow and insidious process. But it certainly can take place. I wTould ask you to turn your eyes to the South American Re publics. They all began their independent economic career with both economic resources and political ideas similar to those of the United States, and it is largely through persistently rising prices that they have reached their present state of social disorder, their social divisions, with a handful of very rich men living by speculation and landowning at one end, and impoverished farmers and laborers at the other, and very little in between except intriguing and am bitious soldiers. This is a grim picture of what can happen to coun tries which allow their currencies to be persistently debased. Some of the Latin American countries have had persistently rising prices ever since the 1880’s. I do not think that any social order can stand prices rising consistently for decades on end. The effects may not be immediate, but they are all the more serious for being delayed. Now, it is wrong to use the word “inflation” for this state of per sistently rising prices, because it conveys the implication that excessive money supply is the root of the trouble. I think that this idea has now been abandoned by nearly all economists. During the last 10 years, while real production has risen 48 percent, money supply has only risen 40 percent. The money supply is being kept down in relation to production, but nevertheless prices have risen by 27 percent, some 2y2 percent per year, which is far too much. Now, there is a quite different view which is held by a number of European economists, that persistent rise in prices is due to overem ployment, excessive demand. I f somehow or other you could check demand even slightly, and by the same token create a certain amount of additional unemployment, then you could check the rise in prices. The principal exponent of this doctrine at present is Professor Phillips of the London School of Economics, who has prepared an interesting set of mathematical equations purporting to relate wage increases to the level of unemployment. There are, however, fundamental defects in his data, because for the 19th century he used trade union unemployment figures, which are far too high, and much above the national average, and for recent years, instead of using the rate of increase of wage earnings, he has used the rate of increase of wage rates, which is far too low. And so he has drawn entirely mistaken conclusions. There is an element of truth in the case which is put by Professor Phillips and his friends. It is quite true that if you drive production and employment too high, you accelerate the rate of wage increase. And it is possible roughly to identify the coefficients controlling this relationship. But the point which these economists miss, the most serious and obvious point, is this: that even when unemployment is quite appre ciable, you still get rapid price increases. W e are getting them now. They pay too much attention to the variable in the equation and not nearly enough attention to the constant or what I call the built-in component. W e now have a built-in component of wage and price increase. And even if we keep unemployment at a seriously high level, we still get the same results. EM PLOYM ENT, G R O W TH , AND PRICE LEVELS 2459 Some people blame the trade unions. They say unions are raising wages too fast. But a point not generally realized is that nonunion wages are rising as fast or faster than union wages, and that profits are rising at approximately the same rate as wages. I submit a table showing that, correctly calculated, labor’s share of the net national product is now about 80 percent, including the labor component of the incomes of independent businessmen. And it is likely to remain at approximately that level. And therefore if wages and profits rise at the same rate, the rate of wages on the whole con trols— no; I shouldn’t say controls, because I do not want to convey that implication. I do not think either wages or profits control the rise in prices. I think the only fair way to describe the situation is to say that there is some more deep-seated force which is pulling up wages, pulling up profits, and pulling up prices simultaneously. Now, I find this force in taxation. And I submit some fresh evi dence to throw light on that. You may remember, Mr. Chairman, in 1952 you presided over a radio debate between myself and Professor Heller, who took the view that high levels of taxation did no economic harm. That is the view held, I am afraid, by the majority of economists. And when we con cluded the debate, we left the matter undecided, and I said, “Well, if I come back after a few years, I am afraid I shall find American prices very much higher than they are now in 1952.” And I am afraid that has been the case. The rise has been persistent, even though we have had two quite sharp business recessions during the intervening years. I first reached this conclusion about 25 percent of the national in come being the safe limit for taxation in an article published in the Economic Journal in 1945. And it is an interesting point that the editor of the Economic Journal who published the article was Lord Keynes, and in addition he wrote me a personal letter in 1944 in which he said that he agreed with my conclusions. In fact, during the last years of his life Keynes went out of his way to say that he himself was not a Keynesian, and he did not agree with the ideas which were being advocated in his name. The advocacy of very high Government expenditure and excessive taxation is an idea which has become economically fashionable during the postwar years, but this idea was applicable only under conditions of very serious recession. I f you turn to the diagram on the last page but one of the tables, the evidence regarding the effects of taxation is set out there. It is a scatter diagram, but it is quite clear which way the points lie. The general trend of the line is from the southwest of the diagram to the northeast. That is to say, in the countries and times where rates of taxation are low, the rates of price increase are low, or indeed may be negative. And when the rates of taxation in relation to national in come become high, the rates of price increase also become high. You may be surprised to see Germany among the very high taxed countries and the very high wage increase countries. The facts about Germany have been concealed, because until very recently Germany had heavy unemployment, but now that she is approaching full em ployment, she is suffering very rapid wage and cost increases. 2460 EM PLOYM ENT, GROW TH, AN D PRICE L E VE L S Yon will see that if you draw a vertical line through the 25-percent taxation position, it will correspond approximately to naught percent per annum price increase. The reasons for this effect of high taxation upon prices are, first of all, that taxation discourages effort, or, alternatively, men have to be paid much more to make the same effort. But even more important, it makes corporations far too ready to spend money on any object other than capital (if the}7 spend money on capital improvements, they are taxed on it). But everything they spend on expense accounts or advertising or bonuses or office maintenance, all the various w&ys in which money can be frittered away, are all, as it were, untaxable money. And most of all in the payment of wages and salaries. I f an employer can pur chase industrial goodwill by giving a wage increase, more than half of the cost of which falls on the U.S. Treasury rather than on his own net profits, he has a very strong incentive to do so. It is true, Mr. Chairman, that a great deal more research is needed into the nature and the effects of different forms of taxation. My suspicion is that the present corporate tax has the most direct eco nomic effect in raising prices, although I think that very high levels of personal tax have the same effect. But I think it is clear that the American economy needs, rather, more investment, particularly more corporate investment, and less money going into the purchase of farms and similar enterprises. I will conclude by saying that a fair-minded economist looking at this diagram, seeing the relation between taxation and prices, may say that this relation is true but that there is some still deeper cause pull ing up taxation and pulling up prices at the same time. This view was expressed to me by a very high official in the Canadian Govern ment, who I must not name, for personal reasons. And he expressed the rather gloomy conclusion that during the last two or three decades there has been a fundamental change in the whole nature of politics, and that politics was now nothing more than a general yielding to pressure groups, and that as a result of that prices went up and taxa tion went up, too. I do not share that pessimistic view. I do not think politics has degenerated as much as my Canadian friend thought. But I do think that largely through lack of knowledge, expenditure on political pro grams has reached a stage beyond reasonable economic limits in almost every country. Switzerland and Japan are among the few exceptions. U.S. Government spending at all levels is at present $129 billion. And to restore price equilibrium, I estimate that as much as $26 bil lion would have to be cut out of spending. That would mean, as I recommended in my booklet “Welfare and Taxation” some years ago, that some very large sectors of expenditure would have to be aban doned by Government and taken over again by individual or volun tary effort. You will not agree, perhaps, Mr. Chairman, w^hen I suggest that there should be no expenditure on farm policy or on housing, or still less, perhaps, will you agree when I say that education should not be such a charge on public funds as it is now; that under most circum stances in the modern world parents ought to be able to pay fees for the education of their children. You may say that that would be hard EMPLOYMENT, GROWTH, AND PRICE LEVELS 2461 on large families, but I am speaking as a father of nine children, my self, who has paid fees for their education. And I think many things which have been traditionally regarded as suitable objects of Govern ment expenditure should now be reconsidered. Thank you, Mr. Chairman. The C h a i r m a n . Thank you very much. Mr. Widnall? Representative W i d n a l l . I would just like to say, Mr. Clark, you said a mouthful. The C h a i r m a n . A remark wThich Mrs, Hyland is supposed to have made to the Queen of Belgium. Representative W i d n a l l . And you certainly proposed some very interesting results for the future, and some of them seem to disagree a great deal with those of other economists. As I understand it, you believe that the greatest inflationary fact of the day is our system of taxation. Mr. C l a r k . Yes. Representative W i d n a l l . Do you feel this is mainly in the income tax field? Mr. C l a r k . Well, excises and similar taxes force up prices di rectly— directly, and once and for all. Income taxes force up prices more insidiously. I think income taxes and corporation taxes may be the more insidious and permanent danger. Representative W i d n a l l . I was interested in your remarks about the interest rate as it applied to triple A bonds and as I suppose it would apply to Government bonds; that there would be no appreciable change in prices if you reduced from 5 to 4% percent, Now, why would you not again have the inflationary factor of pumping a little bit more into the economy, utilizing the existing income tax rates, which would create additional inflation ? Mr. C l a r k . I think the answer is that the lower interest rate would permit more capital construction and yield more production to offset the increased money supply. Representative W i d n a l l . A t the same time, you say they are not taking on depreciation in connection with the capital structure ? Mr. C l a r k . That is so; yes. Internal Revenue does not allow busi nessmen sufficient depreciation for the replacement value of their capital assets. But a lower interest rate would mean rather more investment. Representative W i d n a l l . It is your contention, then, too, that a change in the income tax would provide for more capital investment ? Mr. C l a r k . Certainly. I would rather do it that way, because a reduction of interest rates would be subject to this qualification, that it would encourage the building of inventories perhaps a bit too fast. Representative W i d n a l l . I am very much impressed by the point that you make that you would have considerably higher prices with a reduced workweek. Mr. C l a r k . Oh, yes. That is to say, I have assumed that hourly wage rates were raised to compensate for the shorter workweek. Representative W i d n a l l . And that, of course, inevitably would affect those on a fixed income the most, and we would again get into this vicious circle of the Government then having to appropriate to take care of those who are on pensions, on fixed incomes within Gov 2462 EMPLOYM ENT, GROW TH, AN D PRICE LEVELS ernment, and that would have quite a spiral mg effect, too, as far as inflation is concerned. Mr. C l a r k . Yes. Representative W i d n a l l . Do you feel that structural changes in the economy have important effects on prices and on wages ? Mr. C l a r k . It is not often I want to leave a question unanswered, sir, but I think I should leave that unanswered. It is a very farreaching question. Representative W i d n a l l . Do you in your charts or in formulating these figures allow for structural unemployment ? Mr. C l a r k . No. I think I had better answer your question in the negative. That is to say, I have not seen any important structural effects 011 prices or wages. But I should admit that further research is called for. Representative W i d n a l l . Well, you feel that it does have an im portant effect ? Mr. C l a r k . I t h in k it m ig h t . But I h a v e n o t seen it m y s e lf. Representative W i d n a l l . N o w , in your data applying to countries, on taxation-price increases, you do not have the time periods involved on that chart, do you ? Mr. C l a r k . I have used all the information I could obtain. The time periods for the different countries are somewhat different, but most of the figures up on the right, the high tax figures, are for fairly comparable periods. They are mostly during the last decade. Among the low tax entries, we have some for past periods, but we have some, like Switzerland, for the present time, too. Representative W i d n a l l . Do you believe it would have any mate rial effect if the same periods were used in connection with this? Mr. C l a r k . Yes; I do. Representative W i d n a l l . What do you think would happen with respect to the U.S. rate, as shown in this chart? Maybe that is an unfair question if you do not have your figures here. Mr. C l a r k . Y o u see, during any period a country either has high taxes or low taxes. Any change in tax policy takes some years to have an effect. So you can never compare one country for precisely the same periods. Representative W i d n a l l . Do you have anything in your facts and figures that would show the relationship in the economies of countries that have a high capital-gains tax as against a no capital-gains tax? Mr. C l a r k . N o ; I have done no research on that, I am afraid. But in my opinion capital gains should be taxed at the same rate as in come, because that would make possible a much lower general tax on income. And capital losses should be fully offsettable. Representative W i d n a l l . Would that not tend to sort of deaden the stock market ? Mr. C l a r k . Well, I think the offsetting of capital losses would have the converse effect, sir. You see, at present, I understand, there is no proper offsetting of capital losses. Representative W i d n a l l . I think y o u have made some v e r y im portant points in connection with inflation. Mr. C l a r k . Yes. Representative W i d n a l l . And I particularly paid heed to the fact that you did not feel you would lose from your existing labor force EM PLOYM ENT, GROWTH, AND PRICE LEVELS 2463 materially, but that you would have a great deal of difficulty in re cruiting future employees, if it was known that you were just walk ing into something w7here there really was no future, that provided any material substance for the person. Mr. C l a r k . This will have especially serious consequences with civil servants and police and public officials, I am afraid. Representative W i d n a l l . Suppose the tax increases on personal incomes were used to retire the Government debt. What effect do you think this would have on prices and growth ? Mr. C l a r k . If it were done on a really big scale, it would bring down the rate of interest, which would have a mildly beneficial effect. But I think it would be much better to make a direct tax remission. Representative W i d n a l l . So that money would feed into the econ omy in much greater degree ? Mr. C l a r k . It would, but I think you would lose in the process. Representative W i d n a l l . That is all for the present time. The C h a i r m a n . Mr. Clark, I am not able to understand all of your methods. But if you will turn to your table on prices, productivity, and taxation, which is the third sheet from the end of your supple ment, I notice that your second column, which is the crucial column, is labeled as “the rate of price increase.” But when I go down to the foot note, I found that it is labeled “excess of wage increase over produc tivity increase at normal level of production imports and exports per year.” Mr. C l a r k . Yes. The C h a i r m a n . My question is, Why did you not take prices them selves, which certainly are the most quoted of economic phenomena, instead of getting an inferential price increase based on wage in crease as compared to a productivity increase, with the productivity increase, of course, at least in times past, being one of the less certain of the series, and with the cause of connection of an excess of wage increase over productivity increase upon prices being also uncertain? In other w^ords, why did you not measure prices directly, instead of piling one uncertainty upon another uncertainty? Mr. C l a r k . I find it necessary to work in this way, because in the price indexes of every other country except the United States, import and export goods play a very substantial part. And to conduct this research satisfactorily, it w7as necessary to isolate the economic events within the country. You see, insofar as there is a rise in the price of goods which are sold in world trade, that will enter into the price index of any country, but in differing proportions, according to the extent of the country’s dependence on world trade. The C h a i r m a n . Now, this might be necessary—I do not say that it is—in comparing the United States with other countries. But since foreign trade is, as you say, a relatively small fraction of the total national product of the United States, with w^hich you will get dif ferent results, if in the case of the United States you took the price increases directly, instead of using this method-----Mr. C l a r k . Yes, you would get a result slightly lower than the 3 .0 , for two reasons. The C h a i r m a n . You have two entries for the United States, one for the period 1902-32, which is wTay over at the lower left hand 2464 EM PLOYMENT, GROWTH, AND PRICE LEVELS corner. Then you have another entry to the United States, 1933-58, which is upward and to the right. Which one of those would be lowered ? Mr. C l a r k . The 3.0, the high one, would be somewhat lowered. The C h a i r m a n . H o w much lowered would it be ? Mr. C l a r k . It would be lowered by the effect of farm prices, which have not risen as rapidly as industrial prices, and instead of 3.0, it would come down to about 2.5. The same way for the earlier period. The C h a i r m a n . Would the earlier figure be lowered ? Mr. C l a r k . The earlier figure would be raised. The C h a i r m a n . If the earlier figure is raised and the later figure is lowered, then the causal connection between the increase in the net national income taking of taxes and the increase in prices is cer tainly less striking than represented by your chart. Mr. C l a r k . Yes, b u t n o t very m u c h less. The C h a i r m a n . It would be interesting to get those figures, be cause if the lower figure is raised appreciably, for instance, you have in the period from 1902 to 1932 a decrease in prices in the United States, do you not ? Mr. C l a r k . Yes. The C h a i r m a n . But, historically, looking back upon it, we had the increase in prices up to the outbreak of the World War supposedly caused by the development of the cyanamide process and the more rapid production of gold. And there has been some slowing down of the rate of growth in that period, based on figures I worked out some time ago. So you will forgive me for saying this. Then you have the wartime increase in prices, which was tremendous, a slight fall in 1921, and then stabilization through the 1920’s, and then, only in 192932, the decrease. I think perhaps you may get your decrease— and I am not certain at all if you moved on prices that you would, but to the degree that you take into account the years from 1930,1931, and 1932, which are certainly a result of the recession and not a result of the low level of national taxation, I am dubious about this apparent 1-percent decrease in the rate of prices which you show in the United States over a period of 30 years. On the contrary, I would expect that if you dealt directly with prices you would get an increase even for this 30-year period, and if you stopped your study at 1929, you would get a still greater increase. M r . C l a r k . I h a v e le f t o u t th e w a r y e a r s. The C h a i r m a n . Oh, you have left out 1917 and 1918 ? Mr. C l a r k . Well, I have left out the whole period 1915 to 1920, as abnormal. The C h a i r m a n . Yes, but when you come in again, in 1920, you come in at a higher level. Mr. C l a r k . Yes. Well, I come in again at 1922. And I have also left out farm products. My object was to measure the change in prices for the nonfarm sector of the economy. The C h a i r m a n . I hope you will forgive me if I say this, that I am not certain you should have left those out. And if you were to take prices directly, I think you would certainly find the earlier figure higher and the later figure lower, and in the case of the United States, at least, it would have been hard to establish the fact that the increase EM PLOYMENT, GROWTH, AND PRICE LEVELS 2465 in prices accompanied an increase in the percentage of net national income taking the form of taxes. I might ask the staff, in view of these remarks that I have made, to analyze the price and tax figures directly for the periods and to experiment with different periods and see what alteration you get in Mr. Clark’s figures. Mr. C l a r k . I should add the point that this is meant to represent the rate of price increase in years of normally full employment. I do not put in extreme recession years. And the figure since 1933 I think you will find is not too far off the 3 percent. The C h a i r m a n . N o w there is another point which comes up as to the causes behind the rate of taxation. In the particular objects you selected for the economy, they were what many of use would call welfare purposes. But is not the big increase of national taxa tion primarily caused in those countries by the international situation ? Mr. C l a r k . I am afraid so, yes. The C h a i r m a n . We are all I think very properly concerned about the military threat of Russian communism. Mr. C l a r k . Yes. The C h a i r m a n . And we want to prepare ourselves against it Now if you take our own expenditures, in addition to the nearly $41 billion which we have spent on the military budget, you would have $3% billion on atomic energy and something around $3.3 bil lion on foreign aid, which we would not have engaged in had it not been {a) for the war and (h) for the necessity of building up a democratic alliance against communism. So you get a total of some where around $47 billion, which is directly responsible to military considerations. Then there are various hidden items in the budget, which for one reason or another are not directly chargeable to national defense; so that you certainly get around $48 or $49 billions of war induced expenditures. I have only been here 11 years, but when I came down here 11 years ago we had a defense project of $14 billion, and we were just beginning our experiments in foreign aid. In addition to that we have about $8 billion in interest pay ments on the national debt. That leaves very little for other expendi tures. Of course, we had been carrying on foreign aid prior to that in the form of loans to Britain and France and UNRRA payments, and so on; but the total war expenditures would not have been more than $17 billion, let us say, $16 or $17 billion a year. We have trebled those expenditures. And I think they have been justified in total, because they have given us greater security. I am not one who believes that arms are productive. But I think they are the price we pay in the world as it is for protection and for the preservation of the freedoms which we have. So I wonder if you should not shift the center of your attack from expenditures on education and on health and other objects to the real culprits. By this I do not mean that I am favoring uni lateral disarmament. Quite the contrary, I think there is great danger that the recent visit of Mr. Khrushchev may strengthen the movement for some degree of unilateral disarmament, which would imperil our security. And I do not think this is purely emotional. 2466 EM PLOYMENT, GROWTH, AND PRICE LEVELS I find myself somewhat allergic to your singling these expenditures out as the disturbers of the peace or the disturbers of productivity. Mr. C l a r k . Yes, it is quite true, Mr. Chairman. It is also true that countries like Australia, which are able to show a lower tax level, have done so by evading their due share of the military burden, and the United States has rather done too much. The C h a i r m a n . I know you lived in Australia for many years. Australia is keeping its prices relatively low and has not been meet ing much of a military burden. Mr. C l a r k . That is right. The C h a i r m a n . Y ou spoke of the Canadians. Canada has not been meeting its full military burden ? Mr. C l a r k . Yes. Canada has much higher welfare expenditures than the United States, and a similar rate of taxation. I entirely agree with what you say about military expenditure. It has been a very heavy burden. And I agree with you that it is un avoidable. And it is with some hesitation I suggest that Government welfare expenditures should be reduced. The C h a i r m a n . I think you had better reconsider that. I really do. I think you should have hesitated a little longer. Mr. C l a r k . But on reflection, Mr. Chairman, I would say that the military burden is likely to remain with us for some time, and we do tend to adhere to somewhat out-of-date ideas as to the necessity for welfare expenditure, and in a comparatively wealthy community it might be better, even if there were no military burden, for people to carry more of their own expenses for their pensions and education and so on. The C h a i r m a n . That brings us to the question of whether these welfare expenditures are nonproductive or whether in the long run they are productive. We go in this country upon the theory that we made education available to people. But in the long run it both im proves them as individuals in the intangible values, which are per haps the best product of a civilization, but also in terms of produc tivity ; that you reach people whose parents otherwise would not have the means to give them education or who might not have the desire to give them education, and you enable them to become abler, more productive people, and also finer people. Now anyone who has watched the evolution of the second and third generations of the successive waves of immigration that have beat in upon us realizes this. The first generation has a hard time getting ahead. The second generation does better, in some cases extremely well. The third generation develops to a high level. Personally, I think this process should be continued, that we have large sections of the population which for one reason or another do not have these advantages, the most conspicuous group being Negroes, another group being the Latinos, as we call them, of the Southwest, and the Puerto Kicans, who have come to this country, particularly to New York. And women, for one reason or another, do not get the same ad vantages as men. Now it seems to me that expenditures for education for these people would on the whole be productive. You will have a certain percentage of those who will not advance; but the general result is good. And if I may use a homely analogy. I have always EM PLOYMENT, GROW TH, AND PRICE LEVELS 2467 thought of education as a type of fertilizer, which, spread over the soil of mankind, resulted in greater yield. I do not mean to imply that the quality of education is like fertilizer, in spite of the comments as to what “B” means. Mr. C l a r k . Mr. Chairman, I entirely agree. Apart from its in herent desirability, I think that education has played a very important part in economic development. I attribute the development of Japan to their educational expenditures. And the United States has a much better record than European countries for both private and public ex penditure. The C h a i r m a n . I do not want to raise this question, but some of us have wondered whether the open system of education in this country as compared to the closed system in England, which prevailed until re cently—whether this is not partially responsible for the greater rate of progress in the United States. Mr. C l a r k . Oh, yes. The English system is gravely inadequate. The C h a i r m a n . I think it has improved very much in recent times. But as it existed, say, up to 1870, the time of the Foster Education Act. Mr. C l a r k . But English higher education is almost entirely de signed to produce good civil servants, which it does, and it does not produce good business executives. And American education is the other way around. But I think that with the increasing complexity of technique, educational expenditures will have to go still higher. I do not agree with Professor Smithies if he has advised that the limit has now been reached in educational expenditures. All I am saying is that in a country as wealthy as the United States, a large number of families could reasonably pay for a part of the education. The C h a i r m a n . But some could not. And furthermore, some would not have the desire to provide adequate education for their children, and their children would suffer from the neglect of their parents, and the community would suffer. Mr. C l a r k . I should add, however, regarding university education, that in the United States most students do expect to pay. The C h a i r m a n . I am not speaking of university education. I am speaking of primary and secondary. Now what about expenditures for public health? Is this a sheer waste, or does it, by decreasing the death rate, prolong the years of productive life, and by reducing sickness, increase effectiveness per year of working life ? Mr. C l a r k . Public preventive health measures are entirely desira ble. But that is not the same thing as saying that free medical service is desirable. The C h a i r m a n . We do not have free medical services. Mr. C l a r k . N o . I think the British and New Zealand experience of free medical service has been very adverse. It tends to be wasteful. >The C h a i r m a n . Of course, small sicknesses can grow into large sicknesses. Mr. C l a r k . Oh, yes. The C h a i r m a n . And it is important to check small sicknesses be fore they become severe. Mr. C l a r k . But once you can give your citizens a stable dollar instead of a rubber dollar, they will be, I think, in most cases, willing and able to make their own provision for old age. 2468 EMPLOYMENT, GROW TH, AND PRICE LEVELS The C h a i r m a n . Well, now you are assuming that which your table aimed to prove, that the increase in taxation is responsible for the increase in price. And what I am trying to say is that expenditure for these purposes will increase productivity at the same time and hence will bring into the opposites, on the good side of the price equation, an increase which should equal, partially offset or be equal to or be greater than, the increase in costs. Mr. C l a r k . Well, I am not very optimistic about any striking increase in productivity. I think this 2.3 percent per annum rate seems to be pretty well established. The C h a i r m a n . That is per capita. When I was born, or when I first came to consciousness and read the newspapers, around 1900, I believe the average expectation of life at birth was 43 years. Today it is somewhere around 64 years. Now most of the expansion of that, of course, has been in the fall of the infant death rate. But there has been some increase in the lengthening of the productive years of life. Isow this would mean a greater total increase in productivity, be cause there are more people out of a given population. Mr. C l a r k . Oh, yes; although a lot of that has been taken out in the form of earlier retirement and longer education. The C h a i r m a n . Of course, there we have the question: Is the end of life a mere production of material goods ? Mr. C l a r k . Well, you and I both think not, Mr. Chairman. The C h a i r m a n . N o w there is another factor. Take recreation, which I am getting more and more concerned about. The population is growing. Industrialization is progressing. Yet in this country, at least— I think they have done better in England—there is a shortage of adequate recreational facilities close at hand to our great cities. I think we need to get large areas of nature, beautiful, if possible, water and forest and streams and so forth, under the public park system, so that people can have a chance to spend weekends. What is the pur pose of having a 2- or 3-day weekend if there is no place to go, and what is the purpose of having automobiles if all you do is ride bumper to bumper on the highways, with no adequate place to go to, and the national parks of the West are too far off for the great mass of the population of the country ? This takes lots of money, and I am hav ing many battles on this subject here in Congress, with not too suc cessful results. This, I think, would be a highly productive expenditure. Mr. C l a r k . I enthusiastically agree with you on that point. The C h a i r m a n . I am glad to see that when we get down to details* you are not as much of a public expenditure cutter as I feared at first you were. I am not defending public expenditure itself; I am saying that there are objects of public expenditure which are thoroughly worthwhile, and which cannot be met by merely throwing the burden onto private expenditure. Representative W i d n a l l . If you will yield for 1 minute, I would just like to say that building a park is not going to change the bumperto-bumper situation. They will have to go on top of the other cars to get there. It will only increase the amount of traffic. And until we can get them out into the country properly, I am afraid some of the new facilities would not provide much more entertainment than they get today. EM PLOYM ENT, GROWTH, AND PRICE LEVELS 2469 The C h a i r m a n . I think our public highway system is outrunning our capacity to enjoy life at the end of the highway. Representative W i d n a l l . Dr. Clark, in developing your equations, I assume you presume that Government policies and programs are as outlined today. Now what effects on growth and prices would you expect if tariffs and other protective devices were eliminated, if there were dramatic reductions in the farm subsidies, special tax treatment of oil and mining and things like that ? That is a big question, I know. You could relate your answer to any part of it. Mr. C l a r k . That is one assumption we did not try out, sir. So far as tariff policy is concerned, I think the effective level of tariffs does appear to be coming down, and more imports are coming in, and they are exercising some downward pressure on prices. Representative W i d n a l l . Of course, now you have a reversal of policy by the great labor unions in the country. I notice at their last meeting that they are now coming out for protective devices. The C h a i r m a n . N o w , wait a minute. That may be true of textiles; but I do not think this charge is true of automobiles. And as a matter of fact, I can testify from personal experience that one of the chief sources of support which we have had for low tariffs has been from the unions. Representative W i d n a l l . That has been in the past. Senator Doug las. I just noticed in California recently a resolution was passed which has completely changed the attitude and approach of the A F L CIO toward the protective tariff policy. The C h a i r m a n . I think that was limited to textiles, possibly to plastics. Representative W i d n a l l . N o , I believe it goes beyond that. And I have noticed in my own contact with the various unions that there has been a gradual change as their own industries have been affected by the imports. And this certainly can be true in the automobile industry. The C h a i r m a n . Of course, the chief opposition that we have had to reduction of tariffs has been from the employers in the industries affected, textile employers, chemical employers, the Du Pont Co., and others. I have never believed that we should condemn virtuous men if they take a drink or two and temporarily stray from the straight and narrow path. It is the chronic inebriates who should be subject to scrutiny and treatment. So I do not quite like my good friend and colleague here singling out the group which on the whole has done more to reduce tariffs in this country than any other pressure group. Mr. C l a r k . If I might add a word, speaking as a foreigner, I think the State Department should seek concessions in return. As America moves toward a free trade policy it should seek more con cessions on American goods elsewhere. I think the effect of such concessions would be considerably in creased exports of consumer durables. The U.S. exports of producer durables are already rising rapidly, because they are not restricted so much by the countries as the consumer durables. Representative W i d n a l l . What I have in mind is if the emphasis were more on competitive policy than noncompetitive policy. With 2470 EM PLOYMENT, GROWTH, AND PRICE LEVELS respect to farm subsidies, your farm subsidies tend to create a sit uation where you take out the element of competition as far as prices are concerned. The same thing applies when you lower your tariffs. Now, how material an effect do you think it would have on the economy if some drastic changes w7ere made on these lines ? Mr. C l a r k . I think it would have a considerable effect, but it would be quite slow in developing. But it is one of the assumptions we have not yet tried out. Eepresentative W i d n a l l . Do you have any opinion as to what w^ould happen if you took out all farm subsidies with respect to farm employment? Mr. C l a r k . N o . I h a v e n o t c o m p u te d th a t. Eepresentative W i d n a l l . It would tend to increase consumption, would it not ? Mr. C l a r k . Very little, I think. The U.S. Department of Agri culture holds that nearly all the food markets are saturated. That is to say, consumption per head has reached its maximum, except for fruits and vegetables. Eepresentative W i d n a l l . Would it tend to aggravate the situation of workers leaving the farm and going into industry? Mr. C l a r k . I think if all farm subsidies were removed there would be an acceleration of the outflow of labor from agriculture. But it is going on all the time. There has been something like 3 percent per annum fall in the agricultural labor force for the past 10 years, and it would just go a little faster. Eepresentative W i d n a l l . Has that not largely been due to the economic revolution in machinery that has been taking place on the farms ? Mr. C l a r k . It is associated with it. I would not say that it is due to it. I would say the high wages and good employment in the industrial cities has played a part, too. Eepresentative W i d n a l l . I heard a case the other day where one man used to operate his farm with six employees. He alone now operates three farms with no employees and produces far more than he did originally. Mr. C l a r k . I think very often the relationship works the other way around. The farmer sees that labor is leaving, and he buys machinery in consequence, Eepresentative W i d n a l l . In self-defense ? Mr. C l a r k . Yes. But it is true that the outflow of labor is going on fast, Eepresentative W i d n a l l . Does that mean that labor prices itself out of the market, like elevator operators ? Mr. C l a r k . There is an element of that in it, too, yes, although I would not say it is a very strong element. Eepresentative W i d n a l l . Assuming no change in tax revenues, what changes within the tax structure wTould in your opinion be of benefit to economic growth ? Mr. C l a r k . I w7ould mention first, although it seems off the point, estate taxes, death duties. I think they have a very deterrent effect upon saving by older men. I would say the corporation tax and the higher rates of personal income tax have a deterrent effect to economic EMPLOYMENT, GROW TH, AND PRICE LEVELS 2471 growth. And anything you could do to mitigate those three would be beneficial. Representative W i d n a l l . A s I understand it from some remarks you made earlier, you believe that the total taxation in the country should not be more than 25 percent of the national income. Do you mean that on a Federal basis, or including State and municipal taxes ? Mr. C l a r k . It is a comprehensive figure, including State and municipal and social security payments. Representative W i d n a l l . And you feel this would have a deterring effect as far as inflation is concerned ? Mr. C l a r k . Oh, y es. Representative W i d n a l l . Y ou argue that wages and prices may still rise, even when economic activity and employment are very sub stantially below normal. Would such increases continue were the recession prolonged ? Mr. C l a r k . I think so. We had some experience in the later 1930’s, when, even in the best years, such as 1937, there were nearly 10 million unemployed, wages and prices rose quite rapidly throughout that period. Representative W i d n a l l . Was that not due to growth ? Mr. C l a r k . Well, that period has not been properly analyzed, but there is no doubt that both w^ages and prices went on rising rapidly. Representative W i d n a l l . H o w deep wTould a depression have to be protracted, to affect prices ? Mr. C l a r k . If you attempted to solve the problem by brute force and pushed unemployment up to, shall we say, 12 percent of the labor force, I think you would check prices that wTay. That I would say is out of the question so far as public opinion is concerned. Representative W i d n a l l . Of course, you are always presuming that there would be no Government action in order to try to take care of the 12 percent of the labor force. Mr. C l a r k . However indifferent I may be to political considera tions, I can see clearly that that solution is impossible. Representative W i d n a l l . I am sure nobody would want it that way or try to do it that way. Mr. C l a r k . And I do see the solution lying elsewhere, in changes in taxation. Representative W i d n a l l . If you were to cut expenditures of the National Government and then apply that to the public debt, what would that do ? Mr. C l a r k . That means that you are still keeping taxation at its present level. I think it is the taxation which does the harm. And at a time like the present, I am not in favor of heavy taxation to repay debt. Representative W i d n a l l . But that is the nub of the whole problem, you feel ? Mr. C l a r k . Yes. The C h a i r m a n . Mr. Clark, many years ago, when I studied the theory of taxation, it was felt that while taxes upon margins would increase prices, taxes upon surpluses over and above the amount nec essary to get production, did not increase prices; and while Hobson was not a perfect economist in many respects, in a little book that he wrote, “Taxation in the New State,” he advanced this thesis and it 2472 EMPLOYMENT, GROW TH, AND PRICE LEVELS checked with what I had learned studying under the foremost Ameri can in the field, Professor Seligman of Columbia. Now, excise taxes, which are the predominant method of financing State governments, and from which we get about $11 billion of Fed eral revenue, are of course taxes upon margins and directly operate to increase prices. The income tax we had always regarded as a tax upon surplus, and the corporate tax a tax upon surplus. Now, you are saying in effect that these are not surpluses. I wish you would develop that point further because if they are true sur pluses, then if any reduction in taxes is carried on, it should be upon the excise taxes rather than upon the income and corporate profits taxes. Mr. C l a r k . I think the old doctrine is quite true that a tax falling upon pure economic rents will not affect prices. And it remains true to say that land is the best object of taxation. Unfortunately, the net revenue from land, urban and rural, is a much smaller proportion of national income now-----The C h a i r m a n . I am not proposing the single tax. Mr. C l a r k . N o . That was an unfortunate idea. But what used to be an excellent source of local taxation is much less so now. The C h a i r m a n . Of course the tax on buildings—that is not a tax on surplus. It is simply a tax on raw land values. Mr. C l a r k . Y ou see, as is apparent from the tax on land, there are not many taxes falling on pure economic rents. The tax on corpora tions as it was 50 years ago may have been a tax on surpluses, but I do not think it is now. My evidence for saying that is rather sim ple ; that the posttax rate of return which corporations earn on equity now, is very similar to what they earned 30 or 40 years ago, when taxation was much lower. In other words, the tax is passed on to the customer by corporations. And I think that applies to large ele ments in the individual tax. The taxes on successful professional men are I think now passed on to the customers. The C h a i r m a n . Income taxes ? Mr. C l a r k . I think a good deal of the personal income tax is passed on to the customer; yes. The C h a i r m a n . H o w so ? Mr. C l a r k . I think that professional fees and the profits of non corporate business are much higher than they would be under low tax conditions. The C h a i r m a n . What about the demand for these services with a fixed supply ? You mean the supply is not fixed ? Mr. C l a r k . No. When taxation like this has gone on for some time—you see, this was the question raised by Robertson, when he debated the Colwyn report in 1927. At that time it was held that taxation was not passed on to the customer. And it was Robertson who reached the heart of the problem when he said, “It depends upon the elasticity of the supply of effort.” And as things were in the 1920’s, you could assume that there was a fairly abundant supply of effort still available. Robertson sums it up by saying, aI shall get worried when I see businessmen playing golf on Wednesdays as well as on Saturdays,” I think the time has now come when professional and businessmen are taking it easy. EM PLOYMENT, GROWTH, AND PRICE LEVELS The C h a i r m a n . There to the present tax. u sed to be 2473 long weekends in England prior Mr. C lark . It is a built-in phenomenon in England, yes, but it is a tax phenomenon in America. The C h a i r m a n . The English weekend w a s supposed to run from Thursday night to Tuesday morning ? Mr. C l a r k . Yes. The C h a i r m a n . This question of comparative growth of Russia and the United States is an extremely important one, and I am glad that you have produced these figures of yours, because I think they are a challenge to us to reexamine studies which we have made. I am going to ask the staff to scrutinize once again the comparative figures. There are going to be hearings in November on this subject. It is highly important. I had felt that the Russian claims of a 10 percent growth rate were grossly exaggerated. Mr. C l a r k . Y e s . The C h a i r m a n . Eight percent might be exaggerated, too. But people who are very careful, such as Professor Hoover of Duke Uni versity, testified that they thought their growth rate was approxi mately twice ours, at least since 1953, since the Stalin period. Are you certain on this matter ? Mr. C l a r k . Y o u are familiar with the book, “How to Lie With Statistics,” Mr. Chairman, which I always recommend to students. There is the old art of choosing the base date. The C h a i r m a n . Take 1953. Mr. C l a r k . If you take 1953, you will get the rate slowing dowrn. The whole point is that they did have a very high growth rate in the years of immediate postwar recovery. The C h a i r m a n . Oh, yes, certainly. And I h a v e always felt that should be discounted. Mr. C l a r k . Or alternatively, a Soviet spokesman will take the period-----The C h a i r m a n . Well, where are your figures from—oh, you mean the curve begins to take a lower rate around 1951 or 1952 ? Mr. C l a r k . If you will turn to the fifth page of the tables, you can see where the total product in dollars is set out. The fifth and the sixth pages. You will see that the total product converted to 1950 dollars was growing very fast between 1948 and 1954 and then slowed down in 1955 and 1956. The C h a i r m a n . 1956 certainly shows a slowing down. The rate there is a little less than 3 percent. Mr. C l a r k . Yes. That is exactly what you would expect. After a very rapid postwar recovery, they will get back to their normal rate of growth. And if you turn to the diagram on the last page, the same happened in the 1930’s. After the recovery period, you can see that in the late 1930’s they had gotten on to what you can call a normal trend line. You can approximately draw a straight line between 1913 and 1939 and 1956. You see, I do not believe that Soviet Russia, with her many obvious inefficiencies, has discovered something which was hidden from the rest of the world. Very few countries can do better than the American rate of growth, and when they do it is often through exceptional effort, as in Japan. 38563— 59— pt. &------ 3 2474 EMPLOYMENT, GROW TH, AND PRICE LEVELS The C h a i r m a n . There is a sort of a philosophical conclusion which comes from your figures, more or less of fatalism; that these things will happen, and that by conscious effort you cannot improve them. And this may be the case. It is somewhat alien to the American temperament, which believes that you can always do better. And again, I would like to suggest that the urgency of increasing the rate of growth may not be as great as it would be if the Soviet rate of growth were as has been expected. Nevertheless, the desirability of it is real, in order to improve the standard of life. And again let us take unemployment. We have a present unem ployment rate of 5y2 percent, plus the unemployment within employ ment, which is probably the equivalent of another 1 or 1.4 percent, and you get a total of around 7 percent. It seems to me highly desir able that that should be reduced, that the possibilities of permanent unemployment should be reduced, and if there is any tendency for credit policy or other factors to keep employment at less than substan tially full employment, allowing for seasonal fluctuations, we should do something about it. I personally cannot reconcile myself to a 51/2 to 7 percent unemployment as a permanent affair. You can toler ate it, if this is the peak to which you go up, and then come down; but now the recession has started, as you say, 26 months ago. And while there has been recovery from the standpoint of production and corporate profits, and so forth, it has left us with a very high volume of unemployment. And I do not want to treat the unemployed as built-in stabilizers for a price and production system. I think they are ends in them selves. If you could reduce the rate of unemployment, you would certainly increase the rate of growth. Mr. C l a r k . Yes. I quite agree; although there are separate prob lems, I entirely agree the present rate of unemployment is too high. But how to solve the problem without a further acceleration of the price increase, we have not settled. But on your other question, “Can we get a permanent acceleration of the growth rate above 2.3 percent per man-hour ?— I think it is pos sible. I do not want to appear too fatalistic. The C h a i r m a n . Do you not think that society should devote a good deal of attention to that question ? Mr. C lark . I do. And I think more and better quality education will play a part. The C h a i r m a n . Good. Now we are together. Mr. C l a r k . I failed to point out, Mr. Chairman, that it is my view that if parents paid fees for education, and there were a larger choice of schools and greater competition between them, I think that the quality of education would go up. The C h a i r m a n . And I should say opportunity should be provided for those who wish to have private education to have it. But I do not want to scrap the public school system. Mr. C l a r k . I think some increase in the rate of growth is possible, But I should also add it is always possible to decrease the rate of growth substantially by misgovernment. The C h a i r m a n . I am glad to see you are not opposed to public health or education or recreation. EM PLOYMENT, GROWTH, AND PRICE LEVELS 2475 Representative W i d n a l l . Dr. C la r k , with respect to the figures on Russia, what information do you have about their tax policy over there, the amount of taxation placed upon the people ? Mr. C l a r k . The main burden is indirect taxes, excise taxes. And they are quite unashamed. They put them mostly on food and neces saries. Their income taxation is very light. They believe in having quite extreme differentials of income, Representative W i d n a l l . Is that because of the theory of incentive ? Mr. C l a r k . Oh, yes. When Stalin changed course on this issue in 1931, he said that equality was a bourgeois concept and a Socialist state needed inequality. The C h a i r m a n . Of course, this was a departure from their earlier theory. Mr. C l a r k . Oh, yes. The C h a i r m a n . Their earlier theory, which they held through the 1920’s was one of narrowed differences between economic groups. Mr. C l a r k . Yes. Representative W i d n a l l . Dr. Clark, some people think while we have been going socialistic they have been going capitalistic. That assertion has been made by a number of people. It is quite evident that they have made some substantial changes. Now I think it is pos sible to inherit some property, where originally there was no power of inheritance at all. Mr. C l a r k . Yes. There is a certain a m o u n t o f inherited property. I would say that the most im p o r ta n t change which they have made is that the manager of a state en terp rise does now7 enjoy considerable freedom to dispose of his su rp lu s. He gets a certain amount of his surplus untaxed. He cannot sp en d it on himself, b u t he can use it for additional investment, On the other hand, they are still extremely weak, because they have not got any real concept of the market, Each industrial corporation is supposed to be self-contained, to make as many of its own produc tion components as possible, and buy as little as possible from other corporations. And that does lead to outrageous waste. Representative W i d n a l l . When you say a great deal of their taxes come through indirect taxes, what form is that ? Just sales taxes ? Mr. C l a r k . Y ou can either call it sales tax, or you can call it state trading. The state buys the food and resells it at a very high profit. But they are mostly taxes on the necessaries. Representative W i d n a l l . In connection w7ith lending, over there, when loans are made, what sort of interest rate do they charge ? Mr. C l a r k . That I do not k n o w . The only lending is from the state bank, I think. Representative W i d n a l l . Yes, I believe so. Mr. C l a r k . And I believe it is only short-period lending. Representative W i d n a l l . I had understood that they were going to start some installment buying over there, too. Mr. C l a r k . I had not heard that. Representative W i d n a l l . This may end up by being a curse to them in the end, loading them up more than they would want to be. That is all. 2476 EM PLOYMENT, GROWTH, AND PRICE LEVELS (The following was supplied for the record:) Statistics on Russian man-hours worked and real net product per man-hour N O N A G R IC U L T U R A L L A B O R W age and W age and W age and salary, salary, salary, workers, excluding including agricultural average forces for year 1 wage workers Y ear Millions 11.4 1913 s. 1926.. 1928.. 1929.. 1930.. 1931.. 1932.. 1933.. 1934.. 1935.. 1936.. 1937.. 1938.. 1939.. 1940 4 1948.. 1949. 1950.. 1951.. 1952. 1953.. 1954. 1955.. Millions Millions 8. 5 8. 5 9.9 10. 6 13.0 16.9 20.1 19.5 20. 6 21. 7 2/>. 2 24.' 6 25.4 26. 5 29.0 31.3 33.0 35.8 37.5 38. 9 40.2 41. 9 42. 5 44. 7 10.2 11.6 . 12 2 14. 5 19. 0 22.9 22.3 23.7 24.7 25. 8 27.0 27.8 28.9 31.4 34.2 36.0 38.9 40. 7 42.2 43.7 47.3 48.4 50. 5 Average u rban hours per year 2 N onagricultural m an-hours 2. 850 2.400 2, 400 2, 395 2, 390 2.385 2, 380 2, 050 2,050 2. 050 2, 050 2, 050 2, 050 2, 050 2,225 2, 400 2, 400 2.400 2. 400 2, 400 2. 400 2, 400 2, 400 2, 350 9.9 9.1 10.5 11.2 13.6 17. 5 20. 7 20 .2 21. 4 22. 7 24! 4 26. 0 27.1 28. 5 33.5 34. 8 37. 0 40.2 42.5 43.9 45.2 46.7 46. 5 48.3 Billions 28. 2 21.8 25. 2 26.8 32. 5 41.8 49.3 41. 5 43.9 46. 6 50. 0 53.4 55. 6 58. 5 74.6 83.5 88. S 96. 4 10 2 . 0 105. 5 108. 5 112.2 111.7 113.6 A G R IC U L T U R A L L A B O R Collective farm s Collective Collective m an-years of labor farm s Collective farms E sti Average A gri plus farm s plus m ates rural cultural for private plus wage hours m another T otal plots of in d ivid and per hours salary inclu ding industrial ual years 6 year 7 (billions) private workers peasants w orkers plots 1913 3_. 1926... 1928-.. 1929--. 1930... 1931... 1932... 1933... 1934... 1935... 1936... 1937... 1938... 1939... 1940 41948.. 1949.. 1950._ 1951.. 1952.. 1953.. 1954.. 1955.. 1956.. 36.6 20.8 ’ 381 . 22.2 22.6 24.8 27.3 27.9 32.1 33.2 Footnotes at end of table, p. 2477. 44.2 35.3 36.4 35.5 36.6 ' 46.6 (41. 5) (41. 5) 41.5 (40. 5) (39. 5) 3 (38.0) > (39. 2) 40.4 41.5 51.3 55. 0 55. 0 54.0 52.6 50.4 48.5 47.4 45.9 44.2 43.0 42.3 41.1 2,100 2, 030 2, 030 2 , 222 2,415 2,607 2, 800 2, 800 2, 800 2,800 2,800 2, 800 2,800 2, 800 2,800 2, 800 2, 800 2,800 2,800 2,800 2,800 2,800 2,800 2,800 107. 6 109. 2 111. 7 122 . 1 130. 4 137.1 141. 0 136. 0 132. 7 128.7 123.8 120. 3 118. 5 115.1 130.5 116.2 116.2 116.2 118.4 110.6 106.4 109.7 113.0 116.2 EM PLOYMENT, GROWTH, AND PRICE LEVELS Y ear 1913 s_________________________ ____________________________ 1926__________________________________________________________ 1928________________________________________ _______________ 1929__________________________________________________________ 1930__________________________________________________________ 1931_________ ______________________ _____________ ___ 1932__________________________________________________ ____ 1933__________________________________________________________ 1934__________________________________________________________ 1935__________________________________________________________ 1936__________________________________________________________ 1937__________________________________________________________ 1938__________________________________________________________ 1939__________________________________________________________ 1940 4_________________________________________________________ 1948__________________________________________________________ 1949__________________________________________________________ 1950__________________________________________________________ 1951__________________________________________________________ 1952__________________________________________________________ 1953__________________________________________________________ 1954__________________________________________________________ 1955__________________________________________________________ 1956__________________________________________________________ All m anhours Billions 135. 8 131.0 136. 9 148. 9 162. 9 178. 9 190. 3 177. 5 176. 6 175. 3 173. 8 173. 7 174.1 173. 6 205.1 199. 7 205. 2 212. 6 215. 4 216.1 214.9 221. 9 224. 7 229. 8 2477 Real product, billions of R eal product dollars of 1950 per m anpurchasing hour pow er 42. 0 43.1 45. 6 45. 9 47. 2 47. 7 44.9 48. 7 52. 5 58. 9 64. 9 65. 3 69. 4 79.9 78. 4 85.3 91. 6 96. 8 102. 5 105. 9 111.6 116.4 119. 5 0.309 .317 .307 .282 . 264 . 251 .253 . 276 .300 .333 .374 . 374 .399 .389 .392 . 415 .431 .449 .474 .492 .503 .518 .519 1 Including dom estics and casual laborers som etim es excluded from the figures for the earlier years (see Lorim er, P opulation of the Soviet U nion). 2 Indu strial hours in 1913 were stated (Socialist C onstruction in the U .S.S.R .) to have averaged 9.92 per day b u t Jasny (quoting K uzm inycli-L anin) considers th a t the figure was “ su bstantially sm aller.” T he 48-hour week was introduced in 1917. In 1929 some reduction began in m anufacture b u t no t in building or ooramwve COui’otin of the In stitu te for the S tudy of the U .S .S.R ., M unich, Jan u ary 1959). In 1933 a general sys.*"in of v. '.-king 5 days of 7 hours out of each 6 was introduced, equivalent to a 41-hour week ■^iiapina.:, R e , ' i o < i Econom ics and Statistics, M ay 1954). T he 48-hour week was restored in June 1940. Some sliiriil collections were m ade in 1956. U rban w orkers and em ployees are assum ed to w ork a 50-week year. J Bom idaries of 1921-39. 4 X e w boundaries. 5 Official triulodni are m ultiplied by % to convert to true m an-days, then divided b y 265 to convert to man-year.-;. f- K ‘ UMD'.’ated from 1926 and 1940 on total rural population. 7 T he “ norm al’’ w orking year for the rural population is taken a t 2,800 hours, b u t underem ploym ent of the rural population prevailed to the extent of 25 percent in 1913 and 20 to 25 percent in 1928 (International L abour Review , vol. 3c, p . .W», vol. 27, p. 349). I t is assum ed th a t they have been fully occupied since 1932 (including w ork oil , ri' iito plots). 8 Jo int Econom ic Com.*1:tt',e of Congress estim ate (in their report “ Soviet Econom ic G row th” 1957). T heir figure for 1940 wa.i 4/ .0, and for 1950 was 42.3. 9 Gale Johnson (Journal ol Political Econom y, June 1956) estim ates th a t an up w ard trend of rural em ploym ent a t the rate of 3 percent per ann um began in 1953 after the price and tax concessions to the peasants m ade after S talin’s death. 2478 General sum m ary — U .S .S .R . net national product revalued in billions o f dollars o f 1950 purchasing power 1913 25.26 7.25 1.01 42.0 139.3 2 1939 31.26 7.7 (1.21) 3.9 9.1 (16. 2) 69.4 170.5 1930 1931 1932 1933 1934 1935 1936 1937 1938 28. 07 5.47 1.01 .9 1.07 6.6 43.1 151.4 29.38 6.3 (1.03) .1 (1.1 ) (7.7 ) 45.6 153.8 28. 73 5.8 (1. 05) .4 (1.1 ) (8.8 ) 45.9 155.8 29.13 5.5 (1.06) .4 (1.1 ) (10.0 ) 47.2 157.4 27. 46 5.8 1.08 1.0 (1.2 ) 11.2 47.7 158.1 25. 01 5.8 (1.09) - .4 (1.5 ) (11.9 ) 44.9 158.7 24.96 5.6 (1.11) 2.4 2. 07 (12. 6 ) 48. 7 159.6 25. 65 5.9 (1.13) 3.4 3.1 (13.3 ) 52.5 160.7 27. 56 7.3 (1.15) 4.3 4.6 (14.0 ) 58. 9 162.3 28. 02 7.9 1.17 7.9 5.1 14.8 64.9 165.1 30. 20 7.6 (1.19) 4.8 6.0 (15. 5) 65.3 168. 6 1948 1949 1950 1951 1952 1953 1954 1955 1956 19£7 29. 28 6.9 1.29 9.5 9.6 (21. 9) 78.4 187.1 31.15 8.2 1.32 10.9 11.3 (22. 4) 85.3 183.2 32. 75 9.14 1.38 12.9 12.7 22.9 91.8 186.3 33. 32 11.3 1.44 11.7 15.6 (23.4) 96.8 188.6 34. 69 12.2 1.50 11.9 18.3 (23. 9) 102.5 191.0 36. 18 13.19 1.55 13.0 17.6 (24. 4) 105.9 193.4 36.71 15. 39 1.63 13.6 19.3 (25.0) 111.6 196.1 39.07 16. 49 1.70 14.5 18.9 25.7 116.4 199.0 40.29 17. 94 1.77 15.3 17.6 26.6 119.5 201.9 1940 (new borders) 35. 96 7. 58 1.23 3.7 12.6 18.8 79.9 198.0 17.0 * R evaluation of 1913 income (i.e., value of all products other th a n food and housing): Bm<m nMes All net national incom e (Prokopovich)_____________________________________ 13.06 All n et natio nal incom e excluding agriculture and fishing__________________ 5. 56 A dd G overnm ent expenditure______________________________________________ 8.16 A dd private service in d ustries______________________________________________ 9.26 Exclude incom es earned in tran spo rt and d istrib ution of food______________ 8. 5 C onvert from factor cost to m arket price_____ ______________________________ 10.2 E quals 15,700,000,000 in 1950 dollars. Prokopovicz excluded from his total rents of houses and product of service industries. LEVELS i In terp olated 1929-36, 1938-39, 1948-49, 1951-52, on o u tp u t of cotton and woolen goods, shoes, and paper. 41.79 PRICE Food con sum ption . __ __ ............. .... ----O ther p riv a te consum ption (not housing)1-. . ______ H ousing____ _ _ _ _________ _ _ ----N et capital form ation_____________ _ ------------M ilitary ex p en d itu re.. __ _ ---------E ducation , h ealth, an d ad m inistrative services_____ T o tal p ro d u c t__________ _____________________ P o pulation (millions) _ _ _________________ __________ 1929 EMPLOYMENT, GROWTH, AND Food con sum ption . __ _____________ ________________ O ther private consum ption (not housing) 1 ____ H o u sin g .. _______ _______________________________ _ N et capital form ation____________ __________________ M ilitary e x p e n d itu re ,.___ _______ ___ E ducation , health, and ad m inistrative services_____ T otal p r o d u c t .. ____ _______ ______________ Po pulation (m illions)___________ _______ ______ .. 1928 U .S . statistics on labor’s share o f net national product (nonfarm) [Figures in billions of dollars] 1929 50.2 17.0 7.0 43.2 40.0 31.1 30.4 3.4 - .5 3.9 34.3 85.8 89.8 18.2 8.7 81.1 75.0 52.1 51.1 8.4 3.3 5.1 56.2 74.9 1947 1949 1952 1954 1956 201.2 24.9 19.4 181.8 163.0 128.8 126.0 19.9 8.4 11.5 137.5 84.3 235.0 27.9 24.4 210.6 194.3 140.8 137.9 22.7 9.6 13.1 151.0 77.6 316.2 30.8 30.3 285.9 267.1 195.0 192.2 26.9 12.0 19.9 207.1 77.5 330.6 34.6 34.6 296.0 279.9 207.6 204.9 27.8 9.4 18.4 223.3 79.8 384.0 39.7 43.3 340.7 325.5 241.8 239.0 30.8 10.8 20.0 259.0 79.6 1st half i 1959 435.3 45.4 54. 6 380.7 368.0 273.8 271.0 34.1 11.0 23.1 294.1 80.0 Association of M anufacturers (Econom ic Series N o. 77, “ M ajor Tendencies in Business F inance,” p. 31). T he latter of these 2 sources, w hich indicates the sm aller adjustm ent, is used. t. *“ U .S. N ational Incom e and O u tp u t,” table V II, 9. 6 “ U.S. N ational Incom e and O u tp u t,” table I, 8. 6 “ U .S. N ational Incom e and O u tp u t,” table V I, 1, and 1954 edition, table 14. g| 7 “ U.S. N ational Incom e and O u tp u t,” table I, 8 (“ business and professional” ). 8 R ate of retu rn on equ ity assum ed sam e as for corporations: ratio of equities of cor porations (nonfinancial) and of unincorporated business (nonfarm ) show n in G oldsm ith “ A S tu dy of Savings” for certain years, interpolated an d extrapolated. LEVELS 2479 75.4 17.1 8.0 67.4 61.7 45.0 44.0 6.8 1.4 5.4 49.4 79.9 1940 PRICE i Survey of C u rren t Business, A ugust 1959. 2“ U .S. N ational Incom e an d O u tp u t/’ table 1 , 17. 3 A dju stm ent is necessary to th e official figures for depreciation (not to other com po nen ts of capital consum ption) because th ey are calculated on the historic rath er th a n on th e replacem ent cost of the assets an d therefore overstate the true depreciation cost in tim es of falling prices an d un d erstate it in tim es of rising prices. Correction m u st be m ade b y the ratio betw een indexes of present prices and a w eighted average of the past prices a t w hich the assets were acquired. Such indexes have been calculated b y the M achinery an d A llied P rodu cts In stitu te (see “ P rodu ctivity Prices and Incom es,” published b y the Jo int C om m ittee on the Econom ic R eport, p. 99) and by N ational 72.4 17.4 7.3 65.1 59.8 42.9 42.0 6.5 1.5 5.0 47.0 78.6 1938 GROWTH, AND 96.4 19.1 9.5 86.9 78.8 51.1 49.8 8.8 2.5 6.3 56.1 71.4 1936 EMPLOYMENT, Gross n atio nal prod uct a t factor c o s t3_________________________ C apital consum ption 3 at 1954 prices___________________________ C apital consum ption at current p ric es.. . __________ _____ ___ N e t natio nal incom e a t factor cost. __________________________ N et natio nal incom e at factor cost, excluding farm income 4___ C om pensation of em ployees 5__________________________ ______ C om pensation of em ployees, excluding farm em ployees 6______ Incom e of self-em ployed (not farm ers) 7___________________ Incom e of self-em ployed (not farm ers) of w hich im puted p rop erty incom e 8_ ________________________________________ B alance im p u ted as labor incom e____ _________________________ T otal labor incom e (em ployees an d self-em ployed)____________ T o tal labor incom e (em ployees an d self-employed) as per centage of n et incom e____________________ . ._ 1932 2480 EM PLOYM ENT, GROW TH, AND PRICE LEVELS Prices, productivity, and taxation C o untry Period Produc tiv ity grow th i R ate of price increase 2 A ustralia______________ __ ________ ___ A u stria ,. _____________ __________________ B elgium _______ ______ __ ___ __________ B ritain ..._ _________ ________ . ________ C an ad a___________ _____________________ D en m ark ______________________ _________ F in la n d .__ _____________________ _______ France _ _ . . . _________ . . . ____ _____ G erm an y______ _________________ ________ Ita ly . ___________________________________ Ja p a n _____ ________________ . . . ______ __ N etherlands. _________ ________ . . . ____ N ew Z ealand_______________ ___________ N o rw ay . _______________________________ Sw eden________________ __________________ 1924-31 1954-57 1953-57 1949-57 1923-38 1949-58 1922-35 1953-57 1923-39 1949-57 1930-57 1925-35 1950-57 1927-38 1952-58 1922-35 1950-57 1927-38 1949-57 1927-38 1947-57 1948-57 1923-29 1930-39 1946-58 1931-57 1902-32 1933-58 2.0 2.0 2.0 2.1 1.2 .9 1.9 1.9 1.2 1.2 2.0 2.3 2.3 1.6 1.4 .7 3.8 1.4 2.2 2.2 2.2 2.6 1.4 3.0 3.0 2.1 2.3 2.3 - 1 .5 1.2 6.0 2.5 .5 3.6 - 2 .4 2.1 1.0 5.2 - 1 .2 3.9 5. 5 .4 5.4 —2.7 1.8 - .4 - .4 - .7 2.3 3.2 0 .5 .5 .8 - .9 3.0 S w itzerland...................... . __ ___ _____ U nited S ta te s ..______ ___________ _______ T axation as M arginal percentage price of net increase 3 national incom e 4 0.7 .8 .6 1.2 .5 .7 .6 .6 1.1 1.0 1.1 .3 .6 .2 1.0 .1 .4 .4 .8 .7 .3 .3 .3 .3 .4 .4 15.0 26.9 39.7 21.1 24.0 36.9 17.0 31.2 20.0 28.0 26.0 27.0 40.9 23.0 43.6 25.0 30.0 15.0 30-35.0 16.0 31.0 35.3 13.0 13.0 29.9 20.0 11.0 31.3 1 A verage rate of grow th of real product per m an-hour per year (percent). 2 Excess of wage increase over p rod uctivity increase at norm al levels of production im ports and exports per year (percent). 3 A dditional rate of price increase expected for each 1-percent rise in production above norm al. 4 R ecent data are the average for the period 1952-56. 2481 EM PLOYMENT, GROWTH, AND PRICE LEVELS C o m p a r a t iv e L e v e l s o f T a x a t io n a n d t h e R a t e o f P r ic e I n c r e a s e i n V a r i o u s C o u n t r ie s A US TR IA 1933 - 7 FRANCE #I 9S0 - 57 * D EN MA RK 1949 - 57 GERM ANY 1952 - • BELGIUM 1949 - 57 NEW Z E A L AND 1 9 4 7 - 37 CANADA w 1953 - 5 7 I T A LY 1950 - 57 \ 'S W I T Z E R L A N D SWEDEN .1 9 2 3 -3 9 SWEDEN 9 1946 - 50 JAPAN 1927-38 N EW Z E A L A N O * 1 9 2 7 - 37 AUSTRALIA ' 1924-31 - 2 ITALY „ 1922 - 35 JO 15 20 25 TAXATION 30 35 40 45 AS % N E T N A T I O N A L I NCOME 2 he E conom etric Institute, Inc* 2 3 0 Park Ave. New York 17, N. Y. 2482 EM PLOYMENT, GROW TH, AND PRICE LEVELS PRODUCITVITT PER MANHOUR IN S O V I E T U N ION $ of 1950 Purchasing Power per Manhour — 0.7 — 0.6 0 .2 — j ' I I I I I . I j . ! I I . I I I I . I I I The E conom etric in stitu te Inc. I I I j 2 3 0 Park Ave. I —IQl^___________ I22S______ L23S______ l i S ______ _________ New.Yozktf, J&y* I The C h a irm a n . I have no more questions. Thank you very much. (Whereupon, at 11:55 a.m., the hearing was recessed, to reconvene at 2 :30 p.m., the same day.) AFTERNOON SESSION The C h a irm a n . Gentlemen, we appreciate your coming very much. We continue with forecasts for employment groups and prices. We have our old friend here to testify, Mr. Christ. Would you care to lead off ? EM PLOYM ENT, GROWTH, AND PRICE LEVELS 2483 STATEMENT OF CARL F. CHRIST, UNIVERSITY OF CHICAGO Mr. C h r i s t . I am very glad to be here again, Senator Douglas, and to see you again. One of my special interests is econometric models and another is monetary and fiscal policy. Senator Douglas and Mr. Knowles have asked me to discuss economic analysis and econometric models as a guide to future economic prospects and policy decisions. My five main points will be these: (1) Our present economic growth will probably continue. (2) Available evidence supports the view that free markets, full employment, and a stable price level are broadly speaking compatible. The evidence also supports the view that in the United States since World War II, increases in the price level have been almost entirely due to pressures of buyers’ demands for goods and services, rather than to administered wages and/or prices. (3) Economic analysis tells us a good deal about our economy, and indeed is indispensable for assessing the effects of alternative eco nomic policies. But we are still not in a position to make reliable forecasts, and we still need to know about the size and the timing of the effects of available policies. (4) There will probably be depressions in the future of more or less the same magnitude as those of 1949, 1953, and 1957. I believe it will be possible to prevent any future depression from becoming, or at least from remaining, substantially worse than these, if fiscal and monetary policies are applied according to our present knowl edge. (5) In my view the interest rate ceiling of 4% percent now imposed by the Congress on Treasury bond issues should be removed. Now let me say a word about each of these points. (1) Let us interpret economic growth to mean increasing real in come per capita, including leisure time as a kind of real income. It seems clear that this kind of growth is produced by increases in the quality of labor, increases in the quality and quantity of capital goods per capita, and improvements in productive techniques. Our eco nomic growth appears likely to continue, since there are no signs of diminution in the growth of its ingredients. (2) Concerning the issue of demand inflation versus cost inflation, the following sequence of events is clearly possible: Increases in wages and/or prices by the action of strong unions and/or firms; resultant unemployment, or the threat of it; antidepression policy that makes it possible for the higher wages and/or prices to survive without a depression. But the evidence in the postwar United States indicates that this is not what has been happening here. A recent Ph. D. thesis at the University of Chicago by Yossef Attiyeh covering 1949-57 finds that in all but a few cases, whenever and wherever wage increases occurred they were accompanied by tight labor markets, not neces sarily by heavy unionization; and that whenever and wherever price increases occurred, they were accompanied by high employment, high order backlogs, and pressures of demand against supply, not neces sarily by monopolistic conditions or heavy unionization. These find ings would be expected if the upward movements of wages and prices 2484 EM PLOYMENT, GROW TH, AND PRICE LEVELS were the result of demand pressures, and would not be expected if they were the result of “administered” wages and/or prices. Therefore, in my view, the appropriate policies to control inflation are still monetary and fiscal measures directed at total demand. It is important to insure that prices and wages remain at least as flexible as they have been in the past, first in order that the economy can adjust to moderate doses of monetary and fiscal stringency without general unemployment, and second, in order that relative prices can continue to serve as the signals and the incentive for many desirable reallocations of resources in response to changes in demand and pro duction conditions. (3) Economic analysis tells us that increases in Government pur chases and transfer payments, tax cuts, purchases of securities in the open market by the Federal Reserve, and cuts in bank reserve require ments and the discount rate, are all policies that stimulate the economy, and if used to excess can create inflation. The opposite policies inhibit inflation, and if used to excess can create depression. These things are well established. What is not so well established is howT soon the effects of each of these policies begin to occur after the policy is adopted, how great the effects will be, and how long they will last. Econometric models of the U.S. economy are being improved year by year, and offer hope that fairly reliable short-term economic fore casts will eventually be possible. As yet we do not have models that can safely be trusted to make reliable forecasts and choose correct countercyclical policies. Some implications of this situation will appear under the next point below. (4) Our economy will probably continue to experience disturbances of about the same size as those that brought on the depressions of 1949, 1953, and 1957. Built-in or automatic stabilizers helped to keep those depressions mild. Also, in the first two there were tax cuts at just about the right time, but it must be said that in neither case was the tax cut a deliberately countercyclical act. The 1957 depression was a little deeper and a little shorter than the other two. If we get future disturbances greater than those of 1949, 1953, and 1957, which wTe do not know to be impossible, then we must expect depressions more severe than the three postwar ones. We know that we have monetary and fiscal tools more than strong enough to con trol any depression, except for the danger of doing too much or too little, too soon or too late. The fact that we cannot make trust worthy forecasts means, for example, that if in the course of a boom our forecast is for a downturn 6 months ahead, it would be unwise to enact, say, a tax cut, because the forecast might be wrong, and then the tax cut might only turn the boom into an inflation. It would be better to wait until it is fairly clear that a downturn has occurred. There is still some danger in this strategy, though. Suppose that the need for antidepression policy is not clear until several months after a downturn, that the policy is not applied until some time after that, and that the stimulating effects of the policy do not begin im mediately, or that they last for a long time. Then there is a risk that the result of the antidepression policy will not be so much to al EM PLOYMENT, GROWTH, AND PRICE LEVELS 2485 le v ia te th e d e p r e ssio n , b u t w i ll be m a in ly to a g g r a v a te th e b o o m th a t fo llo w s a n d p e rh a p s t o create u n n ec essa ry in fla tio n . To try to avoid these dangers, it is probably best to rely on policies that do not require accurate forecasting, such as the built-in stabilizers and the more quickly enactable discretionary policies. This would favor monetary policies, as against discretionary changes in Govern ment purchases and transfers, and perhaps as against changes in tax rates. (5) If real threats of inflation are to be met effectively, then either direct controls over investment will be necessary, or market interest rates will have to be high while the threats persist. In my view there is no good reason to have legal ceilings on the interest rates that the Treasury may pay, for their result is only to hamper debt manage ment and render it more difficult to combat inflation. The C h a i r m a n . Thank you very much. STATEMENT OF JOHN E. MEYEE, HAEVAED UNIVERSITY M r . M e y e r . Any discussion of the broad and difficult questions put before today’s panel is best begun with appropriate disclaimers of omniscience. Stating or forecasting the prospects of the economy for the long run, or even the short run, is almost always a difficult and treacherous business. To attempt, in addition, an assessment of the contribution that economic analysis, or more broadly, economic knowledge can make to forecasting and public-policy formation only compounds the problem. At a minimum, these are two reasonably separable topics. Accordingly, part I of this paper is devoted to some general comments on methods while part II presents one econ omist’s forecasts—or best guesses—on the immediate prospects for growth, employment, and prices in the U.S. economy. I Certainly rather evident prefatory remarks almost necessarily must begin any discussion of the potential role of economic analysis in policy formation. The first and most obvious of these is that eco nomic analysis does not provide any simple shortcuts or “higher ob jective grounds” for making the value judgments or social welfare decisions implicit in most economic policy decisions. Elected repre sentatives still must supply these. In the same vein, methods of economic analysis, particularly those pertaining to the forecast of large national income aggregates, are presently in a rather rudimentary form, though undergoing rapid and extensive development. At the present time, moreover, certain impor tant imbalances exist in the development process. The growth of sta tistical methods and formal technique has tended to outstrip the gathering of data and the actual testing of hypotheses about economic behavior. Furthermore, knowledge is cumulative and a good deal more effort has been expended on the empirical testing and analysis of certain types of economic behavior than on other types. The result is that the more ambitious, large-scale attempts to use economic analy sis— for example, in econometric forecasting models—have been marked by very uneven performance and widely varying predictive accuracy. 2486 EMPLOYMENT, GROWTH, AND PRICE LEVELS This unevenness in performance and the cumulative character of economic knowledge is well illustrated by the recent record of the “Klein” aggregative econometric model of the U.S. economy. Two equations in the Klein model, those for consumption and production, consistently perform better than the others, yielding the most accurate forecasts in both an absolute and relative sense. Not unsurprisingly, the empirical measurement of consumption and production behavior have both received considerable attention from the economics profes sion, probably more than almost any other fields of empirical inquiry. Incidentally, much of the study and improvement in the measurement of aggregate consumption behavior was, quite pertinently, instigated by the notorious failures to estimate aggregate consumption in the immediate postwar period. By contrast with this relatively good performance in the consump tion and production sectors, Klein's model lias recorded some rather poor performances in predicting investment, corporate savings, and wage-price adjustments. It might even be argued in some of these cases that the Klein formulations are extremely rudimentary and over look much that already has been learned. Similar inconsistency of performance also characterizes the other important aggregate econo metric prediction models now available, including, I feel, those of the Econometric Institute discussed this morning. The difficulties are well recognized, however, by those constructing the models and the process of improvement in methods, data, and hypotheses continues. Given the resources available for the task, their efforts must be con sidered remarkable, moreover, in both scope and content. Discrepancies in the rate of development of the tripartite base of economic knowledge—models, data, and hypothesis testing— should not, moreover, be a cause for rejecting more formal methods for fore casting and policy purposes. For example, it is sometimes argued that weakness in data make any use of improved, more formally cor rect statistical methods a totally unrewarding, purely formal exercise. In essence this argument is of the weakest-link-in-the-chain character and, indeed, is often expressed explicitly in these terms. Like most arguments from analogy, it is at best diversionary and potentially highly misleading. For example, it might be better to adopt the analogy that economic knowledge is built upon a tripartite base com posed of three legs, like an ordinary household stool. A three-legged stool is clearly a less useful piece of furniture if one leg is shorter than the other two, but not as useless as if it had only two legs. Further more, if the stool is to be a device to be stood upon for reaching articles at heights, there is something to be said for having each leg as long; and substantial as possible. Furthermore, many of the difficulties with formal econometric methods are directly traceable to a malaise that runs through much of today’s economic discussion, whether that be at the formal theoretical or completely informal, pragmatic policy level. This is an uneasiness about the adequacy and appropriateness of traditional economic theories to explain behavior in this modern era marked by the in creasing importance of large labor organizations and corporations. Also, there is probably less than full understanding by economists of the importance and diversity of governmental influences, since large government obviously has developed as a relatively recent response EM PLOYMENT, GROWTH, AND PRICE LEVELS 2487 to the challenges posed by large labor and business institutions. Ex pressions of this uneasiness in the economics profession, and tentative suggestions on how adaptations might be made in the conventional doctrines in order to adapt them to the new developments, constitute a primary if not dominant theme in much of the testimony presented before this committee in the last few years. Traditional analysis has been constructed on the assumption of a highly fragmented, competitive society in which decisions are made by relatively small or microcosmic units. A reluctance to abandon the traditional theory exists because the older models unquestionably have served the economics profession reasonably well and have made economics the one social science which has a reasonably unitary body of theory. Furthermore, it is not at all obvious that competition has completely disappeared from the American scene. There are still large segments of the economy in which the competitive model seems to be a more than adequate description of behavior, particularly if a fairly long time horizon is adopted. One problem, then, is to know when to use the traditional theory and when to abandon it in favor of a more pragmatic, ad hoc approach. Ad hoc pragmatism is dictated by the general lack of new theories to explain the behavior of the new institutions. These new institutions create, in essence, a new sector of our economic society that, while not completely impervious to competitive forces, is at least more in sulated. Fortunately, a new body of hypotheses and knowledge is slowly emerging about behavior in this less competitive sector. Much of what is known about large business and labor institutions and the implications of that behavior has been presented in previous testi mony to this committee. In the next section a few more bits of in formation are offered; in particular, some possible implications of modern corporation investment behavior for growth, employment, and prices. II An often observed characteristic of the modern large corporation is financial conservatism. In particular, these corporations have a strong reluctance to use external financing, preferring to rely instead on retained earnings and depreciation allowances to finance the bulk of their expansion. (This at least seems to be the usual choice of pro fessionally managed corporations whenever left to make their own decisions; such a qualification is necessary because obvious exceptions to the rule of internal financing are the publicly regulated corpora tions in the public utility and transportation fields.) The usual ex planation for this preference for internal financing is that the only serious threat to an established professional management is a financial debacle in which obligations on outstanding debt could not be met. Short of such bankruptcy, the position of the managerial group is almost completely secure. Therefore, group interest strongly dic tates avoiding debt obligations and financial markets to the greatest extent possible. There are certain important exceptions to this rule—besides the publicly regulated sectors already mentioned. For example, in com petitive, rapid-growth situations, a corporation wishing to keep or 2488 EM PLOYMENT, GROW TH, AND PRICE LEVELS maintain its “market shares” will often find it necessary to borrow. Excess profits taxes with peculiar computational bases—that some times can even give rise to interest rates that are effectively negative— create another set of circumstances in which debt may become ex tremely attractive to the corporation. However, even a negative in terest rate may not make debt compulsively attractive. A number of corporations seem to have passed up debt contracts at negative or near zero interest charges; for example, during the Korean emergency period. Seemingly, the only strong, undeniable rationale for acquir ing outside debt seems to be jeopardization of sales and market share position. Indeed, one characteristic of the modern corporation which repeat edly has been observed is that the corporation often seems to place a higher premium on sales or market position than on maximizing profits. Some observers would even go so far as to say that sales maximization has supplanted profit maximization for most large pro fessionally managed corporations. Whether or not this extreme de parture from past behavioral norms is true, it seems reasonably certain that modern corporations have a somewhat w7ider and more complex set of objectives than the simple profit maximizing entrepreneur of traditional economic theory. In terms of investment planning this shift of objectives, taken to gether with a preference for internal financing, gives rise to a pattern of behavior in which investment apparently is initially geared to the level of internally available funds. Departures from this first ap proximation occur mainly in the previously cited competitive, rapidgrowth, and public utilities sectors. Such a hypothesis strongly sug gests that variations in corporation investment outlays will be rather closely geared to variations in corporate retained earnings, particu larly when the economy is not experiencing strong expansionary forces and existing productive capacity is more or less adequate to meet demand. Very recent experience does not seem to contradict such a hypoth esis. During the last 8 months, actual and planned investments have risen as corporate profits have risen. Indeed, it seems highly probable that investment outlays on plant and equipment will next year be very close to or perhaps slightly in excess of those of the 1957 investment boom. Moreover, the current tightening in money proba bly will have only a slight and then belated influence on this invest ment, just as in the 1956-57 period. This follows from the effective insulation of large corporations from the monetary markets. A strin gent monetary policy may have a lagged influence on large corporation investment because of the eventual backing up of reduced activity, induced by tight money, in other sectors of the economy. For exam ple, if, as seems highly probable, the main effect of tight money is to reduce housing starts, public investments, and small business invest ments, the large steel, consumer durable, lumber, and building equip ment corporations serving the housing, public construction and other interest-sensitive or affected sectors, will eventually have a reduction in their own sales and an accompanying reduction in their profits which will eventually retard their investment outlays. At the present time, therefore f rising corporate profits would seem to imply a rising level of corporate plant and equipment in- EM PLOYMENT, GROWTH, AND PRICE LEVELS 2489 vestments, unless the steel strike is unduly prolonged and basically alters current trends. This increase in plant and equipment invest ment may largely offset the decline in housing and public construc tion that seems to have been induced by present tightness in the money markets. Assuming that consumption holds up as well as it has in the immediate past, this would imply that business activity would be at a reasonably high level for the next 12 months at least. I f tight monetary policies persist, however, at some point beyond the 12-month period reductions in housing starts, public construction, and small business investment may begin to be felt in large corpora tions, sales and profits. Barring strong offsetting Government ac tion, or some other unforeseen stimulus to the economy, it seems prob able therefore that we might have a repetition of the 1957-58 type of downturn late in 1960 or early in 1961. The effects of any new upsurge in plant and equipment spending need not necessarily be inflationary. The economy’s capacity to sup ply capital goods was increased, just as most productive capacity in the economy was increased, during the investment boom of 1955-57. Furthermore, while investment outlays next year may wTell reach the 1957 boom level, they seem highly unlikely to exceed it. Accord ingly, the productive capacity in capital goods should be sufficient to meet most capital goods demand expeditiously if not easily. Some evidence exists, moreover, that the pressure on prices during the 1956-57 period was attributable to rather arbitrary increases by large producers selling so-called oligopolistic markets. These price increases have been described as “administered price increases’5 attributable to “underutilized monopoly power.” An interesting question from the standpoint of economic analysis is why such possi bilities to increase profits were not utilized earlier. That is, a need exists to explain wxhat compelling motive suddenly made the “reluc tant oligopolist” undertake full exploitation of his market possibili ties. For at least some large firms during the 1955-57 boom, this com pelling reason may have been a desire to continue or maintain speci fied investment plans without recourse to external borrowing. Thus, as the excess demand in the investment goods sector generated price increases for capital goods, some of the purchasers and potential pur chasers of investment goods may have been tempted to increase their own prices in an attempt to “pass along” some of the increased costs of their investment programs. Certain comments of steel industry spokesmen, especially when making a case for accelerated amortiza tion of steel facilities, would suggest that that industry fell into this category. Such policies could obviously spread price inflation developed in one bottleneck sector more widely than it otherwise would be. In essence, it suggests that during the 1955-57 period, many large cor porations may have tailored their internal flow of funds or retained earnings to the levels of their planned investment outlays. This is, incidentally, just another example of how large corpora tions might defer the impact of tight monetary policies. At the present time the extent or prevalence of unutilized monopoly or oligop olistic power for price increases is probably somewhat less than it was in 1955. Furthermore, the existence today of more modern and larger productive capacity suggests that the desire to maintain investment 38,563— 59— pt. 8------- 4 2490 EM PLOYMENT, GROW TH, AND PRICE LEVELS outlays at all costs and under all circumstances is probably somewhat less than during the 1955-57 boom. Similarly, the productivity effects should not be as undesirable dur ing the next 12 months as they were during the 12 months from mid1956 to mid-1957. The new investment boom, while it may be at the same absolute level as the previous one, will be superimposed on a larger capital goods base and therefore will pose relatively fewer “digestion” and learning problems. Furthermore, many of the pro ductivity increases made possible by the 1955-57 investment boom are still being realized and these should help offset any new “indiges tion” created by the current resurgence in capital outlays. Speculation about long-run price trends is clearly far more uncer tain and problematical than for a shortrun period. In the first place, a good deal depends upon exactly what type of Government policies are chosen over the next few years. Not only must monetary and fiscal policies be predicted, but trends in regulatory, foreign trade, and tax collection policies would also have to be forecasted. Given the uncertainties of political prognosis and the large number of feasible possibilities, any forecast of long-run price trends would be a guess at best. There is also the difficult matter of knowing just how any price increases are to be measured over this longer run period. That is, in the short run technological and quality effects can be assumed to be relatively constant in their impact on price indexes. Over the longer run, especially in an economy and society changing as rapidly as the United States, the seemingly simple problem of merely meas uring price increases can become quite complex and pose several poten tially involvable dilemmas. Looking beyond 1961, serious questions for public policy might be posed by corporation investment behavior being geared to intern ally available funds. Essentially, this kind of investment behavior implies that the growth of productive capacity in our economy will be relatively more rapid in high profit sectors. High profits, of course, often characterize new and rapidly developing industries and to the extent that this is true the diversion of capital resources to those making large profits usually is desirable. On the other hand, high profits can also be attributed to oligopoly or related types of market power. Interfirm rivalry in such situations often neglects price competition in favor of service and new product development. Some of our large corporations now speak, in fact, of “creative de struction” in which the rate of physical obsolescence in many assets, both in producers’ equipment and consumer durables, is greatly ac celerated. The effects of this accelerated obsolescence are oftentimes highly beneficial and contribute to the progress and improvement of our economy. On the other hand, it is not obvious that this will always be the case; for example, it seems rather questionable that the recent reduc tion in the new design cycle for automobiles from 4 to 2 years is a highly desirable development from the standpoint of increasing basic economic productivity. To put the matter more broadly, there are signs that much of our postwar plant and equipment expansion may have been overconcentrated in certain oligopolistic industries. EM PLOYMENT, GROWTH, AND PRICE LEVELS 2491 This impression is only heightened by consideration of the many seemingly urgent investment needs that exist in other sectors of our economy. In the public sector there are the well known needs for urban renewal, schools, highways, and hospitals. Furthermore, many of our publicly regulated industries, particularly those in transporta tion, have enjoyed relatively low profits during the postwar period, their investment rates have been relatively low, and the existing margi nal return on new investments would seem to be comparatively high. Similarly, some of our more competitive industries, like textiles and the manufacture of some small household goods, seem to have been less capable of modernizing in the postwar period, which, in turn, may partially explain the decline in their relative ability to compete in world markets. In sum, while the exemption of a large proportion of the economy’s investment funds from the usual financial tests may not yet have resulted in any serious distortion in the allocation of capital resources, the potential for such misallocation would at least seem to exist. Representative W i d n a l l . Professor Meyer, when you speak about the public sector, are you speaking now about welfare programs ? Mr. M eyer . I am speaking not only of welfare programs just nar rowly defined if you mean what is usually cataloged under the head ing of “Health and W elfare.” I am thinking of the things mentioned this morning by the chairman, such as education, roads, recreation, subsidies of technological developments, which in large measure we carry on now under the defense program. Representative W i d n a l l . I am thinking particularly about this: Sometimes I think with your eye on the Holy Grail or cloud 9 up in the heavens as to what you want for the public, you can destroy the value of the dollar. Is not stability of the dollar most important in order to carry out all of these items in the public sector ? Mr. M eyer. Are you raising the problem raised this morning by Mr. Clark, that inflation will inevitably follow if we continue to tax away more than 25 percent of the national income? Note that he said at the end of his discussion this morning that it is strictly the percentage of taxation that interests him and not necessarily the level of expenditure. If you are raising this problem, I would disagree. I would point out, first of all, that I think that it is more the discrepancy between taxes raised and Government expenditures that counts insofar as inflation goes; that is, it is the size of the Government deficit that is important, and I think it is quite possible with growing national income and so on, that we can finance many programs of increased public expenditure without any serious increase in the tax rates. Furthermore, I am willing to concede, in fact, I would be more than quick to concede, that there are many areas of public expenditure that might be cut out where the marginal productivity is low. One might mention some water resource developments and have some questions about the agricultural programs. On the whole, I suspect our present expenditures on the farm program could be substantially reduced without reducing either the shortrun or the longrun productivity of American agriculture. Representative W i d n a l l . Thank you. I will come back to that later. The C h a i r m a n . Mr. Suits ? 2492 EMPLOYMENT, GROWTH, AND PRICE LEVELS STATEMENT OF DANIEL B. SUITS, PROFESSOR OF ECONOMICS,, UNIVERSITY OF MICHIGAN Mr. S u i t s . The purpose of these remarks is to explore the potential usefulness of econometric models as guides to public policy. The discussion will be divided into two sections. The first part will be devoted to a very brief description of what an econometric model is, and how it works, and examples of forecasts obtained from a model will be presented. The second part will be devoted to an examination of the advantages and shortcomings of the method. The science of economics can be variously defined, but for the present purpose it is useful to think of it as the study of the relation ships among a system of essentially measurable quantities: Prices, costs, incomes, savings, employment, and so on. These relationships derive from the complex behavior and interaction of millions of house holds, thousands of business firms, and hundreds of governmental units. Although, from the standpoint of the individual family or firm, behavior may appear volatile and unpredictable, taking averages over time and over larger numbers results in more stable relationships. In this light we can think of the economic system as, in principle, repre sented by a system of simultaneous mathematical equations. Unfortunately, however, a theoretically complete representation would involve millions of equations—surely hundreds for each house hold and firm. Moreover, each equation would have to be exceedingly complex to reflect the complexities of human behavior, and would in volve numberless variables. We have neither the data nor the computing facilities nor the time and resources to deal with such a vast system. To proceed at all we must simplify. We may treat the millions of households of the Na tion as if they constituted a single entity “the household sector” ; millions of individual incomes become combined into one great na tional income; millions of separate prices become one or two price indexes. Moreover the complex responses of human beings must be simplified so they can be represented by one or two simple equation forms, usually by linear approximations. When we have thus reduced the millions of equations and variables to a handful of approximate relationships we can refer to the histori cal data of the economy and by appropriate statistical methods obtain expressions in numerical form. The representation of the economic system by an actual set of numerical equations involves two kinds of variables. The one kind— the most interesting kind— are those which are determined by the actual operation of the economic system: income, consumer expendi ture, employment, and so forth. The second kind of variable are those that exercise a direct or in direct effect on the course of events but which are not, to any appreci able extent, immediately influenced by it in return. These variables are those whose magnitudes are historically given and change only slowly over time—for example, population— and those which are the result of political decisions: Public expenditures, tax rates, price sup ports, and so on. 2493 EM PLOYMENT, GROWTH, AND PRICE LEVELS In the system of simultaneous equations representing the economy the variables of the first kind represent the unknowns while the vari ables of the second kind are the givens or knowns. This means that a properly prepared and complete model can be used to forecast the level of economic activity by substituting values of the knowns in the equations and solving for the unknowns: Employment, production, saving, etc. Of course, in order to do this, the future course of public policy must be counted as known, and no econometric model can do better than the political acumen of the user permits. On the other hand, however, when alternative policy decisions are embodied in the equations as alternative knowns, the alternative solu tions enable us to compare the implications of the different policy steps. Used in this way, the econometric model can serve as a useful guide to wise policy selection. The research seminar in quantitative economics, a graduate training facility supported at the University of Michigan by the Ford Founda tion, has had 7 years experience in forecasting the American economy with an econometric model. The 20 equation model was originally prepared by Profs. Lawrence R. Klein and Arthur Goldberger and has since been revised several times. The forecasts were prepared for presentation at the conference on the economic outlook held at the University of Michigan each November, and thus constitute a public record of the actual operating success of a forecasting model. The performance is shown in the following table. (The table referred to follows:) Review of forecasts [Figures are in billions of current dollars] 1954 1953 Forecast A ctual Forecast A ctual Forecast 364.3 229. 9 57.1 i 165. 2 72.1 363.2 230.5 50.3 i 172. 8 71.9 361.8 237.8 52 2 i 170.' 4 74.7 360.6 236. 5 52.0 i 171.8 69.5 364.8 235.9 54.1 206.2 76.0 Gross national ^product_____________ _ _ Personal consum ption expenditure,. _ _ Gross private capital form ation_____ __ _ E m ployee com pensation_________________ N onform , nonw aee 2 incom e. _ _ _ __ 1956 G rossfnational p ro d u c t.._ __ Personal consum ption expendi tu re _______ _ _ _ _ _ _ _ . Gross private capital form ation. Em ployee com pensation_____ N onfarm , non wage 4 incom e___ 1955 1958 4 1957 Forecast A ctual 413.2 263. 3 71.8 235.1 93.8 414.7 267. 2 65. 9 241.4 85.3 Forecast A c tu a l3 Forecast 438.8 280.1 69.3 255.6 91.1 434.4 280.4 74.4 254.4 93.1 438.6 285.9 62.7 257.8 90.0 A ctual 437.7 290.6 54.4 253.8 92.6 A ctual 390.9 254.0 60.6 223. 2 78.5 1959 fore cast (1958 prices) 456. 7 295.4 61.2 261.0 103.0 1 P rivate sector only. 2 D ue to conceptual differences the figures for nonfarm , nonw age income are not com parable to D ep art m ent of Com m erce estim ates for years before 1957. 3 Based on estim ates published b y the D epartm en t of Com m erce prior to the revision of Ju ly 1958. 4 Both forecast and actual figures have been revised to conform to the revised series issued b y D epartm en t of Com m erce (July 1958). The table shows a creditable record of performance. The direction of movement in the economy was correctly forecast in each case, and the actual level was predicted fairly accurately. In the case of the 2494 EM PLOYMENT, GROWTH, AND PRICE LEVELS recovery of 1955, however, the model’s performance greatly under estimated the movement of the economy. The model’s forecasts of the GNP as a whole is consistently more accurate than its prediction of the components. The forecasts of private gross capital formation were uniformly the poorest of the figures. The C h a i r m a n . Do I understand the forecast was made in Novem ber for the ensuing calendar year ? Mr. S u i t s . Yes, Senator, that is correct—for the ensuing calendar year. It is clear that while an econometric model gives some promise of success its use is limited. There are many sources of error involved in its construction. In addition to the heroic simplifications already mentioned, we also face the difficulty that at many important points our knowledge of the behavior of individual spending units is entirely inadequate. We do not know, for example, exactly how fast families adapt their living standards and expenditure patterns to changed conditions. Nor are we at all clear as to the factors which influence business in vestment decisions. The greatest need in economic research today is not the study of relationships among global quantities—we have come about as far in that direction as is feasible with our present knowledge. What is required is the study of the details of daily economic life by the careful observation of families and firms. No econometric model is sufficiently accurate or reliable to serve as a sole guide to policy, even on a fairly shortrun basis. And the com pounding that can occur when we turn our attention to the longrun projection of the economy raises the most serious questions about the usefulness of the method in this context. Nevertheless, at the present stage of development the econometric model is a useful adjunct to other methods of analysis and forecasting. If it is nothing else, it is at least the most sophisticated method we now have at our disposal for disentangling the implications of actual or prospective economic events, and for keeping our speculations mutually consistent; potentially it is much more than that. The C h a i r m a n . In looking at the table, is there a possibility that you may understate the gross national product ? Mr. S u i t s . Our forecast for 1959 ? The C h a i r m a n . Yes. Mr. S u i t s . Yes, I think it is greatly understated but I am not sure of the magnitude of the error. The C h a i r m a n . The Economic Indicators on page 2 give the 1958 numbers as 465.75 and 477.8 for the first and second quarters; whereas, your figure for the whole is 457.7. Mr. S u i t s . That is right. Representative W i d n a l l . On the 1955 forecast, looking back at it now^, what do you think was the basic reason for the difference in opinion as to your forecasting and as to the actual ? That is a very marked difference between 364.8 and 490.9. Mr. S u i t s . We thought in retrospect that the difficulty had been in the equation used to forecast gross private capital expenditure. At EM PLOYMENT, GROWTH, AND PRICE LEVELS 2495 that point we revised our model in an effort to improve our forecast of gross capital formation. Representative W i d n a l l . What do you think caused your own fail ure for private capital to go in and do that ? Mr. S u i t s . Our work has been an experiment in the feasibility of using the statistical tools which we now have—together with the guesses which we can make about what the Government is going to do—to forecast the course of the economy. This is a statistical pro cedure. The judgment which we exercise in making a forecast is essentially a political judgment. That is, we try to guess what the course of gov ernmental expenditure will be over the coming calendar year. This means extending the current fiscal year and deciding what will happen during the next period, and so on. If our governmental expenditure and tax extrapolations are reasonably accurate, as they were in 1955, then the failure of this forecast means that either the basic theory on which the equations were based or the statistical procedures which were used in arriving at them were at fault. We then try to find a better theory to see if we can improve. Representative W i d n a l l , Is there not a basic element in this which never shows up on a chart and that is the confidence of the American people, as to their own emotional impulse as to whether the}7 will or will not spend ? Mr. S u i t s . Surely, this is a very important point and it is not in cluded in our equations. Representative W i d n a l l . I do not know how you could. Mr. S u i t s . I would not agree that it cannot be shown on a chart. My colleague, Professor Katona of the Institute of Social Research at the University of Michigan, has as his specialty the investigation of the ebb and flow of consumer attiudes, and I think if he were here he would applaud your comment. However, as you pointed out, this factor is not included in our equation. Representative C o f f i n . Could it be ? Mr. S u i t s . It could be and ultimately must be, but as yet there has not been enough historical material. The C h a i r m a n . Have you attempted to forecast the amount of un employment ? Mr. S u i t s . Yes; we have, Senator. Unfortunately, I have only the forecast for 1958 and for 1959 with me. We forecast a level of unemployment of 4.8 million persons during the calendar year 1958. This was forecast in November 1957 for the calendar year 1958, 4.8 million persons. The preliminary figure—our forecast for 1959—3.4 million. The C h a i r m a n . The average for the year ? Mr. S u i t s . The average for calendar year 1959. The C h a i r m a n . Could it be very much above that ? Mr. S u i t s . I t h in k so. The C h a i r m a n . The record refers to 8 months; 4.7, 4.7, 4 .4 equals 3.6, 3.4, 4.0, 3.7, 3.4. Mr. S u i t s . I think that is correct. Representative W i d n a l l . Will not that material be materially af fected by the steel strike that fans out throughout the economy affect ing other than those on strike ? 2496 EM PLOYMENT, GROWTH, AND PRICE LEVELS Mr. S u i t s . It is conceivable that this would be so. I am not suffi ciently familiar with the way in which the data are prepared to know whether there is an adjustment for this. I would not like to claim that the discrepancy between the actual observed figure and the one which I have forecast is due to such an unforeseen event. The C h a i r m a n . Certainly it would not cause the difference in the first 7 months before unemployment occurred ? Mr. S u i t s . N o; surely not. M r . M e y e r . T h e effect w o u ld be ju s t th e rev erse w h e re th e r e w e re in v e n to r y b u ild u p s a n d so o n . Mr. S u i t s . I would like to say in anticipation—I did not like to declare the extent of error in a forecast before the year is actually over— but in anticipation of a substantial error both in the GNP and the level of employment, we are currently revising our model again. The C h a i r m a n . Did you hear Dr. Clark, who spoke this morning? Mr. S u i t s . Yes. The C h a i r m a n . I do not want to ask you to give away trade se crets, but would you agree with his forecasts on gross national product for 1960? Mr. S u i t s . As to what my model would show compared to his, I cannot say because we have not completed our calculations for 1960. The C h a i r m a n . Do you make your estimates public in November? Mr. S u i t s . Yes. The conference on the economic outlook occurs in the first or second week of November. The C h a i r m a n . I wonder if the staff could get those figures at that time. I don’t know that we should hold up our publication until that time but we could publish those results at some point. You have made a very modest statement and a very scientific state ment. You do not claim any more for your methods than are justi fied and you pointed out the weaknesses. It would be marvelous if we did have accurate forecasts at this time because one could adopt some degree of public policy as well as private policy. (The following was subsequently received for the record:) T he E c o n o m ic Outlook for 1960 a s F o recast b y U n it e d S t a t e s an E c o n o m e t r ic M odel of t h e (By Daniel B. Suits) In earlier testimony before the Joint Economic Committee I indicated the general nature of an econometric model and the way it is used in economic fore casting. At this time I will present the analysis of the outlook for 1960 by use of such a model. The model employed has been newly compiled by the staff of the research seminar in quantitative economics of the University of Michigan, and involves several improvements over earlier versions. In particular the new model involves separation of aggregate consumer expenditure in four categories: expenditure on new and net used automobiles, expenditure on other durable goods, expenditure on nondurable commodities, and expenditure on services. The forecast was prepared for and presented to the Seventh Annual Conference on the Economic Outlook, sponsored by the department of economics at the University of Michigan, November 12 and 13,1959. REVIEW OF 1959 The 1959 forecast is reviewed in table I and compared with the preliminary estimates of the actual outcome for the year. It is important to note that these are prepared from data for the first three quarters only, the third quarter being given double weight. Since the fourth quarter will doubtless be heavily influenced EMPLOYMENT, 2497 GROWTH, AND PRICE LEVELS by the steel strike, the final figures for 1959 may prove to be closer to those fore cast than the preliminary data indicate. The forecast for 1959 of an increase over 1958 of $15 billion (1958 prices) in the gross national product was in the right direction, but was $15 billion short of the preliminary level of $472 billion actually attained. The level o f consump tion expenditure was underestimated by about $11 billion and gross private capital expenditure by $8 billion with an offset of $4 billion in the foreign balance. FORECAST OF 19 60 The forecast of the outlook for 1960 is prepared on the basis of assumptions listed in table II. These represent what appear to be the best judgment at the present moment of experts close to the scene. The assumption of an increase of $3.7 billion (at 1954 prices) in private expenditure for plant and equipment is based on the response of businessmen to the McGraw-Hill Survey of Plans for Capital Spending in 1960-61. (The actual figure differs somewhat from that actually obtained from the McGraw-Hill Survey because of differences in concept.) All computations for the forecast are made in 1954 dollars, and are presented in that form. The detail of the resulting forecast is shown in table III. The important con clusion to be drawn is that, despite an increase in gross national product of $9 billion (1954 dollars), amounting to about $10 billion of current purchasing power, the outlook is still for a rise of 700,000 in the level of unemployment. In other words, the forecast increase in output is at most sufficient to absorb the increase in productivity without increasing the level of employment, not to say reducing the level of unemployment. In order to prevent the level of unemploy ment, from rising would require a substantially larger increase in total output, roughly double the prospective increase, while the increase would need to be three to four times as large as the prospective figure to restore a level of full employment (i.e., to reduce unemployment to a fractional level of, say, 2.5 m illion). The outlook for 1960 is, of course, somewhat clouded by the effects of the steel strike. By reducing inventory stocks below the expected level and postponing production of durables from 1959 to 1960, the result may give 1960 a statistical appearance of being better than it really is, at the expense of 1959. Even under these circumstances, however, it is doubtful that the increase in output would be adequate to restore full employment. T a b l e I.— Detailed review of 1959 forecast Forecast (1958 dollars) Gross national prod uct___ _ _ _ __ -------_.. ______ Personal consum ption expenditure . _ _ _ _ _ .... __ Gross private capital expenditure. _ _ _ _ _ _ _ _ __ R esidential construction __ _ _ P lan t and equipm ent _ _ _ _ _ __ __ _ ______ ___ __ Increase in inventory __ _ ____ G overnm ent expenditure on goods and services plus foreign balance _ E m ployee c o m p e n sa tio n ___________ _ _ _ __ ____ __ __ __ C orporate profits _ _____ __ __ _ _ _ _ _ C ivilian em ploym ent __ _________ __ _________ _ ___ U nem ploym ent __ . __ _ _ _ __ __ ____ 456. 7 295. 4 61.2 17.8 44.0 - .6 100.1 261. 0 47.7 65.9 3.4 A ctual 1 (1958 dollars) 471. 8 306. 0 69. 9 22. 1 43.3 4.5 95.8 273.8 47.6 65.6 3.7 E rror 15.1 10.6 8.7 -4 .3 1 All figures prelim inary as of the 1st 3 quarters of 1959. In calculating these prelim inary estim ates no* allowance has been m ade for the possible effects of the steel strike. 2498 EM PLOYMENT, T able GROWTH, AND PRICE LEVELS I I .—Assumptions for forecast of 1960, increase over 1959 [Monetary values in billions of 1954 dollars] Financial: Aaa bond yield (percent)__________________________________________+ 0.36 Automobile credit terms___________________________________________ (a) Government: Government expenditure, goods and services : F ederal_________________________________________________ 0 State and local__________________________________________$1. 8 ------ '+$1. 8 Government wages________________________________________________ +$1. 0 Wages paid by government enterprises-------------------------------------------- + $ . 2 Government employment (millions of people)_______________________ +. 1 Labor force____________________________________________________________ + .6 Exports________________________________________________________________ C) O ther: Other labor income----------------------------------------------------------------------- + $ . 9 Supplements to wages and salaries-------------------------------------------------+$1. 8 Value added, households___________________________________________ + $ -7 Private expenditure, plant and equipment__________________________+$3. 7 New model stimulus, automobiles_________________________________ -j-$2 .0 1 Unchanged. T able I I I .— Outlook for 1960 [1954 prices] 1959 i 'C onsum ption expenditure: _ _ ______ ______ _________ __ ______________ ______ A utom obiles __ O ther durables _ ______ _ _ __ _________ - _ _ _ _______________ _____ N ondurables _ ____ ____ ______________ - __ _______ _____ _____ Services _ _ __ _____ - ______ _____ __ ______- __ . _______ __________ Gross p rivate capital expenditure: P la n t and equipm ent ____ __________________________________ _______________ R esidential construction _______________ _____ ___ __ - - - - - - __ Increase in inventory __ __ _ _ _ _ _ _ ______ ______________ __ ___ International: E xports _ ___ - __ _________ _________ ____________ I m p o r ts __ ___________ ________ _________ ____________________ ____ ____ G overnm ent _ _ ___ __________ - _ _____ - __________________________ Gross national p ro d u ct_ ___ ______ _ _. __________ ____ ___________ _ ______ P roperty incom e _ ___________ ______________ _ _ ____ _____________________ C orporate profits _ ________ ___ _ _ __ __ __ __ __ _______________ _ D ividends _ _ _ _ _ ____ _______ ________ ________________ ___________________ C ivilian e m p lo y m e n t_____ ____ ______ __ _____ _____________ ___________ _____ U n e m p lo y m e n t--____________________ - __ ______________________________________ 15.7 25.0 136.2 104.7 36.8 19. 9 4.1 23.7 24.5 81.9 423. 5 109.0 43.0 11.7 65.6 3.7 1960 16.7 25.2 138.9 106.3 40.5 19. 7 2. 2 23.7 25.0 83.7 432.0 108.4 42.7 12.2 65.5 4.4 1 P relim inary based on 3 quarters. Mr. S u it s . I often think, given the difficulties, the wonder is that it works at all. The C h a ir m a n . Dr. Clark, I believe, has 87 terms in his equation. How many do you have ? Mr. S u it s . I think a better comparison is in terms of the number of behavioral relations because, of course, one can multiply defini tions indefinitely. One can break the GNP up into finer and finer parts and relate them together. A better comparison is the number of behavioral equations which must actually be estimated by statistical means. There are 15 in our model and in his, if I remember, there are 29. Of course I have never seen the actual equations for his model. The C h a i r m a n . Do you publish your equations ? Mr. S u it s . Yes indeed. The C h a ir m a n . Do you justify the inclusion of various series? EM PLOYMENT, GROWTH, AND PRICE LEVELS 2499 Mr. S u i t s . Yes. The parent model from which this is derived is the Klein-Goldberger model which was published in a volume en titled “An Econometric Model of the United States,” by Klein and Goldberger, and published by North Holland Publishing Co., Amster dam. I should be surprised if there is not a copy in your staff library. Our subsequent revisions of this model have also been pub lished but they are in much-abbreviated form and depend upon fa miliarity with the original for their understanding. I think you will find the Klein-Goldberger volume is perhaps the best exploration and justification for an econometric model that has ever been published. The C h a i r m a n . In November 1956 you forecast the downturn in the summer of 1957 ? Mr. S u i t s . Not the summer of 1957. We forecast only the average for the year. We have never attempted to-----The C h a i r m a n . But it did show some downturn in 1957; did it not? Mr. S u i t s . N o ; the forecast was approximately correct for 1957 and that was still slightly above 1956. The downturn began about the third quarter of 1957. The C h a i r m a n . Then do you count that as a hit or a miss? Mr. S u i t s . Our criterion is the calendar year, and whatever the calendar shows is what we go by. A given average can involve either a rise or a fall or both. The C h a i r m a n . According to this, 1957 was going to be a better year than 1956. Mr. S u i t s . It w a s. The C h a i r m a n . If you take the year as the basis, then people listen ing to your forecasts in November would have said everything is going to be all right but then the bottom began to fall out. Mr. S u i t s . Let me say you would have gotten the same information then, that you will get in retrospect now by looking at the average for the year. It is certainly not true that one gets highly useful information by knowing only what the average will be. The C h a i r m a n . Again, referring to 1929, at the end of 1929 the journals all published articles showing that 1929 was one of the most prosperous years on record. For days the financial pages were full of what a marvelous year 1929 was, and this would be in December 1929 and January 1930 and therefore they argued by inference that the decline which began in October was all an illusion. It was not an allusion, I assure you. Was the decrease that began in 1957 an allusion? What answer would you make to that? Mr. S u i t s . I would say if we had the ability to conduct our analysis on a shorter period basis, say quarterly, it would be much more useful to everybody who uses it, that there is no denying that any kind of average figure—and it makes no difference whether it is average GNP or average hourly earnings—can be greatly misleading if it is not handled with care. The C h a i r m a n . Mr. Clark makes his forecast in terms of quarters. M r . M e y e r . W h i c h , f o r p o lic y p u r p o se s, is p r o b a b ly m o r e u s e fu l. Mr. S u i t s . H o w accurate his forecasts are, I don’t know. We are on record, and I would wait with eagerness to see how he turns out. 2500 EM PLOYM ENT, GROWTH, AND PRICE LEVELS Representative C o f f i n . I understand that Klein is now engaged in designing his new model so he can work on a quarterly basis. Representative W i d n a l l . Have you attempted to forecast gross na tional product and unemployment for 1960 ? Mr. S u i t s . No, sir; we have not done so yet, We will make our presentation before the Conference on the Economic Outlook early in November. I have forgotten the exact date, but I have informed your committee staff. Representative W i d n a l l . I understand that it is all on a yearly basis so you would not have anything on the first quarter or second quarter. Mr. S u i t s . That is entirely correct. Representative W i d n a l l . Professor Meyer, do you feel that interest rates are tighter than they need be ? Mr. M e y e r . That is a very difficult question and any answer would clearly reflect some personal viewpoints and value judgments. On the whole, at this point I would say that I think that the Federal Reserve Board’s timing has been somewhat better than it has been in some preceding and recent history. My reason for this would be that there is clearly, I believe, a fairly strong upsurge underway in investment outlays by manufacturers and other corporations. So, if there was no offset to this increased activity in the capital goods sector, we clearly might have some strong inflationary demands; de mand—pull, if you wish—conditions. To the extent, therefore, that the Federal Reserve policy of high interest rates and tight money works as it seems to have worked in the recent past, by cutting back activity in housing, private and public construction, and small busi ness investment, then tight money may work out now to be a timely offset, a timely offset to what is going on or beginning in private plant and equipment expenditures. Was there a second half to your question? What was the other haif? Representative W i d n a l l . It is a big one, and I know you cannot answer it in 2 minutes. Are they so tight as to reduce savings and capital formulations ? Mr. M e y e r . I should not think that a high interest rate would re duce savings. It might not have much influence on savings, but I would not expect it to be in the direction of reducing savings. As far as the capital formation goes, I would expect the reductions to be in small business, housing, and certain public investments and not in private investments in plant and equipment. Representative W i d n a l l . What does your own study reflect? If the investment rates fall, would the demand also fall ? Mr. M eyer. The point I am making is that we will have this offset to tight money from producers plant and equipment expenditures so that the economy will continue to be at a fairly high level through the next year. In general, I would accept the trends implict in Professor Clark’s estimates, though I might not be quite as bullish as he is, so I suspect the Federal Reserve Board will feel few compunctions to change their policy or reduce interest rates in the next few months. Representative W i d n a l l . Suppose congressional action should at tempt to enforce a lower interest rate. Do you think this will have an adverse or bullish effect on the economy ? EM PLOYMENT, GROWTH, AND PRICE LEVELS 2501 Mr. M e y e r . This gets down to the problem of how you weigh greater employment against price effects. It would be my guess that if Congress were to force a lowrer interest rate and an easier money policy on the Federal Reserve Board, you would probably put greater pressure on prices. On the other hand, you would probably reduce unemployment be low present levels, and here you get into the question of how you value these two developments. I don’t think as a private citizen my value judgments would be any better than any one elses. Representative W i d n a l l . It might create fuller employment, but as far as those on fixed income, it is greater with respect to their own standard of living. M r . M e y e r . It is a matter of how you evaluate these conflicting de velopments. I can see that we w^ould like to have the best of all pos sible worlds, where we would have lower unemployment than we now have and, at the same time, not have any serious erosion of fixed in comes. But given the present array of policies available to public agencies and semipublic agencies it is not clear to me, however, that we can readily achieve these twin objectives in the immediate future. I would agree, however, with Mr. Christ, that I do not think they are incompatible in the long run and with a little bit of thought or a little bit of imagination about the design of better policies. The C h a i r m a n . That is what we are trying to supply. Representative W i d n a l l . I would like to address this question to the panel. I have been here long enough to know that there are some very sincere people down here in Congress regardless of political complexion wTho believe in approaching the same objective from dif ferent directions. Do you have any fairly fixed opinion as to the first things first? Does stability of the dollar mean more in the general economy than possibly starting off on some new program that can capture the im agination of the people and can be very w orthw hile as to objective, like taking care of the older people here, or taking care of the unem ployed there, or recreational facilities? This is a tough question. We have to face it in the Congress, but I think you have really faced up to it yourself in your own studies on it. Where do you think in the overall picture the best contribution could be made affecting more people in the ultimate welfare of the United States through a fairly stable dollar and then going after the objectives, or going after the objectives and forgetting about the stability of the dollar. Mr. M e y e r . I would repeat again, it is not obvious to me that in creased public expenditures on some of the more urgent public needs, such as education, defense, and highways and so on will necessarily lead to increased inflation. It depends on how it is financed. This is what I think is the missing point in Professor Clark’s analysis. You could do these things by diverting resources from other uses. Representative W i d n a l l . When you say by diverting, are you ad dressing yourself to a balanced budget in the end or facing up to expenditures by increased taxation to meet the expenditures ? Mr. M eyek. I would not necessarily to opposed to facing up to the implications of increased taxation if this were necessary to finance certain public programs. I certainly would not like paying addi 2502 EMPLOYMENT, GROWTH, AND PRICE LEVELS tional taxes any more than anyone else but if I were convinced that the long-run productivity in defense and growth of this economy and society wTere dependent on such action, I would be wTilling to take such action. I am not sure it is that desperate. I am just saying I would be willing to face up to the question. Representative W id n a l l . Would either of you care to comment on that ? Mr. S uits . I think that is actually the essence of the matter. There is 110 contradiction between price stability and the accomplishment of any objective, legitimate or otherwise, that we care to set for our selves ; the essence of the matter is that we must face up to the actual problem of financing such projects in a way which will not con tribute to inflation, and this can be done. So, if we think for exam ple that additional hospitals or additional schools or additional teachers are in the public interest, they can be acquired providing we simply face up to the fact that we can have them only if we are willing to pay for them. In other words, so long as we are willing to calculate the cost of the particular project and to provide the tax funds to cover it, I see no contradiction at all between stable prices and accomplishing our public aims. Mr. M eyer. In addition, I think we should exhaust every possible means for more efficient government. I do think there are several areas where money can be saved in governmental operations. Representative W id n a l l . Do you have any suggestions along that line ? Mr. M eyer. I am afraid I would join the majority of economists in opposing the minority lead by one of my colleagues in suggesting that the farm program is an excellent place to begin. I would say that there are several places in the water and river development programs and perhaps minor possibilities in post office operations, where, too, you might go over to a user cost taxation scheme. The C h a i r m a n . Such as increasing the rates on second-class and third-class mail, making the newspapers and magazines pay a little more, as well as direct mail advertisers. This is what many of us have believed for years. We have been trying to get it but instead the administration wants to increase the rates on first-class mail which is not only meeting its costs but giving a surplus. The paradox is the weaklings who are getting the larger subsidies themselves denounce subsidies to others more vehemently than any group in society. Representative W id n a l l . A s I recollect the recommendations of the administration, they wanted to up the second- and third-class mail more than Congress finally agreed upon. They also wanted to increase first-class mail more than Congress finally agreed upon. They also wanted to increase first-class mail more than it was increased. One thing I must say I admire Canada for, at the time they were showing a surplus in the Post Office Department, on 4-cent mail they increased to 5 cents so they could have more money and provide better service all the way through the system. I think that is a healthy thing and I would like to see it done in our system. The C h a i r m a n . The very papers that were denouncing subsidies sent their lobbyists down here to prevent an increase in second-class rates. EM PLOYMENT, GROWTH, AND PRICE LEVELS 2503 Representative W i d n a l l . That is n o t u n u su a l. The C h a i r m a n . It is not unusual but I think it is highly antisocial. Mr. C h r i s t . I would like to speak to Representative WidnalFs earlier question concerning what should get the first priority, main taining a stable dollar or undertaking certain programs that we feel are urgent. I think that the best way to look at it is this: In our country we have available to us a certain amount of resources in the form of minerals in the ground, in the forests, and the people and their skills and the plants and roads and other productive equipment, and these things can be turned to a very wide variety of uses and it is in the power of the country to decide what uses ought to be made of these things. One of the really important questions, I think, for any group of people who have power over economic decisions—this includes an individual for his own purposes, and it includes the Congress for our collective decisions—is what are the things that we really think are the most important to produce with our resources and this is one of the important questions of the day. Having decided this, or perhaps before having decided it, there is a separate question, which is : Shall we finance what we do in such a way that prices rise gradually or rapidly over a period of time, or shall we finance what we do in such a way that prices stay stable ? These are two separate issues. You could, if you wanted to, abolish the schools and still have inflation if you cut taxes enough or printed enough extra paper money and gave it to people and let them spend it. The thing that creates inflation by and large is very large at tempts to spend money to buy more resources than the economy has. If we can keep the total demand for goods from exceeding the total amount of goods that our economy can produce at today’s prices, then we will be able to have stable prices. I think these twTo questions are both important but we can attack them both at once, we don’t have to put one aside in order to handle the other. Representative W i d n a l l . If you do not preserve a certain amount of goods like out of the farm surplus, where it is not playing 011 prices to the extent that it should, there you have a false ceiling or false price that certainly would not reflect consumer demand and supply. Mr. C h r i s t . The farm program holds prices up by having the Federal Government compete with consumers in buying the products of the farm. Mr. M e y e r . We certainly should not say this is the only sector of the economy where artificial impediments stand in the way of market forces. Representative W i d n a l l . Do any of you have any comments on the testimony this morning? Mr. C h r i s t . I was not present. I had planned to be, but by the courtesy of Capital Airlines I barely made it this afternoon. Representative W i n d n a l l . Dr. Clark says he felt that as a result o f the tax program you have the increased taxes to increase profits. That would not be so strong if you lowered the income tax and if you lowered other taxes so that there was, say, a ceiling of 25 percent on all taxes throughout the economy—he felt that this was a greater pressure than labor leaders or management, looking for profits for 2504 EM PLOYMENT, GROWTH, AND PRICE LEVELS the stockholders, and all the way through it contributed more to an inflation in prices than any specific labor pressure or capital pressure. Do you have any comment to make on that ? Mr. C h r i s t . I would be glad to make a comment. I was not here this morning, as I said, but I have been aware of Mr. Clark’s position on this point, He has made statements of this kind several times in recent years. I don’t agree with what he says. As Mr. Meyer said a few minutes ago, it is important to look not only at the taxes which are collected but at the expenditures that the Government makes. Con sider, for example, the way the United States would be under three separate economic policies. Under these three there are taxes at about the level we have now. The first alternative is the United States as it is today with today's expenditures and taxes. In the second example, consider what the United States would be like today if we had these taxes but no Government expenditures at all and the Gov ernment simply took the money and filed it away in the Treasury Department and locked it up and did not spend any of it. In the third, suppose the taxes were the same as they are now but Govern ment expenditures were twice as large as they are now. I think you see where this is going. If there were no Government expenditures, the taxes would repre sent a heavy burden on the incomes of the economy and there would be no expenditures to offset them and this would produce a very large depression. If we were to double Government expenditures with no change in taxes, this would produce a very large inflation. I think we could have an operating economy with taxes at today’s levels or lower or substantially higher and still have stable prices provided that we can balance the demands on resources against the supplies or resources that are available. Representative W i d n a l l . Thank y o u . Could I have your comments on that, too, Mr. Suits ? Mr. S u i t s . I agree essentially with what Mr. Christ has just said. I would like to make just one or two additional points. The first has to do with the data which Professor Clark used in presenting his con clusion. As the chairman pointed out this morning, the data on price increases is of a rather strange sort in that it has involved in it a pro ductivity adjustment. If an expenditure succeeds in raising pro ductivity, let’s say five times, and if prices do not decline correspond ingly, the result in Professor Clark's data, if I understand it correctly, would be an indication of a price increase, _ In other words, instead of merely showing what happens to the pur chasing power of a particular dollar when certain expenditures are made, this involves adjustments that fail to reflect the productivity effect of the expenditures themselves. As the chairman pointed out this morning, education helps in raising productivity. But if prices don’t decline just as fast, then the education expenditure seems to be raising prices even though we are all better off, and the actual buying power of our incomes has gone up. The second point is this: There are a few other countries and a few other times that I would like to call attention to. Germany, after World War I, Hungary after World War II—in the Hungarian in flation after World War II, the price index went from a figure of 1 in 1929 to a figure, if I remember correctly, of 10 to the 14th power, or EM PLOYMENT, GROWTH, AND PRICE LEVELS 2505 some such number. It was a number that only astronomers are fa miliar with. Of course that is an extreme example, but there is also Poland and Austria and Russia after World War I, and a number of others. If we examine those economies where prices were rising at an astronom ical rate, wre find a low proportion of taxes in ratio to income. The difficulty was that the proportion of expenditure to income was very high. The result was simply a force pump on the price system. Again, it is not the level of taxes which is the issue, but the level of taxes in ratio to what the Government is, in fact, spending. If we tax and are willing to tax sufficiently high, there is no reason why we can’t expend any reasonable proportion of our income. The C h a i r m a n . Of course, there are peculiar years. For instance, this last fiscal year in this country, where the deficit was $121/£ billion, but had substantial price stability during the year— and I think what happens then is that the increase of Federal expenditures made good a decrease in private investment, M r. M fig u re s ? eyer. C o u ld I m a k e one m o r e te c h n ic a l c o m m e n t on h is Just going through his list of observations quickly, it would appear to me he might have a bias because of a failure to control on the cyclical factor. In particular, there seems to be a tendency for the cases of high taxation and rapid price increase to be measured in terms of price increases over a span of years between a trough of a cycle and a peak of a cycle. This raises a question of whether he has an equal number of boom and an equal number of depression years represented in his various observations. One could also ask what is the mechanism by which Professor Clark’s deleterious effects of taxation are felt. Apparently, it is something like the following: That high taxes tend to destroy in centives to increase productivity, which would imply in turn that we should observe a relationship between productivity in his table and the tax rate, and it is not clear that such a relationship exists. Representative W id n a l l . I think the point he was making in con nection with it, though, was if you were earning X dollars and you knew that $52 w7as going to the Government regardless, you could still utilize some of that $52 that was going to the Government if you passed it out in wage expenses, business expenses, and the like, and they tend to create a flame in the economy that would not normally be created if you had a tax situation where there was more incentive to try to accumulate and fan it out in a different direction. I do not know if this is an exact analogy or not, but it is in the monopoly area of a barber for personal service, where he can arbi trarily increase his price from $1.50 to $1.75 to $2. You have to get a haircut so you have no choice in the matter. In this case the public is caught in the horns of a dilemma because the large corporation or the business is using tax structure to spend moneys in a way to keep from paying the Government taxes. This is the point I think he was trying to make. I do not know if I quite understood it but this is as I tried to grasp it, Mr. M e y e r . Perhaps he was trying to make two points. Your point would be that if you had a high level of taxation, you tend to reduce the incentives for capital accumulation as well as reducing, as 38563— 59— pt. 8— 5 2506 EM PLOYMENT, GROW TH, AND PRICE LEVELS Clark pointed out, the incentives to increase productivity, his refer ence to 2 days of golf rather than 1 day of golf, being symptomatic of this concern. This would mean that we should observe your reformation of his hypothesis and see if a relationship exists between taxation as a percentage of national income and rates of capital accumulation. He has not provided us with all of the needed data and I am not sure just from a quick glance if we would observe any pronounced relationship between these two series on a cross section basis for dif ferent countries. We might and we might not. I am just not certain. Mr. S u it s . I think there are two sides to the question of the in fluence of taxation on incentives. Since paying taxes is extremely unpleasant, it is easy for us to think in terms of its effect in dis couraging effort. On the other hand, to a person who feels that a certain level of life and a certain standard of consumption is suitable to him, an increase in the tax rate may actually induce him to work harder in order to maintain his living standards in the face of the extra taxes he has to pay. Surely, both of these things are found in our society. The ques tion is which, on balance, is the most important. Representative W id n a l l . At the same time, if you have a man on a fixed income, a pensioner, he has no way of avoiding the perils of taxation and while the person can utilize all these things, I have heard about a dramatist the other day who creates one play a year but he is being paid over a 15-year period. He used to be paid just for 1 year but he finds that 90 percent of his earnings go to the Federal Government so, therefore, it is paid over a 15-year period. Anybody on a $10,000 a year salary would love to have the same situation but still their brains and effort earning $10,000 a year salary— and I can’t quite see this differential which has now been given. To me, it destroys incentive. This man can produce 10 plays a year. He can do a better job if he is a genius than he is doing now. He becomes lazy in connection with it, and he is also trying to avoid taxation and successfully doing it. The C h a i r m a n . William Shakespeare never wrote 10 good plays a year ? Mr. S u it s . I would like to comment that while I do not know your friend, the playwright, I do know many university professors who are doing industrial consulting, writing and similar activities, when they, too, would much rather be out playing golf. They feel they simply need the income. In other words, to the extent that the tax burdens were removed some of them might go play golf on Wednesdays. The incentive effect cuts both ways. We have to recognize that it does and the question is, whose incentive does it retard and whose promote ? Representative W id n a l l . In theory it is the greatest taxation form in the world, but in practice it provides the greatest inequity. Mr. S u it s . The way the taxes are worked out in theory, of course, is an entirely different issue. The C h a i r m a n . I would like to try out three thoughts which occur to me and connect themselves with the demand of the administration. It seems to me that very frequently fiscal policy is taken for granted EMPLOYMENT, GROWTH, AND PRICE LEVELS 2507 and some say that not too much can be done about it and credit policy is virtually the sole source that you must use. The administration says that if we wish to restrain inflation, the proper policy is to raise the interest rate, while you are discouraging investment from the capital goods sector or diminishing the rate of installment selling and so forth, and when one says yes, but this wil.1 increase unemployment and the reply is made, well, you have to choose and we choose stability. The bank authorities from 69 countries are meeting here in tlii^ country right now. As I walked throug the Sheraton Park today,, the lobby was crowded, and they are virtually unanimous in support of this policy and similarly, financial writers and banks like it because it means higher interest rates for them. Forming in my mind has been the possible alternative to these possibilities. Some of us believe that we could by plugging tax loop holes, get approximately $3 billion added revenue to the Government without changing the basic rates. This would be done by closing the oil depletion allowance loophole, by withholding on dividends and in terest at the source; by repealing the dividend credit; and by stopping excessive business expense account deductions. We could also effect economies of approximately $3 billion govern mental expenditures. We could reduce farm payments by at least $1 billion, particularly that portion that goes to the big farmers who do not need the money, We could save another billion in the military without reducing combat efficiency by having contracts drawn more carefully and by better handling of surplus. We have a military stock in warehouses of about $42 billion. The military have conceivably more than they would need for any type of war, and then another billion that could be picked up on increased rates in second and third class mail, reducing business subsidies and so forth, so we would go get $3 billion on the budget and plug loop holes for a total of $6 billion. Then, this $6 billion could be devoted to the following four purposes in approximately even proportions: One quarter for greater combat efficiency in the military; one quarter for what we would call welfare expenditures; one quarter for tax reduction of two types— (a) reduction of the rate of taxation in the upper brackets from 91 down to, say, 70 percent with some grading down below this, and a reduction in excise taxes at the same time; and the final point, a reduction in the national debt which would be around a billion and a half a year. If a billion and a half a year in Government securities are retired by the Government; and if, in addition, the Federal Reserve in ex panding the circulating media, did so not by lowering the reserve ratios but by open market purchases; and if it did not confine its purchases to bills as it does now but also bought some longtime securities matters— say, about a half a billion a year—so you get a total of combined purchases of Government securities of around $2 billion a year—would not this lower the interest rate on Government bonds and hence, the general level of interest rates so it would not be necessary to raise the maximum of 4*4 percent on the longtime Gov ernments ? As the interest rate was lowered, the basic interest rate was lowered. 2508 EMPLOYMENT, GROWTH, AND PRICE LEVELS This would, of course, induce a greater volume of investment than would otherwise be the case and hence reduce unemployment. Mr. C h r is t . You have just written a book. There are many parts to what you have said, but to begin with the question about the interest rate ceiling—please understand me. I do not want the Congress to raise the interest rate. This is not the idea. I want the Congress to raise the interest rate ceiling, to take it away-----The C h a i r m a n . Just as practical men we know perfectly well if you increase the ceiling, the Treasury will increase the interest. Mr. C h r is t . Sure, they will raise the coupon rate right now, be cause the market yield is already high, but I think in 6 months or a year or 2 years interest rates will be lower. I am not in favor of hav ing high interest rates all the time but I am in favor of having inter est rates do what they must to maintain stability. The C h a i r m a n . Do what they must do if nothing can be done about fiscal policy, but suppose we could make these changes in fiscal policy with sufficient rapidity, would you still say you should raise the interest rate in order to check inflation and enable the Government to get adequate distribution of its security between short and long term? Mr. C h r is t . I think so, but let me for a moment try to examine your suggestion and see whether or not I understand. There are many parts with which I would certainly agree. I think we can stand to plug a lot of tax loopholes and I think it would be good to have the rates at the very high end of the income tax scale not quite so as they are now, and so on. There are no objections from this side of the table, I am sure, to improving the efficiency of many of the operations that you men tioned— and in your earlier book as well. But the main thing in your suggestion that relates to interest rates is this question of letting the Federal Reserve buy securities so as to raise bond prices in the market and thus, push interest rates down. The C h a i r m a n . That is only one-quarter of the total. That would be, say, half a billion a year but a billion and a half would be retired from the budget itself. Do you think I am advocating pegging Government bonds? I want to make it clear I am not. I am merely saying that as the money supply is normally expanded, it should be done by open market operations. Mr. C h r is t . Good. I agree with you on that. What I understand you to be suggesting here is that during a period such as the present one when interest rates are threatening to go high you would like to try to prevent them from going high by a monetary policy of buying bonds-----The C h a i r m a n . Three-quarters by fiscal policy. Mr. C h r is t . May I try to finish this sentence? I am sorry but I do not seem to say it in such a way that you can see that there is more coming. As I understand it, there are two parts to what you are suggesting: One is to use fiscal policy mainly for the anti-infla tionary effects; and the other is to use monetary policy with the other EM PLOYM ENT, GROWTH, AND PRICE LEVELS 2509 hand, so to speak, to prevent interest rates from rising during this boom time when it is necessary to try to restrain the economy. The C h a i r m a n . It is easy to come to that conclusion but that is not what I was proposing. I said we can expect a growth in the national income, and to get stable prices, then presumably the quantity of credit should be increased in approximately the same proportion. How will that increase in the quantity of credit take place? I can tell you what the Federal Reserve Board wants to do. They want to reduce the percentage of reserve requirements from the pres ent average of around 16, down to an ultimate of about 10. This would mean no longtime increase in the purchase of Govern ment bonds by the Reserve. I would say, Would it not be better for the Reserve to expand the circulating medium by open market opera tions rather than by lowering reserve ratios? and it does it at the present rate of 3 percent per year. This would be $5 or $6 million a year. They should not be limited to purchasing bill only, so that in addition to the billion and a half that the Treasury could retire, the Treasury could retain $500 to $600 million more. So the net volume of purchases of Government securities would be $2 billion a year. Wouldn’t that have an effect in lowering the interest rate on Gov ernments and, therefore, lowering the general interest rate and, there fore, expand the volume of investment ? Mr. C h r i s t . I am not sure about this, but I think the interest rates on Governments would end up about the same if the money supply were increased by permitting lower reserve requirements on bank re serves as it would if it were done through open market operations. It seems to me this is right because what we are going to do in total is to let the total money supply increase about the same in both cases. We are going to let it increase say, about 3 percent or 4 percent a year, or something like this, to keep up with real output of the coun try. This means that at a given level of prices people can have about the same amount of purchasing power held in the form of money in either case. If the Federal Reserve buys Government bonds on the open market, this is a form of putting money into circulation and taking securities away from private hands, not involuntarily, but the securities are removed from private hands and money is put in their place. The C h a i r m a n . They g e t paid for it and there is no confiscation. Mr. C h r i s t . That’s right. If banks expand their loans as they pre sumably will do if reserve requirements are lower, then something that is very nearly the same kind of thing happens. Money is put into the hands of the public and the public gives to the banking sys tem as a whole securities in return. Now, these don’t happen to be the Government securities that were outstanding. They happen to be newly outstanding promissory notes of corporations that happen to be borrowing from the banks. This is my first answer and I would like to think more about it before I take a stand about it. The C h a i r m a n . What about the billion and a half of purchases af fected through the Budget or the Treasury and not by the Reserve ? What would you say about that ? Mr. S u i t s . Senator, if I understand your proposition correctly--to go back to the original statement—it involves the reduction of ex penditures of one kind by $3 billion and the increase of expenditures 2510 EM PLOYMENT, GROWTH, AND PRICE LEVELS by another kind by $3 billion so that we can ignore, I think—this is purely-----The C h a i r m a n . Yes, the global effect. Mr. S u it s . Aside from any particular effect that this might have to increase Government efficiency, and so on, we have increased our tax revenue by $3 billion. Now, if I understand correctly, by lowering excises and adjusting the upper bracket income tax, we reduce the tax yield by a billion and a half so the net impact of what we are doing is increasing taxes by one and a half billion. The C h a i r m a n . Tax receipts, not tax rates. Mr. S u it s . We increase tax receipts by one and a half billion and we use that to pay off a portion of the national debt. Several aspects of this are quite clear: First, that the tax receipts of the Government will expand whereas the Government expenditures will not. To this extent, surely, this will help hold in check the rise in prices. To the extent that the retirement of Government debt by the Treasury repre sents a form of expanded demand or reduced supply of Government bonds, this is surely going to encourage a rise m their price, or, in other words, exert a downward pressure in the interest rate. How much pressure, of course, is a completely open question, but other things being equal, surely this would tend to hold down or reduce the rate of interest. Whether the effect would be any different via a direct expansion of bank reserves or via Federal Reserve purchase on the open market rather than by Government retirement of debt, again, I think is an open question. But given the reduction in the rate of interest, the question is, What impact would it have ? I must confess that while I have spent a good many years investigating the relationship between the rate of interest on the one hand and the real phenomena in our economy, I have never yet succeeded in satisfying myself that I saw any connection. I do not say there is no connection. As a matter of fact, I find it impossible to believe that there is no connection, but I must say I have never succeeded in demonstrating it, and to my knowledge, neither has any body else. Mr. M e y e r . It would seem to me that a crucial question here is what the effects on prices would be. The working assumption is that you would have offsetting effects on total demand or, hopefully, even some increase in total demand, so the question is whether demand would be reallocated from bottleneck areas to areas where there is excess capacity or vice versa. At this point we must enter into some speculation about the prob able effect of lower taxes and interest rates. It would be my guess that the main effects would be to increase consumer durable and hous ing expenditures somewhat. That would result from a combination of the tax reduction and the probability of somewhat lower interest rates. Now we get into the question of whether housing and consumer durables are areas of excess or deficient productive capacity, given present demands. It would be my guess that these would be two sectors in the next year that might stand some increase in activity without running into any serious capacity bottlenecks. EMPLOYMENT, GROWTH, AND PRICE LEVELS 2511 If this is true, then the scheme might work out quite well and have real merit. I would not want to press it too far, however, and I would raise the question of whether it might not be better to be on the safe side under the present boom conditions, and postpone that $1.5 billion for tax reduction until a less expansionary period. Dr. F r u c h t . I am trying to get clear the numbers the Senator used. Take the $3 billion saved by closing loopholes. That $3 bil lion we are going to reduce the Government debt by $li£ billion and we are going to reduce taxes by $1^ billion. One might think that we had a standoff here because the funds obtained by closing loopholes would presumably at the present time be spent upon capital formation. The C h a i r m a n . N o . One of the big loopholes is this very item of business expense accounts about which Mr. Clark spoke this morning, and our aim would be to discourage luxurious expenditures by busi nessmen to diminish the business of New York nightclubs and tickets to “My Fair Lady” and suites in the Sheraton Park, Sheraton-Carlton, and other hotels in Washington, and thus reduce the lavish business expenditures on luxury. I am not a patron of the nightclubs but in spite of the columns of the nightclubs, I do not regard the nightclubs as the center of culture and the arts. Maybe you have been able to discover cultural qualities at some club that I do not know about. Dr. F r u c h t . I will extend my proposition to the effect that any saving of this kind would be a saving and it would presumably represent resources that could become available for capital formation. Does the Senator have any notion as to the relative proportion of the $3 billion that this would constitute ? The C h a i r m a n . Roughly, we thought getting $800 million from excessive business expense, and $400 million from the depletion allow ance, and then withholding of taxes on dividends at the source, which would make some stocks less attractive than now. Representative W i d n a l l . It would also make Government bonds less attractive. The C h a i r m a n . No, not quite; and then removal of the 4-percent dividend credit, which would make them more attractive. There are other items we can throw in—splitoffs and spinoffs, pension plans, stock option plans, and so forth. Dr. F r u c h t . Would the panel grant t h a t to the extent that the t a x saving, the revenue saving, were accomplished a t the expense of in vestment by corporations, that the piling of these funds into the Treasury and their subsequent use, say, for the reduction of the Government debt would not constitute on that balance an increase in the actual savings and presumably would not have any effect on the interest rate? If I reduce a potential investment and then replacement within the market these funds for another investment, the same or another invest ment, need it follow that there is any shift in the interest rate ? Mr. M eyer. I don’t think it would happen that way, following out your own hypothesis. To the extent that the funds came from a reduction in corporate retained earnings, which I think you are sug gesting, in order for your offsetting effect to be felt it would be neces sary in turn for firms to go out and secure more funds in the market 2512 EMPLOYMENT, GROW TH, AND PRICE LEVELS in order to maintain their investment programs. My own studies would suggest that they would probably reduce their investment outlays rather than turning to outside sources, so the net effect would be depressive on private plant and equipment investments. This would be a good thing if we are going to say this is a strong expansionary area, perhaps leading to some price inflation because of demand pull forces. I don’t think demand will be that strong in this sector, but if there is a candidate sector for potentially unfavor able price influences, it would appear to be in the producers’ durable sector. Dr. F r u c h t . Y ou would think this would reduce the interest rate? Mr. M e y e r . I think so, but I must add that after hearing the off sets, I am not so sure that after you weighed in the expense effects against the depletion effects and so on that you would necessarily have a reduction in corporate earnings. It would also depend on what the rest of the Government was doing, because it is entirely possible that if you make the quotas stricter or some other action is taken to protect, say, the petroleum industry that the effect of the tax change would be offset. The C h a i r m a n . It would seem to me that the depletion allowances are the most striking examples of where incorrect tax policies causes misapplication of resources because it is pretty clear what is happen ing has caused an undue investment in the oil and gas industry. Mr. M e y e r . I would agree, but I was pointing out that it would be possible for the Government to offset the change in depletion allow ances if they were to extend the present import quota policy. Or they could use other kinds of semi or actual Government controls to restrict output and increase profits in the face of a reduction of deple tion allowances. The effectiveness of all this would depend on what you assume about the elasticity of demand for petroleum. The C h a i r m a n . Thank you very much. I wish you gentlemen would continue to think of this and if you have any thoughts send them along. Can you raise the interest rate by action of the monetary authorities and inadequate vigilance by the fiscal authority or cooperation by the fiscal authority to the point where it is in excess of the marginal productivity of capital and thus shut off investment? Mr. C h r is t . I think this is possible in principle. If the central bank is in a very strong position, standing over the money market, and if it makes money very easy, sets rates of return on Government bonds or permits rates of return on bank loans to be 1 or 2 percent, this can produce a big inflation. And contrariwise if the central bank puts the screws on very tightly, it can set the rates of return on Government bonds through its open-market operations very high, 6 or 8 percent, 10 percent. The C h a i r m a n . Therefore, businesses will say, we cannot earn this-----Mr. C h r is t . So we will not produce. The C h a i r m a n . Would you agree wTith that ? Mr. C h r is t . While I agree this is possible, I do not think that is what is happening just now because there still are firms that are go ing to the market to borrow at the current rate and the investment seems to be quite high. EM PLOYMENT, GROWTH, AND PRICE LEVELS 2513 The C h a i r m a n . It is not that all investment would be shut off, but suppose the volume of investment is less than the volume of saving? Mr. C h r is t . It we got a volume of planned investment less than the volume of planned saving, then I think you would see the results in a decline in total activity in the very near future. It is important for the central bank not to raise rates that high, but I think in order to main price stability it may sometimes be necessary to permit the rate of interest temporarily to go rather high because there is tem porarily, from time to time, a very large demand for borrowing in order to invest in new projects and this is what I think has to be held in check. Mr. S u it s . I think I have nothing to add. The effects of changes in interest rate as I have studied them, are exceedingly nebulous and speculative, especially speculative in the intellectual sense, and I know that much effort is spent in devising and studying these policies. I must say I do not understand what their impact really is. The C h a i r m a n . I spent a year of my life trying to see if there was any statistical evidence between changes in the interest rate and the rate of saving or investment. I must admit I could not find one but this does not mean as you say there is none. But, a priori, I wxould tend to believe the higher interest rate, the greater the saving, but certainly, the less the investment. Mr. M e y e r . Yes; and it is certainly pertinent here that you are asking us to speculate on an increase in interest rates that would carry us beyond our historical experience, so the fact that we have not ob served statistical relationships in the historical sense might not serve us in evaluating your proposal. On the whole, if you raised interest rates high enough, sooner or later something would have to give. The C h a i r m a n . Thank you very much. (Whereupon, at 4:35 p.m., the hearing in the above-entitled mat ter was recessed, to be reconvened on Tuesday, September 29, 1959, at 10 a.m.) EMPLOYMENT, GROWTH, AND PRICE LEVELS TUESDAY, SEPTEMBER 29, 1959 C ongress of t h e U n it e d S t a te s , J o in t E c o n o m ic C o m m it t e e , Washing ton, D.G. The committee met, pursuant to recess, at 10 a.m., in Room P-63, the Capitol, Hon. Paul H. Douglas (chairman) presiding. Present: Senators Douglas and Bush; Representatives Reuss? Curtis, and Boggs. The C h a i r m a n . The committee will be in order. The economics profession of the country suffered a great loss yester day in the death of Prof. Sumner Slichter of Harvard. Mr. Ulman has prepared a memorial in tribute to him which I shall ask him to read. STATEMENT OF LLOYD ULMAN, PROFESSOR OF ECONOMICS AND IN DUSTRIAL RELATIONS, UNIVERSITY OF CALIFORNIA, BERKELEY, CALIF. Mr. U l m a n . Mr. Chairman, I should like to pay tribute to Prof, Sumner H. Slichter of Harvard University, whose death occurred the day before yesterday. His work in general and in the area before us for discussion today has stimulated scholars, Government officials, and representatives of virtually all private groups in this country and abroad. His students and colleagues in his profession suffer fhe irreparable loss of one of the most profoundly original, bold, and prophetic thinkers on economic and social problems this country has ever produced. The entire community stands deprived of a man who served it to the limit of a very great capacity, not only by sub jecting many of its institutions and policies to fearless and construe tive criticism but above all by practicing without compromise in his personal career the individualism which so many in this age have been content to preach. The C h a i r m a n . Thank you very much. If I may amplify the statement by Mr. Ulman, I want to join with him in deploring the death of Mr. Slichter. I have known Mr, Slichter for 40 years, and found him always to be intellectually honest and with a very keen mind. He was, I think, a very helpful influence both in the field of economic thought and in the field of public policy , The country has been the richer because of his life and is the poorer because of his death. Senator B u s h . Mr. Chairman, I would just like to say a word my: self. I noted with deep regret this morning the death of Professor Slichter, who quite recently appeared before this commitee. He was 2515 2516 EMPLOYMENT, GROWTH, AND PRICE LEVELS a very stimulating and indeed an exciting witness, as he often has been, before a congressional committee. He was a most unusual man, an honest and straightforward thinker. Even if one did not agree with him at all times, one was bound to be influenced by the depth o f the sincerity o f his convictions in matters pertaining to the economic life of this country. I am sure that economists and those who are in terested in our economic welfare everywhere will join in deep regret over the death of Sumner Slichter. The C h a i r m a n . While Professor Slichter spent the last 35 years o f his academic life at Harvard, he originated in the Middle West and in the State o f Wisconsin, where his father wras a distinguished mem ber of the faculty of the University of Wisconsin and from which Professor Slichter graduated in the days before World War I. I know that Congressman Reuss has certain comments and tributes which he wishes to pay. .Representative R e u s s . Yes; thank y o u , Mr. Chairman. I would like to associate myself with what you have said and what Senator Bush has just said about this fine and thoughtful man who is no longer with us. His roots were indeed very deep in Wisconsin. He partook of the heritage of economic thinking which has made the name of the University of Wisconsin so illustrious in the history of economic thought, As Senator Bush has said, his was a lively mind. He never hesi tated to attack sacred cows. It was my pleasure just recently to be a member of the American Assembly at Arden House in New York, with Professor Slichter and others, and there again he brought a set of lively insights to the study of what has on occasion been called the dismal science, To his family, to his brothers in Wisconsin, and other members of his family, I want to express my deepest sympathies. He was a great man and we are going to miss him sorely. The C h a i r m a n . We will begin the discussion this morning on the subject of the economic significance of collective bargaining: aWages and Income Distribution,” by Edward C. Budd, of Yale. STATEMENT OP EDWAED C. BUDD, ASSISTANT PKOFESSOR OF ECONOMICS, YALE UNIVERSITY Mr. B u d d . Thank you, Mr. Chairman. Certainly, no one questions the important position currently occu pied by organized labor and the institution of collective bargaining. In attempting to evaluate the effects of collective bargaining on the wage structure, wage levels, and income distribution, however, it seems to be important to recognize that we live in an economy which is only partially unionized. After all, less than a third of all em ployees, and an even smaller proportion of the labor force, are mem bers of unions. In such a world as this, economists have generally recognized that unions, particularly the stronger ones, can through collective bar gaining obtain increases in wages (or in fringe benefits) greater than those which would have resulted in their absence from external market forces or may be successful in resisting wage reductions that might otherwise have occurred. EM PLOYMENT, GROWTH, AND PRICE LEVELS 2517 In cases of this sort, economists have been inclined to argue that the action will be reflected eventually in reduced employment in' the area involved. Employers will be encouraged to substitute other in puts for the labor whose cost has risen, and the prices of the products produced by the labor will tend to rise because of the increase in costs, leading to some reduction in output and hence employment, the mag nitude of these effects depending on the possibilities of substituting other inputs for labor and on the el* sMcity of demand for the indus try’s product. Workers released IV: li i heir jobs because of the rise in wages will be forced to seek oil:, a less remunerative employment in nonunionized labor markets, producing downward pressure on nonunion wage rates as employment in such sectors expands relative to the union sector obtaining the wage increase. In periods of generally expanding employment, of course, employ ment m the latter sector need not necessarily fall, simply expanding less than it otherwise would. Insofar, therefore, as collective bargain ing does achieve the objective of higher wages for union members^, economists have predicted that the impact would be primarily on the structure of wages and on the distribution of employment among industries, occupations, and regions. ••: With Henry Simons, they have tended to agree, if somewhat reluc tantly, that collective bargaining makes high wages higher and low wages lower, and limits the extent of employment in unionized areas. Attempts at empirical verification of this traditional model haive not met with much success. By and large, investigators have found that wage rates in the unorganized areas have increased just about as much as in the union sectors, or, if not, that the differential move ments can be accounted for more satisfactorily by some other variable, such as changes in employment or productivity. Generally speak ing, those sectors which have experienced a relative rise in wages ar’e those which have seen a more rapid expansion of employment and productivity and which are also, surprisingly enough, concentrated in their market structure. The picture is by no means clear cut, and there are important exceptions. There is some evidence, for example, that unions were more successful than unorganized workers in resist ing wage cuts during the early thirties. Newly organized unions also appear to have some initial impact on the wage structure, with the differentials remaining, but not widening, in subsequent periods. The behavior of wages in manufacturing during the recovery period culminating in 1937 may be an example of this sort. In this connection it may be instructive to see if we can find evi dence of emerging differentials in earnings and employment at rather aggregative levels in our economy. In figure 1 (the data for which are given in table 1 )1 have made a rough division of industries given in our national income accounts into strongly and weakly organized sectors, and have compared average annual earnings (including sup plements, such as fringe benefits and employers’ contributions to social insurance) for each sector on the assumption of constant employment weights for each industry in order to eliminate the effect on average earnings of employment shifts from lower to higher paying industries. You will note that from 1934 on (the period in which unionism increased greatly in importance) average annual earnings in the strongly organized sector have, with few exceptions, increased by a 2518 EM PLOYMENT, GROWTH, AND PRICE LEVELS greater proportion than in the weakly organized sector. A measure of the change in this differential is provided by the index shown by the solid line in the lower part of figure 1. From table 1 you can see that whereas average earnings in the union sector exceeded those in the nonunion sector by about 9 percent in 1929, the figure had increased to 41 percent in 1958. My fellow economists on this panel will undoubtedly be primed with objections to comparisons of this sort, and I will take the time to point out a couple. For one thing, I have included supplements to earnings. These incorporate not only fringe benefits but social secu rity taxes as well, which are of greater importance in the union sector. For another, the statistics are of necessity in terms of average annual earnings instead of hourly wage rates and are affected by variations in hours worked, in premium pay, and in the composition of employment within the individual industries making up the total. Even more important is the behavior of relative employment. The proportion of employment accounted for by the union sector, expressed as an index, is shown by the dashed line in the lower half of figure 1. You will note that the two indexes seem to move pretty much together, both in the shortrun and over the period as a whole. This behavior is consistent, not with the hypothesis of a differential effect of unionism on earnings but rather with a more traditional view: the expansion of employment, occasioned by shifts in the composition of demand and by the more rapid expansion of productivity, in sectors in wThich unions happen to be important has served to bid up earnings relative to thos^ sectors experiencing a lower rate of expansion. In economics, statistical demonstrations cannot in and of them selves be entirely conclusive, and the data are still consistent with the hypothesis that union wage raising power prevented a more rapid and more extensive expansion of employment than that which has occurred. Furthermore, there are periods in which the earnings differential and relative employment move in opposite directions, for example 1953 to 1957, and in which the competitive analysis is hard pressed for an explanation. For myself, I still remain somewhat skeptical of the actual impor tance of unions in creating and widening differentials in earnings. Perhaps there are good reasons for supposing that aside from the initial impact of union organization the effect of unionism on the wage structure may be small. For one thing, the bargaining power of the union is offset, at least partially, by the bargaining power of the employer. Both must take account of the market forces which still operate on wage levels even with the existence of collective bargaining. For another, the proper contrast is between unionized and nonunionized labor markets, not between unionized and purely competitive mar kets. Even in the absence of unions, market forces operate only in directly and slowly on wage levels. Employers do not simply auction off jobs at the factory gates to the lowest bidder; each employer must have a wage (and personnel) policy, and he must recognize its effect on the willingness of those already on his payroll to give him a good day’s work for a day’s pay. Probably there never wTas a day when unorganized employees failed to possess some bargaining power (other than the power traditional EM PLOYMENT, GROWTH, AND PRICE LEVELS 2519 theory has always recognized—their option of looking for employ ment elsewhere). Without some such underpinning at this, theorists will always be hard pressed to explain the existence of positive wage rates in periods of severe unemployment. We are, I believe, coming to the view that there is not as much difference between union and nonunion situations, at least in their wage aspects, as has been sup posed. Certainly Simons exaggerated the differences that can rea sonably be expected to emerge. The impact of unionism on the distribution of income between labor (in the form of employee compensation) on the one hand and property income (including rent, interest, and profits) on the other is open to even more question. The popular view seems to be that collective bargaining, if successful in raising wages, should also serve to raise labor’s share of the total pie. There are, however, a number of missing links in the argument. Even if we ignore differences be tween union and nonunion sectors and imagine the wage level to be raised by one big bargaining agreement, in all probability part of the increase will, because of increased costs, be offset by increased prices, leaving a much smaller rise (if any) in real wages. Even if profit margins are initially reduced, businessmen will attempt a substitution of capital for labor. When the existence of sectors not sharing in the postulated rise in wages is recognized, there may be further interindustry repercussions which may help to mitigate the initial impact on profit rates. Even within the frame work of existing productive techniques, the final effect on the prop erty share is uncertain, depending as it does on the offsetting effects of changes in profit rates and in the amount of capital. Furthermore, the rise in wages may stimulate businessmen to search for innovations that will reduce their labor costs relative to the cost of capital, thus tending to offset any initial tendency for the share of employee compensation to rise. Let us, however, accept the popular view of the matter and set aside all of the above complications which are of interest primarily to economists. What does the evidence suggest ? I call your attention first of all to table 2, which shows the share of employee compensation in income originating in corporations. I have selected this particular concept for several reasons. First, a larger proportion of union strength is in the corporate sector. Second, the nonlabor share is more clearly a return to property ownership than is the case when self-employed proprietors are included. Finally, the data are available quarterly, at least since 1946, a fact which permits us to exclude the share-increasing effects of recessions, an exclusion which is not possible when annual data are used. (For example, the fall in labor’s share from 1953 to 1955 indicated by both tables 3 and 4 arises from the fact that the 1954 recession started in the second half of 1953.) An examination of the share in net income will, I believe, indicate that at least from 1948 on (when postwar reconversion and inflation problems were largely things of the past), there has been a rather steady rise, interrupted only by the Korean war, when labor’s share fell (just as it did in both world wars). If we take the share as exclusive of the compensation of corporate officers, a small rise can be found between 1929 and 1948. 2520 EM PLOYMENT, GROWTH, AND PRICE LEVELS Before the hasty conclusion is reached that the popular view of the matter is vindicated, I would like to point out three further facts. First, the gradual rise in labor’s share in net income has been going on for a long time, possibly since the turn of the century, and at least before World War I. I have assembled some of the relevant data in table 3. In order to obtain the data for comparisons over longer periods, it is necessary to broaden the income concept to include income originating in un incorporated enterprises as well as corporations. (I have continued to exclude the Government, for its contribution to income is measured at labor cost alone.) Labor’s share has, of course, risen for a number of reasons, some of them only remotely related to the question of bargaining power. For example, the proportion of those who are self-employed, in con trast to those working for wages or salaries, has tended to fall over time; sectors in which the share of employee compensation is high have grown relative to those in which it is low. The two most important examples of the latter are agriculture, where the wage share averages about 15 percent, and residential housing, where it is virtually zero. Column (2) of table 3 represents an attempt to exclude the effect of the falling proportion of selfemployed and of the declining importance of agriculture ; column (3), in addition to these two adjustments, shows the effect of assuming a constant share for rent. Since the base for these adjustments is 1953, the share in this year is the same for all three columns. The impor tant point to note is that even with these adjustments the upward secular drift in the share is not eliminated, and the upward movement is apparent in periods in which it would be difficult to maintain that unions were of much importance. Second, the behavior of labor’s share in strongly organized sectors of the economy does not appear to be materially different than in the weakly organized sectors, The relevant data are shown in figure 2 (and table 4), which have been prepared on the basis of the same classification as that used in figure 1, with the exception of real estate and domestic service, which have been excluded from the nonunion sector for reasons that I do not have time to justify here. Within either the prewar or the postwar period, the share of em ployee compensation appears to have increased more in the nonunion than in the union sector. The only period in which the union seg ment seemed to enjoy an advantage was during the years of World War II, and for this, wartime price controls and reconversion prob lems are probably more relevant considerations than union pressure, particularly since wages were largely under the control of the Gov ernment. Third, you may already have noted, in figure 2 and in tables 2 and 4, that the upward drift in labor’s share since 1948 is much more evi dent in net income (exclusive of depreciation) than it is in gross in come. Capital per unit of labor and per unit of real output has been rising since World War II, with a consequent rise in the ratio of de preciation charges to gross income, a development accentuated in 1954 by the change in the tax law permitting the use of depreciation methods which involve more rapid writeoffs of capital assets. EMPLOYMENT, GROWTH, AND PRICE LEVELS 2521 Since the proportion of social security and indirect business taxes to gross business product has also risen somewhat since 1948, the “ squeeze” on net property income (if that is what it is) can be in terpreted just as plausibly as a depreciation and tax squeeze as a wage freeze. In fact, if such a squeeze were not to develop, prices would have had to rise relative to unit labor costs during the past 10 years. In any case, the fact that labor’s share has risen in the postwar period is quite consistent with the idea that unions have had little impact 011 distributive shares, since the significant increase in the ratio of capita to labor can easily account for the observed results. I want to make one final, and necessarily brief, comment on the gen eral level of money wages. The view is becoming increasingly com mon, as evidenced by the current popularity of the “cost-push” infla tion hypothesis, that the primary impact of union bargaining strength is on the money wage level rather than on the wage structure. Wage increases obtained in union sectors, so the argument goes, are trans mitted, by some form of “contagion” or sympathetic pressure, to the nonunion areas of the economy, with the result that the entire wage level is pushed up. If the wage increases themselves do not succeed in generating the increase in money aggregate demand necessary to support them with out the emergence of unemployment, then, it is argued, the Govern ment, by means of either monetary or fiscal policy, will make up the difference. I myself have serious reservations to this whole approach to the problem of inflation, if that is what we have had in recent years. For one thing, it seems to me absurd to argue that considerations such as these have weighed at all heavily with those responsible for our monetary and fiscal policies in the past few years. Even more important, the proponents of this idea have not explained, at least to my satisfaction, how the increases resulting from collective bargain ing are transmitted to the unorganized sector. There are, to be sure, firms contiguous to union areas who will grant wage increases in order to forestall possible threats of unionization. This argument does not, however, seem plausible for the great mass of unorganized firms; in any case, the external market pressures on nonunion wages will probably on balance be in the opposite direction. Hence the major effort of union pressure ought to appear as a con tinuous widening of wage differentials rather than as an upward movement of the wage level. Further, it is extremely difficult to obtain any empirical verification of this thesis. Comparisons such as those I made earlier between union and nonunion sectors would obviously not be valid, and the rate of wage increase, or the relation between wage and price increases, cannot, in and of themselves, tell us whether the source of the diffi culty is “cost push” or “demand pull.” I would be the first to concede that we do now have a satisfactory theory of the determination of the money wage level. If, however, the particular version of the cost-push argument which lays the blame at the doorstep of collective bargaining is valid, I would be rather pessimistic about finding any rate of permanent unemployment, toler able or intolerable, which would prevent a rise in money wages, or a rise in wages greater than the increase in productivity, to use the 38563— 59— pt. 8.------ 6 2522 EM PLOYMENT, GROWTH, AND PRICE LEVELS currently popular formulas for determining the rate of money wage increases consistent with stable prices. T a b l e 1 . —Average annual earnings ( including supplements) of, and proportion of those engaged accounted for by, strongly and weakly unionized sectors of tile private domestic economy, selected years A Y ear B C D Average annual earnings U nion E F Percent of engaged i n - A Y ear N on R atio U nion N o n union ( B - C ) union 1929___ $1, 584 $1,453 1934___ 1,192 1,123 1937___ 1,474 1,279 1939___ 1,494 1,294 1946___ 2, 707 2,226 1947___ 2, 998 2,427 1948___ 3,261 2, 598 1949___ 3,337 2, 661 1950___ 3, 572 2, 793 1.090 1.061 1.152 1.155 1.216 1.235 1.255 1.254 1.279 41.7 37.2 40.0 39.2 45.9 46.4 46.4 45.0 46.0 58.3 62.8 60.0 60.8 54.1 53.6 53.6 55.0 54.0 B C Average annual earnings U nion 1951___ 1952___ 1953___ 1954___ 1955___ 1956___ 1957___ 1958___ D 3,294 4,163 4, 402 4,507 4, 753 5,027 5,282 5, 468 E F P ercent of engaged i n - N o n R atio U nion N o n union (B -f- C) union 2,940 3,065 3,199 3,319 3, 457 3,603 3, 753 3,882 1.335 1.358 1.376 1.358 1.375 1.395 1.407 1.409 47.5 47.7 48.2 46.8 47.0 47.0 46.9 44.9 52.5 52.3 51.8 53.2 53.0 53.0 53.1 55.1 N o t e . —Strongly unionized (union) sectors include m ining, m anufacturing, contract construction, tran spo rtation , com m unications, and public utilities. W eakly unionized (nonunion) sectors include agriculture, forestry and fisheries, wholesale and retail trade, finance, insurance and real estate, and services. Average annual earnings for each sector include supplem ents and were obtained b y com bining average ann ual earnings for the above industries on the basis of fixed (1953) em ploym ent weights. T otal engaged in production includes proprietors of unincorporated enterprises as well as full-tim e equivalent employees. Source: C om puted from d ata in D epartm en t of Com m erce, “ U.S. Incom e and O u tp u t,” 1959; “ N ational Incom e,” 1954 edition; and “ Survey of C u rren t B usiness,” Ju ly 1959. Classification into strongly and w eakly unionized sectors is based on data in “ M onth ly L abor R eview ,” M ay 1947, and H . M . Levinson, “ U nionism , W age T rends, and Incom e D istributio n,” A nn A rbor, 1951. 2523 EM PLOYMENT, GROWTH, AND PRICE LEVELS T a b l e 2 . —Share of employee compensation in gross and net income originating in corporations, selected periods N O N R E C E S S IO N P E R IO D S Y ear and qu arter 1929 _____________________ 1937 ............... ....... 1946 _______________________ _____ _________ 1947 1948 _______________ 1950: 2 d ___ ___________________ 3d and 4th _____________ 1951: 1st and 2d.__......................... 3d and 4 th _______________ N et income Gross income Percent Percent 75. 7 80. 8 81.3 79.3 76.6 76.0 74.1 75.5 76.1 68.4 73.4 76.4 74.0 71.2 70.0 68.6 69.7 69.9 Y ear and quarter N et incom e Gross income 1952: Percent Percent 1st_______________________ 77.6 71.1 4 th ______________________ 79. 0 72.3 72.1 1953: 1st and 2d______________ 78.9 1955_________________________ 79.3 71. 5 1956_________________________ 81. 5 73.1 82.2 1957: 1st, 2d, and 3 d _________ 73.0 1958: 4 th ____________________ 81.8 72.3 1959: 1st________________ _____ 72.1 81. 5 R E C E S S IO N P E R IO D S 1949____________ 1950: 1st________ 1952: 2d and 3d. 1953 : 3d an d 4th 77.9 77.6 79.3 80.4 71.7 71.1 72.4 73.5 1954__________________________ 1957: 4 th _____________________ 1958: 1st, 2d, and 3 d ._ _____ 82.1 84.0 85.2 74.1 74.2 74.9 N o t e . —B etw een 1929 and 1937, on the one hand, and the postw ar years on the other, the share of com pen sation of corporate officers in corporate incom e fell by 1.7 percentage points. H ence the share of nonofficer com pensation rose by 1.7 percentage points m ore betw een prew ar and postw ar years th a n is indicated by th e shares above. Recession periods are defined to be those quarters in w hich incom e originating in corporations w as less th a n in some preceding qu arter. N onrecession periods are those quarters in w hich income originating w as equal to or greater th a n th a t of any previous quarter. Q uarterly d ata for incom e originating in corpo rations are not available prior to 1946. N et corporate profits have been adjusted to reflect depreciation a t reproduction cost rath er th a n original cost. Indexes used for the adju stm ent are from the M achinery and Allied P rodu cts In stitu te. Source: C om puted from d a ta in D epartm en t of Com m erce, “ U.S. Incom e and O u tp u t,” 1959, table 1-12 and 1-14. Since quarterly estim ates of corporate depreciation charges are not available, the annual estim ates (from table VI-18) were interpolated on the basis of quarterly estim ates of total capital consum ption allow ances (table 1-18). T he estim ates for the 1st q u a rte r of 1959 were extrapolated from the 4th q u arter of 1958 on the basis of total corporate profits and private em ployee com pensation. 2524 EM PLOYMENT, T a b l e 3 . —Share GROWTH, AND PRICE LEVELS of employee compensation in private domestic income for selected peacetime, nondepression years Y ear A ctual share A djusted share (2) (1) 1900____________________________________________ 1910_______________________________ ____________ 1920________ ___________________________________ 1929____________________________________________ 1937____________________________________________ 1948___________________________________________ 1953____________________________________________ 1955____________________________________________ 1956____________________________________________ 1957____________________________________________ Percent 40 3 47.8 56.0 58.2 62.6 62.3 67.7 67.2 69.1 70.0 Index of ratio of capital to labor (1953=100) Percent 55.3 55.7 60.4 61.4 67.3 65.1 67.7 66. 7 68.1 68.9 Percent 61. 8 60.9 62.6 65.3 68.7 64.5 67. 7 66.7 68.1 69.0 N o t e .— T he shares for 1937 are affected b y the existence of substantial u n e m p lo y m e n t in that year. 69 75 82 94 91 88 100 123 P riv ate dom estic income is equal to national income m inus income originating in G overnm ent (com pen sation of G overnm ent em ployees), from abroad, and from interest on consum er debt. P ro p erty incom e is n et of estim ated inventory valuation adju stm ent (the difference betw een depreciation at original and at reproduction cost). T he “ adjusted share (1)” in private dom estic income has been adjusted toexclude the effects of th e declin ing proportion of those engaged in production who are self-employed and of the changing im portance of farm incom e relative to nonfarm incom e. These adjustm ents were m ade by com bining the shares of em ployee com pensation in farm ing, and in the nonfarm sector, for each year on the basis of the 1953 im portance (in incom e originating) of each of these tw o sectors, and by adjusting em ployee com pensation in each in d u stry so as to reflect the 1953 proportion of employees to self-employed workers. T he “ adjusted share (2)” in private dom estic income has been adjusted in the same m anner as for (1), except th a t the proportion of ren tal income (inclusive of interest paid by the real estate sector) to private dom estic income has been assum ed constant at the 1953 proportion. T his adju stm ent elim inates from (1) th e effect on the share of changes in the im portance of rental income. Source: From 1929 on, com puted from the national income publications of the D epartm en t of Com m erce referred to in tables 1 and 2. P rior to 1929, the figures are based on prelim inary and un pu blish ed estim ates of the author, and are conceptually com parable w ith the estim ates from 1929 on. Indexes used for esti m ating the depreciation valuation adjustm ents are from the M achinery and A llied P rodu cts In stitu te, “ C apital Goods R eview ,” N o. 29, and from R. W . G oldsm ith, “ A S tu dy of Saving in the U nited S tates," vols. I and III. From 1929 on, the indexes used do not differ m aterially from those published b y the D ep art m en t of Com m erce for m anufacturing establishm ents (U.S. Incom e and O u tp u t, table V-12). EM PLOYM ENT, G R O W T H , AND PRICE 2525 LEVELS T a b l e 4 . — Share of employee compensation in net and gross income originating in strongly and weakly unionized sectors of the private domestic economy, selected years [Percent] Y ear Share in net income U nion 1929____________________________________________________ 1934____________________________________________________ 1937__________________________________ - _____________ _ 1939____________________________________________________ 3946 _____________________________________ _____ 1947____________________________________________________ 1948____ _______________________________________________ 1949_____ ______________________________________________ 1950____________________________________________________ 1951 ______ __________________________________________ 1952___________________________________________________ 1953____ _______________________________________________ 1954___________________ ____________________________ 1955 ________________________________________________ 1956 __________ ____________________________________ _ 1957 _____________________________________ _______ 1958___________________________________________________ 71.9 85. 5 76.6 77.2 77. 7 75.6 72.6 73.3 70.6 70.8 73.9 75.4 77.0 74.1 75.8 76.8 78.9 N onunion 56.6 67.5 59.1 61.0 50. 5 53.9 53.2 54.5 55.0 54.4 55.2 56.7 57.4 56.9 57. 7 58.3 58. 7 Share in gross incom e U nion N onunion 0) 0) 0) 0) 0) 0) (0 0) 74.0 71.7 68.8 68.8 66.5 66.7 69.2 70.3 71.3 68.0 69.4 70.1 71.3 48.3 51.0 50.1 50.8 50.7 50.6 51.2 52.5 53.0 52.6 53.1 53. 4 53.8 1 N o t available. N o t e .—Strongly and w eakly unionized sectors (“ u n ion” and “ nonunion” ) have the same definition as in table 3, except th a t households (which includes dom estic service and interest on consum er debt) and real estate have been excluded from the nonunion sector. Em ployee com pensation has been adjusted to elim inate the effect of changes in the proportion of those engaged in production w ho are self-employed. Because of lack of data, property income has not been adjusted for the difference betw een depreciation com puted on an original cost and a reproduction cost basis. W hile this omission does not m aterially affect com parisons am ong postw ar years, it tends to un derstate the rise in the share of em ployee com pensation in net income betw een the prew ar and the postw ar period. In order to exclude the effect on shares of shifts in the im portance of the industries w ithin each of the 2 sectors, industries were com bined on the basis of fixed (1953) income w eights. H ad this procedure not been followed, the share rise in the nonunion sector in the past 10 years w ould have been significantly greater th a n th a t indicated above. Source: Same as that for table 1. 2526 EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS FIGURE AVERAGE ANNUAL EA R N IN G S IN STftON&LY AND W E A K L Y RATIO SCALE X AND RELATIVE EMPLOYMENT UNIONIZED SECTO RS OF THE PRIVATE ECONOMY J __ I__ I__i 47 MS V9 Source; same 4S figure X - _____ SO si y' 5XS3SHsFsT5lT8 E M P L O Y M E N T , G R O W T H , AND PRICE FI GORE SHARE OP EM PLO YEE LEV ELS 2527 Z COMPENSATION IN GBOS5 AND NET NOTIONAL INCOME CffJ FACTO ft COST) IN 5TROIM&LY AND WEAKLY UNIONIZED onion Sector includes mining rwdrK»^ciurina, catw+rucfion, tnjnspoHi+icn, commonleatioits 6t>d public oilfih'es. Non-union Sector includes ^na/lh/ne, forestry o-ftslwrtes, wholesale*fetailtrw^ ^»nence,mfur«f*a, And services (<tu.losr*& of real estate domahc /cryice) Source: compute* W n daU in y-y. incom e^d o ^ ^ 0 ^ N#J0»rtl iM gjgt HS*t Editionf and Survey of Corrent 8<*sinessf J*t«j, M5*i The C h a i r m a n . Thank you very much, Mr. Budd. Mr. Hildebrand, will you continue the discussion ? Senator B u s h . Mr. Chairman, I do not wish my request to have all these papers read stand. I bow to the chairman’s wish because they are a little too long for that. On the other hand, I would be glad to listen to them if the others want them. The C h a i r m a n . The next one runs to 20 pages and would normally take about 40 minutes. Senator B u s h . I only thought Mr. Budd’s paper was rather brief. The last few pages were charts. I withdraw any request of that nature. The C h a i r m a n . Mr. Hildebrand, do you think you can summarize in 10 or 12 minutes % 2528 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS STATEMENT OF GEORGE H. HILDEBRAND, PROFESSOR OF ECO NOMICS AND DIRECTOR, INSTITUTE OF INDUSTRIAL RELATIONS, UNIVERSITY OF CALIFORNIA, LOS ANGELES, CALIF. Mr. H i l d e b r a n d . Yes. Although prices as a whole have been stable for over a year there continues to be great public and official concern about inflation. In large part the cause is believed to start with a cost push rather than a demand pull. In turn this thinking has led to two major conse quences. First, that the labor unions and collective bargaining are mainly responsible for inflation and, second, that the cure for inflation is tight money and a tight Federal budget. I dissent from both conclusions. Overemphasis upon inflation today has led to a policy of restraint upon demand. This has checked investment in two ways: By reduc ing the incentive to expand and improve plant and equipment, and by denying us the economies of continued high volume output. This, in turn, has reduced the average rate of growth in productivity and output, making cost pressure more effective in raising prices. Since there is some cost-push, part of which originates in wages, we should look to other correctives in place of restraint of demand. I do not intend thereby to suggest an irresponsible monetary or fiscal policy. However, not to look for other strategies is to accept too much un employment and idle capacity, restricting our rate of growth. There is a school that contends that collective bargaining as a whole has no real influence on wages, that the power of unions as a whole is illusory save in the political field. This stretches credulity too far for it means that unions never raise wages more than they would have gone up anyway under the pull of demand. This indeed would be miraculous if true. It would be more accurate to say that unions do influence wages in two major ways. First, they add to the rise of wages in their own spheres of in fluence, as indicated by several facts: the unbroken rise in manufac turing w^ages since 1945, the enormous growth of fringe benefits, gains in union security and working rules, and elimination of several North-South industrial differentials. Second, unionism is probably even more effective in checking gen eral cuts in wage rates during recessions. In both respects there is a wage push, and it bears some responsi bility for a cost push. However, no one so far as measured the net effect o f collective bargaining upon inflation. To this extent the issue remains moot, although I think it prudent to assume that collective bargaining has had some influence. I think, however, it would be easy to overdo this inflationary impact of unionism, for several reasons. Collective bargaining embraces but one-third of all wage and salary workers. Furthermore, there are reasons apart from unionism for the wage levels in our growing economy: the slowing of the rate of increase o f the labor force, the relative decline in the size o f the pro ductive age group, and the rise in the relative importance of the highly paid technical and professional workers. Other qualifica tions also must be made. EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2529 During expansions, increases in wage costs occur with the wage drift, apart from the effects of newly negotiated increases. Prices will be under upward pressures from both sources, and this is not all. The strategic middle link between wages and prices is the pro ductivity of labor. During 1956 and 1957, and early 1958, the annual rate of increase in labor productivity in the private nonfarm sector dropped very sharply, causing increases in wage costs to exert more upward leverage upon prices. I do not think this change can be attributed to increased featherbedding or tightening working rules. A more obvious reason is the failure of the economy to grow more rapidly and more steadily in recent years because of an unfortunate combination of tight mone tary and fiscal policies with the much-heralded automatic stabilizers. Both of these forces have held back investment. Finally, it must be pointed out that unit wage costs are not the only element in cost, nor wages the sole determinant o f price. I refer here to the behavior of salaries since 1953, which have risen more rapidly than manufacturing wages, the rise of interest payments and o f rents, and also increases in depreciation charges. This review of the wage level question suggests two main con clusions: (1) Although collective bargaining has enhanced the up ward drift o f wages to some extent it has not been the only influence; (2) this net contribution of collective bargaining together with its effectiveness in checking declines in wages has contributed something to the rise o f the postwar price level, but at most its influence has been minor. Now, next I would like to consider the impact of collective bargain ing upon the distribution of income. I shall ignore here the effects o f Federal tax and transfer legislation which, in my judgment, have been the major factor in the reduced inequality of personal incomes. Unions have had some political force here but only as a part of a larger movement. The several studies of the relative shares of labor and of property in the national income have all shown that the proportion going to employee compensation has risen moderately in trend for the past several decades. Explanations of this upward trend attribute little or no importance to collective bargaining; instead, they stress the rapid increase of Government employment, the decline in the pro prietorship industries, the rise of more intensive employee industries, and until recent years, Government restrictions upon rents and rates o f interest. Two reasons account for the apparent negligible influence of col lective bargaining here. On the organizational side the power re quirements for the unions are simply too great, while the incentive to do so is too feeble. It is too easy to win w^age increases on other grounds. On the economic side, private enterprise has proved so far quite capable of keeping ahead of the business agents, despite considerable pressure from wage costs. Management generally has successfully responded with laborsaving innovations, equipment, processes, and the location and organization of work. This has permitted money in real wages to rise without encroachment upon profits. 2530 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS Third, I would like to consider the impact of collective bargaining upon wage structure, which is a problem with multiple dimensions. So far as the relationship of wages among industries is concerned, numerous studies indicate that at most collective bargaining has played a minor role, or, in Professor Keynolds5terms, no seismic dis ruption has been brought about. The question is, W hy? Why has there been no cataclysmic dis turbance in the structure of these differentials? Partly this may be due to a masking effect exerted by prosperous conditions which ex tend wage gains throughout the system. More fundamentally, the explanation lies in power requirements and union motivation. Collective bargaining in a given industry is largely shaped by economic factors peculiar to that industry, its product markets, pro ductivity, skill requirements, and its relation of labor to total cost. These conditions vary from case to case, and they are decisive. On the side of occupational wage structure, there is no doubt that a relative compression has been underway. The relative rates for skilled, unskilled, and semiskilled work have been converging for many years. This trend primarily reflects market forces and centers upon changes in the relative supplies of and demands for skilled and unskilled labor, the broadening of education, decline of immigra tion, and, until recently, the inroads of mechanization upon the de mand for skilled workers. I would conclude that the unions have never had the objective of equalizing all job rates within their areas of effectiveness and that this objective is not likely to emerge in the near future. It is also true that there has been some narrowing of the NorthSouth differential although no studies have been made of this ques tion in recent years. In general, I think the forces that have brought about this nar rowing have reflected in economic influences such as persistent high employment, the outward migration of southern labor, and the south ward migration of capital. Government has also helped with mini mum wage and collective bargaining legislation. The unions, how ever, have been successful in certain industries, such as basic steel, chemicals, petroleum refining, rubber, glass, meatpacking, and bi tuminous coal. Last, the unions have wrought, I think, their most significant im pact on wage structure by designing job classifications and job rates within plants and making more uniform wage rate relationships be tween plants in comparable labor and product market areas. I would say that the success of the unions here has been one that is marked and also clearly of general social benefit. As a general influence, I would hold that collective bargaining has probably exerted more impact in the domain of wage structures than in inflation or the distribution of income, and that in turn its greatest impact upon w^age structure has been at the plant and local product and labor market area. The matter of implications for public policy arises as a result of this brief and admittedly inadequate review. The persistence of creeping inflation and the findings of the Select Committee on Improper Activities in the Labor or Management Field EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2531 together have brought about a substantial change in public opinion toward unions in general and collective bargaining in particular. I would say that today collective bargaining is on trial. Any or ganization capable of profoundly affecting the welfare of all society must stand public scrutiny, and the labor movement is no exception. That this public scrutiny may well lead in turn to new legislation in new areas of public regulation is equally true. Here, however, a word of caution is in order. The question before the Joint Economic Committee in these hear ings is the impact of collective bargaining upon wages and prices. Corruption, violence, and unethical practices here and there within the labor movement, while important on their own terms, are a sepa rate matter and should not determine our judgments in the field of wages and prices. The latter issues should be decided on their own merits. Prudence dictates, I believe, that we conclude collective bargaining has contributed to the rise in the wage level since W orld W ar II. For the same reason it is safe to assume that collective bargaining bears some responsibility for creeping inflation. However, it has not been the main factor in the rise of the price level since 1945. Turning now to remedies, most of these seem to me far worse than the disease. Government wage control, assuming that it would work at all, would upset all dimensions of our present wage structure, mak ing it rigid and sclerotic when it ought to be responsive to change, and crippling seriously our voluntary bargaining system. Since that structure has not undergone serious distortion so far, intervention of this kind is not worth the price. I am also opposed to attempts to atomize collective bargaining through reform of the antitrust laws to outlaw industrywide bargain ing. Industrywide bargaining is neither widespread nor growing to day and it has not been shown to create unique inflationary pressures. Quite the^ contrary. It checks strikes by raising their cost to the union while it permits both sides to take a broader view of their industry. As for dissolution of marketwide unions as such, I think it would increase political rivalries and intensify wage pressures, working greater hardships upon marginal employers. This brings me to monetary and fiscal restraint. To work effectively in today’s economy, where over broad areas wages and prices are quite rigid downward and yet move upward under sellers’ pressure even when demand is adverse, restraint re quires too much unemployment and idle capacity. What this policy has brought to us is a series of weak booms broken at frequent inter vals by recession. There is good reason to think that the cumulative result is a slower rate of growth than we ought to have because the inducement to invest is not adequate. Thus, although gross product turned upward again in the fourth quarter of 1958, investment and durable equipment by the second quarter of 1959 was running 7 per cent behind the same period in 1957. In making this criticism, I do not suggest we go over to wild mone tary expansion. Rather, I propose that we take a calculated risk with a somewhat more liberal policy in the monetary field, coupled with changes in the taxation of corporate profits and in depreciation allow ances that would stimulate more investment of a productivity creating 2532 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS type. Productivity would also be helped under this approach by increased use of existing plants where economies of volume could be obtained in greater measure. Much good could also be accomplished by starting a cutbock of the enormous subsidies to agriculture, divert ing them to tax savings and to more productive fiscal uses and at the same time permitting abandonment of the present uneconomic support-price system. There are also clear advantages in encouraging rather than ob structing competitive imports as a means of policing more effectively the price and wage policies of some of the more highly concentrated industries. The same purpose would also be served by making it easier for new competitors to enter certain fields. Although collective bargaining is under increasing criticism today, and has never been without its seamy side, I wish to affirm my faith in it as a workable and basically desirable social and economic insti tution. I do not think that it should be made to bear a major share o f the guilt for postwar inflation, nor do I think that proposals to remake collective bargaining in the name of fighting inflation are even workable in their own terms. The system of industrial relations we have slowly and painfully developed in this country has generally worked well because it is suited to our pluralistic social order. It has proved to be both elastic and constructive, hence a successful adaptation to our economic en vironment. I therefore voice a conservative plea for caution in urging that we do not enter lightly upon drastic proposals to alter it. It is the radical who voices these proposals, although he may act from conservative premises. Before adopting such ideas, it is well to re member that in Europe the alternative to weak unionism and collec tive bargaining is a strong Socialist movement. Economics is a practical as well as a theoretical discipline. The economist has the obligation to give advice to Government, where, as in investigations of this kind, Members of the Congress have done our profession the honor of seeking its counsel. In turn, the advice we supply should be well founded in evidence and scientific analysis, so that any legislation that may issue from this inquiry can be based upon knowledge rather than speculation regarding the truth. Since there is much that is still not known regarding the impacts of unionism upon the American economy and much that is only sus pected rather than firmly established, it would be dangerous indeed to recommend action in areas about which the profession is still uncertain. For this reason, any recommendations I may wish to make about policy will mostly be negative. Since the end of World War II we have experienced a measure of inflation. Although the rise in prices and wages in the United States has fallen well short of that in most of the leading countries of Europe, the upward movement here has been marked and persistent enough to cause concern. Even though the issue o f inflation is now being greatly overworked to the neglect of more pressing problems o f growth and higher employment, it does demand attention. How ever, this does not mean that present tight money and fiscal policies are necessarily correct. This depends upon the importance we attach to price stability as an end in itself, and upon the confidence we have EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2533 that creeping inflation is demand-induced and can be stopped without sacrifice of adequate growth and employment. The initial postwar inflationary surge during 1945-48 coincided with a great wave of strikes from which large wage increases were gained. This was also the Nation’s first experience with a broadly based and powerful labor movement in peacetime prosperity, in an economy freed of price and wage controls. This was the setting in which was born the idea that inflation was caused by a cost push from wages instead of excess demand from too much money. It was but a short step to conclude that the unions were the under lying cause, hence that they must be controlled in some way. So arose proposals to ban industrywide bargaining, or to break up marketwide unions into local company organizations, or to create enough unemployment to policy wage and price movements. After 1948 the inflation died down and the whole question of unionism re ceded in importance. Even with the brief upward surge of prices and wages during the first year of the Korean war, the inflationary role of unionism commanded little attention. As the economy moved into decline after the third quarter of 1957, prices and wages continued to rise, and concern with creeping infla tion again began to assert itself. In professional discussions as well as public opinion, two unseemingly opposed theories of inflation have dominated the stage. On the one side is the traditional “ demand pull” explanation, and on the other the newer “ cost push” version. We are expected to choose between the two, and Congress and the administration to shape remedies accordingly. This is too neat a choice and well overlooks the possibility that the two forces may be intertwined, aided by other influences not easily identified with either. Even more, “ proof” of either theory usually rests upon factual generalizations drawn from the behavior of very broad aggregative statistics, such as gross measures of wholesale and consumer prices throughout the economy and average hourly earn ings for manufacturing as a whole. Only by recourse to the details of economic structure is reliable knowledge to be acquired. The advocates of demand pull contend that the origin of all in flationary movements lies in an excessive creation of money and speed of its turnover, which draws up the demand for commodities and labor, raising prices and wages directly, and through induced in creases of income, indirectly as well. On this view, the labor unions usually turn out to have supposedly illusory power: They can raise wages in local labor markets here and there, but usually no more than wages would have risen anyway under the pull of labor demand. Apart from a few craft unions that have monopoly power over the supply of labor, trade unions are largely ineffective for raising wages and completely so for pushing up the entire wage level. Instead, their real power is political. By exploiting the popular notion of full employment they drive the Government into monetary and fiscal policies that are excessively expansionary. Working through demand, these policies draw up wages and prices and we have inflation, for which the unions get the credit or the blame. I f acted upon, this analysis calls for tight monetary and fiscal policies, and perhaps the sacrifice of full employment as a goal. Apart from 2534 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS legislative attack upon certain restrictive union rules affecting work and employability, the demand-pull approach suggests that little need be done about wage setting under collective bargaining, al though a few of its academic proponents would like to break up the unions into company-by-company organizations in the spirit of older ideas of atomistic competition. Those who uphold the cost push include some who believe that the push is centered in union-made wages, and others who include admin istered prices imposed in the corporate sector and under Federal price policy for agriculture. In essence, the cost-push doctrine holds that prices and wages are no longer effectively policed by competition, but are raised by sellers even when elasticities and movements of de mand are adverse to them. Moreover, the power that sellers now possess enables them to keep prices and wages from falling in reces sions, so preventing correction of excessive upward movements during expansions. In this view the market power of sellers is now so wide spread that they can dictate the whole national levels of prices and wages. By contrast, the demand-pull version continues to insist that most sellers can only fix a few local prices and wages, while the overall average levels continue to be determined by the quantity o f money and the speed of its turnover. The difference between the two views is, of course, monumental. On the side of causation, the demand-pull school sees inflation as starting with money, while the cost-push group points to price and wage mak ing by sellers, making money passive. On the policy side, the advo cates of cost push must first choose between drastic proposals to con trol the wage-setting power of unionism and acceptance of creeping inflation as tolerable for the pursuit of growth and high employment. Often they are led into proposals for Government control of wages and prices or prior review of proposed increases. Quite generally they conclude that tight monetary and fiscal policies alone are almost wholly inappropriate for checking sellers’ inflation, because these policies would require amounts of unemployment and of idle produc tive capacity that would be intolerably large from the standpoint o f public opinion and our international interests and commitments. In my judgment, the problem of inflation today is greatly exag gerated, and the responsibility of unionism for it even more so. In saying so, I do not wish to suggest that inflation is a thing to be taken lightly or that the unions have had nothing whatever to do with it. Since 1945 wholesale prices have risen about 74 percent, consumer prices 62 percent, and hourly earnings of nonsupervisory production workers in manufacturing 118 percent, at simple average annual rates o f 5.2, 4.4, and 8.4 percent respectively. These simple facts show that inflation has been going on, although there has been no significant change in the wholesale price level, and only a slight rise in consumer prices, for more than a year past. Mention of these global figures proves nothing, of course, about causation, since the same facts can be used to support either theory. A t the same time, these measures refer to net movements over a 14-year period. The big surges in the wholesale price level occurred during 1946-48, 1950-51, and 1956-58. The first two cases favor the demand-pull doctrine, because they were accompanied by large in EM PLOYM ENT, G R O W T H , AND PRICE LEVELS 2535 creases in the supply of money. However, the last one coincided with monetary and fiscal stringency for much of the time, terminating in recession during late 1957 and all of 1958. Accordingly, it does not fit well into a demand-pull explanation. On the side o f manufacturing wages, it must be noted that average hourly earnings have risen every year since 1945, despite declines in overall business activity during late 1945 and early 1946, in 1949, in 1953-54, and in 1957-58. These facts indicate the ability o f the unions in manufacturing to bring about going annual increases in wage rates every year, not to mention rapidly expanding fringe bene fits whose value is not included in these figures. The story is similar for the other sectors in which unionism is strong—construction, transportation, and mining. In no single year since World W ar I I have hourly earnings fallen, despite three inter vening recessions, a fact that attests to the ability of modern unionism to enforce the “ no-cuts” doctrine. Equally important, however, increases in hourly wages in manu facturing have been much larger during times of business expansion than during times of decline. During 1950-51 they rose by an annual average of 9.4 cents an hour (6.7 percent) ; during 1955-57 the aver age annual rise was just under 9 cents (4.8 percent) ; while between June 1958 and June 1959 they advanced 12 cents (5.7 percent). By contrast, they rose only 5 cents in 1949 (3.7 percent), 4 cents in 1954 (2.3 percent), and 6 cents in 1958 (2.9 percent)—all years of recession. The inference accordingly suggests itself that the level and direc tion o f change of total business activity, and o f employment-unemployment in particular, does influence the rate at which increases are made. However, at no time has recession cut deeply enough to stop all advance in the manufacturing wage level, much less to bring about a general reduction in wage costs. I suspect that the unemployment rate would have to rise to at least 10 percent, and to remain there for some time, for such a decline to occur. Certain reservations are necessary concerning these figures. They do not include the cost of certain types of fringe benefits that are becoming increasingly important. Hence they understate changes in wage costs. Furthermore, they are limited to production and maintenance workers in manufacturing, although they are often wrongly used to indicate movements in the wage level of the entire economy, including important sectors such as government, agricul ture, and services, where unionism is usually weak. Also, these figures tell us nothing about wages and salaries for overhead labor in manufacturing. Finally, increases in hourly earnings are not a direct measure of increases in basic job wage rates, because they in clude overtime and premium pay, incentive earnings, promotions, and merit increases under existing wage scales, and probably shifts of workers to higher wage firms and industries. Even if basic job rates were not increased at all, average hourly earnings would nonetheless rise during business expansions, because of theses other influences. This makes it all the more difficult to determine the actual movement of pure wage rates, and all the more dangerous to attribute the whole of any increase in hourly earnings to the w^age-fixing activities of unionism. 2536 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS At the same time, this phenomenon o f “ wage drift” originates from changes in output and wrorking hours in the economy rather than from newly won demands of unions. If, then, we were to adopt a strategy of indirect wage control to check inflation, it follows that output would have to be severely constricted, at a heavy cost in un employment, if that strategy is to succeed. With these preliminaries on record, I propose now to consider the impacts of unionism upon some vital areas o f the economy, and then to follow wTith some recommendations regarding policy. 1. Impact of collective bargaining upon the wage level.—Do unions raise wages in the markets with which they are involved? O f course they do. The more important question is whether they raise wages more than they would have risen under the influence of increasing demand or decreasing supply. The demand-pull school suggests that in general the unions add nothing, but to accept this contention is to stretch credulity beyond the breaking point. The most that need be conceded is that aggregate demand must be high enough to prevent a major depression if the upward pressure of unionism is to be effective. To this extent, the demand and cost sides are linked together. The reasons for crediting or debiting unionism with some net con tribution to the rise o f the wage level are good ones. Since the war the unions generally have won enormous fringe benefits for health, welfare, and pension plans. In vital segments of corporate industry they have achieved the union shop. They have wiped out regional differentials in some important industries in the South. Each year, whether it was one of recession or recovery, they have brought off annual increases in wage rates, together with improvements in other benefits. At no time have general wage reductions been attempted, even with 7 percent unemployment in 1958. In view of these broad and diverse manifestations of bargaining power, it would strain the imagination too much to be asked to conclude that w^ages and supplements to wages on average have been increased through collective bargaining at most only by as much as would have occurred anyway in a com pletely nonunion economy. Particularly is this so wiien it is recalled that between 1950 and 1958 there were 2 years in which gross demand actually fell (1954 and 1958) and three others in which it rose only moderately (1952, 1956, and 1957). At no time since 1952 has total demand pressed seriously against productive capacity, which makes it quite difficult to reconcile recent wage behavior with the demand-pull theory, all the more so when it is recalled that corporate profit margins showed a tendency to shrink during 1956-58 and no tendency to widen after 1950. However, no one has yet determined how much has been the net contribution of collective bargaining to the upward movement of wages, although there is reason to doubt that it has been large. Paradoxically, the influence of unionism may be greatest when it is least evident or suspected, and least when it seems most apparent, as Clark Kerr suggests. During recessions wage gains are generally less, but collective bargaining shores up the structure against decline, preventing in unobtrustive fashion the downward adjustments in the E M P L O Y M E N T , G R O W T H , AND PRICE LEV ELS 2537 wage and price levels that would be necessary to offset preceding ad vances during years of expansion. Even if unionism added nothing to these advances, its imposition of much downward rigidity would impart some upward movement to the trend of wages and prices. But since it does add something to the advances as well, its net overall impact is greater. And at such times, when wages are most likely to rise anyway, unionism is most vulnerable to blame for inflation. Yet it is easy to overdo the inflationary impact of unionism. Col lective bargaining today embraces but one-third of all wage and salary employees, hence its direct weight upon the wage level is correspondingly reduced, although its wage gains do tend to spread competitively to the nonunion sectors. Furthermore, there are rea sons apart from unionism for the wage level to rise in our growing economy—the slowing in the rate of increase in the labor force; the relative decline in the size of the productive age group; and the rise in the relative importance of highly paid technical and professional workers. Other qualifications are also in order. During expansions, in creases in wage costs occur with the wage drift, apart from the ef fects of newly negotiated increases. Prices will be under upward pressure from both sources. But this is not all. The strategic mid dle link between wages and prices is the productivity of labor, for the quality of product over which wage costs must be spread is equally decisive with increased wage costs themselves for raising unit labor costs. During 1956-57 and early 1958, the annual rate of increase in labor productivity for the private nonfarm sector dropped very sharply, causing increases in wage costs to exert more upward leverage upon prices. There is no evident basis for attributing this sag in produc tivity gains to a sudden tightening of working rules or increased featherbedding by unions generally. A much more obvious reason lies in the failure of our economy to grow more rapidly and more steadily in recent years, because an unfortunate combination of tight monetary and fiscal policies with the much-heralded automatic sta bilizers has checked the growth of aggregate demand. In turn this has held back investment in new capacity and in productivity-gener ating improvements. Finally, it must be pointed out that unit wage costs are not the only element in costs, nor wages the sole determinant of price. Sala ries for overhead labor, whose relative importance is increasing mark edly, and for officials have been rising faster than wages for nonsupervisory workers. These represent costs over which the unions have little direct influence. Since 1953 interest payments have risen by 50 percent, from rising interest rates and increased credit financing. Depreciation charges have also been advancing, partly from liberalizing legislation in 1954. And last, there are important sectors of industry and agriculture where prices are subject to upward manipulation by sellers or by Government in behalf of sellers. These prices, like wages under collective bargaining, are markedly less sensitive to classical competitive forces. Clearly, their upward 38563— 59l—!pt. S— 7 2538 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS movement since the war should not be attributed solely to the move ment of wages, much less to union wage policies alone. This review of the wage-level question suggests two main conclu sions. One, although collective bargaining has enhanced the upward drift o f wages to some extent, it has not been the only influence work ing in this direction. Second, this net contribution of collective bar gaining, together with its effectiveness in checking declines in wages, has contributed something to the rise in the postwar price level, but at most its influence has been minor. Many other forces have been at work both from the side of demand and of supply. Since the problem o f creeping inflation admits of no simple answer, by the same token it cannot be overcome by efforts to control wages alone. 2. Impact of collective bargaining upon the distribution of in come.— Since this question centers upon collective bargaining, it necessarily excludes the political activities of unionism in behalf of higher tax progressivity, rent control, and depressed rates of interest in the early postwar years, and increased transfer payments of a redistributive type. Therefore I shall note only that tax-transfer legislation in the past quarter century has effected a considerable reduction of inequality in personal incomes, and that the labor move ment has generally supported the measures by which this develop ment has been attained. Only because this legislation enjoyed a broad basis of popular support outside the ranks of organized labor has its enactment proved possible. The several studies of the relative shares of labor and of property in the national income have all shown that the proportion going to employee compensation has risen moderately in trend for the past several decades. Cyclically, labor’s share declines during expansions and rises during contractions. Explanations of the upward trend in labor’s share attribute little or no importance to collective bargaining. Rather, they stress the follow ing: the rapid increase of Government employment, the decline of the proprietorship industries, the rise of more labor-intensive “ employee” industries, and until recent years, Government-depressed rents and interest rates. Significantly, this longrun rise in labor’s relative share has not encroached upon the share of corporate profits before taxes—a useful test of the redistributive power of collective bargaining. However, crude data for 1955-58 do indicate some ad verse movement in profits and profit margins, for which the 1957-58 recession and continuing cost pressures were mainly responsible. It is sometimes argued that the bargaining power of unionism exerts a concealed effect upon relative shares because it operates to prevent labor’s share from falling. This claim presumes that capital invest ment is mainly a competitive substitute for labor, operating unless offset to reduce the demand for labor and thus to depress wages. While machines and men do compete in specific situations* in the large they are complementary to each other. So far no evidence has been forthcoming to show that the competitive relationship dominates, hence that investment is highly labor displacing in nature. In the main, investment raises the demand for labor in the economy as a whole. Two reasons account for the apparently negligible influence of col lective bargaining in raising labor’s relative share. On the organiza- E M P L O Y M E N T , G R O W T H , AND PRICE LEV ELS 2539 tional side, the power requirements are simply too great while the incentive to do so is too feeble. It would be extremely difficult for unions in the American environment to win control over the employ ers’ freedom to innovate, to determine the size of their work force, to adjust price, and to dispose of profits, while it would be equally hard to bring about Government price control or a fair shares tax policy in peacetime. W hy make the attempt when these ideas enjoy little support among* the rank and file and when more conventional bargaining policies have long proved entirely capable of providing so rich an economic yield ? On the economic side, private enterprise has so far proved quite capable of keeping ahead of the business agents. Despite consider able pressure from wage costs, management generally has successfully responded with labor-saving innovations in equipment, processes, and in the location and organization of work. These changes have been adequate to prevent a longrun fall of returns to capital, preserving the inducement to invest. Yet they have not gone far enough to create serious unemployment or falling real wages. Instead money and real wages could rise without encroachment upon profits. A flexible and growing system of competitive private enterprise has proved not only compatible with vigorous collective bargaining, but the best guarantee of its own survival. 3. Impact of collective bargaining upon wage structure.—Regard ing wage differentials among industries, numerous studies suggest that at most collective bargaining has played a minor role in chang ing them. In Lloyd Reynolds’ words, no “ seismic disruption” has been brought about. The low-wage industries of earlier decades are still at the bottom today, and similarly for the high-wage ones. While some observers do think that new unions have a wage advan tage for a time and that unionism obtained relative gains in the twenties, when Government policy did not intervene for the un organized and employment was not full enough to cause competi tion to work in their favor, this is the most that has been conceded professionally to collective bargaining. Rather, the forces found decisive for shaping the structure of interindustry wages have been traditional economic ones : The rate o f improvement of labor productivity in the industry, the degree of market concentration among a few large firms, the skill mix and sex ratio of the work force, and the ratio of labor to total costs. Why has collective bargaining apparently wrought no cataclysmic disturbance in the structure of these differentials? Partly the rea son may be the masking effect exerted by generally prosperous con ditions and high employment since 1942. Temporary wage gains in unionized industries then tend to be matched elsewhere through induced shifts in labor supply, generally rising demands for labor, and nonunion employers’ fear of unionization. More fundamentally, the explanation lies in power requirements and union motivation. Collective bargaining in a given industry is largely shaped by economic factors peculiar to that industry: its product markets, its productivity, its skill requirements, and its ratio of labor to total cost. These conditioning elements vary from case to case. Neither the employers nor the union can afford to overlook 2540 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS them when they are adverse. While new unions show more aggres siveness, interindustry comparisons usually have little value as a bargaining symbol to the rank and file save in markedly inflationary times. By contrast, the older unions usually develop a profound although unexpressed respect for economic restraints. Thus these restraints operate to sort out industries as well as unions, perpetuat ing well-established differences. Turning to occupational wage structure, there is no doubt that a relative compression has been underway in the United States for many decades, while the same phenomenon has been noted for many other countries. This trend has primarily reflected market forces, which have centered upon changes in the relative supplies of and demands for skilled and unskilled labor—the broadening of educa tion, the decline of immigration, and, until recently, the inroads of mechanization upon the demand for skilled workers. During wartime, Government wage control favored the lower-paid groups. The emergence of large-scale industrial unionism had a similar effect during markedly inflationary periods between 1’941 and 1953, wThere settlements favored flat-rate increases to keep up with the cost of living. More recently, however, the craft severance move ment within these unions has checked leveling tendencies, marking a return o f structural problems to wage determination. It seems quite safe to conclude that the relative compression of occupational differentials since World War I lias been more the effect o f market forces than of equalitarian ideas. At no time has compres sion gone far enough to create serious shortages of skilled people or surpluses o f the unskilled, as has occurred abroad. With the rise of automation and other new technologies requiring increased numbers of technical and professional employees at the expense of the semi skilled, occupational differentials may well begin to widen. I f so, the organizing goals of the unions are not likely to be thwarted by insistence upon equalitarian ideas that have little basis of popular appeal. Turning to wage differentials among regions, relatively little re search has appeared since the early postwar years. Studies at that time revealed that by industry averages the South ranked lowest, and that by skill grades this was also true, with the disadvantage largest for unskilled workers. Within the South, the relative wage premium for skill was greater than elsewhere, as might be expected. However, the southern differential has been narrowing for about a quarter century in its various dimensions relative to other regions. Partly, this has reflected economic forces—persistent high employ ment, the outward migration of southern labor, and. the southward migration of capital. Government has helped with the minimum wage and legislation to promote collective bargaining. Particularly in industries with national or interregional product markets, unionism has been successful in wiping out or reducing the North-South differ entials— in basic steel, chemicals, petroleum refining, rubber, glass, meatpacking, and bituminous coal. Today the southern differential persists mainly in local-market industries, in those that are weakly organized, and in those centered in rural areas where abundant sup plies of cheap farm labor are still available. EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2541 The last dimension of wage structure affected by collective bargain ing is that of the plant and of interplant relationships within the same labor market or industry. There is impressive agreement among spe cialists that one of the great contributions of collective bargaining lies in the development of a more orderly structure of jobs and o f job rates within the industrial plant. The price of each type of work accord ingly becomes standardized, while rate relationships among jobs are made more responsive to employee ideas of equity and fairness. Be cause the parties themselves have intimate knowledge of the facts and of the interests at stake, they can by negotiation achieve a more satis factory wage structure than could be had in any other way. As alter natives, Government regulation and the classical competitive market are very poor substitutes. Wage regulation is necessarily rigid and insensitive because it de pends upon decisions by outsiders who are circumscribed by external rules and who are not intimately informed. The competitive market, most notable for its imperfections where labor is involved, affords no readymade structure of jobs and of job prices, and so is quite irrele vant as a practical instrument for the detailed design of a plant wage system, granting that market rates must be used as peg points in the formation and adaptation of that system. The very inadequacy of competition within local labor markets affords nonunion employers diverse degrees of discretionary power for setting rates from plant to plant. Typically the result is much diversity of rates for comparable jobs within the same industry and labor market. Here the union, in its effort to “ take wages out of com petition,” actually simulates the processes that competition would provide if it were more effective—making more uniform among plants the rates for comparable jobs and, as a labor exchange, in creasing the mobility of junior workers. In some cases, notably the United Steelworkers, interplant wage structure as a whole has been made more uniform by means of a joint evaluation plan. At the same time, the movement for wage uniformity has on occa sion been sacrified to protect employment in weaker firms. Moreover, it should not be concluded that collective bargaining ends all competi tion for or among workers. Nonwage differences among firms still affect hiring and turnover as well as morale and efficiency. Uniform wages do not mean equal wage or total costs, because differences in quality o f labor, equipment and technique, and managerial efficiency do persist. Finally, firms continue to compete over shares in the product market, which in turn pits plant work forces against each other. As a general inference I would hold that collective bargaining has had more influence in the diverse domain of wage structure than in affecting either inflation or the distribution of income. The rela tively decentralized American system of collective bargaining makes unionism most effective in shaping wage structures within plants and among firms within common product markets or common local labor markets. By contrast, it continues to be markedly less effective in changing interindustry and occupational wage relationships, and for the same reason. 2542 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS 4. Implications for public policy.—The persistence of creeping in flation and the findings of the Select Committee on Improper Activi ties in the Labor or Management Field together have brought about a substantial change in public opinion toward unions in general and collective bargaining in particular. That any organization capable o f profoundly affecting the welfare of all society must stand public scrutiny goes without saying. The labor movement is no exception. That this public scrutiny may well lead in turn to new legislation and new areas of public regulation is equally true. Here, however, a word of caution is in order. The question before the Joint Economic Committee in these hear ings is the impact of collective bargaining upon wages and prices. Corruption, violence, and unethical practices here and there within the labor movement, while important on their own terms, are a sepa rate matter and should not determine our judgments in the field of wages and prices. The latter issues should be decided on their own merits. Prudence dictates, I believe, that we conclude that collective bar gaining has contributed to the rise of the wage level since World War II. For the same reason it is safe to assume that collective bargain ing bears some responsibility for creeping inflation. However, it has not been the main factor in the rise of the price level since 1945. Turning now to remedies, most of these seem to me far worse than the disease. Government wage control, assuming that it would work at all, would upset all dimensions of our present wage structure, mak ing it rigid and sclerotic when it ought to be responsive to change, and crippling seriously our voluntary bargaining system. Since that structure has not undergone serious distortion so far, intervention of this kind is not worth the price. I am also opposed to attempts to atomize collective bargaining through reform of the antitrust laws to outlaw industrywide bargain ing or to dissolve marketwide unions into company-by-company or ganizations. Industrywide bargaining is neither widespread or grow ing today, and it has not been shown that it creates unique inflationary pressures. Quite to the contrary. It checks strikes by raising their cost to the union while it permits both sides to take a broader view of their industry. As for dissolution of marketwide unions as such, I think it would increase political rivalries and intensify wage pressure, working greater hardships upon marginal employers. This brings me to monetary and fiscal restraint. To work effectively in today’s economy, where over broad areas wages and prices are quite rigid downward and yet move upward under sellers’ pressure even when demand is adverse, restraint requires too much unemployment and idle capacity. What this policy has brought to us is a series of weak booms, broken at frequent intervals by recessions. There is good reason to think that the cumulative result is a slower rate of growth than we ought to have, because the inducement to invest is usually not adequate. Thus, although gross product turned upward again with the fourth quarter of 1958, investment in pro ducers’ durable equipment even by the second quarter of 1959 was still running 7 percent behind the same period in 1957. In making this criticism, I do not suggest that we go over to wTild monetary expansion. Rather I propose that we take a calculated EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2543 risk with a somewhat more liberal policy in the monetary field, coupled with changes in the taxation of corporate profits and in depreciation allowances that would help to stimulate more investment of a pro ductivity-creating type. Productivity would also be helped under this approach by increased use of existing plant, where economies of volume could be obtained in greater measure. Much good could also be accomplished by starting a cutback of the enormous subsidies to agriculture, diverting them to tax savings and to more productive fiscal uses and at the same time permitting abandonment of the present uneconomic support-price system. There also are clear advantages in encouraging rather than ob structing competitive imports, as a means of policing more effectively the price-and-wage policies of some of the highly concentrated indus tries. The same purpose would also be served by efforts to make it easier for new domestic competitors to enter certain fields. Although collective bargaining is under increasing criticism today and has never been without its seamy side, I wish to affirm my faith in it as a workable and basically desirable social and economic institution. I do not think that it should be made to bear a major share of the guilt for postwar inflation. Nor do I think that proposals to remake collective bargaining in the name of fighting inflation are even work able on their own terms. The system of industrial relations we have slowly and painfully developed in this country has generally worked well because it is suited to our pluralistic social order. It has proved to be both elastic and constructive, hence a successful adaptation to our economic environment. I, therefore, voice a conservative plea for caution in urging that we do not enter lightly upon drastic pro posals to alter it. It is the radical who voices those proposals, al though he may act from conservative premises. Before adopting such ideas, it is well to remember that in Europe that alternative to wreak unionism and collective bargaining is a strong Socialist move ment. The C h a i r m a n . Mr. Ornati. STATEMENT OF OSCAR A. ORNATI, ASSOCIATE PROFESSOR OF ECO NOMICS, AND DIRECTOR OF LABOR EDUCATION CENTER, NEW SCHOOL FOR SOCIAL RESEARCH, NEW YORK, N.Y. Mr. O r n a t i . Sir, I would like to read my paper. I am here at the invitation of the committee to present my views as to the economic significance of collective bargaining in general and as to certain specific practices in particular. I will present a limited number of considerations on— 1. The general impact of trade unions and of the process of collective bargaining on the general wage level and on wage differentials; 2. The responsiveness of wages to changes in the level of un employment; and 3. The implication of long-term collectively bargained con tracts with automatic cost of living and productivity adjustments. 2544 E M P LO Y M E N T, G R O W T H , AND PRICE LEVELS I . T H E IM P A C T O F T R A D E U N IO N S A N D C O L L E C T IV E B A R G A IN IN G O N T H E GENERAL W AGE LEVEL Wage setting in an economy characterized by the presence of sub stantially unionized sectors has long concerned us. Views on this topic are varied and there does not exist a clear-cut consensus of opinion on how precisely trade unions affect the economy. The ques tion o f the unions’ impact on wage setting is here viewed in the par ticular context of the committee’s inquiry which, although not limited to it, emphasizes the problem of inflation. A preliminary observa tion on inflation is, therefore, in order. A P R E L IM IN A R Y OBSERVATION ON IN F L A T IO N Data on the recent movement of prices are available to the com mittee and they have been discussed and analyzed in previous hear ings, With other witnesses that have appeared in front of this com mittee, and with a large group of economists that have analyzed the problem, I feel that much ill-informed talk about inflation has placed the problem in improper prospective, aroused unjustified fears, and led to ill-advised policy decisions. Talk about inflation on the part o f both unions and management, but primarily management, has given Americans an inferiority complex on the subject of inflation that the facts do not warrant. The discussion of the impact of trade unions and collective bargaining on the economy is seriously vitiated by the extent to which Americans are unaware of the actual facts of price increases and particularly of the fact that the rate of price increases is diminishing. In the decade 1939-48, the Consumer Price Index, which reflects changes in the general price level, rose by 73 percent. From 1948 to date, it rose by slightly less, than 21 percent. In the first 5 years of the decade 1948-58, which ended in December 1953, the Consumer Price Index rose by 11.3 percent ; in the 5 years ending December 1958, it rose by only 7.7 percent. Between December 1958 and July 15, 1959, the Consumer Price Index rose by 0.97 percent, which means that between December 1953 and today, prices rose by a total of 8.7 percent, or at a rate of about 1y2 percent per year. When the rise in the Consumer Price Index in the United States is compared with the rise in prices in the rest of the world, as is done in table I below, the limited nature of our inflation is even more obvious. Thus, even though our economy has developed several insti tutions and forces—and trade unions are among them—that tend to produce a slowly rising price level, the reality of our inflation is much less serious than important Government officials, the press, and cer tain businessmen would have the country believe. Those who see inflation here as a Trojan horse filled with union leaders are thus wrong, as the “ horse” just is not really there. EM PLOYM ENT, T a b le G R O W T H , AND PRICE 2545 LEVELS I .— R is e i n th e C o n s u m e r P r i c e I n d e x in 1 6 in d u s t r ia l c o u n t r ie s i n E u r o p e a n d N o r t h A m e r ic a , 1 9 4 8 -3 8 a n d 1 9 5 3 - 5 7 A ustria_____ _ _ _ __ ___ B e lg iu m ,._______ ___ ___ C anada, ___ D enm ark___ ___ _ _ _ _ _ F in lan d _____________________ F ra n c e ,-, __ _ ___ _______ G reat B rita in ._ ___ ___ __ Ireland, _ __ __ _____ _ 1948-53 1953-57 100.0 5.3 19.0 23.5 56.3 06. 7 29.9 26.6 12.0 7.0 6.0 16.0 20.0 6.0 16.0 12.0 1948-53 Ita ly __________ N etherlands___ N orw ay_______ S pain_________ Sw eden_______ S w itzerland___ U nited S tates, , W est G erm any 16.3 28.2 35. 1 26.6 29.9 4.2 11.1 7.5 10.0 8.0 12.0 23.0 13.0 5.0 5.0 6.0 Source: C ited b y Prof. Sum ner Slichter in his address to the 15th A m erican As sem bly, in “ W ages, Prices Profits, an d P ro d u ctiv ity ,” p. 169. T H E Q U E STIO N S T H A T NEED TO BE AN SW E RE D Disagreement among economists on their evaluation of the impact of trade unions and collective bargaining is due to the complexity of the problem. This is, in part, because of the intermingling of ideol ogy and group self-interest and, in part, because of the interdepend ence o f economic forces which obscure identification of the union’s effect. The basic difficulty probably lies in the fact that the question o f the union’s impact is essentially in the “ iffy” category. Indeed, a complete answer requires knowing what the level and structure of wages would be if trade unions were not in existence. This, obvi ously, cannot be determined empirically. A ll available studies on the economic impact of trade unions are therefore based, implicitly or explicitly, on comparisons which cannot be carried out with any de gree of precision. In economic matters, all one can do is call upon experience, and experience is different from experiment. On the problem of the trade union’s impact on wages, “ there is only experi ence ; and the knowable reality from this experience is little more than conjecture. To speak with full assurance in this area is to speak from prejudice or from ignorance or both.” (Cf. Clark Kerr, “ The Impact of Unions on the Level of Wages” in “ Wages, Prices, Profits, and Productivity,” the American Assembly, Columbia University, June 1959, p. 99.) The impact of the union has most often been analyzed by— (1) Studying how union wages have moved in comparison with nonunion wages; (2) Studying wage movements in periods of considerable or ganization and comparing them with the wage movements in periods of limited or no organization; and (3) By comparing, at the same point in time, the wage move ments of countries with well-organized labor movements with the wage movements of countries with weakly organized labor movements. In addition, in the present context, the question of why wages have, since 1950, risen more than prices and productivity must also be answered. T H E A V A ILA B L E EX PE RIEN CE 1. The results of studies which compared the movement of Ameri can union and nonunion wTages are, in my judgment, inconclusive. (For a summary of the pertinent literature see G. H. Hildebrand, 2546 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS “ The Economic Effects of Unionism” in “ A Decade of Industrial Relations Research,” Harper & Bros., 1958; L. G. Reynolds, “ The Impact o f Collective Bargaining on the Wage Structure o f the United States” ; and Clark Kerr, “ Wage Relationships—The Comparative Impact of Market and Power Forces” in “ The Theory of Wage Deter mination,” New York, Macmillan, 1957.) Indeed, in surveying wage movements in the United States from the turn of the century to 1956, one can conclude that, with the exception of the period 1936-37 and the situation of the construction and printing trades, there is no evi dence of gains in union over nonunion wages. Although no extensive study of the movement of union and nonunion wages from 1956 to date has been completed, there is sufficient evidence to suggest that in this period nonunion wages have risen faster than union wages. Certain valid preliminary observations on this develop ment have already been presented to the joint committee. (Cf. Peter Henle, “ Price Movements in Recent Years” in “ Commentaries on the Relationship of Prices to Economic Stability and Growth,” joint committee print, Oct. 31,1958, 85th Cong., 2d sess.) The fact that union wages have not risen faster than nonunion wages is, on the other hand, no evilence that unions have no impact on wages. Workers in nonorganized sectors of the economy may benefit from wage increases due to the general impact that trade unions have on the economy. The nonorganized sector may, so to speak, have played the “ follow the leader” game. Argument following approximately the line here suggested has been ably presented to this joint committee by Fackler of the chamber o f commerce (Walter D. Fackler, “ Commentaries on the Relationship o f Prices to Economic Stability and Growth,” joint committee print, Oct. 31, 1958, 85th Cong., 2d sess., pp. 95-58) and earlier by Prof. Abba Lerner (Abba Lerner, “ Inflationary Depression and the Regu lation of Administered Prices” in “ The Relationship of Prices to Economic Stability and Growth,” joint committee print, Mar. 21,1958, 85th Cong., 2d sess., pp. 257-268). They argued— reaching opposite policy conclusions—that the trade union impact had a definitely in flationary “ spillover” effect on wage in the nonorganized sector. Such an effect, although not easily demonstrable in statistical terms, un doubtedly exists. I f such a development is viewed as undesirable— and I do not so view it—let us be quite clear as to where the blame lies. It is not the union that is the thorn but the free market through which the “ spill over” operates. However, history shows that, with a free market, forces initiating wage increases in at least a portion of the economy always exist. 2 Comparing the wage movements in the United States for the periods between 1900 and 1910 and between 1947 and 1957, two periods of comparable economic growth but very different degrees of organ ization, shows that while money wages grew faster than productivity in both periods, wages grew faster in the second period than in the first. In 1958 and 1959 wages have continued to grow, but only half as fast as in the second half of 1957. Assuming a constant increase in productiv ity for both periods, Prof. Clark Kerr calculated the price impact of the greater comparative wage increase of the second period at one-half of 1 percent per year (Clark Kerr, op. cit., note 1, p. 101). I find it 2547 E M P L O Y M E N T , G R O W T H , AND PRICE LE V ELS impossible to ascribe even such a limited push on prices entirely to trade union action, even though the trade unions have contributed to the greater tightness of the labor market of the second period to which the cause for the greater rise in wages must be ascribed. 3. The postwar wage experience with wage movements o f some Western European countries is summarized in table II. In all seven of the countries included in the table, manufacturing workers’ earn ings increased faster than in the United States, a point to be remem bered when analyzing the problem of foreign competition in the American market. Although comparisons o f such broad aggregates as reported in table 2 are, at best, approximate, the table shows no correlation between wage movements and the strength of national labor movements. What the table underlines is that forces aside from union action contribute to wage increases. T a b le II.— In d e x e s 1 o f h o u r ly m o n e y e a r n in g s i n m a n u f a c t u r in g i n s e le c te d c o u n t r ie s , 19 4 6 -6 -7 [1950=100] 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 F ran ce___ ____________ G erm an y. _____ ________ H olland _ ______ __ _ Ita ly _____________________ N orw ay ___ ________ Sw eden. _ ____ ____ U nited K ingdom ___ _____ U nited States___________ 37 70 81 79 74 79 74 53 73 87 71 87 85 87 84 81 82 92 94 92 93 93 92 91 94 92 99 94 96 96 95 100 100 100 100 100 100 100 100 128 113 108 no 114 121 110 108 148 122 110 115 127 144 118 114 152 127 113 118 133 150 125 120 162 130 132 122 140 156 132 123 174 139 136 129 148 168 143 130 187 152 150 138 159 183 155 135 202 166 169 165 141 Sources: Intern atio nal L abor Review , Statistical Supplem ents; U .N , Statistical Y earbooks. W AG E M O V EM E N TS IN T H E 1 9 5 0 ’ S The salient facts of the w^age and price movements o f the 1950’s are: (1) Wages rose faster than prices, (2) The rise in compensation per man-hour was nearly twice as large as the gain in real output, and (3) Labor costs per unit of output have risen by almost any measure o f labor costs. These facts, which show that whatever caused the limited price in crease did w7ork through wages, coupled wTith the publicity given many collectively bargained wage increases, has led to the identifica tion of the trade union as responsible for whatever general rise in price levels has occurred. This observation, in turn, has led to the notion that we have been suffering from an inflation of the “ cost push” variety rather than the classical type of inflation in which wages are pulled up by demand. It might be worthwhile to clarify the essence of the difference be tween a “ cost push” and a “ demand puli” inflation. In simple terms, asserting that we are suffering from a “ cost push” inflation means that manufacturers are paying higher wrages and are charging higher prices because they had to, while asserting that wre are suffering from a “ demand pull” inflation means that manufacturers are paying higher wages and charging higher prices because the consumer wanted them to. What is important is that whether they had to because of 2548 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS labor pressure or were told to because of consumer pressure, Amer ican manufacturers are charging higher prices and consumers are paying these slightly higher prices without grumbling. Demand for labor in the 1950’s remained at a higher level through out and, although we have not had large general labor shortages, there exists ample evidence of localized labor shortages and of short ages in certain categories of labor that led employers to offer gen erally higher wages to obtain additional labor. As will be explained below, existing shortages were often strengthened by an increased fragmentation of the labor market brought about by the wide ac ceptance o f certain employer hiring and promotion policies and by certain practices agreed upon by labor and management through col lective bargaining. There is no doubt in my mind that in certain cases, and the fullfashion hosiery and hat industries may be given as examples, em ployers were most unhappy at having to pay higher wages and were pushed into doing so. In other cases, employers found it profitable to bid more for labor either to get additional workers or to get work ers of certain particular skills or characteristics. Thus, it seems to me that the situation since 1950 can be summarized as being one in which “ excess demand for labor pulled up wages in the sectors in which trade unions are weakest, and trade unions pushed up wages in sectors where excess demand was weakest” (see James S. Dusenberry, “ Underlying Factors In The Postwar Inflation” in “ Wages, Prices, Profits and Productivity,” the American Assembly, Columbia University, June 1959) and price increases can easily be passed on, T H E ECON O M IC IM P A C T OF T H E COLLECTIVE B A R G A IN IN G PROCESS Studies relating to the impact of the bargaining process rather than the existence or pressure of trade unions have brought out cer tain facts that help clarify the overall economic impact of the cur rent pattern of payments for labor services. Recent labor market studies have indicated the extent to which the labor market is inef fective and sluggish; how the workers’ ability and willingness to shift freely among employers, industries, and geographical areas in response to economic inducements is decreased with the spread o f col lective bargaining. There is no doubt in my mind that the widely accepted policy of promoting from within, tight seniority systems, more careful hiring, formal systems of wage and salary administra tion, discrimination in hiring on the grounds of race or color, etc., have reduced horizontal labor mobility and consequently created “ artificial” bottlenecks which have led to wage increases. In this way, collective bargaining as an institution has put into motion forces that contribute to raising the general level of wages and has brought about an allocation of human resources which in many re spects is undesirable. It is in this area that new policies aimed at increasing mobility should be considered. The list of possible actions in this area is long and this is not the place for me to expand upon the subject. Lest I be misunderstood, however, two observations are in order: 1. Collective bargaining as an institution and the contents of col lective agreements have evolved under management and union EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2549 pressures. Both are thus responsible for the impact of collective bargaining. Furthermore, personnel practices restricting labor mar ket mobility are as common in the nonorganized sector of American industry as they are in the organized sector. They affect production and nonproduction workers alike. 2. The bargaining methods and policies of the trade unions vary markedly among trades, industries, regions, and localities. There exists no uniform or single union wage or bargaining policy. Nego tiations, both on management and the union side, are for the most part decentralized and sectors, districts, and even local and plant union groups or management tend to develop specific demands and bargaining patterns appropriate to their own situation* Consequently, general preachments and gratuitous advice on the advantages or disadvantages of centralized or decentralized bargain ing structures are ineffective and misleading. Most proposals advo cating internal union reform or the abolition of industrywide bar gaining to reduce their alleged “ inflationary” effect are thus wide of the mark and may, in fact, have a diametrically opposite impact. The desire, implicit in the recently passed labor law, to make unions more responsive to rank and file pressure, while justifiable on its own terms, cannot be passed off on the grounds of its contribution to a lessening of demands for wage increases. On the contrary, if the purpose of the bill will be fulfilled, and a leadership more responsive to the immediate preferences of the workers comes into being, we can expect demands for larger wage increases and more strikes. The his torical work of Mr. X Iman, in the explanation that he provides of the development of the present type of union government, is cited in support of my prediction. (See Lloyd TJlman, “ The Rise O f The National Trade Union,” Cambridge, Harvard University Press, 1955, pp. 203-301.) TRADE U N IO N S . COLLECTIVE BA R G A IN IN G , AN D W AG E STRU CTURES Numerous forces, including a general egalitarian philosophy widely accepted in our society, have contributed to a marked nar rowing of differentials in earnings which has been occuring since the turn of the century. There is no doubt in my mind that both the overall ideological “ push” of unionism and specific union demands have contributed substantially to the narrowing of differentials that has taken place. I I . T H E R E S P O N S IV E N E S S O F W A G E S T O C H A N G E S I N T H E L E V E L O F UNEM PLO Y M EN T In the last 10 years, the country passed through three periods in which the aggregate demand for labor decreased. Along with each fall in the demand for labor, wages continued to increase. Table I I I below summarizes the changes in employment, prices, and wages that took place in 1949, 1954, and 1958. This experience, as well as the characteristics of the labor market that are discussed below and to which reference has been made earlier, points clearly to the fact that reliance on tolerable levels of unemployment to achieve price stability is an impotent weapon. 2550 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS The facts summarized in table I I I underline that wage rates are more influenced by the employed than by the unemployed. Minor shifts in the size of the pool of the unemployed do not seem to affect the fear o f unemployment of currently employed workers, They are, to a substantial degree, isolated from the market by seniority rules, “ bumping” arrangements and, most important, the specificity of the production function of industrial processes which, from the employer’s point of view, makes the worker currently employed on a given job the best, if not the only, man for it. T a b l e III. — E m p lo y m e n t , p r ic e s , a n d Y ear 1948_________________ __________________________ 1949____________________________________________ 1953__________ _________________________________ 1954 ________________________________ 1957 . ___ ________________________________ 1958 __ ______________________________________ w a g e s in 1 9 4 9 ,1 9 5 4 , a n d 1 9 5 8 E m ploy m en t in thousands 59,117 58, 423 61,945 60, 890 65, Oil 63,966 Percent u n em ployed Percent change in C onsum er Price Index 3.8 5.9 4.9 5.6 4.3 6.8 - 0 .9 - .3 + 2 .7 Percent change in average hourly com pensation + 2 .7 + 2. 9 + 3 .0 When partial curtailments take place within a plant, a few workers are laid off. These are generally new workers in the labor pool or in the materials-handling group. More often than not, they had been earning only the entrance rate. Downward adjustments in job classi fications and possibly in rates follow. After the adjustments are completed, there is little to hold back workers who have kept their job from asking and getting wage increases. Their market is not the broad market conceived by the economist which includes all the em ployed and the unemployed workers of a given area or skill. Their market is delineated by the plant’s own boundaries which provide the worker with most of the opportunities which may exist. Wage rates within the plant remain at the same level or go up. Employer policy to pay prevailing rates and union pressure that he do so further nulli fies whatever pressure an increase in the number of hands at the gate might theoretically create. Unless further shifts are expected, that is, unless the workers and the unions expect substantial unemployment, a shift to the left of the demand curve for labor is accompanied by a parallel shift to the left— not the right— of the labor supply curve. Under these conditions, there appears little reason, if any, to assume that limited increases in unemployment will lead to dampening in wage pressures. A policy Ibased on the expectation that “ tolerable” levels of unemployment would curtail wage pressures would thus not only be intolerable and immoral but entirely erroneous. I I I . T H E E C O N O M IC IM P A C T O F C O N T R A C T S W I T H A U T O M A T IC C O S T -O F L IV IN G A N D P R O D U C T IV IT Y A D J U S T M E N T S The development of long-term contracts with automatic cost-ofliving and productivity adjustments, allegedly pioneered in the 1948 G M -U A W agreement" and now covering about 5 million workers, produced considerable debate on their expected inflationary effect. EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2551 The debate is unjustified either by the “ novelty” of the agreement or by its impact. Escalator clauses have a long history and their impact is neutral. The results of the two available empirical studies of the GM development agree in showing no substantial impact. The GM-type contract, in the long run, has provided unions bound by it with the same rate of growth in wage rates as that obtained by workers in unions that do not have it. Comparing the actual rate of growth in wages and a theoretical distribution based on a correction for cost of living and productivity in automobile and aircraft on the one hand and rubber and steel on the other, Silver found “no marked difference” (cf. Bette Silver “ Escalator Clauses—Their Effects Upon Union Wages” in Labor Law Journal, November 1958, p. 859). The escalation effect of the GM-type contract was found to be most effective in the 1950-51 period which was characterized by rapid price increases. In this period, contracts of the GM type “ led” negotiated wage adjustments without exceeding them. When the tempo of price increases slowed down, escalator type agreements did not influence negotiated wage increases. Between early 1953 and the end of 1955, negotiated wage increases exceeded automatic wage increases; while after 1956, the presence of automatic wage increases facilitated the upward movement of negotiated wage increases. (See Benson Soffer, “ The Effects of Recent Long-Term Wage Agreements on General Wage Level Movements: 1950-56” in Quarterly Journal of Economics, February 1959, pp. 36-60.) In the face of rapid price upswings, GM-type agreements reduce whatever “ wage stickiness” effect the very presence of collective agree ments may introduce. But it is wrong to assert that contracts of this type generally add to inflation. On the other hand, it has been asserted that in periods of stable prices, the presence of this type of agreement produces smaller increases in wTages than the normal process of collective bargaining. (See Lloyd G. Reynolds, “ The General Level of Wages” in New Concepts in Wage Determination, New York, McGraw Hill, 1957.) While a priori inferable, this has not taken place as contracts with automatic increases have been generally renegotiated in periods of stable prices to bring their yield in line with payments agreed upon through negotiations. Long-term con tracts of the GM type can therefore properly be called “ wage parity” agreements at best (Soffer, op. cit., p. 54). I would like to begin by noting, and I believe that I am in agree ment with the others here, that talk about inflation on the part of unions and management has given, generally, Americans an inferior ity complex on the subject of inflation that the facts in no way war rant. The discussion of impact of trade unions and collective bar gaining is seriously vitiated by the extent to which we are not aware of the actual price movement, particularly of the fact that the rate of price increases is diminishing. The limited nature of the price movement in the United States is underlined when we compare, and I have submitted in my testi mony data to that effect, the price movement of the United States with that of the industrial countries of Europe and North America. The committee will notice that the only countries where the consumer price index has risen less than in the United States are Switzerland and Belgium. 2552 E M P L O Y M E N T , G R O W T H , AND PRICE LEVELS Having disposed of this preliminary observation, I would like to underline or, to make some comments, as to the reasons why there is as much disagreement among economists and the public in general in the evaluation of the impact of trade unionism and collective bar gaining on prices. This is due to the complexity of the problem. It is also due to the intermingling of ideology and group self-interest and in part due to the interdependence of economic forces which obscure identification o f the union’s effect. The basic difficulty of the problem lies in the fact that the question o f the union’s impact is essentially in the “ iffy” category. Indeed, a complete answer requires knowing what the level and structures of wages would be if trade unions were not in existence. This obviously cannot be determined empirically. Nevertheless, certain questions are always asked in analyzing the impact of trade unions. These are: How have union wages moved in comparison with nonunion wages; studying wage movements in periods of considerable organization and comparing them with wage movements in periods of limited or no organization, and by comparing, at the same point of time, the wage movements of countries with well organized labor movements with the wage movements of countries with weakly organized movements. I have submitted testimony as to the first point and I will not try to summarize it here. I will move on to summarize the wage movements in the United States in periods of high organization with those of periods of not so extensive organization. Comparing the wage movements in the United States for the peri ods between 1900 and 1910 and 1947 and 1957, two periods of compara ble economic growth but very different degrees of organization, shows that while money wages grew faster than productivity in both periods, wages grew faster in the second period than in the first. In 1958 and 1959, wages have continued to grow but only half as fast as in 1957, particularly in the second half of 1957. Assuming a constant increase in productivity for both periods, Prof. Clark Kerr calculated the price impact o f the greater comparable wage increases of the second period at one-half of 1 percent a year. I would like to associate myself with that finding. But I find it impossible to ascribe even such a limited push on prices entirely to trade union action, even though trade unions have contributed to the greater tightness of the labor market of the second period to which the cause o f the greater rise in wages must to a substantial degree be ascribed. In my testimony I have submitted data on the movement of money earnings in selected countries of Europe, and again we find that wages in the United States have risen less than in any of the major industrial countries of Europe. The salient facts of wage movements since 1950 appear to be the following: Wages rose faster than prices; the rise in compensation per man-hour was nearly twice as large as the gain in real output, and labor costs per unit of output have risen by almost any measure of labor costs. These facts show that whatever caused the limited price increase did work through wages, a fact which, coupled with the publicity of trade union action in certain well, or overly reported, disputes have EM PLOYM ENT, G R O W T H , AND PRICE LEVELS 2553 led us to discuss at great length the problem of the cost-push versus the demand-pull inflation. It might be worth while to clarify the essence of the difference between a cost-push and a demand-pull inflation. I am fully aware, and my colleagues at this table will agree with me, that what follows is a simplification of the problem but I believe it to be a simplification worth while making. In simple terms, asserting that w^e are suffering from a cost-push inflation means that manufacturers are paying higher wages and are charging higher prices because they had to, while asserting that we are suffering from a demand-pull inflation means that manufacturers are paying higher wages and charging higher prices because the consumers wanted them to. What is important is that whether they had to because of labor pressure or were told to because of consumer pressure, American manufacturers are charging higher prices and consumers are paying the slightly higher prices without grumbling. Demand for labor in the 1950’s, remained at a higher level through out and, although we have not had large general labor shortages, there exists ample evidence of localized labor shortages and of shortages in certain categories of labor that led employers to offer generally higher wages to obtain additional labor. As will be explained, exist ing shortages were often strengthened by an increased fragmentation of the labor market brought about by the wide acceptance of certain employer hiring and promotion policies and by certain practices agreed upon by labor and management through collective bargaining. Thus it seems to me that the situation since 1950 can be summarized as being one in which excess demand for labor pulled up wages in the sectors in which trade unions are weakest and trade unions pushed up wages in the sector where excess demand was weakest. Studies relating to the impact of the bargaining process rather than the existence or pressure of trade unions brought out certain facts that helped clarify the overall economic impact of the current pattern of payments for labor services. Recent labor market studies have indicated how the workers’ ability and willingness to shift freely among employers, industries, and geo graphical areas in response to economic inducements is decreased with the spread of collective bargaining. There is no doubt in my mind that the widely accepted policy of promoting from within, tight seniority systems, more careful hiring, formal systems of wage and salary administration, discrimination in hiring on the grounds of race or color, have reduced horizontal labor mobility and conse quently created artificial bottlenecks which have led to wage increases. The C h a i r m a n . Just a question at that point. Do I understand you to say that there is greater discrimination in hiring on the grounds of race and color now than formerly ? Mr. O r n a t i . N o , sir. Did I sa y th a t ? The C h a i r m a n . There is an implication from that sentence. Mr. O r n a t i . N o , sir. I f th a t im p lic a t io n is in th e r e I d id n o t mean it. The C h a i r m a n . What did you mean by that ? Mr. O r n a t i . I said there is no doubt that some of these policies including discrimination in hiring on the grounds of race and color lead to creation of artificial bottlenecks. 38563— 59—>pt. 8-------8 2554 EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS The C h a i r m a n . Y ou say “ have reduced horizontal labor mobility,” which implies, I would think, such a comparison. Mr. O r n a t i . Mr. Chairman, if there is an implication here that discrimination has increased, I have no ground on which to base that contention. Representative C u r t i s . Mr. Chairman, as I read this paper, the point is that where you have organized labor this is one of the artifi cial barriers that tends to be increasing. The C h a i r m a n . That is what I wanted to find out. Representative C u r t i s . Y ou asked the question whether it is in the entire country. The C h a i r m a n . No; I meant in the field of union bargaining. Representative C u r t i s . For instance, in my own community in St. Louis, Mo., in building trades, a Negro cannot be a carpenter or brick layer. It is purely the result of unionism. In sectors where organ ized labor does not exist a Negro can be a bricklayer or carpenter. I understood what he meant, and I agree with it. The C h a i r m a n . Let us find out what he meant, first. What did you mean ? Mr. O r n a t i . Mr. Chairman, my statement points at a whole series o f hiring and promotion practices which both management and unions have helped develop. Among these I find the practice of discriminat ing on the grounds of color to be a practice that creates artificial bottlenecks. The C h a i r m a n . I think there is no doubt about that. Then there are two questions. First, has this increased over time? Second, are unions responsible on the whole for such discriminations ? Mr. O r a n a t i . I am not competent to answer with precision either of the two questions. I am aware that trade unions in some areas have tried to reduce the extent of discriminatory hiring on the basis o f race and creed. I am also aware of the fact that in many trade union areas discrimination on the grounds of color continues to exist. The C h a i r m a n . I think you will generally find in the mass pro duction industrial units there is less discrimination on the ground of color than before. There has always been a considerable degree o f discrimination in the field o f craft unions in the building trades. Mr. O r n a t i . Yes, I agree with that. Representative C u r t i s . I wonder about that. Has anyone ever made an objective study? There is a lot of lip service given to it, Mr. Chairman, on the part of some of these people. But when it gets down to what they actually do, I wonder. The C h a i r m a n . I think if you go to some of the local meetings of the steel, autos and chemicals and coal-----Representative C u r t i s . I have the situation in St. Louis where the discrimination is the other way. White men cannot get jobs running the little locomotives. The C h a i r m a n . First you say there is greater discrimination of Negroes in unions, now you say that unions discriminate against whites. Representative C u r t i s . When you have discrimination, it can build another way. That happens in a company that employs mostly Negro labor. That is how these things can develop. It is the rigidity of unionism that has produced it. E M P L O Y M E N T , G R O W T H , AND PRICE LEV ELS 2555 Mr. O r n a t i . Mr. Chairman, may I call attention to the order in which I have listed these various practices, and would you kindly note that discrimination in hiring on the grounds of rftce or color although followed by an “ et cetera” is the last item. The C h a i r m a n . There are apparently three views of what you said: Our own views, Congressman Curtis’ views, and my views. Mr. O r n a t t . In this way, collective bargaining as an institution has put into motion forces that contribute to raising the general level of wages and has brought about an allocation of human resources which in many respects is undesirable. I would like to be permitted to talk on the topic of responsiveness o f wages to changes in the level of employment. In the last 10 years the country passed through three periods in which the aggregate demand for labor decreased. Along with each fall in the demand for labor, wages continued to increase. Table I I I o f my prepared statement summarizes the changes in employment, prices, and wages that took place in 1949, 1954, and 1958. This experience, as well as the characterization o f the labor market that are discussed below and to which reference has been made earlier, points clearly to the fact that reliance on tolerable levels of unemploy ment to achieve price stability is an impotent weapon. The facts summarized in table I I I underline that wage rates are more influenced by the employed than by the unemployed. Minor shifts in the size of the pool of the unemployed do not seem to affect the fear of unemployment of currently employed workers. They are, to a substantial degree, isolated from the market by seniority rules, bumping arrangements and, most important, the specificity of the production function of industrial processes which, from the em ployer’s point of view, makes the worker currently employed on a given job the best, if not the only, man for it. When partial curtailments take place within a plant, a few workers are laid off. These are generally new workers in the labor pool or in the materials-handling group. More often than not, they had been earning only the entrance rate. Downward adjustments in job classifications and possibly in rates follow. After the adjustments are completed, there is little to hold back workers who have kept their job from asking and getting wTage increases. Their market is not the broad market conceived by the economist which includes all the employed and the unemployed workers of a given area or skill. Their market is delineated by the plant’s own boundaries which provide the worker with most of the opportunities which may exist. Wage rates within the plant remain at the same level or go up. Employer policy to pay prevailing rates and union pressure that he do so further nullifies whatever pressure an increase in the number of hands at the gate might theoretically create. Under these conditions there appears little reason, if any, to assume that limited increases in unemployment will lead to dampening in wage pressures. A policy based on the expectation that tolerable levels of unemployment would curtail wage pressures would thus not only be intolerable and immoral but entirely erroneous. I concluded my written testimony with certain observations that I was asked to prepare on the impact of contracts with automatic cost 2556 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS o f living and productivity adjustments, and generally I feel that they are noninflationary. The C h a i r m a n . The final paper is by Prof. Lloyd Ulman of the University o f California. STATEMENT OF LLOYD ULMAN, PROFESSOR OF ECONOMICS AND IN DUSTRIAL RELATIONS, UNIVERSITY OF CALIFORNIA, BERKELEY, CALIF.— Resumed Mr. U l m a n . Thank y o u , Mr. Chairman. There has been a difference of opinion among economists—perhaps only among economists—as to whether the negotiation of wages under collective agreements covering slightly over one-third of the Nation's nonagricultural employees has constituted one of the forces contrib uting independently to the upward movement of wages and prices in the period following the end of World War II. Attempts to test this hypothesis have thus far failed to uncover any single satisfactory yardstick. Comparisons between union and nonunion wage move ments, comparisons of movements in w^ages, prices, productivity, and unit labor costs, the presence or absence of reported shortages o f labor, of wage “ glides” or “ slides” above contractual rates, or of a high degree of unemployment—each has failed as a sufficient criterion in itself because none can provide evidence which is inconsistent with either (a) the proposition that collective bargaining has contributed to wage and price inflation or ( b) the proposition that recorded increases in wages and prices have been due solely to other causes—such as con sumer spending out of accumulated liquid assets in the immediate postwar period, the growth in population, and increased Government spending on defense and farm price support programs, I should like to dwell briefly on certain characteristics of our wagesetting institutions, which might aid us in evaluating the statistical measures and to assess their relationship to either established public policy or proposed changes therein. These characteristics are (1) de centralization of union power and freedom of employee choice, (2) decentralized bargaining in concentrated industries, (8) some other aspects of the industrial environment of trade unions, and (4) reac tions in the nonunion sectors. Under the first heading, we shall try to explain the propensity of American unions to push so vigorously for higher money wages; in the second and third areas, we shall explore some of the opportunities and obstacles which affect their ability to raise the wages of their members; and in the fourth, we allude briefly to some forces which tend to mag nify and others which tend to offset the impact of union-won wage increases. 1 . D E C E N T R A L IZ A T IO N O F U N IO N P O W E R A N D F R E E D O M O F E M P L O Y E E C H O IC E Where, as in some foreign countries, collective bargaining is highly centralized and central federations of trade unions are influential in the determination of wage rates in particular industries, it has been at least possible for both employer and union groups to be influenced (restrained) by the probable impact of their decisions upon the eco EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2557 nomic welfare of the entire country. Moreover, negotiated wage changes in such centralized systems frequently reflect the mediating influence o f governmental authority; the labor movements in such countries depend heavily for the advancement of the interests of their membership on their associated labor parties and upon what the Webbs termed the “ method of legislative enactment.” American unionists, on the other hand, have, for a variety of rea sons, placed much more emphasis on self-help through collective bar gaining; and this has resulted in a weak central federation—weak in the sense that it exerts no control over the bargaining activities of the national unions affiliated with it. As a result, union bargaining policies are made by a large number of units—there are, for example, over 180 national unions in the United States— 110 one of which is able to formulate its own decisions in the light of the economic re quirements of the country as a whole. To be sure, the degree of decentralization of authority within the trade union community was and is limited. Among the older unions emphasis on collective bargaining decreed that with the widening of labor and product markets and the growth of multiplant firms—and their equivalents in construction and transport—power should pass, in varying degrees, from local unions up to the internationals. In some of the newer CIO unions, which were organized by the parent federation, power rapidly passed down to the national level. Not infrequently the national unions used their power to restrain sub ordinate locals from pressing for wage increases or working rules which would put their employers at a serious competitive disadvan tage or, on the other hand, which would put their members “ out of line” with members in other locals. And if the national official was in no position to see the economy as a whole, he was better able than the local membership to see his occupational or industrial jurisdic tion as a whole, including the limits to what the traffic would bear. Moreover, the national unions agreed in principle to limit de facto decentralization in one important area. It was agreed that each na tional union should exercise “ exclusive jurisdiction” over some defined occupational area, This rule was frequently honored in the breach by rivals for jobs or members, but, even so, such rivalry often took the form of competitive underbidding of wages to “ organize the em ployer” rather than of competitive raising of wages to win the favor o f their employees. However, modern public policy, reaffirmed in the Taft-Hartley Act o f 1947, and again in the legislation recently enacted, has tended to weaken some of the self-imposed limitations upon decentralization. By attempting to place the question of bargaining representation within the sole and protected domain of the individual employees in the bargaining unit, it placed a premium on interunion competition for members which, in view of the employer’s enforced neutrality, has most frequently taken the form of competitive bidding up of wages and other standards of employment, rather than of underbid ding. Partly in an effort to cope with the problems presented by this public policy, the federations and many unions have entered into no raiding agreements, and some unions have even embarked on joint bargaining ventures; but very serious jurisdiction rivalries still con tinue. And even where no rival union is in sight, the union on the 2558 EM P LO Y M E N T, G R O W T H , AND PRICE LEVELS premises is kept on its toes by the existence of the employees’ legally available alternative of choosing “ no union.” In placing certain restrictions on organization picketing, the Labor-Management Reporting and Disclosure Act of 1959 may rein force the efforts of the A F L -C IO and of some of its affiliates to re duce jurisdictional competition. But insofar as the new law is ef fective in making the union leaders more responsive to the member ship— whether by striking at such abuses as “ sweetheart agreements” and “ conflicts of interest” or by strengthening democratic pro cedures within unions—it may strengthen the wage-rising pro pensities of these organizations. On a previous occasion, I sug gested that reform of union government might entail some economic cost in terms of higher wages and prices, but I omitted two highly relevant points. The first is that the above argument assumes that the members themselves continue to seek money wage increases as ardently as they have in the past. The second is that the analysis does not, of course, imply that the cost outweighs the benefit to the entire community from further implementation of the national labor policy o f the past quarter-century. 2 . D E C E N T R A L IZ E D B A R G A IN IN G I N C O N C E N T R A T E D IN D U S T R IE S A t the end of the war not a few observers predicted the widespread extension of industrywide bargaining. Some feared that industry wide bargaining would increase the bargaining power of unions, since each employer would be certain that his competitors would settle with the national union on terms no more favorable than his own, every employer’s effective will to resist would be sapped. Now it might appear that these observers, like the legendary first baseman o f the Brooklyn Robins, were looking in the wrong direction, al though it is true that multiemployer bargaining on a localitywide or regional basis has grown since the war, especially where the firms involved are small, mobile, confronted with high labor-cost ratios, and very competitive. (Frank C. Pierson, “ Prospects for Industry wide Bargaining,” “ Industry and Labor Relations Review.” vol. 3, No. 3, April 1950, pp. 340-361.) But multiemployer bargaining in the manufacturing sector of the economy actually declined in coverage between 1947 and 1956, while comp any wide bargaining by large multiplant firms grew more popu lar in the postwar period. In large-scale industries unions were fre quently able to achieve under company-by-company bargaining many of the same results by way of “ taking labor out of competition” that were generally claimed for industrywide bargaining; these some times included the reduction or elimination of interplant and even interfirm wage differences and the negotiation of uniform increases in wages. Furthermore, companywTide bargaining has held certain advantages for the unions over industrywide bargaining. In the first place, in such oligopolistic industries as autos, aircraft, rubber, and glass, unions have frequently been able to avoid industrywide strikes. Se lective striking places a union in a stronger financial position than industrywide striking. Nor does it stiffen employer resistance to union demands, for the employer can be virtually as certain under EM PLOYM ENT, G R O W T H , AND PRICE LEVELS 2559 industrywide bargaining as under company-by-company bargaining that the national union with which he deals—or an eager rival union—■ will negotiate with his important competitors the same settlement (in terms of cost) which he will make himself. On the contrary, employer resistance tends to be sapped by the wliipsawing potentiali ties of the piecemeal approach; fear of losing one’s historic “ share of the market”—either to existing competitors or, as in the case of the automotive components industries, to do-it-yourself customers—makes any and every firm in the industry especially reluctant to take a strike when important rivals are not shut down. Decentralized bargaining has also benefited stronger and more effi cient firms by permitting them some freedom of action. Under indus trywide bargaining, they might be prevented from taking a long strike because of the inability of the weaker firms to hold out against the union. Conversely, under industrywide bargaining, they might be obliged to shut down because the high-cost firms are unable or unwilling to pay the price of peace. (Under decentralized bargain ing General Motors was not prevented from peacefully concluding a long-term agreement with the Auto Workers in 1950 while Chrysler was on strike.) But when larger or more efficient firms are more determined to resist union demands, it would appear that the com pensations o f decentralization to employers are outweighed by its disadvantages. Greater attempts by employers to present a united front in the automobile negotiations and during airlines strikes (when an income-sharing pool was formed) in the recession of 1957-58 and the recently announced strike-insurance plan of the railroads suggest that industrywide bargaining, at least on a de facto basis, might yet live up to the early claims of popularity made on its behalf. Should such attempts be discouraged by legislation to outlaw indus trywide bargaining or to fragment the national unions ? Supporters o f these proposals might point to the extremely high wage increases which have been registered under industrywide bargaining in even a declining industry like coal or, on a de factor basis, in steel. More over, fragmentation of the national unions might weaken their present components financially, although it might be noted that the Steel workers do not have a regular system of national strike benefits. And in certain highly competitive, nongrowth industries like the garment trades the elimination of industrywide bargaining and/or the dismantling of the national unions therein would in all proba bility be effective in reducing union bargaining powder. But in the latter industries wages have not risen very rapidly at all; and neither national unions nor elaborate apparatuses of indus trywide bargaining have been able to prevent the emergence of non union competitors with lower labor costs in industries where entry is so easy. On the other hand in less competitive industries, character ized by larger, more profitable, and fewer units of enterprise, it is extremely doubtful whether the proposed changes would accomplish much beyond the elimination of some of the smaller and weaker firms by locals presently chafing under such restraints as existing national union authority has been able to impose. We must recall that many existing inequalities in labor costs are due to differences in efficiency— in work rules rather than in basic wage rates, where the latter are negotiated by the national union while the former are the jealously 2560 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS guarded prerogatives of the locals whose dubious successes have varied inversely with the ability of the employer to pay—or resist, As for the firms of average financial strength and operating efficiency, they would presumably be confronted by piecemeal pressure from companywide unions uncoordinated by national authority, but anxious to equal one another in securing higher standards of employment, Thus it is questionable whether most employers in large-scale indus tries would gain more bargaining strength than they would lose by the extension of the antitrust laws to the areas of industrywide bar gaining and union structure. (For a similar conclusion proceeding from a different line of analysis see John T. Dunlop, “ Policy Prob lems: Choices and Proposals,” in Wages, Prices, Profits and Produc tivity, the American Assembly, 1959, pp. 146-147.) On the other hand, if, as we shall presently argue, the degree of competition in product markets is an important determinant of union bargaining potential, more vigorous enforcement of antitrust legislation in prod uct markets would be a more effective, although indirect, means of curbing the ability of some of our largest national unions to push up wages, costs, and prices. Measures to liberalize international trade would have the same effect, and efforts by both business and union groups to secure increased protection from increasing competition from abroad should be firmly rebuffed by the Congress and the execu tive branch of the Federal Government. Indeed, it may not be unreasonable to expect that increased com petition in product markets, apart from limiting union power, might tend to divert some of the employer response to union pressure on costs from the channel of price increases and into the channel of increasing efficiency. 3 . T H E IN D U S T R IA L E N V IR O N M E N T While it is generally agreed that unions in the building trades, printing trades, railroads, and other so-called crafts have great bar gaining strength, the bargaining power of the so-called industrial unions in the manufacturing sector has been called into question. In some important respects, the environment of the latter has indeed tended to restrict their ability to raise wages. Under the impact of shifts in consumer demand to services and products of “ tertiary” industries and of labor-displacing innovation, the demand for produc tion labor in manufacturing has not kept pace with the demand for nonagricultural labor as a whole. And partly because the level of man ufacturing wages is relatively high, no overall stringency in the labor markets involved has developed to provide the unions with a favoring tailwind. Moreover, the services of semiskilled production workers are not as essential to the production process or, perhaps, as subject to restriction in supply as are the services of skilled unionists. And in some of the industrial jurisdictions labor-cost ratios are high—■ although, on the whole, labor-cost ratios in industrial-union jurisdic tions are at least as low as those in industries organized by the crafts referred to above. However, other characteristics of some of the major industrial union jurisdictions tend to compensate for these restraints on union bargaining power. We have already referred to high ratios of con EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2561 centration. The existence of formidable technological or pecuniary barriers to entry of new firms can provide insulation against the emergence of nonunion competition and compensate for a union’s in ability to restrict the supply of labor in a semiskilled field. Moreover, the greater technical substitutability of semiskilled pro duction workers is neutralized to some extent, for industries making the substitute, e.g., machine tools, are also organized, so that their labor costs and prices move upwTard in some correspondence with the wages of the union workers in question. And w^hile the displacement of union labor due to technological innovation is indeed characteristic of industrial jurisdictions, unions have not been prevented thereby from pushing up wages. On the contrary, it would appear that some industrial unions have exploited above-average rates of growth in output per man-hour, which might otherwise have tended to depress wages by reducing demand for labor, in negotiating wage increases part of which were thus not reflected in higher unit costs to the em ployers. This, together with the ability of firms in concentrated indus tries—and especially in industries for whose products no close substi tutes existed—to raise their prices wdth relative impunity, helps to explain how wages in the highly unionized manufacturing sector, where the demand for labor lagged, were able to keep up with wages elsewhere in the economy—including those service trades wThere the demand for labor was very strong and which are lightly organized. To the extent that wage increases have helped to prevent prices from falling in industries w'here demand for labor has been growing slowly but productivity has been growing rapidly, the influence on the aggregate level of prices has been inflationary. This has prompted the Chairman of the Council of Economic Advisers to urge upon unions and managements in such industries a policy char acterized both by the negotiation of w^age increases belowr increases in productivity and by price reductions. Whether managements in administered-price industries would have been willing to reduce prices if they had been able to negotiate smaller wage increases in the past is problematical; the steel industry’s negative response to suggestions that it cut prices if it could negotiate a standstill agreement in wages was disappointing. On the other hand, certain spokesmen from in dustrial union groups have issued statements in the past which con tained proposals not too dissimilar from Professor Saulnier’s. The prospects of success, while hardly overwhelming, should nevertheless warrant continued public pressure. I firmly believe, however, that such pressure should stop well short o f governmental controls over wages and prices. In my opinion, the damage, both economic and political, from creeping controls would outweigh by far the damage from creeping inflation. 4 . T H E S P L A S H A N D T H E R IP P L E *. S E C O N D A R Y E F F E C T S O F U N IO N -W O N W A G E IN C R E A S E S The total impact of collectively bargained wage increases upon the overall levels o f wages and prices depends upon how and to what extent the wages of nonunion workers are affected thereby. Most observers and practitioners in the field of industrial relations believe that, just as certain “ key” bargains are copied more or less faithfully 2562 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS by other unions and managements, so union-won wage increases “ ripple out” over the nonunion sectors. In this connection I shall refer very briefly to three mechanisms which can generate such sec ondary effects and which have been discussed in the literature. The first two indicate how collectively bargained wage increases could result in increased demand for nonunion labor and in a rise in nonunion wrages and prices. In the first place, if prices in unionized industries rise relative to prices in the nonunion sector, consumers would tend to substitute non-union-made goods for unionmade goods to some extent (J. R. Hicks, “ Economic Foundations of Wage Policy,” the Economic Journal, vol. L X Y , No. 259, September 1955, at pp. 397-398; L. Ulman, “ Marshall and Friedman on Union Strength,” Review of Economics and Statistics, vol. X X X V I I , No. 4, November 1955, at pp. 395-399; A. P. Lerner, “ Inflationary Depression and the Regulation of Administered Prices,” The Relationship of Prices to Economic Stability and Growth, Joint Economic Committee, 85th Cong., 2d sess., Washington, 1958, at pp. 265, 266). In the second place, if bargained wage increases result in higher payrolls and total expenditures by the firms involved, part of the increase in wage in comes will be spent on products and services in the nonunion sector (J. R. Hicks, “ The Instability of Wages,” the Three Banks Review, March 1956, No. 29, at pp. 12-14; S. H. Slichter, “Labor Costs and Prices,” in “ Wages, Prices, Profits, and Productivity,” op. cit., at pp. 172-176). Third, employers of nonunion labor raise wages in their own plants in order to avert a decline in employee morale and effi ciency and, in some cases, to keep unions out of their plants (S. H. Slichter, “ The Current Labor Policies of American Industries,” Quarterly Journal of Economics, vol. X L III, May 1929, pp. 393-435; Lloyd G. Reynolds, “ The Structure of Labor Markets,” New Y ork: Harper, 1951, pp. 219, 220; J. S. Duesenberry, “ Business Cycles and Economic Growth,” New York: McGraw-Hill, 1958, pp. 305, 307; and “ Underlying Factors in the Postwar Inflation,” in “Wages, Prices, Profits, and Productivity,” pp. 83, 84; see also the testimony of Prof. H. P. Minsky, University of California, Berkeley, before this com mittee). The magnitude of the secondary effects of collectively bargained wage increases depends in part both on the mobility of labor between the union and nonunion sectors and on the size of the unionized sector. The nonunion wage and price raising mechanisms referred to above could be offset by a sufficiently large increase in the supply of nonunion labor, if the bargained wage increases resulted in a suffi ciently great reduction in employment and if enough redundant workers applied for jobs in the nonunion sectors. However, union jobs are high wage jobs and their consequent attractiveness, combined with nontransferability of skills, limit the willingness and ability o f union jobholders to move. Evidence of such immobility is found in the prevalence of sharework and makework rules and the opposition which they arouse among employers. Such rules not only increase costs directly; they help to insure that the secondary effect of union won increases in wrage rates will be inflationary rather than offsetting. Moreover, the fact that one out of every three nonfarm wage earners is a union member should suggest that the overall contribution of collective bar EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2563 gaining to the upward movement of wages and prices is appreciable. If, as I believe, the processes of inflation in the postwar period have not all been of the so-called classical variety, they could not be neu tralized by sufficiently vigorous monetary or fiscal policies without reducing employment and slowing down the rate of economic growth. For while such policies could remove the inflationary increases in demand generated by collective bargaining, they could not roll back to an equal extent the simultaneously induced increases in costs. 5 . C O N C L U S IO N S A N D P R O S P E C T S Although one cannot tell exactly what would have happened to wages and prices in the absence of collective bargaining, it is nec essary for responsible Government officials to make an educated guess the nature of which will presumably affect certain economic policies which they will recommend and formulate. In view of some of the characteristics of trade unions, collective bargaining, and product and labor markets to which I have referred, it is, in my opinion, reasonable to assume that collective bargaining has contributed to the upward movement in wages and prices in the postwar period. I also conclude that policies designed to decentralize bargaining within the limits contemplated by their proponents would not, on the whole, reduce this upward pressure on costs and prices, although poli cies designed to reduce barriers to entry in product and factor markets might have this effect. Moreover, I believe that, while policies de signed to restrain aggregate demand could reduce the pressures of collective bargaining, they would also reduce our rate of economic growth and would result in politically unacceptable levels of unemployment. In this connection it is relevant to note that the only years in which hourly earnings in manufacturing, exclusive of over time, rose by 3y2 percent or less were all years of relatively high unemployment— 1949-50, 1953-54, 1954-55, and 1957-58. In 1949, the average rate of unemployment was 5.9 percent; in 1954, 5.6 per cent; in 1955,4.4 percent; and in 1958, 5.8 percent. But while I am not sanguine about the efficacy of these policy changes, there is some reason to hope that the problem, instead of intensifying in the future, as some believe, might diminish to some extent. In the first place, continuation of the growth of foreign competition should increase the inclination and the ability of Ameri can employers to resist union demands—provided, however, that Con gress and the administration resist their pleas for greater protection. In the second place, some of the best organized sectors of the nonfarm labor force—notably production labor in manufacturing and mining—have been shrinking relative to the total; and, to date at any rate, the unorganized frontiers have presented formidable obstacles to the growth of unionism. Finally, unions and managements themselves have been developing some institutions which, while increasing costs, might be less infla tionary than equivalent straight wage increases. Increases in pri vate pension funds fall in this category. Provision for severance pay— which has been gaining greatly in popularity in recent years— and other financial provisions for employees displaced by techno logical change are a desirable alternative to immobilizing make-work rules and probably outweigh any tendency which they might have in 2564 EM PLOYM ENT^ G R O W T H , AND PRICE LEV ELS discouraging innovation in tlie first place. Even cost-of-living esca lator clauses may be placed in this category. To the extent that they permit wage increases generally to respond more promptly to price increases, they prove more inflationary than negotiated wage increases (Benson Soffer, “ The Effects of Recent Long-Term Wage Agreements on General Wage Level Movements,” Quarterly Journal of Economics, vol. L X X I I I , February 1959, pp. 36-60). But, by making it unnecessary for unionists to attempt to discount anticipated price increases or to “ make up” for past price increases in current wage settlements, they tend to rule out a self-defeating process which could generate successively greater wage and price in creases. Thus, even if the escalators have contributed to the inflation ary creep, they might also help to prevent it from breaking into a gallop. Mr. Chairman and members of the committee, my written state ment to this committee was concerned with certain characteristics o f our wage-setting institutions. These are: (1) decentralization of union power and freedom of employee choice; (2) decentralized bargaining in concentrated industries; (3) some other aspects of the industrial environment of some of our largest unions; (4) reactions in the nonunion sectors. I attempted under the first heading to account in part for the propensity of American unions to push so vigorously for higher money wages; in the second and third areas, to explore some of the opportunities and obstacles which have affected their ability to raise the wages of their own members; and in the fourth, to identify some forces which tend to magnify and others which tend to offset the im pact o f union-won wage increases. In this summary I shall concern myself primarily with the conclu sions emerging from my statement which relate to certain public pol icies, enacted or proposed. In conclusion, I should like to point to some developments which, if they persist, might tend to reduce the contribution of collective bargaining to price inflation in the future. 1. our n a t io n a l labor p o l ic y For the past quarter century, the Federal Government has at tempted consistently to establish and implement the basic principle that the question of bargaining representation lies within the sole and protected domain of the individual employees in the appropriate bar gaining unit. This has placed a premium on interunion competition for members which, in view of the employers’ legally enforced neu trality, has most frequently taken the form of competitive bidding up o f wages and other standards o f employment rather than competitive underbidding to organize the employer which was more prevalent prior to the passage of the Wagner Act than it has been since that time. Moreover, even when no rival union is in sight, the union on the premises is kept on its toes by the existence of the employees’ legally available alternative of choosing no union in representation or de certification elections. Insofar as the new Labor-Management Reporting and Disclosure Act of 1959 further implements the basic national philosophy, whether E M P L O Y M E N T , G R O W T H , AND PRICE LEVELS 2565 by striking at such abuses as sweetheart agreements and conflicts of interests or by strengthening democratic procedures within unions, it may heighten the wage-raising propensities of these organized groups. In particular, measures designed to make union officers and especially officers o f national unions more responsive to the wishes of the mem bers may weaken the ability of national unions to restrain the sub ordinate locals in their bargaining activities. Two cautionary remarks should be made in this connection. The first is that the underlying argument assumes that the members themselves continue in the future to seek wage increases and other economic benefits as ardently as they have in the past. The second is that the economic cost to the community, however great or however small, entailed by our national labor policy, is surely outweighed by the benefits flowing from the system of industrial democracy which that policy has helped to establish. 2 . P R O P O S A L S TO P R O H IB IT IN D U S T R Y W ID E B A R G A IN IN G OR TO F R A G M E N T N A T IO N A L U N IO N S Employees in some industries have recently been making more de termined attempts to present a united front in negotiations and strikes. Should such attempts be discouraged by legislation to outlaw industry wide bargaining or to fragment the national unions ? Supporters of these proposals might point to extremely high wage increases which have been registered under industrywide bargaining in even a declining industry like coal or, on a de facto basis, in steel. Moreover, fragmentation of the national unions might weaken their present components financially—although it might be noted that the steelworkers’ national union does not pay regular strike benefits. And in certain highly competitive nongrowth industries like the garment trades, the elimination of industrywide bargaining or the dismantling of national unions would in all probability be effective in reducing union-bargaining power. But in the latter industries wages have not risen at all rapidly and neither national unions nor elaborate apparatuses of industrywide bar gaining have been able to prevent the emergence of nonunion com petitors where entry is so easy. On the other hand, in less competitive industries characterized by larger, more profitable, and fewer units of enterprise, it is extremely doubtful if the proposed changes would accomplish much more than the elimination of some of the smaller and weaker firms by locals presently chafing under such restraints as exist ing national union authority has been able to impose. We must recall in this connection that many existing inequalities in labor costs are due to differences in efficiency, in work rules rather than in basic wage rates, where the latter are negotiated by the national union while the former remain the jealously guarded prerogatives of the locals. As for the firms of average financial strength and operating effi ciency, they would probably be confronted by piecemeal pressure from company unions uncoordinated by national authority but anxious to equal one another in securing higher standards of employment. 2566 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS 3 . L E G A L E N F O R C E M E N T O F P R O D U C T C O M P E T IT IO N Thus, it is questionable whether most employers in large-scale indus tries would gain more bargaining strength than they would lose by the extension of the antitrust laws to the area o f industrywide bar gaining and union structure. On the other hand, since, as I already have implied in my reference to the garment trades, the degree of competition in product markets is an important determinant of union bargaining strength, more vig orous enforcement of antitrust legislation in product markets would be a more effective, although indirect, means of curbing the ability of some of our national unions to push up wages and prices. Measures to liberalize international trade would have the same effect. Growing efforts of both industry and union groups to secure more protection from competition from abroad should be firmly re buffed by the Congress and the executive branch of the Federal Government. Indeed, it may not be unreasonable to expect that increased com petition in product markets, apart from limiting union power, might tend to divert some of the employer responsibilities to union pressure on costs from the channel of price increases and into the channel of increasing efficiency. 4 . E N C O U R A G E M E N T O F W A G E R E S T R A IN T A N D P R IC E R E D U C T IO N S I N L A R G E -S C A L E IN D U S T R IE S To the extent that union-won wage increases have helped to pre vent prices from falling in industries where demand for labor has been growing rather slowly but productivity has been growing rap idly, the influence on the aggregate level of prices has been inflation ary. The Chairman of the Council of Economic Advisers has urged upon unions and managements in such industries a policy charac terized both by the negotiation of wage increases below increases in productivity and by price reductions. Whether managements in administered-price industries would have been willing to reduce prices is problematical; the steel industry’s negative response to suggestions that it cut prices if it could negotiate a standstill agreement was disappointing in this respect. On the other hand, certain spokesmen from industrial union groups have issued statements in the past which contain proposals not too dissimilar from Professor Saulnier’s. The prospect of success should warrant continued public pressure. I firmly believe that such pressure should stop definitely short of gov ernmental controls over wages and prices. In my opinion, the damage (both economic and political) from creeping controls would outweigh by far the damage from creeping inflation. 5 . R E L IA N C E U P O N F IS C A L A N D M O N E T A R Y R E S T R A IN T S I conclude in my written statement that unions have been an inde pendent force in raising wages and prices, not only in their own juris dictions but in the unorganized sectors as well. (Here union work rules play a role in magnifying the impact of negotiated wage in creases.) If, therefore, the processes of inflation in the postwar period have not all been of the so-called classical variety, they could not be EM PLOYM ENT, G RO W TH , AND PRICE LEV ELS 2567 offset by sufficiently vigorous monetary or fiscal policies without reduc ing the rate of increase in employment and output. Such policies might reduce the inflationary increases in demand generated by col lective bargaining but they could not roll back to an equal extent the simultaneously induced increases in costs. Would the requisite level of unemployment be “ tolerable” ? That depends in part on what level of unemployment the public, whose preferences are registered every 2 years at the polls, is willing to tolerate. In this connection it is relevant to note that all the years in which hourly earnings in manufacturing, exclusive of overtime, rose by 3y2 percent or less were years of relatively high unemploy ment: 1949, 5.9 percent; 1954, 5.6 percent; 1955, 4.4 percent; 1958, 5.8 percent. It is extremely doubtful whether the majority of the voting public will be willing to tolerate as “normal” unemployment rates in the neighborhood of 4y 2 to 6 percent in order to prevent prices from rising as rapidly as they have in the past. SO M E FU T U R E PRO SPECTS While I do not believe that proposals to outlaw industrywide bar gaining, further decentralize union authority, or rely solely upon fiscal or monetary policy can accomplish their avowed objectives, there is some reason to hope that the problem, instead of intensifying in the future as some believe, might diminish to some extent. In the first place, continuation of the growth of foreign competi tion should increase the inclination and ability of American employ ers to resist union demands—provided, however, that Congress and the administration resist their pleas for greater protection. In the second place, some of the best organized sectors of the nonfarm labor force, notably production labor in manufacturing and mining have been shrinking relative to the total. To date, the unor ganized frontiers have presented formidable obstacles to the growth of unionism. Finally, unions and managements themselves have been developing some institutions which, while increasing costs, might be less infla tionary than equivalent straight wage increases . Increases in private pension funds fall in this category. Severance pay, which has been gaining greatly in popularity in recent years, and other financial provisions for employees displaced by technological change are a desirable alternative to immobilizing make-work rules and probably outweigh any tendency which they might have of discouraging inno vation in the first place. Even cost-of-living escalators may be placed in this category. To the extent that they permit wage increases generally to respond more promptly to price increases, they prove more inflationary than nego tiated wage increases. But by making it unnecessary for unionists to attempt to discount anticipated price increases or to make up for past price increases in current wage settlements, they do tend to rule out a self-defeating process which could generate successively greater wage and price increases. Thus, even if the escalators have contrib uted to the inflationary creep they may also have helped to prevent it from breaking out into a gallop. 2568 E M P L O Y M E N T , G R O W T H , AND PRICE LEV ELS The C h a i r m a n . Thank you very much. The discussion will be opened by Congressman Reuss. Representative R e u ss . Mr. Chairman, I would like to raise a ques tion which has been inherent in what you gentlemen have been talking about, although I do not think you were asked to comment on it directly, and that is the implications for our international trade posi tion o f collective bargaining and wage increases. It used to be in the recent past the fact that American real wages were vastly higher than that of most other industrial countries was not a serious detriment to our international position. There was the difference in productivity of this country and competitive countries, and the abundant nature of the post-World W ar I I market. However, recently these things have tended to level out. Productivity in many other industrial countries is approaching our own. Therefore, wre have a problem. Therefore, I would like any suggestions that any of you may have for American policy to meet the problem brought on by the fact that our real wages are high and our productivity advantage is lessening. Specifically, to what extent, if at all, by diplomatic representations, by work in the international labor organization, or by any other means can we help to bring about higher wage standards elsewhere ? Is that desirable? To what extent will foreign competition, by stiffening the backbone of management, tend to make our own wage standards more compatible with international trade? To what extent is it too hopeful to expect any competitive lowering of American w^age standards ? Finally, is there anything we can do about it ? I f anyone has any thoughts on this subject, I would be very glad to hear them. Mr. H ild ebr an d . I do not think any careful study has been made to indicate that wage pressure from collective bargaining has specifically injured our foreign trade, though I do not think we can overlook that possibility. From what we hear, the steel industry has claimed that it has lost domestic sales to Japan, Italy, and some other countries as the result of price competition. We also hear this in connection with electrical equipment for powerplants, and so on. I think the safest judgment at this point is that any impacts of wages have been spotty in relation to prices. I would like to say in this connection that some of the lost sales abroad and increased im ports at home can be more obviously traced to demand and product factors. To illustrate this, consider the fall in sales of American automobiles abroad, which has been sharp and persistent. It goes directly to the kind of automobile we are making rather than to its cost, although that undoubtedly has some influence. This also holds for increased auto imports. Further, the decline in sales of coal to Europe last year was directly connected with the leveling off of the European boom rather than to price problems on this end. I would venture that probably the areas of greatest potential pres sure would be in fabricated products. There is no doubt that European automobiles, Japanese cameras, and Swiss watches, are increasingly coming in, and these cause pres sure and concern for domestic producers and unions. However, I think that this is a good thing on the ground that if the domestic E M P L O Y M E N T , G R O W T H , AND PRICE LEV ELS 2569 producers want to survive under the system of competition that they advocate daily then they have to act like free enterprisers and find ways to become more efficient. From that point of view, the pressure is to the good rather than to the bad. Representative R eu s s . Does anyone care to comment further ? Mr. O r n a t i . I f I may add a point. I am in general agreement with the comments of Professor Hildebrand. I would like to note that to my experience labor costs throughout Europe—and I have particularly in mind Italy and France—have been increasing much more markedly than wages have been increasing in these countries. The nonwage cost part of the labor cost package in Italy is almost twice as large as the wage cost package. Therefore, I think what ever advantage has been recently accruing to European producers must be ascribed primarily to the fact that their plant is being re newed at a faster rate than ours. Representative C u r t is . Do you mean percentage labor costs ? You mean percentagewise, the percentage of the cost that goes to labor when you say increase ? Mr. O r n a t i . That is right. The striking thing is the extent to which Europe is renewing its industrial plant. It seems to me that is the crucial part. I do not think I am in disagreement with you on that. I think that is the crucial part in the explanation o f why we have been losing markets. The C h a i r m a n . I f Congressman Reuss will yield, may I ask, is this true for Germany and Japan? Is it not true that since the war the rate of increase of wages generally has been appreciably less than the increase in productivity and relative to efficiency in the countries, the German wages are lower and the same thing is true in Japan ? As I understand, this has been partially true because of the past large unemployment, because of the East Germans who have been coming into West Germany. It may be that as they have been absorbed into industry, wage rates will move up more rapidly now. Has this not given us trouble in Germany and may it not continue to give us some trouble and particularly so in the case of J apan ? What you say about France and Italy is probably true. What about those two countries, where most of the complaints come from as a matter of fact ? Mr. H ild ebr an d . I think the evidence indicates you are quite cor rect. The German unions have followed a policy of wage restraint more or less continuously since 1948 or 1949. The German achieve ment in rebuilding modern plants of advanced type has undoubtedly also helped. The record indicates that Germany has penetrated the world mar ket in a truly miraculous way in the last 5 years. To that extent it represents some competition with ourselves, particularly in third countries. The C h a i r m a n . In times past when these factors were brought up, we would say that the gold movement would ultimately result in the equalization of prices. The gold standard is not as operative gen erally as it was. How long do you think it would take for gold move ments and exchange alterations to produce a situation where labor costs per unit of output or total cost per unit of output would be 38563— 59— pt. 8------ 9 2570 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS approximately constant in various countries? We used to say in the long run this will adjust. How long does the run have to be? Mr. U l m a n . Maybe we will all be dead before that happens. The C h a i r m a n . Is it 2 miles? Is it a marathon? Is it a tread mill ? Is it a cross-country race ? Mr. O r n a t i . I would like, Mr. Chairman, to go at it a different w^ay. The equilibrating force, as far as I am concerned, stems pri marily out of the expected increases in income of these countries, which will raise the propensity to import of these countries. I see in that area the equilibrating force rather than in the longrun movement of gold. Mr. H il d e b r a n d . I would like to throw in the suggestion that none o f these countries, including our own, is any longer willing to abide by the logic of the gold standard but instead will adopt internal poli cies to try to offset the effects of adverse movements in gold and for eign exchange. Probably the British are the most notable example o f this, so much so that Professor Hicks has suggested that none o f these countries operate on a gold standard any longer. They now operate on what he calls a labor standard. By this he meant that the wage-push is the base of the cost-push and the cost-push is the central fact in the rise of prices abroad, and at home, too, and from this it follows that what you do is to adjust every time these cost-pushes in the different countries throw things out of line by restricting imports if they work to your disfavor. That seems to be the current approach. While I would not go all the way with Professor Hicks I think that the Americans can take some comfort in the fact that other countries also have rising prices and wages. I f you wish to call this a cost-push, all right. In any event it is not a case of the United States alone moving far out of line with the rest of its trading partners. Representative R e u s s . I would just comment that I cannot be quite as optimistic about the effect of competition on this matter as perhaps Professors Hildebrand and Ornati are. The differential between wage levels of the industrial countries and ours seems to me so great that while it may indeed have a back-stiffening influence as far as future increases in wage rates are concerned here, I cannot see as a practical matter that it is going to produce much of an equilibrium by driving down American wage rates. Indeed I do not think it would be a good thing if that happens. This, however, does not bring me very much closer to the conclusion as to what our policy ought to be. Mr. U l m a n . May I add, sir, to your comment, that if a future stiffening prevents the situation from deteriorating to any greater extent, that would leave us where we stand now, which is with a current surplus on merchandise account. In other words, we are exporting more in the wav of merchandise than we are importing right now. It is the trend, I think, that is causing concern rather than the actual situation at present. Representative R e u s s . Yes. However, if you take our foreign aid obligations as a semipermanent facet of American life, and what we have done with the W orld Bank and the Inter-American Bank make it look that way, you overtop your merchandise account and produce the problem that we are talking about. EM PLOYM ENT, Mr. U lm an . Yes. aid very effective. G R O W T H , AND PRICE LEV ELS 2571 That is one of the things that makes our foreign Representative R e u ss . Yes. Although put all together it pro duces the imbalance of payments which does need a solution in tho long term. Mr. O r n a t i . I would like again to reemphasize the point that competition is not operative on the wage side. Two things must be remembered. Generally, wages are rising throughout the world faster than they are rising in the United States. More important, and this is what I tried to say before, total labor costs; that is, wages plus costs for social security and other supplementary payments which most governments require the employer to enter into, are growing even faster. The advantage of some foreign producers is on the side of capital improvement, better merchandising, better producing practices, et cetera. The C h a i r m a n . Is this true in Japan and Hong Kong, for ex ample ? Mr. O rnati. I do not know, sir. I have not studied the situation in Hong Kong. The C h a i r m a n . The British textile and clothing industries are complaining very bitterly about goods coming both from Japan and from Hong Kong, which have this privileged position, and we are getting more and more complaints about Japanese competition in plastics as well as textiles and clothing. Mr. O r n a t i . Certainly the outstanding fact about the Japanese situation is the improvement in the products themselves. Representative R e u ss . I will conclude by saying that I would cer tainly hope that the rest of the industrialized world would grow up to the structure in the sense that you are talking about, but I must say there is a lot o f airspace and I hope it will grow fast enough. The C h a i r m a n . I f you have an exploited labor force in these areas, with low labor cost per unit of output, and where equilibrating forces are not allowed to operate, you may have a permanent difficulty, Mr. B u dd . I just wanted to make one comment, a distinction be tween the general level of money and real wages and money and real wages in particular industries, I think so far as the first topic is concerned, monetary factors, such as changes in the exchange rates, are the proper solution. I f we do find we are being undersold all around, it simply requires a change in our exchange rate relative to those of other countries. I think the other problem that is of some concern here is the degree to which other countries can undersell us in particular items, such as optical goods or Swiss watches. I think the only answer one can give there is that either the industries that are competing with these foreign imports must find ways of reducing their real costs or that* if they cannot do this, we really should not be producing these items but it would be better for us to import our watches or our optical goods and have our labor employed in other industries where real wages are as high if not higher. Representative C u r t is . I wanted to clarify one thing before I do any questioning. When you say that labor costs had increased, I understood you to say that you were referring to the percentage o f labor cost and the 2572 EM P L O Y M E N T , G R O W T H , AND PRICE LEVELS price, but when you answered Senator Douglas with regard to Ger many, productivity had increased more rapidly than labor costs. Would we not have the reverse there? The percentage of labor cost had not increased. In fact, it probably had declined. I just wanted to get that straightened out in my own mind. When you said labor costs increased in Western Europe, what were you referring to? Mr. O r n a t i . T o both the actual level of money wages and to the payments that employers in Europe pay to a large series o f funds held by the state, such as the social-seeurity group, and to a large number of nonwage items called fringe benefits. Representative C u r t is . Y ou were not referring necessarily to whether or not the wage element in the cost of a product had in creased. I asked my question whether you were referring to the per centage increase in the labor cost as a percentage of product price. Mr. O r n a t i . The proportion of labor as a part of the total has also been increasing. Representative C u r t is . That is the point I was getting at. In light of what Chairman Douglas asked about Germany that would not be true if productivity had exceeded the increases in labor. Would you not, in that case, have a lower percentage of labor costs in relation to the product? Mr. O r n a t i . Actually, sir, I have not recently looked at the specific figures and the percentages in the two groups. Representative C u r t is . My point has been clarified and I have it clear in my own mind. Professor Budd, in your tables you adjusted for several things. The one thing that I wondered whether you had given any thought to is what I would regard as the labor mix, referring to unskilled labor, semiskilled, and skilled, that goes into this labor cost, I am looking particularly at table 3, from 1900 to 1957. Do you not think there has been an increase in the amount of labor cost in the skilled areas and less in the unskilled and semiskilled? Mr. B u d d . In general, it would be impossible to correct the figures here. About the only thing one could hope to do is to make correc tions in the industrial composition. Representative C u r t is . I think it is very important to know be cause from the standpoint of real cost of manufacture, it makes a lot o f difference whether a manufacturer’s labor force consists of very highly skilled personnel or whether it goes down and there is a lot of unskilled, particularly as our recent unemployment seems to have been in the area of unskilled and semiskilled. Mr. B u dd . I might add here that it is not at all clear to me how a shift in the composition of unskilled to skilled would necessarily af fect the share of employee compensation in the total. I would agree that in recent years there appears to have been a rise in the number o f salaried workers and particularly of research personnel and so on, end that has been a larger proportion of the total cost. Representative C u r t is . The reason why I suggest it is a very per tinent thing is that if that is the reason for the increased labor cost— and I suspect for a good bit it is—it calls for a different kind of action than if it were in just the fact that labor itself had increased its per c e n t a l of the actual share. EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2573 Mr. B udd . I do not think that is as obvious in table 3 as it is in figure 1, where you find the differential movements in average annual earnings between the union and nonunion sector. I would suspect that this widening differential we find here may in part be due to the fact that the manufacturing industries which are in cluded in the union area are hiring a somewhat larger proportion of skilled personnel than they did before, research personnel, salaried workers and so on, and that does affect the average. So it will tend to lift the ratio in the union sector. By how much, we cannot deter mine from the figures by which I calculated this. Representative C u r t is . I was not bringing this out in criticism o f the presentations you made because I appreciate all these papers very much. I was trying to get at a factor which, to me, seems quite im portant. Mr. H il d e b r a n d . May I comment on that, Mr. Chairman? Ignoring for the moment the question of its impact upon relative shares, the trend at least has been to favor the rise of the semiskilled operator, the assembler and so on, in industry at the expense of the old-fashioned unskilled laborer. Even he, too, operates a machine in construction trades and so on. Along with that same trend you have some decline in traditional skills in industry. You might say there occurred a concentration in the middle brackets of skill. It has been noted in the last 4 or 5 years there has been a change in the other di rection. With automation and related new technologies there is a rise in the demand for the professional and the technician and a new kind of skilled worker. This so-called overhead labor, which is largely nonunion, has had an impact on the overall labor costs of manufac turers and other types of enterprises. A t the same time, it seems to have something to do with the decline or the growth in productivity in 1956, 1957, and 1958. Putting this new type of staff to work did not immediately increase product. This will be delayed. So it has that significance. So far as the unions are concerned, their problen. is to organize these new people. My guess is that you are not going to see much more compression of occupational differentials up the skill ladder but perhaps a widen ing once more. Representative C u r t is . I am happy to hear that prediction because that is what I was going to get around to. I would pose this question. As we continue automation— certainly economic growth is coupled; in that to me—are we not almost bound to continue to increase what we have been calling the percentage of labor share? Let us take the big arguments of the farmers. They are claiming that they are getting so little for their raw material. They get so little for the wheat and the loaf of bread costs so much, forgetting all about what we havei been doing to that wheat before it actually gets to the consumer. It just seems to me if we are going to continue the economic growth.in our society we are going to have a continual increase in this area. I am perfectly relaxed about it but, if it is so, I would like to know it, or, if someone does not think it is so, I would like to know. Mr. H il d ebr an d . I f I may answer, I would suggest that I think there is a possibility of some upward shift in the share going to em ployee compensation from this. I also think you have to weigh against that that the rise cannot go too far without cutting into returns to 2574 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS capital so mucli that it would defeat itself as a process of technological change. Representative C u r t is . Or you have the other side, because just as this is coming about, so the amount of capital required to employ these workers is going up. I have seen figures of $20,000 per employee. Mr. H ildebr and . The very process would have the effect of also economizing on labor, making it more productive and thus offsetting the rising tendency you are talking about. It is a sort of two-sided affair. Representative C u r t is . No, I think your productivity will increase, but as your productivity increases, is not labor share bound to be greater and must it not be, because the amount o f actual raw material you start with is going to have more done to it and therefore the cost will go up ? I do not know. It depends, naturally, on what you need in the way of investment to have the capital machines necessary to put this into effect. A great deal of increased productivity, using it in a very broad sense, in my judgment does not depend on machines. It may be just a new method of doing things, or a new idea, which may or may not require capital, but we know certainly requires labor. The next point I was coming to has been answered to some degree. It has been brought out in Mr. Ornati’s paper. I w^ant to get this clear. I happen to agree fundamentally that a strong labor movement is very essential if the private enterprise system is going to work. So in my criticisms here, if they seem ad verse, it is in that context. As Mr. Ornati points out, it seems to me the labor union has been one o f the great holdbacks in this business of job classifications and relating wages scales to different skills. You have indicated, Mr. Hildebrand, that possibly it is going to spread again. I wonder if you would comment on that, Mr. Ornati, in relation to the unions. Are they not fighting that spread ? Mr. O r n a t i . I see two tendencies and I have in my paper empha sized the more important one that has led to a narrowing of differen tials I think we now are seeing within the unions— and I give here an example, the U A W —a tendency the other way. This is reflected in their own organization and in the creation of separate bargaining units for separate groups of workers. I have in mind the skilled workers. The same situation has taken place in New York Transit, where Mr. Quill has for so long bargained for the less skilled worker that he now finds himself in trouble with his more skilled worker. Representative C u r t is . In the UAW , did not a skilled group actual ly have 40,000 or 50,000 people break off from the union in protest ing the effect of this compression ? Mr. O r n a t i . I see the unions reacting to take administrative ac tions that will make it easier for them to bargain toward the widen ing rather than the compressing of differentials. The longrun effect has been essentially one of compressing the differential. Representative C u r t is . The other point there is mobility of labor. Again it seems to me that the unions tend to oppose labor mobility, and not just the geographical mobility but mobility from one skill to another because they lose jurisdiction of the individual union man. E M P L O Y M E N T , G R O W T H , AND PRICE LEV ELS 2575 Mr. O r n a t i . I would agree with you, sir, but I would prefer to broaden your statement, if I may, in terms of this not being a process in which the unions are the only ones involved but rather it being a joint process wherein management and the unions, if one can use the word “ blame,” should be equally blamed. Representative C u r t is . I am not trying to lay blame really as much as expose it. I can see where management tries to retain its personnel and keep it intact. I think management is interested in the mobility, at least in switching skills. The reason I pose the question is that I do not think it is inherent in unionism that they should have to oppose mobility of labor. I think there has been that tendency up to date. Mr. H il d e b r a n d . May I comment on this point, Congressman Curtis ? Representative C u r t is . Yes. Mr. H il d e b r a n d . I think it is definitely a moot issue whether union ism has really reduced mobility of labor. I know Professor Ornati, my colleague, suggests that it has. I would only like to say that, as in recent study by Professor Ross of the University of California, if you take voluntary-quit data going back to before the First World War, wre do not find any change for comparable levels of economic activity in the turnover of labor. So the new feudalism that people talk about may not be a fact. I f you take the craft unions, which generally organize horizontally, they promote the horizontal mobility o f labor by bringing buyers and sellers together. In the industrial case, it is true you have much less turnover between employers by em ployees after you pass the very junior people who are still so-called job-shoppers. But in those contexts this has always been the case apparently for industrial workers. Once they work into a company they stay with it. The mobility we talk about is vertical up the skill ladder. When you look at that, the unions are not interested in denying—the industrial unions at least—their people opportunities to rise. When they do insist upon generally is that the seniority or length o f service yardstick be used instead of merit, or at best they will concede some compromise. I f that is so, it would be generally an incorrect inference to say that they block advance up the ladder of skill in the plant, but it would be correct to say that they try to control it according to a standard of equity. Representative C u r t is . Would you not say that although they may not admit they do that, by resisting constant classification and reclas sification and a relation of skills to wage scales they in effect are doing that? I think that is what is bringing about this upheaval. Mr. H il d ebr an d . I think it is correct to say when many jobs are at stake the unions will resist change. This will, in turn, check the adaptation of the men’s skills to new job opportunities. However, there is prospect for accomplishing those changes more easily if shift is made to some sort of temporary severance pay or retraining ar rangement. There has been a good deal already accomplished in certain industries in this direction. Mr. O r n a t i . I f I may, I am not in disagrement with what Pro fessor Hildebrand has said, but the comparison is not so much in terms o f whether the mobility, be it vertical or horizontal, has in creased over time— and I suspect that it has—but whether this mo 2576 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS bility has been fast enough in terms of the needs of our economy. M y looking at the labor movement since 1953 does not lead me to say that there has been less mobility, but it does lead me to say that there has not been enough mobility. Although I have not, inasmuch as I had not so been instructed, made an policy proposals in my paper, I would ask for permission to mention some areas in which I think the committee may think of studies about policies which may improve mobility. May I, Mr. Chairman ? The C h a i r m a n . Yes. Representative C u r t is . My time has run out. I wonder if you would add to the thought that you think management tends to resist this mobility, because it would be my impression that management would tend to want to encourage it or allow it to go as fast as possi ble. I may be wrong. Mr. U l m a n . Frequently when management wants to hire people they are all for greater mobility. When it comes to a question of people leaving their firms, then we hear talk of excessive turnover or labor pirating. Eepresentative C u r t is . That is mobility. They resist losing their men, I agree. Mr. O r n a t i . I would like to go at the same point Mr. Ulman went at differently. It seems to me that one of the greatest drawbacks o f mobility— and I think that the academic profession and the labor specialists in the universities have contributed to this— can be found in what I will, for shorthand, label the tendency of management to hire “ our own type of man.” The concern that management has had in recent years with smooth, good human relations has led to a reduc tion o f mobility. As long as you want a smooth, quiet, Musakoriented work force, you are not going to have very much mobility. The economy requires more mobility and possibly less happiness. I would suggest that the committee would gain by studying various specific acts to make movement of workers from job to job easier. One of these would be to make the cost of moving from one job to another tax deductible by the worker, as it now is when moving costs are paid for by the employer. I would also suggest, and this requires a good deal of study of ad ministrative problems, that making pension plans vested in the indi vidual and transferable within the Federal system would be a great gain. Along with that, I think we can do much better in the area of the operation of the USES, and I would suggest, as an area of study, the possibility of requiring employers, within certain limits, to hire all new workers through the U.S. Employment Service, The C h a i r m a n . Is that not going too far ? Mr. O r n a t i . It might be. I propose it as an area of study. The C h a i r m a n . It might require them to give notice to the employ ment service of all vacancies. You might set up a job monopoly if you require hirings that way. Mr. O r n a t i . It might. Yam aware of countries in which an equiva lent agency has become a job monopoly where it has been used for political purposes. EM PLOYM ENT, G R O W T H , AND PRICE LEVELS 2577 The C h a i r m a n . That was the common tendency under fascism in both Italy and Germany. Mr. O r n a t i . Exactly. Mr. H il d e b r a n d . It is still true in Italy. Mr. O r n a t i . It is. The C h a i r m a n . Senator Bush is not here. Mr. Levinson, of the staff, has a question which he would like to ask. Mr. L e v in s o n . I would like to ask the panelists if they could clarify or extend a bit on their feelings about the issue of so-called cost-push versus demand-pull in the labor area. I believe in general that Mr. Budd has considerable skepticism as to whether or not unionism really lias had any impact. Mr. Hildebrand sort of feels that it has, but it has been relatively minor in the postwar period. Mr. Ornati feels that perhaps the pressures have been greatest where employment has been weakest, though I am not quite sure what his feeling is. Mr. Ulman feels that unions have had an impact. In particular, I would like to put the question this way. In view of the fact that there does seem to have been upward movement in wages even during the recession periods of 1949, 1954, and 1958, and in view of the fact that there seems to have been upward movement in wages during the whole period from 1955 through 1958, even though there has not been any significant pressure on the demand side in the labor market, how would the panelists react to this phenomenon ? Can this be explained as being part of a demand phenomenon, or do you feel that it is a cost-push ? How is this a minor element in the picture ? I would appreciate your comments on it. Mr. U l m a n . In my own summary, Mr. Levinson, I concentrated on the area of collective bargaining because that is the subject before the committee. In my written statement, I indicated my belief, and it is a strong belief, that collective bargaining has been an independent factor among others making for inflation. I would not, therefore, contrast cost-push with demand-pull. I regard the inflationary tendencies which we have experienced as mixed, with both sets of factors operating. I make no claim that collective bargaining is the strongest of these factors. Professor Slichter thought so. I know of no way of disentangling the different contributing forces. I used the word “reasonable” in my paper to characterize by posi tion. Mr. Hildebrand said “ prudent” in his. In the face of nonquantitative evidence in the labor market, some of which you have summarized, it is indeed reasonable to presume that there has been an appreciable contribution on the side of collective bargaining. I think the evidence which you cited as far as continued upward movements of wages in the postwar recessions are concerned is persuasive. Although we must realize that there were recessions in the 1920’s in which hourly earnings also rose, they were not, to my knowledge, relatively as severe as were the recessions in the postwar period. As far as the upward movement in 1955 to 1958 is concerned, I would agree it is difficult to explain this in the pure demand-pull 2578 EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS hypothesis. I do not like to contrast the term “ demand-pull” with the term “ cost-push.” Perhaps Professor Lerner’s terms “ sellers’ inflation” and “ buyers’ inflation” are better. Professor Eckstein, on the staff o f this committee, has advanced a thesis concerning the role played by the growth in investment pressing against certain bottlenecks and capacities. A t the same time this does not explain satisfactorily to my way of thinking the increases in wages which have taken place because it was capital which was the limiting factor. I do not think that in many cases labor was the limiting factor. For that reason I believe that Professor Eckstein, who unfortunately is not here to defend himself, did include both collective bargaining and the bottleneck in his analysis. There is very little evidence of stringency in many o f the labor markets in which wages have increased most rapidly. For example, in the entire period 1947 to 1958, I know of no category in which hourly earnings have risen more rapidly than railroads, in which they have more than doubled in the face of a drop in unemployment of 27 percent. In bituminous coal mining, hourly earnings rose by 84 per cent in this period, while employment fell by over 50 percent. In nondurable manufactures, a good part of which are organized, earn ings rose by 65 percent and employment fell fractionally—by about 2 percent. In steel, an industry of some interest right now, hourly earnings doubled between 1947 and 1958, although employment dropped by 9 percent. I see no stringency in such labor markets, I would agree with you. Mr. H il d eb r an d . I would like to associate myself fully with Pro fessor Ulman’s observations. It seems to me that it would be a mistake to say we have to choose between two explanations of the rise in prices and wages since W orld W ar II, one which says they are due to demand and the other to cost. I think the two factors have worked together. Indeed, I suspect there are other factors intermediate between wages and prices that do not fit either explanation particularly well. For that reason, I would contend there are both demand and cost factors at work. We have had high employment continuously since 1942. While for re cent years, since Korea, the monetary side does not seem an effective explanation, at least the quantity of money, certainly the velocity of turnover in money, has had something to do with it. I f you take the period after 1952, we do have certain things that suggest that some thing new may be at work. One is the apparent failure of profit margins relative to sales to widen in any sharp or clear way, which we would expect with a demand-induced inflation. Another is the fact that in certain years since 1952 we have had very obvious excess capacity. I think the Board of Governors of the Federal Reserve System estimated for 1958 excess capacity in certain key industries o f 35 percent. So the level of employment and the volume of unemploy ment are not very good indicators of how much slack there may actually be in the economy. But in the years of 1954 and 1958 we actually had a fall in gross demand, whereas in 1952, 1956, and 1957 the rise o f gross demand was very moderate indeed. So that 5 years out o f the 8 since 1951 do not indicate great pressure on the side of demand. E M P L O Y M E N T , G R O W T H , AND PRICE LEVELS 2579 On the other side, we have this wage behavior that Professor Ulman has cited, and no evident shortage o f people in some of these industries where wages have risen. Therefore, it is? indeed difficult to attribute this phenomenon to something working on the demand side. When I said “ prudent to admit this,” I meant we should admit that unions can exert upward pressure in certain industries, at least, beyond what happens in their absence. On the other side, they do have ability to resist downswings in wages and the two forces together mean something net for the wage level, and beyond the wage level, something net for the upward movement of prices. How much, I do not know, and I do not think anyone knows, but I feel very strongly that the influence on prices has been minor, and even more strongly that the danger of price inflation is grossly exaggerated today. Mr. B udd . I would like to take a few moments to clarify my own position on the matter of cost-push versus demand-pull inflation. My particular comment was related to the effect of collective bargaining on the supposed cost-push. My objection there concerned the manner in which the wage increases get transmitted to the nonunion sector. I still feel there is no evidence or theory to justify the belief that every time the union sectors get a 10-percent increase in wages the nonunion sectors automatically follow by 7, 8, or 9, or 10 percent, So if the unions are bidding up the wage level, it has to show up to a large extent in the widening of union-nonunion wage differen tials. The evidence does not seem to be clear cut there. Secondly, in trying to account for why prices have risen, if they have, since there seems to be some question about the adequacies of our price indexes to reflect the real price increase, it is possible to de velop a model, and a study that was prepared for this committee by Professor Schultz that I was reading on the way to this hearing points out th at if you have general w age rig id ity dow n w ard, and this seems to be true o f both nonunion and union sectors, since even nonunion employers are reluctant to cut wage rates, and if you grant the fact that both union and nonunion wage rates are sensitive to upward changes in demand, and you get a very substantial shift in the com position of output as you got in 1955 through 1957 with a very sub stantial increase in investment demand, then these increases in wage« and in prices that are occasioned by the increase in demand in one sector are not counterbalanced by wage reductions elsewhere. Hence, you will find an upward rise in the wage and the price level, but it is difficult to take this on and say it is evidence of a cost-push in the sense that some autonomous element on the cost side, such as unions, are doing the pushing. Mr. O r n a t i . Mr. Chairman, it seems to me this committee is well prepared in the field of economics and we need not rehash familiar arguments. A t the risk of reading myself out of the profession, I would like to make some unorthodox and possibly trite comments. The first is that it seems to me that the cost-push demand-pull analysis cannot be cracked open in an aggregate sense. It seems to me that there are clearly certain areas, certain labor markets, and certain in dustries where one can note an increase in costs that stem essentially 2580 E M P LO Y M E N T, G R O W T H , AND PRICE LEVELS from the union side and that have been transferred into price in creases. An example of this would be the hat industry. Here we deal with very broad aggregates, the manufacturing sec tor, and we call that a unionized sector, versus, in brief, the nonmanu facturing, and we call that the nonunionized sector. I would like to point to the reality of places such as Binghamton, N.Y., and Roches ter, N.Y., which are two essentially industrial manufacturing centers which are not organized at all; nevertheless, in both these centers the wage level is relatively high. It is comparable to the wage level of other centers. We must ask the question why are wages high in these particular areas ? Without ever having done a detailed study, and without want ing at this point to be cross-examined on this particular detail, I ’would argue that the level of wages in Binghamton is in no small part determined by the presence of the IBM company which is a highwage-paying concern even though it is nonunionized. ' The C h a i r m a n . Also, they have one of the most interesting shoe companies in the country, the Johnson Shoe Co., which has always tried to maintain a differential in wages raid working conditions over the industry as a whole. Mr. O r n a t i . That is right. The same applies for the city of Rochester where the influence of the Eastman Kodak Co. on the total operation of the labor market can not possibly be discounted, both in terms of the kind of wTorkers they select and attract and the kind of wages they pay. Nevertheless, Rochester can be, in spite of the good wTork of the Amalgamated, labeled as a nonunion town. • What I am trying to say, Mr. Chairman, is that I feel ill at ease in the debate as long as we keep it at the level of aggregates. . Mr. IJlman . Mr. Chairman, in my written statement I refer to three channels whereby union wage increases may be transmitted into wage increases in nonunion sectors, all of which have been referred to in the professional literature. To summarize these very briefly : (1) if prices of union-made goods rise relative to prices of nonunion goods, this exerts an upward pres sure in the demand for labor in the nonunion sector indirectly; (2) if bargained wage increases result in higher payrolls and total expen ditures by the unionized firms involved, part of the increases in wage incomes would be spent on products and services in the nonunion sector; (3) and this phenomenon has been referred to here at least by indirection, employers of nonunion labor frequently raise wages in their own plants in order to avert a decline in employee morale and efficiency and in some cases to keep unions out of their own plants. Now, all of these channels may be offset, or neutralized as it were, *if sufficient labor leaves the union sectors where wages have been raised and moves over to the nonunion sectors, thus contributing to an increase in the nonunion supply which would offset the increase in demand for nonunion labor. There have been studies, and we referred to the results of some of these studies, concerning immobilities involved in industrial labor. I would like to mention two factors again. One of these factors con cerns the relative level of union and nonunion wages, at least in some important sectors. In the manufacturing sectors, wages are high. EM PLOYM ENT, G RO W T H , AND PRICE LEVELS 2581 These jobs, from the viewpoint of many workers, are good jobs. They do not want to leave these jobs. For that reason, I mentioned rather elliptically the role of make-work rules and local work practices, which tend to restrict the mobility of labor from union to nonunion sectors and thus prevent any increase in the demand by nonunion employers from being matched by a sufficient increase in the supply. As far as the downward rigidity of money wages is concerned, X would agree that employers, at least in large-scale industries where competition is not extremely keen, have always been reluctant ‘to reduce wages of their employees. So at least in the short run the labor market has not operated as might be predicted by a very simple, naive, classical theory of behavior. However, it is generally agreed by stu dents of this subject that the downward rigidity in wages and prices1 has been greater recently than it used to be before the war. This might be due in part to changes in industrial employment practices; but I think it only reasonable to expect that, since one difference be tween the situation now and the situation before the war is the much greater extent of unionization, at present the unions have exerted a definite effect in this area of behavior. Mr. Chairman, I recently moved from the State of Minnesota where they had an old farmer who did not believe in the existence of giraffes. One day a circus came to town and it featured a giraffe. The farmer took a look at the beast and walked all around it and his reaction was: “ There ain’t no such animal.” The C h a i r m a n . We are nearly at the end. I do not think we should prolong the substantive discussion much further but I understand that Congressman Curtis has a minor question he would like to raise. Representative C u r t is . I do not know that this is generally true, but many unions, when they elect their officers, say the only people who can vote are those who are actually employed at the time. I think this is fairly general. I am just curious what sort of effect that produces if the union officers, this being so, will then tend to represent those who were on the jobs themselves. And the absentees, not having a right to vote, how might their interests be represented ? Maybe there is no economic impact. The C h a i r m a n . You mean that those who have not yet entered the industry or plant are members of the union and can not vote. Representative C u r t is . N o . Say there have been cutbacks. Then there is a vote. Your union officials are going to pay more attention to the ones who surely have their votes—I guess it is sort of like being on a committee. I f you are on the Ways and Means Committee and are within the first 10, regardless of the switch in the political control of the Congress, you keep your job; but if you are Nos. 11, 12, and 13 and it switches, you are out. I do not know that it has any serious economic impact. I have just been thinking about that and have been ever since I realized that that was so. The C h a i r m a n . Ever since you acquired status on the Ways and Means Committee and realized that your job was fast. Representative C u r t is . There is one other point and that is the impact of the labor unions on the utilization of our older citizens. There is a resistance on the part of the unions for us to raise the amount that people can earn and still retain their social security 2582 E M P LO Y M E N T, G R O W T H , AND PRICE LEV ELS benefits, that resistance has been primarily from the labor leaders; they do not like to say that publicly, but I can assure you that they are the main core o f opposition to it. I do not think that is inherent in labor organizations. I do not see why it would be although I can see the reason for wanting to move the older citizens out so that the jobs are available for those coming in. But I do not think that is inherent in labor organizations. That is the other question I wanted to pose for any one who would like to supply a comment. Mr. U l m a n . Perhaps with reference to your last point, Congress man, we might again refer to one of Professor Slichter’s favorite remedial policies. Perhaps it would be a good idea to give em ployers of older workers some financial incentive to employ them by reducing the contributions which such employers might have to pay into the social security fund. Representative C u r t is . I am on that subcommittee and have been for a number of years and I am particularly interested in that subject. The C h a i r m a n . I f there are no further questions, we thank our friends for coming. The next meeting of the committee will be tomorrow morning at 10 o’clock. (Whereupon, at 12:30 p.m., the committee recessed to reconvene at 10 a.m., Wednesday, September 30,1959.) EMPLOYMENT, GROWTH, AND PRICE LEVELS W EDNESDAY, SEPTEMBER 30, 1959 C ongress of t h e U n ited S ta te s , J o in t E conom ic C o m m it t e e , 'Washington, D.C. The committee met, pursuant to recess, at 10 a.m., in room P-63, the Capitol, Hon. Paul H. Douglas (chairman) presiding. Present: Senators Douglas and Bush; Representatives Curtis and Reuss. The C h a i r m a n . The committee will be in order. Gentlemen, we appreciate your coming. We have a very distin guished panel this morning: Mr. Kahn of Wayne State University, an old friend of ours, George Taylor, distinguished arbitrator and student of industrial relations, and Mr. Stieber of Michigan State University. Mr. Kahn, I take it you are to lead off, so will you start ? STATEMENT OF MARK L. KAHN, WAYNE STATE UNIVERSITY Mr. K a h n . Thank you, Mr. Chairman. Today’s inquiry by this committee into the impact of nonwage col lective bargaining practices on economic growth is part of a wide spread reexamination of the consequences of unionism now taking place in the United States. This reexamination is the result, in large measure, of public concern over the possible inflationary effects of collective bargaining and about disclosures of corrupt activities within some unions, and of an increasing concern on the part of many com panies and industries for sustaining profitable and competitive operations. One thing is clear: many employers have decided that there exists today a good tactical opportunity to insist at the bargaining table on substantial revisions of working rules and conditions which they consider onerous, and which, incidentally, they have in many cases tolerated for years. This accounts for the present emphasis on such matters by the steel and railroad industries and is at least a major facet of the “ hardening of attitudes” on the part of management noted by Arthur Goldberg and others in recent months. In this statement—which is prefatory to the panel discussion that will occupy most of this morning’s session—I wish first to make some observations concerning the difficulties inherent in any attempt either to isolate or to generalize concerning the economic consequences of collective bargaining on nonwage matters, and then to suggest why, in my judgment, the careful study of individual cases and much for- 2583 2584 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS bearance are desirable aspects of public policy in connection with working rules which are alleged to inhibit productivity. The growth of unionism since the 192(Vs creates a temptation to attribute causation to collective bargaining where little or none may actually exist, particularly by succumbing to the logical fallacy, “ After this, therefore because of this.” Thus, a prevailing view that seniority rights and pension plans have seriously immobilized our labor resources is yielding to the judgment that they merely reinforce the inertia of long-service workers. (Arthur M. Ross, “ Do We Have a NewT Industrial Feudalism?” American Economic Review, Decem ber 1958, pp. 903-914.) See Herbert S. Parnes, “ Research on Labor Mobility” (New Y ork: Social Science Research Council, Bulletin 65, 1954) for an excellent appraisal and survey of labor mobility research. Collective bargaining may even promote the flexibility of our labor force in times of national emergency, once guarantees are furnished that workers will not forfeit acquired seniority status upon shifting to vital tasks, since the unorganized worker has less assurance of such protection. Again, craft unions may reinforce occupational immobil ity, but it is clear that skilled workers with specialized training and experience tend to be occupationally immobile in any case. It is dffiicult to distinguish developments in employer practices or in the terms and conditions of employment which are the result of collective bargaining, or which exist (possibly but not necessarily) because of the unionized example in nonunion contexts, from develop ments which would in any case have occurred. Thus, while seniority rights are established only by collective agreement, many nonunion employers utilize seniority as a criterion for preference. (Manage ment Record, September 1949, reports the findings of a National In dustrial Conference Board survey of 110 nonunion companies with well-defined seniority systems.) Work standards vary among un organized as well as organized employers, and I am not aware of any reliable evidence which makes it possible to suggest that the average unionized worker supplies either more or less effort to his job than the average unrepresented worker. Unorganized workers are known to restrict output. Fringe benefits for workers do not appear only in collective bargaining agreements, and would have continued to de velop in the absence of collective bargaining, although, no doubt, with different timing and perhaps with different emphasis. I do not mean, by these observations, to suggest that we should not be concerned about the impact of collective bargaining practices on economic growth. I merely want to stress that causation should not be assumed. It is also important to examine collective bargaining practices in light of the particular technological and economic environment in which they have evolved and to which they respond. Valid generali zations are useful and important, but must be developed with caution. I would like to illustrate this point by reference to the subject of management rights and functions. There is considerable variation among industries in the scope of union activity and therefore in the areas reserved to unilateral managerial determination. Should unions participate in such activities as product pricing, controlling entrance to the industry by new firms, supplying workers for jobs, conducting research for product improvement, undertaking promotional cam EM PLOYM ENT, G R O W T H , AND PRICE LEVELS 2585 paigns to stimulate customer demand for the product, or regulating the relocation of plants? There are unions which do each of these things. While some unions negotiate only relatively narrow agree ments (and utilize the grievance procedure to police alleged employer violations), there are—at the other extreme—unions which actually organize or effectively promote and direct the organization of em ployers and which control major “ managerial” decisions in their indus tries. There is a full range of intermediate examples. It is a reasonable premise that the central institutional purpose of American unions is to protect and promote the job-related interest of the members. The diversity in practice appears as a response to variations in the economic, technological and legal status of the indus try and in the characteristics of the employees who are represented. It does not appear to be haphazard, or primarily attributable to the ideology or militancy or ability of the particular union and its lead ership. For example, industries characterized by a large number of small and intensely competitive firms tend to be relatively recep tive to utilizing the union as an instrument for promoting stability and limiting competition. On the other hand, where an industry is composed primarily of a few large multiplant corporations, em ployers have little reason to lean on unions, are more concerned about maintaining bargaining power in relation to the union, and tend to resist strongly efforts by the union to expand the scope of its activities. I think it may be a valid generalization, Mr. Chairman, to say that the greatest concern on the part of employers about the infringe ment of managerial prerogatives has tended to come from employers in industries where, looked at from the outside, managerial preroga tives appear to have been invaded the least. Yet, these employers may have proper cause for concern in many instances. It is no wonder that unions and managements have been unable to agree on a list of functions which all managements should be en titled to retain on a unilateral basis. (The President’s National Labor-Management Conference, Nov. 5-30, 1945, U.S. Department of Labor, Division of Labor Standards, Bull. No. 77.) A list appro priate for one industry might be incongruous if applied to another. The key question, it seems to me, is whether or not a particular man agement successfully retains the authority it requires for the effective conduct of its responsibilities. This is, I think, increasingly recog nized by managements. As Professors Brown and Myers have put it : * * * it seems to us that there have been marked changes in management philosophy [concerning management prerogatives]. Put briefly, the changes reflect a shift from the concept that management’s decisions are unchallenge able simply because they are management’s decisions, to the proposition that, in the interests of efficiency, it is better that certain ypes of decisions be made by management with minimal interference or control by the union. (Douglas v. Brown and Charles A. Myers, “ The Changing Industrial Relations Philosophy of American Management,” proceedings of the Industrial Relations Research Association, December 1956, p. 93.) What is important, then, is that managements insist, at the bar gaining table, upon the retention of essential discretion and authority. It is my opinion— and this is merely an impression—that most em ployers have been able to achieve such retention. There will always be legitimate differences of opinion on borderline matters, of course, 38563— 59— pt. 8— 10 2586 EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS as well as instances in which a management yields too much. In any event, no one can be better equipped than the parties themselves, who live by their own decisions, to bargain the matter out. The outsider, with preconceived notions of what functions should be shared or un shared, may not judge wisely. A similar caveat is applicable to other nonwage aspects of collec tive bargaining. Permit me to illustrate with a brief reference to seniority. There is no basic difference of opinion between unions and managements on the desirability of giving consideration to seniority as a criterion for allocating job opportunities among orgnized em ployees. It is true, of course, that badly designed seniority systems can lower efficiency and raise costs in three major ways: first, by so neglecting the “ ability” criterion as to place employees in jobs they can’t handle well, or as well as other employees; second, by causing excessive internal turnover—the “ bumping” problem; and third (chiefly as a consequence of the first two), by damaging employee morale— by delaying, for example, the promotion of proficient work ers or by displacing hitherto contented workers from their customary jobs. On the other hand, well designed seniority systems promote efficiency by reducing external turnover, partly because employers are impelled to adopt better hiring practices and to make careful use of probationary periods, and partly because high seniority status does reinforce, as noted earlier, the long-service employee’s natural reluc tance to move. Another consequence, which may be more significant than many employers realize, is the development of a more versatile work force as a result of bumping provisions. Finally, seniority can improve morale by establishing rights to jobs, letting workers know “ where they stand” and offering additional protection against arbi trary or discriminatory treatment. Seniority provisions must be tailored to the particular industrial setting and be responsive to changes in the structure upon which seniority rights have been based. As in connection with management functions, provisions which fit one industry or company or plant may be unworkable or prohibitive in cost in another setting. Again, it is the parties themselves who can best balance the sometimes conflict ing issues of equity and efficiency in a workable fashion. It is un fortunate that, in the development of collective bargaining, many unions insisted upon rigid seniority application to protect the union, in the absence of any union security provision, against the conse quences o f possible company discrimination against active members. Seniority rights, once established, are peculiarly difficult to change because any amendment must be to someone’s disadvantage where relative rights are involved (Mark L. Kahn, “ Seniority Problems in Business Mergers,” Industrial and Labor Relations Review, April 1955, p. 364). Nevertheless, the secure union which is no longer fearful o f management efforts to destroy it will be more willing to recognize the need for adjustments. Finally, I should like to mention Professor Slichter’s cogent observation that even where an employer is compelled by seniority commitments to retain a large proportion of less efficient older men (assuming that they are in fact less efficient in the particular case) there may be offsetting gains for the community: “ From the standpoint of the community, however, prolonging the effective working life of employees is desirable because the important thing is not how EMPLOYMENT, GROWTH, AND PRICE LEVELS 2587 much a man produces per hour or day, but how much he produces in a lifetime (Sumner H. Slichter, “ Union Policies and Industrial Management,” Washing ton, D.C., Brookings, 1941, p. 161). (A major revision of this valuable book is nearing completion.) This is a useful reminder that many aspects of collective bargaining should be examined not alone from the vantage point of the employer or the union, but also from the perspective of community well-being. This is especially necessary where social costs are involved in em ployer decisionmaking. Permit me to conclude these remarks with some observations about railroad working rules. The railroads do not provide a healthy en vironment for collective bargaining. Public concern for the avoid ance of work stoppages has led, under the Railway Labor Act, to considerable intervention by the Federal Government through media tion, emergency board hearings and recommendations, and even Presi dential intercession. We cannot permit the right to strike to exercise its usual catalytic effect. Strict seniority rules (having their origin in pre-union-shop days), the occupational basis of representation, the decline in total employment, the national level of bargaining, the rapid pace of technological innovations, and industrial reorganiza tion by way of consolidations and abandonments, have made the modernization of agreements difficult to achieve. Moreover, since the working rules are peculiar to the industry and are most complex, outsiders serving on Presidential emergency boards have been loath to recommend significant changes. On the other hand, it is important to remember that the total compensation of road service employees is less than one-sixth of total railroad payrolls, and it is only in connection with these operating brotherhoods that the railroads are now making serious complaints about the working rules. A competent student of these rules has concluded that while much is obsolete in the structure of compensation for operating em ployees, one must be very cautious about concluding that any par ticular rule has make-work aspects. (Morris A. Horowitz, “ The Rail roads’ Dual System of Payment: A Make-Work Rule?” Industrial and Labor Relations Review, January 1955, pp. 177-194.) Incidentally, as a result of this, there occur many inequities in rel ative earnings of different kinds of employees depending on the cir cumstances in which they find themselves, the particular equipment which they operate, and so forth. My own reading of analyses by Horowitz and others leads me to suggest that the working rules are badly in need of revision, but that the railroads may be overstating their purely make-work consequences. (Other articles by Horowitz: “ Make-Work Effects of Railroad Constructive-Allowance Payments,” Labor Law Journal, May 1955; “ Wage Guarantees of Road Service Employees of American Railroads,” American Economic Review, December 1955; and “ Some Effects of Labor Costs upon the Railroad Industry,” Labor Law Journal, February 1957.) The Association of American Railroads recently urged President Eisenhower to establish a special commission to study railroad work ing rules. As of the present time, he has not done so. My judgment is that such a study commission would provide desirable information as a basis for collective bargaining, but that a substantial time would be required for its investigations to bear fruit. Consequently, I 2588 EM P LO Y M E N T, G R O W T H , AND PRICE LEVELS believe that such a commission should be appointed, but after the present round of bargaining lias been concluded and not before. In closing, permit me to reemphasize these major points. First, that we must be careful about attributing causation to collective bargaining merely because certain practices exist under collective bargaining. Second, that from a public policy standpoint particular nonwage practices must be appraised from the vantage point of their impact on the community as well as on the parties. The results will not necessarily be the same. Finally, that collective bargaining is characterized by diversity and is intimately rooted in the particular environment in which it has evolved. Consequently, only the parties can be thoroughly cognizant of all of the considerations involved in their practices. While the Government cannot leave collective bar gaining unregulated, and must sometimes intervene in specific situa tions in order to safeguard the public interest, such intervention should proceed on the cautious premise that outsiders cannot readily judge the necessity of wisdom of the conent of labor agreements, par ticularly in the area of working rules. Where intervention fails to consider the imperative necessities of the situation, as in the case of the unwise and largely ineffective ban on the closed shop imposed by the Labor-Management Relations Act of 1947, such intervention creates more problems that it solves. The C h a ir m a n . Thank you very much. Professor Taylor, we are very happy to have you with us. You have probably as long and distinguished a record in this field as any American, and therefore we will listen to your statement with great interest. STATEMENT OF GEORGE W . TAYLOR, PROFESSOR OF INDUSTRY, WHARTON SCHOOL, UNIVERSITY OF PENNSYLVANIA THE ECONOMICS OE WORKING CONDITIONS Mr. T a y lo r . In business practice, notably since the 1920’s and later on in labor law, it has been recognized that the contract of em ployment has three principal dimensions— wages, hours, and working conditions. They overlap. For example, many a work rule has been established in lieu of so many cents an hour in a general wage increase. Yet each dimension has its own direct influence upon pro duction and costs as well as upon the status and well-being of em ployees. “ The economic effects of working conditions and the per sonal hopes and desires of individual employees are uniquely involved in the one dimension.” Consequently, issues in this area are frequently more difficult to resolve than are disputes over wages and hours, par ticularly as respects work rules over the manning of equipment, for instance, which determine whether or not some employees keep on their regular jobs or become unemployed. Those employees have a personal concern that is not dispelled by a vista of general economic progress. Working conditions include work rules—such as layoff by sen iority— governing the job rights of employees and the manner in which services of employees are directed and utilized—such as the right of appeal from disciplinary action. In addition, there are pro- EMPLOYMENT, GROWTH, AND PRICE LEVELS 2589 visions like pensions, unemployment pay, and medical benefits, which are colloquially termed “ welfare benefits.” They all have an im portant bearing upon that variable—contribution to production which employees make during the hours for which wages are paid. The direct cost of providing a specified set of working conditions, it is commonly recognized, has an offset difficult exactly to measure, be cause of the incentive to more effective production effort which they provide. Beyond these economic considerations of costs and pro duction— difficult enough to measure—lies the growing demand of employees for conditions of work—such as minimum wages and steady work—designed to prevent what is construed as their ex ploitation on behalf of the consumer and in the interest of economic progress. The manner in which labor resources are utilized also involve human considerations in our kind of democracy. Questions about the balancing of such values with the urgent needs of economic progress comes to a head in the determination of working rules. The economics of working rules is a complex and elusive subject. Even rough measurements indicate quite clearly, however, that sig nificant cost and production variations between plants and indus tries are occasioned not only by differences in formally stated working conditions but by the alternate ways in which work rules are admin istered in day-by-day operations. The manner in which real men actually live and work together in their daily endeavors cannot be neatly specified in a set of work rules. This reality is central to per sonnel administration and the application of labor agreements. The import of working rules has received little attention and less emphasis in general economic analysis. Perhaps that is because one cannot easily generalize in this area or treat working rules en masse. The subjective judgments which are involved are also an obstacle. A measure of job security may be derived from certain seniority practices or from an asserted craft jurisdiction. Is it assumed that the morale of employees is thereby increased and consequently their output; or that this is an encouragement to employees to give less than a full productive effort because that can be done with relative impunity ? One’s philosophy about the nature of man has much to do with the answer. We do know from experience that the answer may also vary as between the introduction of a new working condition and the elimination of one that has long been in effect. Even the with drawal of an obsolete working rule can evoke strong emplo3^ee re sistance, particularly w^here this would involve a change in employee job security or the loss of a vested interest. The job of adjusting working rules to keep them from becoming obsolete is among the most arduous tasks of industrial relations managers and union leaders. I know of instances where, despite an assumption of full responsi bility by the union leader, the prospect of a slow and exacting proc ess for getting employee acceptance of change in working rules has been a major factor in a management decision to locate a new plant in a new area. There is another characteristic of working rules which should be mentioned. Their consequences often cannot be even reasonably an ticipated when they are introduced. Work rules are not self-effectu ating. As an illustration, I had occasion, several years ago, to com pare the results of a work rule clause that was identical in the labor 2590 EMPLOYMENT, GROWTH, AND PRICE LEVELS agreements of two companies. The clause provided simply that all available overtime would be shared equally among the employees in each department. A t one company, it worked out that management exercised its own discretion in making the assignments to overtime jobs but relating its appraisal of the fitness of employees to do a par ticular job with the overall requirements of the clause. The over time accounts of employees were then balanced every 6 months. The other company, under strong local union pressure, prepared a weekly list showing each employee’s share in overtime to date. Regardless o f his fitness, the employee lowest on the list was accorded the right o f assignment to the next overtime job. Sometimes two men had to be assigned—the man lowest on the list and another man to instruct him in the proper performance of the job. As between the two companies, there was a difference o f about 20 percent in the labor cost of output during the overtime hours as well as a less measurable difference in quality of output. The real problem lay not particularly wTith the work rule itself as in the lack o f a cooperative labor-management relationship. Even if the clause in question were eliminated, the competitive positions of the two com panies would unquestionably reflect the comparative advantage which the one has in its better relationship with the employees and the union. There is no shortcut approach or gimmick that will overcome all work rule difficulties. And when “ work rules” become a major issue in industrywide bargaining, it is more than likely that some of the companies will have no real problem at all in this area and that, as among the others, the seriousness of the management stake will vary all over the lot. Some historical perspective may be helpful. The working con ditions dimension was not widely appreciated by American manage ment until the early 1920’s. In earlier years, discipline and the threat of it was more generally looked upon as the primary means of insuring a maximum productive effort by the employees. Much approval was given in those days to the remark, an oft-quoted classic, about “person nel policy” in which a foreman epitomized it all by saying “I fire one a week to keep the fear of God in their hearts”. The all-sufficiency of discipline in this regard was appraised as erroneous. During World War I, when the need to “conserve labor” and the importance of employee morale to production became apparent, a discharge would decrease the work force with no replacements in sight and even loaf ing on the job tended to be tolerated. These points of view were reinforced in the postwar years when immigration was reduced and the need increased for employees in the rapidly expanding industries. In manufacturing, average quit rates in the early twenties were so high that the cost of training replacements became a matter of management concern. There was a growing awareness of the short comings of the negative discipline policy and a surge toward more positive ways of eliciting a full productive effort. From the manage ment standpoint, the interest in working conditions—and their rela tive importance vis-a-vis discipline—is, of course, related to the state of the labor market. Primarily with a view to reducing employee turnover and increasing productive morale, the working conditions approach was formalized by management in the 1920’s. EM PLOYMENT, GROWTH, AND PRICE LEVELS 2591 Personnel administration emerged as a distinctive branch of management. Defense against the threat of unionization was un questionably a factor in the use of the personnel movement. Some writers at the time called this movement a “welfare offensive” against unionism. This was doubtless a part of the picture but surely there was much more. I recall the rationale of a personnel manager seek ing Board approval of a 1-week vacation with pay for all employees. This would really not increase labor costs at all. A vacation after a long stretch of steady work would help the employee do a better job in the months ahead. Besides, he would be less likely to quit or to join a union. A widely used slogan of the 1920’s in support of personel administration was: “It Pays.” I suggest that in this rationale lies the probable source of the term “fringe benefit” which in earlier days designated a direct outlay for the benefit of employees wThich, however, would not increase labor costs and might even decrease them. Should criteria for establishing working conditions be based on such a rationale or should a certain code of work rules be approvable in consideration of certain inherent rights of employees? While there are work rules which are too costly in terms of related employee rights which are quite peripheral—such as a paid holiday on one’s birthday—the basic issues seem to be between the economics of work ing conditions and the employee demand for status and security. It was clearly the widespread adoption of collective bargaining which brought new considerations into the employment contract, not only as respects the determination of wages and hours but, most notably in manufacturing, of working conditions as well. Now, the working rules needed by management as conceived through the per sonnel administration route had to be reconciled with the hopes and desires of the employees as represented by the union. (It is not implied that every management viewed personnel administration as an exercise in the arithmetic of labor costs, or were unaware of the human considerations involved. To a large extent, however, com petitive necessities were an inhibiting force.) In bargaining, the unions sought many a working rule as a matter of employee right— for steady employment and job security, for example—and as a furtherance of what was considered essential to “human decency” in a democracy such as ours. The changed approach may be illustrated by a further reference to the working conditions: vacations with pay. It has now become common for employees to accumulate vacation credits week by week. According to a schedule, they receive vacation pay when tenure is terminated even after a relatively short-term service. The idea is to insure the vacation as a need of the individual quite apart from the employee’s contribution to future production of the company paying for the vacation. A notable shift in rationale has occurred. In this connection, one additional observation may be made. As already noted, the emerg ence of the personnel movement in the 19205s carried an element of insurance against unionism. There are doubtless instances under collective bargaining where working conditions are broadly related to union security. It can be argued that the collective-bargaining way of establish ing working conditions extends the original personnel administra 2592 EM PLOYM ENT, GROWTH, AND PRICE LEVELS tion concept to its logical conclusion at least as respects the employee morale factor. I f a union position does reflect the hopes and desires of employees about fair and equitable treatment, they should be more willing than ever to carry out their production responsibilities. In deed, many beneficial results have come from collective bargaining as respects the extension of the personnel idea to “ backward” com panies and industries. In this connection, too, it should not be over looked that in all democratic countries, including our own, there has been a marked acceleration in the expectations of employees as to what constitutes their fair and equitable treatment. On the other hand, there is a widespread concern about the high cost and the limited usefulness of some working conditions developed under collective bargaining during an extended period of business prosperity and conditions of tight labor markets. There is some concern now lest employees have tended to become too immobilized; that is, that the rate of labor turnover is too low in the face of declining factory employment. Moreover, some work rules would seem to single out a few employees for preferred treatment, rather than insure the equitable treatment for all. This is, of course, a mat ter of subjective judgment. Reference is to such devices as the comprehensive bumping provi sions used in some seniority systems by which a layoff of, say, 10 percent of the employees in a department may entail, at great cost m the production of goods, job reassignments and demotions for over half the total work force. It is easy to question whether the results in job security justify the costs to company and employees as well. This is not at all to say that job security is unimportant or that seniority has no valid uses. I f present working conditions—■ such as the bumping—were eliminated altogether, new ones would have to be re-created. Even if the labor market became very loose, a reversion to pre-1920 concepts about the employment relationship is scarcely conceivable. Current arguments about working rules are grounded in differences about the extent to which their development is a function of per sonnel administration— a part of “ management’s right to manage”— and the extent to which this is a function of collective bargaining. The general public has an obvious interest in the arguments—it doesn’t seek goods and services provided by exploited labor, but it doesn’t want to pay for nonessential or absolete working conditions as additions to cost of production. Here is really the nub of the question—how are nonessential or unduly costly working rules to be eliminated and how are the obsolete ones to be modernized ? The need for more efficient production and lower costs is but a part of the case. We come up to the question of how established con ditions are to be given up by employees. Far more is involved than whether or not the decision is to be made by collective bargaining. Proposed changes have to be sold to the employees whose interests and vested rights will be adversely affected—indeed, whose jobs are often at stake. This is touchy territory. There was recently a union-management agreement by which sick leave eligibility for the first day of illness was eliminated because its abuse by some employees resulted in a heavy cost. The savings helped increase the amount of a general wage increase for all employees. EM PLOYMENT, GROWTH, AND PRICE LEVELS 2593 The employee reaction was little short of violent. An outside ob server might have lauded the union for its statesmanship; the em ployees affected charged the union with having entered into a “ sweet heart contract.” It cannot be assumed that the problem of effectuat ing such changes can be solved if the decisions are made by manage ment alone instead of through collective bargaining. Indeed, union cooperation in making such changes has been helpful in many situa tions. The hard fact of the matter is that the current working condition situation is comprised of a myriad of unique, individual problems which have to be dealt with in the particular rather than as a general argument. Success in handling these problems will be dependent upon the ability of decisionmakers to devise ways and means for ameliorating the impact of any changes upon the employees affected. Otherwise, the rate of change will be impeded whether or not col lective bargaining obtains. It’s too much to expect employees to become enthusiastic about changes in working rules, even the obsolete and outmoded ones, which are designed to make their services un necessary. The anticipated problems in this area, with automation developments and the like, are of such concern to employees that they will, in most cases, keenly feel the need for union representation. It seems to me that this union function is highly important and most difficult. Mr. Chairman, it seems to me we have not recognized in this country to the extent that the labor contract is three dimensional. There are wages and hours, and then we have this working condition element which determines the manner in which labor services are utilized. They can be utilized in such a way as to augment production or to impede production. It has always seemed to me that this lack of attention to working rules has a historical base, and that is the role of discipline in our industral area. I think before World War I, and perhaps up until the twenties, it was assumed that you run a factory by paying wages for a certain number of hours, and you exercised discipline in order to see to it that full productive contribution was made. This emphasis on discipline started to weaken when we had to conserve labor during World War I and when discharging some body was not always a good answer to getting more production be cause you could not replace him. We had this during World War II, of course, and in all of these periods of short labor. I think, too, with the decline of immigration we had to begin to conserve labor. Lo and behold, out of that there emerged not from unions but from manage ment the so-called personnel movement. The idea w^as in personnel by introducing certain conditions of employment you would induce people to give a better productive effort instead of threatening them by virtue of discharge. I remember during the twenties when the personnel movement was in its infancy and just developing, the idea was something like this. I f you gave a man a vacation after 51 weeks of work, he would recharge the batteries. He would be a better man for the next 51 weeks of work in the plant. The slogan of the personnel movement was “ It pays.” Incidentally, just as an aside, I think this is how this word “ fringes” came into being. It was a collateral wage payment. You would give the employees some money. It would not increase your cost of pro 2594 EMPLOYMENT, GROWTH, AND PRICE LEVELS duction. I don’t think that yon can just say that this personnel movement which used work rules as an inducement to supplement discipline was accepted unanimously in American industry. Indeed, you still had many linemen who said, “ Good old discipline is the way to do it,” and the personnel movement was continuously battling with line people. I think it is fair to say, too, that part of the introduction of work rules before unions was to gain union assurance in some cases. Man agement was saying as workers are organized, they can seek better job security, better status and more dignity. There was this element o f union insurance in this. The point is that you have to look at some work rules not just as a cost of production but as a means of increasing productivity. It is very difficult to measure. Take this question of seniority that was mentioned earlier. It can go to a point where it is an impediment to production. I was at a plant a few years ago where the oldtimers had instituted the seniority to such an extent that you had to line up by order of seniority to punch out your time cards, the idea being that the fellow who got out of the parking lot first had an advantage, as indeed he did, and this should accrue to the man with seniority. It goes in those directions that are doubtful. Yet I have no question about the fact that a proper use of seniority— proper use of it— can be a real contribution to productivity. There wrere some automobile companies in the twenties that followed this kind of practice, which they now tell you with chagrin, when a model changeover came along, everybody would be laid off and would be hired back after a model changeover as new employees. This was a great insecurity in the lives of these people. I clo not think that contributed to good production, the fear of losing a job or being demoted after the newT model got in operation. This is one reason why in some areas you find an overemphasis on seniority, the fear o f this early lack of protection. I don’t think you can look on work rules as exclusively an impedi ment to production at all. Properly developed and properly utilized as a tool, they can augment production. I think a real shift in approach to work rules came with extensive collective bargaining. It became the rule then that certain of the work rules, manners in which labor services are utilized—that is what we are talking about— and certain social risks that are involved where a matter of right in a community such as ours is involved, so irrespective of cost consequences certain rights were assumed to be due. This is what the unions introduced. For example, the restric tion on management’s right to discharge capriciously wTas introduced. Such breakthroughs in recent years as payment for people not work ing when they get laid off, a supplement to the Federal-State unem ployment system. These became matters of rights. I think these points get to be arguable, where the rights are vital and where they are not. Part of this we are arguing about at the time. I think there are some of them you can say are rights that are very doubtful indeed. I am thinking of the payment of wages to a person for a day off on, his birthday. I think you can question whether this is a right that needs to be established in collective bargaining or elsewhere. Some one was telling me about a labor union in New England, I think, EM PLOYMENT, GROWTH, AND PRICE LEVELS 2595 who had a clause in a labor agreement that each female employee would be entitled to 2 days off a year to get a permanent wave. Only in America, I guess, could we talk about such things. I would not think these were vital rights, but there are some vital rights in here that are at the other extreme. I think the freedom from discharge except for a showing of cause is a vital right. I think none of us at this stage would say whether a person kept his job or not should depend upon unilateral management action. We have now the no cost increase element of working rules. W e liave this subjective question about the rights of employees. Then we come into a third category which is bothering us right now. I would call them peripheral or obsolete working rules. I think the peripheral working rule would be that which I mentioned about a holiday with pay on your birthday. These obsolete ones, however, are the ones that are giving us great pause. They come into being for a purpose and they become a vested right of employees. They involve job security eventually and the employees resist them. The real question is whether we can afford two things: The peripheral working rules, and whether or not we can compete with other coun tries, whether we can compete in our own industries by what is a slow process that is inherent in collective bargaining in changing obsolete working rules. It seems to me this is a fundamental question. I have had occasion to work on this a bit in Pennsylvania. We are concerned with the loss of industries out of Pennsylvania and the lack of our industrial growth. We have had a question where there are obsolete working rules in a plant. Seniority that is no longer working well, or this other question which I think is our biggest oper ating problem, the breakdown of some of these incentive systems, and the deterrence to production that some of these outmoded incentive systems have brought about is one of our major operating problems in American industry. I have been at many of these situations in Pennsylvania where the union leaders understand this well. They would like to avoid these matters. They talk with management about how best you can get rid of these obsolete working rules, but don’t underestimate the ground swell demand of employees for their re tention, especially since a change would result in the layoff of a group o f people. You cannot expect rank and file people to get enthusiastic over modernizing an obsolete work rule by a vista of broad economic progress if they are going to lose their job in the consequence of it. I think the union leaders have a very difficult spot in this particular situation, especially with recent legislation. You emphasize that they are to represent the rank and file of the workers, and very f requently it is difficult to distinguish between labor statesmen and a sweetheart contract in the eyes of the rank and file. I f I might just give you an illustration of this difficulty, I recently sat with Dave Cole and Horowitz with the New York City Board of Transportation where we were to advise with respect to a unit. This union did a wonderful statesmenlike act, I thought. There had been introduced a sickness-benefit program, and before the union ever came in, it was provided that a man w^ould get his sick benefits on the first day of illness. This is a costly kind of business. There is some indication that it was introduced to dissuade the workers from really joining the union a long time ago. Since it got to be 2596 EMPLOYMENT, GROWTH, AND PRICE LEVELS abused. The union figures we can do something statesmanlike here about this thing. They said, “What is the cost of this first day and we will spread it out over all the employees in the next general wage increase?” The men were up in arms. I never saw a reaction such as occurred. We had a hearing with about 5,000 members of that union, and they said this union engaged in a sweetheart contract, and they used the term. I think this aspect of working rules indicates that the change o f them involves not just a dictatorial decision by a union leader or by management. It requires a careful working out of the problems which workers have in consequence of changes in rules. Unless this element is introduced you will not add to production simply by changing the working rules. I f I might just take a moment or two to add one other character istic of this working rule problem, when a labor agreement gets estab lished and has all these formal clauses in it, two things happen, in my experience. The clause itself gets worked out quite differently in each plant. This is really wThere collective bargaining gets individual. In deed, there are many items not included in the formal labor agree ment at all which are worked out by local agreements. A whole series o f local agreements which supplement the formal agreement. I f I might give just one illustration of the first, about how a formal clause in the agreement works out entirely different, this I mention in the statement: I think it is a significant thing. A few years ago I was working with two companies which had an identical clause in their agreement, and it was simply this. Overtime work should be shared equally between all the men in the department. This seems like a simple clause. It should not result in much difficulty in apply ing it. One of the two companies which was on its toes and had good relations with the union said, “What we will do on this, manage ment will assign as it sees fit. At the end of every 6-month period, we will balance accounts and see whether we shared.” It worked very well. The other company did not have good relations with its employees, and they battled how you apply this clause. Out came a local agreement that they would have a list and on this list every man would have a cross before his name the number of overtime hours shared. Whenever an overtime job occurred the man lowest on the list got it. It was the same clause. The difference in the over time work in these two companies was over 20 percent in terms o f how this clause got effectuated. You could not meet this problem by a change in terms of the agreement. It stems right back to the working relationships in the shop. I don’t know anything about what is happening in steel except what you read, but I will venture to say that a lot of these companies don’t have any working rule problem at all. In another company they will have a tough one on seniority and another tough one on incentive. It is a particularized series of problems that take place. Dealing with them in general has some implications which I would like to allude to in just a moment. Working rules are where the individual participates very closely in his collective bargaining, as in this case that I mentioned. Not just the top union leader who does it. Then these local agreements which are all over the lot, sometimes quite different than the terms EM PLOYMENT, GROWTH, AND PRICE LEVELS 2597 o f the national agreement itself. Day by day a shop steward works with a foreman and he says, “ There is a layoff problem, let us work it out this way.” They will either put it down in a written agree ment or a practice will develop. This is not an en masse job. When it is approached en masse, you say here is a working rule problem, it raises a fundamental question. The only way you can make a big change en masse is to say these areas get removed from the scope of collective bargaining. No longer do you bargain about all of these areas. Management now resumes the unilateral right to act with respect to them. A ll the workers suddenly see these local agreements being undermined. A local agreement will affect whether they hold a job or get laid off, or how they get promoted. This is direct, vital, personal to these individuals. I give this background and give this nature of the problem because I think this is the reality of the prob lem of working rules. I think the workers of this country will not give up their right to participate in various ways in the operation of these working rules. This is a vital condition to them. This is why you get many of these labor disputes. The working rule problem is much more vital to the rank and file than whether the wage increase is X or X plus Y. This is the way they live and work together that is involved. Perhaps this is enough of this kind of approach at this point and maybe we can have some discussion. The C h a ir m a n . Thank you very much. Mr. Stieber. STATEMENT OF JACK STIEBER, LABOR AND INDUSTRIAL RELA TIONS CENTER, MICHIGAN STATE UNIVERSITY Mr. S tie b e r . Mr. Chairman, I have a statement here which deals with the nonwage aspects of collective bargaining and their effects on productivity, labor mobility and economic growth. In the interest of conserving time, I am going to omit various portions of the state ment, trying to cover at least the major points in these three areas. The C h a ir m a n . The statement will be printed as a whole. (The statement referred to follow s:) Statem ent on N o n w a g e A s p e c t s o f C o l l e c t i v e B a r g a i n i n g : E f f e c t s o n P ro L a b o r M o b il it y , a n d E c o n o m ic G r o w t h d u c t iv it y , (By Jack Steiber, Labor and Industrial Relations Center, Michigan State University) The subject allocated to this panel is a broad one. Nonwage aspects of collec tive bargaining would seem to include all contractual provisions except those pertaining directly to rates of pay. For purposes of this discussion, I have divided the nonwage provisions found in a typical eolective bargaining agreement into five broad classifications and indicated the kinds of subjects covered under each classification.1 This listing is, of course, designed to be illustrative rather than exhaustive. 1. Structural and procedural: Scope and purpose of the agreement, pro cedures governing the handling of grievances and arbitration, prevention of strikes and lockouts, reopening clauses. 1 With some modification these classifications are taken from Llovd G. Reynolds “ Labor Economics and Labor Relations,” Sd edition. Englewood Cliffs: Prentice-Hall, 1959. 2598 EMPLOYMENT, GROWTH, AND PRICE LEVELS 2. Status and rights of the union and management: Union recognition, rights and responsibilities of the parties, union security, union representa tives’ activity during working hours. 3. Job tenure and job security: Layoffs promotions, transfers, and the application of seniority. 4. Work schedules, work speeds, and production methods: Determination of the standard workday and workweek, overtime pay, holiday pay, vaca tion schedules, machine assignments, assembly line speeds, time standards and work quotas under incentive systems, crew size, work methods, intro duction of new machinery, working conditions governing health, safety, heating and lighting. 5. Off-the-job security: Pensions, health and welfare plans, supplementary unemployment benefits, severance pay, and other clauses which try to ease the workers’ lot upon retirement or loss of work for other reasons. Rather than consider the economic significance of provisions in each of the classifications, I shall address myself directly to the questions which are of major interest to this committee. What are the effects of the nonwage aspects of collective bargaining on productivity? On labor mobility? On economic growth ? IM P A C T ON PR OD U CTIV ITY The contract provisions which have often been considered as having negative effects on efficiency are those dealing with seniority, production methods, work rules, crew size and others which tend to reduce or limit average output per man-hour. The effects of seniority might be considered as indirect while the so-called make-work policies directly affect productivity. American management generally believes that efficiency of operations would be increased if more regard were given to ability and less to seniority in promo tions and layoffs. While, under most contracts, management may promote a junior employee where he has greater ability, supervisors are loath to risk going to arbitration on the issue of seniority versus ability. This reluctance stems from a belief that arbitrators tend to overemphasize the objective factor o f seniority at the expense of the subjective factor of ability; a tendency of arbi trators to put the burden of proof of greater ability on management; and a growing collection of arbitration opinions that a company must be able to prove that a junior man is “ head and shoulders” above the senior man in ability to warrant his promotion, even under a clause which says that seniority will govern in promotions only when ability, merit, and capacity are equal. In layoffs, seniority rule— especially under a system of plantwide seniority—requires man agement to go through a costly bumping policy and wind up, at a time when efficiency of operations is particularly important, retaining men not on the basis of their relative worth to the enterprise but because of their continuous length of service in the plant. Finally, management decries the lack of incen tive for workers to put forth their best efforts in the hope of promotion because of the importance attached to seniority. While not readily measurable, the impact of strict seniority provisions is undoubtedly in the direction of decreasing efficiency in a particular company. However, it is probably not as adverse to efficiency as depicted by management because of the growing importance of machine-paced as opposed to man-paced jobs making ability less important; the likelihood on many jobs of finding a high correlation between length of service and ability; and the increase in efficiency which may result from the greater security provided by seniority. There is also some reason to question the competence of management to assess accurately rel ative ability as between employees to do a given job. This skepticism would appear to be justified by the findings of Prof. James Healy of the Harvard Business School who investigated 58 arbitration awards in which the arbitrator had set aside management’s decision to promote a junior man over a senior employee on the basis of superior ability. Out of 46 replies, the senior employee was found to have proved himself able to do the new job within a short period of time in 29 cases. Furthermore, in 22 of the 29 cases, supervision expressed doubts whether the junior employee originally favored by management would have done any better on the job.2 Even if we acknowledge that individual companies might increase their over all efficiency if not tied down by seniority provisions, this does not mean that 2 James J. Healy, “ The Factor of Ability in Labor Relations,” ch. I l l in Arbitration Today, edited by Jean T. McKelvey. Washington : Bureau of National Affairs, 1955. EM PLOYMENT, GROWTH, AND PRICE LEVELS 2599 the national economy would benefit. This is just another case where the whole does not equal the sum of its parts or to reverse a more modern aphorism, “ What’s good for a particular company is not necessarily good for the United States.” A company by laying off older workers and retaining more productive younger men would be improving its own efficiency and profit position at the expense of the economy generally. For such older workers would probably have difficulty finding alternative employment opportunities and if reemployed would undoubtedly be less productive in a new work environment than they were in their former jobs. Viewed in this context, seniority may be regarded as a method of distributing the less productive employees more or less equitably among employers, rather than concentrating them among the lower paying firms who can probably least afford a further drag on their overall efficiency. Before leaving the subject of seniority it is well to point out that continuous length of service is also important in nonunion companiees. A National In dustrial Conference Board study of 110 nonunion companies in 1950 found that over 95 percent gave consideration to length of service in layoffs and over 70 percent considered seniority in promotions and rehires.3 On the other hand, outside the United States, seniority is given much less weight even in unionized companies. Teaching in Great Britain this spring, I found management and union representatives quite surprised at the important role played by seniority in the United States. Management officials particularly found it difficult to understand how the high productivity which they almost unquestionably accept as characteristic of American industry could go hand in hand with this apparent disregard of merit and ability in deciding which employees to promote or to retain in layoffs. The absence of predetermined rules to follow may serve as a partial explanation for the relative frequency of work stoppages in Great Britain when layoffs do occur. Make-work policies of unions have been classified into nine categories by Slichter in his “ Union Policies and Industrial Management” written 20 years ago. While the emphasis among these categories may have shifted they still represent a comprehensive grouping of ways in which unions may attempt to increase employment (1) limiting daily or weekly output; (2) limiting speed of work; (3) controlling quality; (4) requiring time-consuming methods; (5) requiring unnecessary work; (6) regulating crew size or requiring unnecessary men; (7) requiring that work be done by a particular skilled trade; (8) pro hibiting employers or foremen from doing production work; (9) retarding or prohibiting use of machines or labor saving devices.4 It is unnecessary to discuss in detail the above listed restrictive practices. They all share in common the objective of requiring a greater input of manhours per unit of output than would be required in their absence. In order to place them in proper perspective, I should like to make the following points regarding restrictive practices. 1. Make-work practices are not an invention of trade unions. They are found among unorganized as well as organized wrorkers; in offices as well as in plants; among professionals as well as among lesser educated groups; and even among the executives of the same companies that inveigh loudly against make-work practices in their plants. Who has not heard of Parkinson’s law which leads to empire building even while actual productive work is decreasing? Or what knowledgeable and honest teacher would deny that part (not all, but part) of the opposition of teachers to using TV to help solve the teacher shortage may be explained by the same “lump of labor” theory that professors of economics are so busily disproving in their classrooms?5 Unions may be blamed for formalizing and enforcing restrictive practices but not for inventing them. 2. Economics textbooks tell us that make-work rules and policies are more prevalent among craft unions than among unions organized along industrial lines. Presumably this is so because skilled workers have more to fear from deskilling technological changes and loss of work to other crafts or to lesser skilled workers. But today we find that it is in the industries organized by large industrial unions that the greatest battles are being fought over so-called re strictive practices: steel, meatpacking (in two companies), oil refining, nonferrous metal mining, glass. How account for this sudden interest of employers 3 “ Seniority Practices in Nonunionized Companies,” National Industrial Conference Board, Studies in Personnel Policy, No. 110, 1950. * Sumner H. Slichter, “ Union Policies and Industrial Management,” W ashington: The Brookings Institution, 1941, ch. VI. 5 See, for example, article on “ Teachers and TV,” Wall Street Journal, Sept. 9, 1959. 2600 EM PLOYM ENT, GROWTH, AND PRICE LEVELS in mass production industries in practices which have been prevalent for years in their industries without exciting much active opposition? Employers explain it in terms of increased competition, return of a buyers’ market, and lower profit margins (in 1958). Perhaps this is so, but it is only one side of the equation. It is significant that the drive against make-work rules has come to the fore shortly after the most serious recession of the postwar years. Many of the workers who are now insisting on strict adherence to “ local working conditions,” “ working rules,” less productive methods, etc., were unemployed or on a short workweek in 1958. Many of their friends and co workers are still unemployed despite soaring production indexes and improved profits. They do not have to hark back to the thirties (although many of them can and do) to rationalize actions to “make the job last” as long as possible. 3. Behind make-work rules, restrictive practices and opposition to techno logical change is the fear of unemployment and economic insecurity. Workers and unions may agree that limiting output is bad for the company, for the economy, and even for their own long run employment opportunities. These are things that they have been told and that intellectually make sense. But the individual knows, often from personal experience, that changing technology and high productivity may mean less work for him next week, next month, or next year. Understanding the reasons behind featherbedding and make-work rules should give us a clue as to how to deal with them—in short, like any good doctor we should try to treat the underlying cause of the illness, not the symptom. Recent agreements in meatpacking and in the Pacific-coast longshoring in dustry would seem to have recognized this distinction between symptoms and underlying causes of workers* anxieties over automation. Originally suggested by Armour & Co., contracts have been signed between the two unions in the meatpacking industry, the United Packinghouse Workers and the Amalgamated Meat Cutters and Butcher Workmen, and eight companies setting up “ auto mation funds” to be financed by the companies. In Armour the company will contribute 1 cent for every hundredweight of meat products shipped up to a maximum of $500,000. The fund is to be used to “ cushion whatever unemploy ment may arise through the introduction of automation.” 8 On the west coast the Pacific Maritime Association and the International Longshoremen’s and Warehousemen’s Union have set up a mechanization fund under which dockw orkers will receive a share of the savings from the introduction of labor-saving devices. In return the union has promised not to oppose the introduction of new machinery.7 Other companies and industries interested in reducing union opposition to automation and changing restrictive contractual provisions might well consider this approach rather than attempting to achieve their objectives by persuading unions and workers that make-work policies are “bad” for the economy and “ wrong” as a matter of principle. 4. It is ironic that restrictive work practices should be such an important issue in the United States of all countries. During the postwar years, some 68 British productivity teams, made up of management technicians and operatives from various industries, have visited the United States to learn our secrets of increased output and productivity. Graham Hutton summarized the reports prepared by these teams in a book entitled, “ We Too Can Prosper.” He w rote: “ Team after team noted the interdependence in American industry of the ‘sense of camaraderie and freedom of expression based on mutual respect,’ the readiness to change working assignments and conditions, the willingness of unions to conform to new methods or apply new machinery (with hard bargain ing for due reward), and the all-pervading belief in the need to raise pro ductivity.” 8 On the other hand, J. A. Livingston, a journalist who writes on economic subjects, compares American workers’ attitudes toward productivity unfavorably with those of workers in the Soviet Union on the basis of a report by Prof. Emily Brown of Vassar.9 I understand from other studies of the Soviet system of production that they too have their problems of achieving maximum effort and productivity.10 But if Soviet workers are more amenable to technological 6 Business Week, Sept. 19, 1959, pp. 56-58. 7 Business Week, Sept. 5, 1959, p. 100. 8 Graham Hutton, “ We Too Can Prosper,” New Y ork : The Macmillan Co., 1953, pp. 145—146. 9 The State Journal, Lansing, Mich., Sept. 21, 1959. 10 See, for example, Joseph S. Berliner, “ Factory and Manager in the U.S.S.R., Cam bridge : Harvard University Press, 1957. EM PLOYMENT, GROWTH, AND PRICE LEVELS 2601 changes and more cooperative in efforts to maximize production, I suggest that the relative unemployment rates in the two countries may be part of the expla nation. In this regard it is significant to note that American companies using the union-management cooperation system known as the Scanlon plan, which rules out layoffs resulting from increased productivity, have been highly successful. 5. Finally, so-called make-work rules, featherbedding and other restrictive practices may, on investigation, turn out to have a sound basis. Like so many other issues which seem to be quite clear on the surface, there are usually two sides to the argument over specific make-work practices.11 Slichter’s admonition of 1941 still holds true today : “ It is not always easy to determine when a union is ‘making work.’ There are some clear cases, such as those in which the union requires that the work be done twice. But the mere fact that the union limits the output of men, or controls the quality of the work (with effects upon output), regulates the size of crew or the number of machines per man, or prohibits the use of laborsaving devices does not in itself mean that the union is ‘making work.’ In such cases it is necessary to apply a rule of reason and to determine whether the limits are unreasonable. Opinions as to what is reasonable are bound to differ, but failure to apply a rule of reason would be to accept the employers’ requirements, no matter how harsh and extreme, as the proper standard.” 12 As in the case of seniority, it is probable that certain contractual provisions pertaining to work speeds, methods, and other limitations on production have an adverse effect on productivity. But there are balancing factors which must be considered before closing the books on this subject. On the credit side, we must assess the positive effect on productivity of the increased security given workers' by such contract provisions as seniority, pensions, sub, severance pay, the grievance procedure, and negotiated funds to study and decrease the dele terious effects of automation. Like their negative counterparts, the monetary value of these clauses cannot be estimated, but there is little question that high worker morale, increased security and increased productivity go together.13 Even more important are the pressures that increased production costs exert on management to increase efficiency. In addition to substantial wage increases during the postwar years, costs of other contractual benefits have also risen. The Secretary of Labor in his background report on the steel indus try stated that fringe benefits in manufacturing came to 29 percent of straighttime average hourly earnings. The cost of fringe benefits is probably more directly attributable to unions than wage increases which are more respon sive to labor market pressures. Unions must take much of the responsibility for the increasing cost of fringe benefits and are entitled to some of the credit for measures taken by management to raise productivity in order to meet added costs of production. It is useless to try to assess the net consequences of unions on productivity, faced as wTe are with opposing influences neither of which is measurable. This, however, should not deter efforts to minimize and, if possible, eliminate those union practices and policies which tend to decrease productivity, while at the same time retaining the positive effects of other provisions which operate in the opposite direction. IM P A C T ON LABOR M O B IL IT Y In a free competitive economy we depend on the mobility of capital and labor to achieve the optimum allocation of those resources among producing units. It has been claimed that contractual provisions which tie workers more closely to their jobs and companies decrease labor mobility and thus hinder economic growth. The provisions most often cited in this connection are those dealing with seniority, pensions, supplementary unemployment benefits and others which relate benefits to length of service. 11 See, for example, Morris A. Horowitz, “ The Make-Work Effects of the Railroad’s Constructive Allowance Payments,” Labor Law Journal, May 1955, pp. 331‘-334. 12 Slichter, op. cit., pp. 165-166. 13 I am aware of studies with contrary findings. For a complete list of citations pro and con see Harold L. Wilensky, “ Human Relations in the Workplace : An Appraisal of Some Recent Research,” ch. I ll in “ Research in Industrial Human Relations,” edited by Conrad M. Arensberg and others, New Y ork : Harper & Bros., 1957. 38563— 59— pt. 8------ 11 2602 EM PLOYMENT, GROWTH, AND PRICE LEVELS This committee last March heard from Stanley Lebergott regarding the long term factors limiting labor mobility. They included such revered symbols of our culture and our time as the American home, motherhood, education, national defense, individual security, enlightened personnel policies, and Gov ernment efforts to maintain the family farm. Also mentioned was the end of large-scale immigration. If unions, through collective bargaining, have added to these immobilizing factors, they are in very good company. However, it is not at all clear that the nonwage aspects of collective bargaining have had any significant effects on the mobility of labor. Ten years ago, when negotiated pension plans were proposed and then won by unions in several major industries, they were deplored by management, Govern ment, and academic economists as an undesirable development because they would chain workers to their employers and inhibit labor mobility. Even labor leaders and economists did not seriously dispute this, though they justified their demands on the ground that private plans seemed the only way to increase the miserable benefits then being paid under social security. In the last few years the pendulum seems to have swung in the other direc tion. Recent studies point out that labor turnover or mobility (depending on whether you prefer the “ bad” word or the “ good” one) has always been con centrated among younger workers—a fact which should have been apparent even in 1949. Workers who stand to lose the most from changing jobs—in terms of seniority and accrued pensions—are in higher age brackets "where turnover has always been relatively low. Arthur Ross, after an intensive examination of quit rates over the last 45 years, concluded that “ little evidence can be found for the proposition that labor resources have become immobilized and a new industrial feudalism has been created because men can no longer afford to quit their jobs.” 14 In another study, Robert Tilove found that though private pension plans tend to restrain mobility, other influences such as those mentioned earlier are much more im portant. Tilove even suggests that, in the future, pension plans may contribute to the increase of mobility by freeing workers from physically demanding jobs at an early enough age to move to lighter jobs.15 IM P A C T ON ECONOM IC GRO W TH Economic growth is the result of a combination of productivity gains and increases in the quantities of labor, capital, and other inputs used in produc tion. Productivity increases have been estimated to account for about half of the rise in total real output of the American economy since the turn of the century.16 The nonwage contractual provisions discussed earlier would appear to have had only minor effects on productivity and mobility of labor, the avenues through which they might have had some influence on economic growth. Much more important to economic growth, both in the past and in the future, are such factors as changing technology, training, education, and size of the labor force, the rate of investment, research and development, managerial skill, and hours of work. The last-named factor—hours of work—is the avenue through which unions have probably had the greatest influence on economic growth in the past. It may also be the most important factor through which unions are likely to influence economic growth in the future, though with somewhat different results. Since 1890 we have seen a reduction in the average workweek in manufactur ing from about 60 to 40 hours. Unions have been a major force in achieving this change, both through collective bargaining and legislation. Despite this reduction of one-third in the average workweek, total output has increased substantially. It has been estimated that productivity increases during the last 50 years have been shared in a 60-40 ratio as between income and leisure.17 Today unions tend to favor a further reduction in the workweek. The results of such a reduction, should it occur, may not be as favorable for economic 14 Arthur M. Ross., “ Do We Have an Industrial Feudalism ?” American Economic Review, December 1958, pp. 903-920. 15 Robert Tilove, “ Pension Funds and Economic Freedom,” the Fund for the Republic, New York, pp. 20—28. 16 John W. Kendrick, “ Productivity Trends : Capital and Labor,” Review of Economics and Statistics, August 1956, pp. 248-257. 17 See Clark Kerr, “ The Prospect for Wages and Hours,” ch. 6 in “ U.S. Industrial Rela tions : The Next Twenty Years,” edited by Jack Stieber, East Lansing: Michigan State University Press, 1958. EM PLOYM ENT, G R O W T H , AND PRICE LEVELS 2603 growth as in the past. Historically, the shorter workweek has gone a long way toward paying for itself by speeding np the average tempo of work, and by reducing accidents, wastage, and absenteeism. Snch favorable effects on output per man-hour are less likely to result from future reductions in hours worked per week. While one cannot say with certainty what is the maximum-output week under a given technology—and it will vary among industries and occu pations— evidence suggests that it is probably not less than 40 hours per week in most industries.18 This is not to say that a reduction in the workweek is un desirable. Productivity increases will continue to occur as a result of factors other than the number of hours worked per week. They may even be accel erated by increasing automation. The benefits of these increases in output per man-hour may be taken either in the form of increased real output or increased leisure, or— as in the past—in some compromise between these two. The shorter workweek can be had—but only with some sacrifice of additional income. I f unions and their members are prepared to choose more leisure over more income, the shorter workweek is a logical policy to follow. If, however, the shorter workweek is designed to serve as a remedy for unemployment, there are other ways of achieving this goal (as brought out in other hearings before this committee) which are more consistent with the objective of a growing national economy. Mr. S t i e b e r . The subject allocated to this panel is a broad one. Non wage aspects of collective bargaining would seem to include all contractual provisions except those pertaining directly to rates of pay. For purposes of this discussion, I have divided the nonwage provisions found in a typical collective bargaining agreement into five broad classifications and indicated the kinds of subjects covered under each classification. (W ith some modification these classifications are taken from Lloyd G. Reynolds, “ Labor Economics and Labor Relations” , third edition. Englewood Cliffs: Prentice-Hall, 1959.) This listing is, of course, designed to be illustrative rather than exhaustive. 1. Structural and procedural: Scope and purpose of the agreement, procedures governing the handling of grievances and arbitration, prevention of strikes and lockouts, reopening clauses. 2. Status and rights of the union and management: Union recogni tion, rights and responsibilities of the parties, union security, union representatives’ activity during working hours. 3. Job tenure and job security: Layoffs, promotions, transfers, and the application of seniority. 4. Work schedules, work speeds, and production methods: Determi nation of the standard workday and workweek, overtime pay, holiday pay, vacation schedules, machine assignments, assembly-line speeds, time standards and work quotas under incentive systems, crew size, work methods, introduction of new machinery, working conditions governing health, safety, heating, and lighting. 5. Off-the-job security: Pensions, health and welfare plans, supple mentary unemployment benefits, severance pay, and other clauses which try to ease the workers’ lot upon retirement or loss of work for other reasons. Rather than consider the economic significance of provisions in each o f the classifications, I shall address myself directly to the questions which are of major interest to this committee. What are the effects of the nonwage aspects of collective bargaining on productivity ? On labor mobility ? On economic growth ? 18 P. Sargent Florence, “ The Economics of Fatigue and Rest,” New York : Henry Holt & Co., 1924 ; “ The Health and Efficiency of Munitions Workers.” London, Oxford University Press, 1940 ; BLS, Hours of Work and Output, Bulletin No. 917, 1948. 2604 EM PLOYMENT, GROWTH, AND PRICE LEVELS IM P A C T O N PRO DUCTIVITY The contract provisions which have often been considered as hav ing negative effects on efficiency are those dealing with seniority, production methods, work rules, crew size, and others which tend to reduce or limit average output per man-hour. The effects of seniority might be considered as indirect while the so-called make-work policies directly affect productivity. American management generally believes that efficiency of opera tions would be increased if more regard were given to ability and less to seniority in promotions and layoffs. While, under most con tracts, management may promote a junior employee where he has greater ability, supervisors are loath to risk going to arbitration on the issue of seniority versus ability. This reluctance stems from a belief that arbitrators tend to overemphasize the objective factor of seniority at the expense of the subjective factor of ability; a tendency of arbitrators to put the burden of proof of greater ability on manage ment ; and a growing collection of arbitration opinions that a company must be able to prove that a junior man is “head and shoulders” above the senior man in ability to warrant his promotion, even under a clause which says that seniority will govern in promotions only when ability, merit, and capacity are equal. In layoffs, seniority rule—especially under a system of plantwide seniority—requires management to go through a costly bumping pol icy and wind up, at a time when efficiency of operations is particularly important, retaining men not on the basis of their relative worth to the enterprise but because of their continuous length of service in the plant. Finally, management decries the lack of incentive for workers to put forth their best efforts in the hope of promotion be cause of the importance attached to seniority. While not readily measurable, the impact of strict seniority pro visions is undoubtedly in the direction of decreasing efficiency in a particular company. However, it is probably not as adverse to effi ciency as depicted by management because of the growing impor tance of machine-paced as opposed to man-paced jobs making ability less important; the likelihood on many jobs of finding a high corre lation between length of service and ability; and the increase in effi ciency which may result from the greater security provided by seniority. There is also some reason to question the competence of management to assess accurately relative ability as between employees to do a given job. This skepticism would appear to be justified by the findings of Prof. James Healy, of the Harvard Business School, who investigated 58 arbitration awards in which the arbitrator had set aside management’s decision to promote a junior man over a senior employee on the basis of superior ability. Out of 46 replies, the senior employee was found to have proved himself able to do the new job within a short period of time. Furthermore, in 22 out of the 29 cases, supervision expressed doubts whether the junior employee orig inally favored by management would have done any better on the job. (James J. Healy, “The Factor of Ability in Labor Relations” ch. I l l in Arbitration Today, edited by Jean T. McKelvey. Washington: Bureau of National Affairs, 1955.) Even if we acknowledge that individual companies might increase their overall efficiency if not tied down by seniority provisions, this EM PLOYMENT, GROWTH, AND PRICE LEVELS 2605 does not mean that the national economy would benefit. This is just another case where the whole does not equal the sum of its parts or, to reverse a more modern aphorism, “What’s good for a particular company is not necessarily good for the United States.” A company by laying off older workers and retaining more productive younger men would be improving its own efficiency and profit position at the expense of the economy generally. For such older workers would probably have difficulty finding alternative employment opportuni ties and if reemployed would undoubtedly be less productive in a new work environment than they were in their former jobs. Viewed in this context, seniority may be regarded as a method of distributing the less productive employees more or less equitably among employers, rather than concentrating them among the lower paying firms can probably least afford a further drag on their overall efficiency. Before leaving the subject of seniority it is well to point out that continuous length of service is also important in nonunion com panies. A National Industrial Conference Board study of 110 non union companies in 1950 found that over 95 percent gave considera tion to length of service in layoffs and over 70 percent considered seniority in promotions and rehires (“Seniority Practices in NonUnionized Companies,” National Industrial Conference Board. Studies in Personnel Policy, No. 110,1950). On the other hand, outside the United States, seniority is given much less weight even in unionized companies. Teaching in Great Britain this spring, I found management and union representatives quite surprised at the important role played by seniority in the United States. Management officials particularly found it difficult to under stand how the high productivity which they almost unquestionably accept as characteristic of American industry could go hand in hand with this apparent disregard of merit and ability in deciding which employees to promote or to retain in layoffs. The absence of pre determined rules to follow may serve as a partial explanation for the relative frequency of work stoppages in Great Britain when layoffs do occur. Make-work policies of unions have been classified into nine cate gories by Slichter in his “Union Policies and Industrial Management” written 20 years ago. While the emphasis among these categories may have shifted, they still represent a comprehensive grouping of ways in which unions may attempt to increase employment: (1) Limiting daily or weekly output; (2) Limiting speed of work; (3) Controlling quality; (4) Requiring time-consuming methods; (5) Requiring unnecessary work; (6) Regulating crew size or requiring unnecessary men; (7) Requiring that work be done by a particular skilled trade; (8) Prohibiting employers or foremen from doing produc tion work; (9) Regarding or prohibiting use of machines or labor-saving devices (Sumner H. Slichter, “Union Policies and Industrial Management,” Washington: The Brookings Institution, 1941; ch. V I). who 2606 EMPLOYMENT, GROWTH, AND PRICE LEVELS It is unnecessary to discuss in detail the above listed restrictive practices. They all share in common the objective of requiring a greater input of man-hours per unit of output than would be required in their absence. In order to place them in proper perspective, I should like to make the following points regarding restrictive practices. 1. Make-work practices are not an invention of trade unions. They are found among unorganized as well as organized workers; in offices as well as in plants; among professionals as well as among lesser educated groups; and even among the executives of the same com panies that inveigh loudly against make-work practices in their plants. Who has not heard of Parkinson’s law which leads to empire building even while actual productive work is decreasing? Or what knowledgeable and honest teacher would deny that part—not all, but part— of the opposition of teachers to using T V to help solve the teacher shortage may be explained by the same “ lump of labor” theory that professors of economics are so busily disproving in their classrooms. (See, for example, article on “ Teachers and T V ,” Wall Street Journal, Sept. 9, 1959.) Unions may be blamed for formaliz ing and enforcing restrictive practices but not for inventing them. 2. Economics textbooks tell us make-work rules and policies are more prevalent among craft unions than among unions organized along industrial lines. Presumably this is so because skilled workers have more to fear from deskilling technological changes and loss of work to other crafts or to lesser skilled workers. But today we find that it is in the industries organized by large industrial unions that the greatest battles are being fought over so-called restrictive prac tices : steel, meatpacking—in two companies— oil refining, nonferrous metal mining, glass. How account for this sudden interest of em ployers in mass production industries in practices which have been prevalent for years in their industries without exciting much active opposition ? Employers explain it in terms of increased competition, return of buyers’ market, and lower profit margins, in 1958. Perhaps this is so, but it is only one side of the equation. It is significant that the drive against make-work rules has come to the fore shortly after the most serious recession of the postwar years. Many of the workers who are now insisting on strict adherence to “ local working condi tions,” “ working rules,” less productive methods, and so forth, were unemployed or on a short workweek in 1958. Many of their friends and coworkers are still unemployed despite soaring production in dexes and improved profits. They do not have to hark back to the thirties— although many of them can and do—to rationalize actions to “ make the job last” as long as possible. 3. Behind make-work rules, restrictive practices, and opposition to technological change is the fear of unemployment and economic insecurity. Workers and unions may agree that limiting output is bad for the company, for the economy, and even for their own longrun employment opportunities. These are things that they have been told and that intellectually make sense. But the individual knows, often from personal experience, that changing technology, and high productivity may mean less work for him next week, next month, or next year. Understanding the reasons behind featherbedding and EM PLOYMENT, GROWTH, AND PRICE LEVELS 2607 make-work rules should give us a clue as to how to deal with them in short, like any good doctor wre should try to treat the underlying cause of the illness, not the symptom. Recent agreements in meatpacking and in the Pacific coast longshoring industry would seem to have recognized this distinction be tween symptoms and underlying causes of workers’ anxieties over automation. Originally suggested by Armour & Co., contracts have been signed between the two unions in the meatpacking industry, the United Packinghouse Workers and the Amalgamated Meat Cutters and But cher Workmen, and eight companies setting up “ automation funds’* to be financed by the companies. In Armour the company will con tribute 1 cent for every hundredweight of meat products shipped up to a maximum of $500,000. The fund is to be used to “ cushion what ever unemployment may arise through the introduction of automa tion” (Business Week, Sept. 19, 1959, pp. 56-58). On the west coast the Pacific Maritime Association and the Inter national Longshoremen’s and Warehousemen’s Union have set up a mechanization fund under which dockworkers will receive a share of the savings from the introduction of laborsaving devices. In re turn the union has promised not to oppose the introduction of new machinery (Business Week, Sept. 5, 1959, p. 100). Other companies and industries interested in reducing union oppo sition to automation and changing restrictive contractual provisions might well consider this approach rather than attempting to achieve their objectives by persuading unions and workers that make-work policies are bad for the economy and wrong as a matter of principle. 4. It is ironic that restrictive work practices should be such an important issue in the United States of all countries. During the postwar years, some 66 British productivity teams, made up of man agement technicians and operatives from various industries, have visited the United States to learn our secrets of increased output and productivity. Graham Hutton summarized the reports prepared by these teams in a book entitled “ We Too Can Prosper.” He wrote: Team after team noted the interdependence in American industry of the “ sense of camaraderie and freedom of expression based on mutual respects,” the readiness to change working assignments and conditions, the willingness of unions to conform to new methods or apply new machinery (with hard bar gaining for due reward), and the all-pervading belief in the need to raise productivity (Graham Hutton, “ We Too Can Prosper,” New York: The Mac millan Co., 1953, pp. 145-146). On the other hand, J. A. Livingston, a journalist who writes on economic subjects, compares American workers’ attitudes toward pro ductivity unfavorably with those of workers in the Soviet Union on the basis of a report by Prof. Emily Brown of Vassar (the State Journal, Lansing, Mich., Sept. 21,1959). I understand from other studies of the Soviet system of production that they too have their problems of achieving maximum effort and productivity. (See, for example, Joseph S. Berliner, “ Factory and Manager in the U.S.S.R,” Cambridge: Harvard University Press, 1957.) But if Soviet workers are more accepting of technological changes and more cooperative in efforts to maximize production, I suggest that the relative unemployment rates in the two countries may be part of the explanation. In this regard, it is significant to note 2608 EM PLOYM ENT, GROWTH, AND PRICE LEVELS that companies using the union-management cooperation system known as the Scanlon plan, which rules out layoffs resulting from increased productivity, have been highly successful. 5. Finally, so-called make-work rules, featherbedding, and other restrictive practices may, on investigation, turn out to have a sound basis. Like so many other issues which seem to be quite clear on the surface, there are usually two sides to the argument over specific makework practices. (See, for example, Morris A. Morowitz, “ The MakeW ork Effects on the Railroad’s Constructive Allowance Payments,” Labor Law Journal, May 1955, pp. 331-334.) Slichter’s admonition of 1941 still holds true today: It is not always easy to determine when a union is “ making work.” There are some clear cases, such as those in which the union requires that the work be done twice. But the mere fact that the union limits the output of men, or controls the quality of the work (with effects upon output), regulates the size of the crew or the number of machines per man, or prohibits the use of laborsaving devices does not in itself mean that the union is “ making work.” In such cases it is necessary to apply a rule of reason and to determine whether the limits are unreasonable. Opinions as to what is reasonable are bound to differ, but failure to apply a rule of reason would be to accept the employers’ requirements, no matter how harsh and extreme, as the proper standard ( Slichter, op. cit., pp. 165-166). As in the case of seniority, it is probable that certain contractual provisions pertaining to work speeds, methods, and other limitations on production have an adverse effect on productivity. But there are balancing factors which must be considered before closing the books on this subject. On the credit side, we must assess the positive effect on productivity of the increased security given workers by such con tract provisions as seniority, pensions, SUB [supplementary unem ployment benefits], severance pay, the grievance procedure, and nego tiated funds to study and decrease the deleterious effects of automa tion. Like their negative counterparts, the monetary value of these clauses cannot be estimated, but there is little question that high worker morale, increased security and increased productivity go to gether. (I am aware of studies with contrary findings. For a com plete list of citations pro and con see Harold L. Wilensky, “ Human Relations in the W orkplace: An Appraisal of Some Recent Research” ; ch. I l l in “ Research in Industrial Human Relations,” edited by Con rad M. Arensberg and others, New York: Harper & Bros., 1957.) Even more important are the pressures that increased production costs exert on management to increase efficiency. In addition to sub stantial wage increases during the postwar years, costs of other con tractual benefits have also risen. The Secretary of Labor in his back ground report on the steel industry stated that fringe benefits in manufacturing came to 29 percent of straight time average hourly earnings. The cost of fringe benefits is probably more directly at tributable to unions than wage increases wdiich are more responsive to labor market pressures. Unions must take much of the responsi bility for the increasing cost of fringe benefits and are entitled to some of the credit for measures taken by management to raise pro ductivity in order to meet added costs of production. It is useless to try to assess the net consequences of unions on pro ductivity, faced as we are with opposing influences both of which are not measurable. This, however, should not deter efforts to mini mize and, if possible, eliminate those union practices and policies EMPLOYMENT, GROWTH, AND PRICE LEVELS 2609 which tend to decrease productivity, while at the same time retaining the positive effects of other provisions which operate in the opposite direction. IM PA C T O N LABOR M O B IL ITY In a free competitive economy we depend on the mobility of capital and labor to achieve the optimum allocation of those resources among producing units. It has been claimed that contractual provisions which tie workers more closely to their jobs and companies decrease labor mobility and thus hinder economic growth. The provisions most often cited in this connection are those dealing with seniority, pensions, supplementary unemployment benefits and others which relate benefits to length of service. This committee last March heard from Stanley Lebergott regard ing the long-term factors limiting labor mobility. They included such revered symbols of our culture and our time as the American home, motherhood, education, national defense, individual security, en lightened personnel policies, and Government efforts to maintain the family farm. Also mentioned was the end of large-scale immigra tion. If unions, through collective bargaining, have added to these immobilizing factors, they are in very good company. However, it is not at all clear that the nonwage aspects of collective bargaining have had any significant effects on the mobility of labor. Ten years ago when negotiated pension plans were proposed and then won by unions in several major industries, they were deplored by management, Government, and academic economists as an unde sirable development because they would chain workers to their em ployers and inhibit labor mobility. Even labor leaders and econo mists did not seriously dispute this, though they justified their de mands on the ground that private plans seemed the only way to increase the miserable benefits then being paid under social security. In the last few years the pendulum seems to have swung in the other direction. Recent studies point out that labor turnover or mo bility— depending on whether you prefer the “bad” word or the “good” one—has always been concentrated among younger workers, a fact which should have been apparent even in 1949. Workers who stand to lose the most from changing jobs—in terms of seniority and accrued pensions—are in higher age brackets where turnover has al ways been relatively low. Arthur Ross, after an intensive examination of quit rates over the last 45 years, concluded that— little evidence can be found for the proposition that labor resources have be come immobilized and a new industrial feudalism has been created because men can no longer afford to quit their jobs (Arthur M. Ross, “ Do We Have an Industrial Feudalism?” American Economic Review, December 19*58, pp. 903920). In another study, Robert Tilove found that, though private pension plans tend to restrain mobility, other influences such as those men tioned earlier are much more important, Tilove even suggested that, in the future, pension plans may contribute to the increase of mobility by freeing workers from physically demanding jobs at an early enough age to move to lighter jobs. (Robert Tilove, “Pension Funds and Freedom,” the Fund for the Republic, New York, pp. 20-28.) 2610 EMPLOYMENT, GROWTH, AND PRICE LEVELS IM P A C T O N ECO NO M IC GROW TH Economic growth is the result of a combination of productivity gains and increases in the quantities of labor, capital, and other imputs used in production. Productivity increases have been estimated to account for about half of the rise in total real output of the Ameri can economy since the turn of the century. (John W. Kendrick, “ Productivity Trends: Capital and Labor,” Eeview of Economics and Statistics, August 1956, pp. 248-257.) The nonwage contractual provisions discussed earlier would appear to have had only minor effects on productivity and mobility of labor, the avenues through which they might have some influence on economic growth. Much more important to economic growth, both in the past and in the future, are such factors as changing tech nology, training, education, and size of the labor force, and the rate o f investment, research and development, managerial skill, and hours o f work. The last-named factor—hours of work—is the avenue through which unions have probably had the greatest influence on economic growth in the past. It may also be the most important factor through which unions are likely to influence economic growth in the future, though with somewhat different results. Since 1890 we have seen a reduction in the average workweek in manufacturing from about 60 to 40 hours. Unions have been a major force in achieving this change, both through collective bargain ing and legislation. Despite this reduction of one-third in the aver age workweek, total output has increased substantially. It has been estimated that productivity increases during the last 50 years have been shared in a 60-40 ratio as between income and leisure. (See Clark Kerr, “ The Prospect for Wages and Hours,” ch. 6 in “ U.S. In dustrial Relations: The Next Twenty Years,” edited by Jack Stieber, East Lansing, Michigan State University Press, 1958.) Today unions tend to favor a further reduction in the workweek. The results o f such a reduction, should it occur, may not be as favor able for economic growth as in the past. Historically, the shorter workweek has gone a long way toward paying for itself by speeding up the average tempo of work, and by reducing accidents, wastage, and absenteeism. Such favorable effects on output per man-hour are less likely to result from future reductions in hours worked per week. While one cannot say with certainty what is the maximumoutput week under a given technology—and it will vary among in dustries and occupations—evidence suggests that it is probably not less than 40 hours per week in most industries. (P. Sargent Florence, “ The Economics of Fatigue and Rest,” New York, Henry Holt & Co., 1924; “ The Health and Efficiency of Munitions Workers,” Lon don, Oxford University Press, 1940; BLS, “ Hours of Work and Out p u t” Bulletin No. 917, 1948.) This is not to say that a reduction in the wrorkweek is undesirable. Productivity increases will continue to occur as a result of factors other than the number of hours worked per week. They may even be accelerated by increasing automation. The benefits of these in creases in output per man-hour may be taken either in the form of increased real output or increased leisure, or, as in the past, in some EM PLOYMENT, GROWTH, AND PRICE LEVELS 2611 compromise between these two. The shorter workweek can be had, but only with some sacrifice of additional income. I f unions and their members are prepared to choose more leisure over more income, the shorter workweek is a logical policy to follow. If, however, the shorter workweek is designed to serve as a remedy for unemployment, there are other ways of achieving this goal, as brought out in other hearings before this committee, which are more consistent with the objective of a growing national economy. In this connection I would also mention the chairman of this com mittee’s bill to do something about depressed areas. I think this would be another way of trying to assist employment in those areas that have chronic unemployment. The C h a ir m a n . Thank you. I will add my personal thanks for your last sentence. Congressmen C u r t is . Representative C u r tis . I was interested in your statement on the British observers’ comments on our seniority rule. I have always found that when you have a seniority rule, it is not because anyone thinks it is particularly good, but it is the fact that when you try to put something else into its place, you run into even greater problems. You have said that they were surprised that they did not use a system based more on ability. I wonder what the European or the British system is. They profess to pay attention to ability, but I suspect if they don’t have the productivity that we have, and apparently they do not, probably their system does know a great deal of nepotism among other things. I don’t know. Have you examined that and found out what they mean when they say that they would have a system that has more relation to ability ? Mr. S tie b e r . I did not make any intensive examination of either any industry or a large number of plants. These were classes which were attended by management people and also foremen in industry in some cases, and union people. I do not want to give the impression that seniority is not at all important in Great Britain or other coun tries. Actually, it varies with industries in Great Britain. For example, in the steel industry in Great Britain they have a rather strict observance of seniority worked out between the union and management. However, I think it is fair to say that by and large seniorit}^ is not as important as it is here, Foremen, for example, were aghast at the idea that because o f seniority considerations, they would not be able to select the best man for promotion. Representative C u r t is . What standard would they apply ? Mr. S tie b e r . According to them they would apply the standard of ability in most cases. However, there are other considerations which would come into play in negotiating with the union. For example, when a layoff was going to occur, through joint consultation, which is more common in British industry than in American industry, and is required by law in Government-owned industries, they would sit down with the union and decide on the order of layoffs. They would consider seniority as one factor, but in other cases they might con sider, for example, a man’s family responsibilities. They might con sider other unique characteristics of that particular plant, and em ployers would stress ability. I had the impression from talking with employers that they felt that very often their arguments on the ground of ability were accepted by 2612 EM PLOYMENT, GROWTH, AND PRICE LEVELS the unions. As to the results of this, I think that productivity has increased in Great Britain. There are other problems that they have with regard to obsolescent machinery, being an older country, and their industries being older in some cases, that we don’t have, But they do tend to regard the United States as being the acme of achieve ment in this area of productivity. Representative C u r t is . There is one other thing. You quote Pro fessor Slichter’s rule of reason, as far as determining this make-work thing. That is fine. But reason for what? What reason are you talking about, or is he talking about, when he says that it is not a make-worth thing ? Mr. S t ie b e r . Let me give you an example of what I think he is referring to. For example, let us say that there are 12 men employed on a particular series of related operations. There is a crew of 12, The company may feel that through some reorganization o f that particular process this crew could be reduced to 10 men and the job could be done just as efficiently. I do not think that we, sitting here, or anybody who does not know that particular operation extremely well, and even people wiio do, mind you, could make the judgment as to whether it is a 12-man job or a 10-man job. Reasonable men might differ on this. Obviously if 10 men are doing the job, they are going to have to put out somewiiat more effort than 12. However, it may very well be that even in putting out this additional effort they would not be overexerting themselves, and they could very well do the job. However, this is not the kind of thing on which you can accept one side’s opinion. I would no more accept the union’s opinion that this is a 12 man job and only a 12 man job than I would accept the company’s opinion that this job should be done by 10 people. This is what I think Professor Slichter means, that we have to apply a rule o f reason. Take the speed of an assembly line, to give you another illustration. What is the appropriate speed ? Representative C u r t is . What I think is this. He has built him self a straw man, or you have in quoting him, that the rule of reason does not prevail in the beginning. I thought that is the basis on which any decision would be made. O f course, people differ as to where you might end. That is what management would be looking for and labor might be arguing the other way. What other rule would have been applied ? Mr. S tie b e r . What was the rule of reason 10 years ago may not be a rule o f reason today. I think this is the short answer. In other words, when the working practice—and this, I think, is one of the issues in dispute in the steel industry— a working practice that was established 15 or 20 years ago, based on conditions existing at that time, may not be an efficient working practice today. It may very well be that some changes in that practice would be reasonable. You ask what other rule may be applied? The problem is that in some industries there are contractual provisions which restrict the parties in departing from a rule that was established some time ago, except by mutual agreement. Representative C u r tis . I have one other question. Mr. Kahn, when you were referring to railroads and their situation, I began speculating as to whether we don’t have a different situation in labor- EMPLOYMENT, GROWTH, AND PRICE LEVELS 2613 management relations when we get to these utilities or regulated in dustries that is quite a bit different from businesses that are in a competitive area. Taking railroads as an example, the fact that today there is real competition in the transportation field which per haps did not exist in 1890 or 1900 when we began developing these regulations for labor-management relations. Possibly today with the competitive picture, the need for this much Government surveillance in the field of labor-management does not carry as much reason. I am pointing my question to this. In the utility, regulated industry don’t we have a different labor-management situation in your judgment? Mr. K a h n . We certainly have the basic problem, Mr. Curtis, that, the cost of the stoppage to the community is charactistically much greater in relation to the costs borne by the parties. Representative C u r t is . I was not thinking so much from that angle. That is obvious from the social standpoint. I was thinking really from the angle of collective bargaining. The fact that if you have competition there are a number of economic checks that are imposed both on labor, which is interested in jobs and management, which is interested in survival, which you don’t get in a utility. Mr. K a h n . I agree with your general proposition, sir, completely. The problem is that much railroad service is still essential, even though it is much more highly subject to competition now than it used to be. I think this is the reason why we have this awkward and difficult problem. These alleged make-work rules, and let us just call them working practices and customs, developed in an era in which railroad had much more monopoly of transportation than they do today, and in many cases they developed out of efforts to prevent management from discriminating against union members or against workers generally for reasons of which workers would not approve, in an era when unionism as such had no legal protection and no form of union security device. Now the environment has changed. The problem, as I mentioned in my paper, is that because in a sense the right to strike is severely inhibited—we are just not going to stand for a major railroad stoppage of any duration— neither party has the kind of pressure to live up to the responsibility of working out a satisfactory solution to these problems that exists in a situation in which a work stoppage can be permitted to continue for a long period of time, and the parties them selves suffer more than the community, and therefore feel they better get around to solve this problem, or the consequences will be rough for all concerned in the industry. Let me give you just one illustration. There was a very long strike at the Chrysler Corp., I believe in February and March of 1950. I heard one observer shortly after the strike was over say that there was nothing in the settlement which a competent arbitrator could not have determined and handed down in the event that the parties had agreed to arbitrate this dispute, and all this economic waste and this bitter dispute could have been avoided. But the answer, of course, is that an award handed down by an arbitrator, even one based on omnis cience and remarkable insight and wisdom and identical to what they would have agreed on themselves would not have been satisfactory to the parties because it would not have been worked out by themselves 2614 EMPLOYMENT, GROWTH, AND PRICE LEVELS as the only kind of solution they were willing to live with on a mutual basis. This is the kind of economic test in which we cannot permit the railroads and the railroad brotherhoods to engage. Representative C u r t is . Thank you, Mr. Chairman. The C h a ir m a n . Congressman Reuss. Representative R eu ss. Mr. Chairman, my question to the panel is suggested by certain tentative conclusions which seem to emerge from what you three gentlemen have been telling us. Work rules are a somewhat difficult thing to pin down. Neither management nor la bor’s absolute view of a particular work rule may be right, you sug gest. Work rules have something to do with increasing or de creasing productivity, and hence are related to the problem of national growth. Finally, Mr. Stieber, I think it was, pointed out some of the British experience with productivity teams that were reported on by G rah am Hutton. As I recall, the Queen Mary treatment was given to teams of management, labor, and sometimes government people who came over here and saw what was going 011 in our plants, which suggests that those who set up those teams thought it was humanly possible to make value judgments as to which work rules were by and large such as tended toward increased productivity and which were not. My question is, Do any members of the panel think that it would be an idea worth considering for teams of Government, labor, and management in this country to take a look at some specific work rules in specific industries and to try to come to tentative conclusions as to whether those work rules are such as tend to increase produc tivity or tend in the other direction, bearing in mind about things that have been said about leaving the collective bargaining process the actual decision in a particular employer-employee relationship as to what work rules to adopt. Does anybody care to comment on that ? Mr. T a y lo r . May I comment on that ? Representative R e u ss. Yes. Mr. T a y lo r . I would think that we could do well if we had some benchmarks developed out of situations where people have done good collective bargaining so that people would be more generally aware of what would be good practice in a particular area. I f I might allude back to the seniority discussion a bit, I am disturbed about our inability to analyze this work rule problem in reality. There are some unions and some industry that would not touch seniority. They don’t want it. The building trades, for obvious reasons. They are market oriented. Unless you want to call jurisdictional ques tions a work rule. To them seniority is reflected in their jurisdic tional rules, that is, maximizing job employment. In the men’s clothing industry, women’s clothing industry, they get security by division of work. Mr. Chairman, you know in the men’s clothing industry if you tried to say people’s status depends on their seniority, they would throw you out. This is not adapted to their problem. They divide work. It has been my observation, with a few illustrations, seniority is used in the mass production industry for those employees called semiskilled, where it is very difficult to discern differences in ability EM PLOYMENT, GROWTH, AND PRICE LEVELS 2615 between these people. I think seniority has a great usefulness in the mass production industry, semiskilled, in two ways. It is an orderly way by which you apportion jobs that anybody can do, or most people can do reasonably well with a minimum of experience on the job. Indeed, I find nonunion companies who tell their fore men to use ability. They can’t do it. They will get in bad with the people they supervise if they try to discern differences in ability. I would think you would not use it in the mass production industry, the tool and die, the highly skilled people, where you can make measurable differences in ability. I think this leads up to this thing. I f we would say here is a seniority practice, it has usefulness to pro duction if used in these ways and not in others, and if there could be some illustrations such as the National Planning Association did in its series of causes of industrial peace, if you have a series of type studies and say how seniority has been used in particular companies to help productivity, and give some horrible examples about how it is used to interfere. I f I might give an illustration in the last, I get disturbed when some of these bumping rules are very high on seniority. Depart ment A has to lay off 10 percent of its people. People bump all down the line and 80 percent of the people get reassigned. In other words, you demote 80 percent of the people in order to preserve the job rights of 10. I think this gets to be a very questionable kind of use of seniority, and also for the well-being of the people. I don’t see how you can do this en masse. There are a whole series of these particular problems. I f we could learn the uses of a tool like seniority and its abuses, and put some benchmarks up for the bargainers, I think the idea would have greater merit. Representative R e u ss. Perhaps you can tell me, I am not aware that anything very like what we are talking about is going on. It certainly is not being done under any public authority. I don’t know that Brookings or NPA or the National Industrial Conference Board or anybody is doing this, although NPA did a very valuable though limited series on labor peace. I think they had four or five different industries. Mr. T a y lo r . I think you are right. I got very much interested in this myself at the University of Pennsylvania. I don’t think any body has ever tackled this question of working conditions as to the economics, where it is useful and where it is not useful. As I indi cated, this is because of this discipline. You get a lot of people who don’t want to develop what you call industrial jurisprudence and the economics of industrial jurisprudence. Things get tough and you approach it in one of two ways. Let us reinstitute good disci plinary action and things will be OK. Senator Curtis’ point about the rule of reason, I think one of the big battles that is going on today is whose rule of reason should apply. Whether it is manage ment’s rule of reason, or a rule of reason, not the union’s but one that is worked out by collective bargaining. I think we are committed to developing rule of reason by collective bargaining in this country, which does not presuppose either the union’s or management’s idea on this point should prevail. 2616 EM PLOYM ENT, GROW TH, AND PRICE LEVELS Representative R e u s s . When you talk about benchmarks, it seems to me what you are saying is that collective bargaining participants ought to have some guideline as to what is in the public interest. We do assume that the collective bargainers will take the public issue into account. Mr. T a y lo r . I think it is right. Representative R e u ss. It wrould be well if they knew what the public interest was. Mr. T a y lo r . We assume in collective bargaining the union will represent the workers’ interest and management, management’s. They will reconcile it by talking this out. We have had failures in collective bargaining. It has not always worked right. We are striv ing to make it a better process. I think people would be guided in their bargaining if you could say here is what is accepted good prac tice in this area. I think it would be an instrument of persuasion that would have great value. Don’t forget these union leaders have a tough job when they go to the rank and file. Every instrument o f persuasion you can give them would be helpful, as well as the sort o f thing that was mentioned earlier about the Armour approach, Indeed, in addition to these more recent ones, this coal situation in bituminous coal had this approach very early. There you had great technological advance in our coal industry. I should not praise that, but the War Labor Board had great battles with Mr. Lewis, and we usually lost them. But when you look at what they have done in that industry by way of ameliorating this technological change upon the people affected as necessary to get these people to go along, you get a guide even before Armour and some of the others. I think the road is clear. I think what we are talking about is whether you can get benchmarks which would be an instrument o f per suasion to the people. I think it would be very helpful in that respect. Mr. K a h n . I think, too, sir, if I may follow up on this point, that this is an area in which employers still have a great deal to learn. I am referring to what might be called the problem o f the administra tion o f labor relations. Employers have learned a great deal about personnel administration. But time after time in situations that I encounter, where there are specific problems in adjustment, they occur because a company overlooked the problem as long as it had no ecomonic pressure to do something about it. As a result o f overlooking the problem, the situation became one to which the men became accus tomed. What you are used to doing is normally considered the right way to do something. May I give just one illustration o f that: A large brewery in Detroit installed some 10 years ago a pair o f rather formidable machines called palletizers. They are about 11 feet high and 25 feet long and 8 feet wide, and they take cases o f beer off moving assembly lines and very neatly arrange them in patterned layers on wooden pallets and deposit them someplace else for fork lift trucks to take away. When these machines were introduced 45 men were displaced from their jobs. The company at the outset said: “ We are not quite sure how these machines will operate, but we will be conservative and we will use four men to keep an eye on them.” As is so typical under automation, if these machines do their work properly no labor is involved. It is a question of people being around in case something goes wrong. EM PLOYMENT, GROWTH, AND PRICE LEVELS 2617 The company started off with four men and never sought to reduce that number for about 9 years. Then one day, 9 years later, virtually no changes having been made in this pair of machines, the company decided that one or two men would be enough. I suggest to you this decision should have been made about 1 or 2 months after the ma chines started operating, and that at the outset the company should have said to the union and to the four men involved: “ We are tem porarily operating with four men. We are going to have an evalua tion of this in 30 days and we will then decide on an appropriate complement,” But they waited 9 years. Naturally there was resist ance. There is a sequel to this story. This case went to arbitration. The arbitrator ruled that three men were the right number. But he also said that if the company gave a little more study to the problem, put in a few gadgets and rearranged work locations it could undoubtedly reduce the number of men to two within a short period of time. A year later I ran into a representative of this company (this was 2 weeks ago) and I said, “ What has happened ? Have you got the number of men down to two yet?” He said, “ No, as a matter of fact, it is still going along with three, just as the arbitration award indicated.” I asked him why and he said, “Well, we decided not to make a fuss about it because in a little while we are introducing a new palletizer machine, just one, with a larger capacity than the pair of old machines, which we are selling to Mexico, and which will obviously require no more than one man to operate.” The moral of the story, you see, has two parts: the problem of being lax about administration in these areas until something unwise be comes customary, and, on the other hand, the fact that eventually in our rather progressive society new developments often occur which make some of the old problems obsolete as problems. The C h a ir m a n . Senator Bush. Senator B u s h . Mr. Chairman, regarding this matter of seniority, I suggest to you and my colleagues in the House that members of the Senate and even members of the House approach this subject rather gingerly. The C h a ir m a n . I thought I did. I happen to be an opponent of the seniority rule. But Speaker Rayburn says the longer one stays around here, the better he likes seniority. Senator B u s h . I was going to comment that I came here some years ago with the determination that we ought to do something about seniority. This was a very bad thing. I have seen it settle so many problems in the Senate that I am a willing victim o f the argument of Mr. Taylor and others regarding the uses and the convenience of it in connection with labor-management matters. I thought Mr. Taylor made a very interesting and convincing argument for it, which surprised me, The C h a ir m a n . I can well understand the pleasure which the bipartisan coalition has in the seniority rule. Mr. T a y lo r . Sometimes you seek the least worst approach to a problem in industry. Representative C u r t is . Y o u mean the coalition between the south ern and northern Democrats ? 38563— 59— pt. 8------ <12 2618 EM PLOYMENT, GROW TH, AND PRICE LEVELS The C h a ir m a n . N o , between the conservative Republicans and southern Democrats. Senator B u s h . Anyway, I think you will agree we better approach this gingerly, all of us. You gave, Mr. Taylor, pretty good illustra tions of where the seniority rule works very well. Would you give an illustration where it would not work so well ? Mr. T a y lo r . The one I mentioned. For myself and my own standards, I think it is a doubtful use of seniority when in order to reserve job rights of a relatively few people on layoff, and you ump and really demote 80 percent of the work force. I think this is giving undue preference to the man with the whiskers as they say in some of these industrial establishments. I think the cost to the people involved might be too great on this business. I have seen this sort of thing, too, which I think is an abuse in industry, that is. Seniority has different uses. One for promotions and one for layoffs for a temporary period of unemployment. These are quite different operations actually. In some cases if you have what we call plantwide seniority, and the company is required to pull men back in order o f seniority, they might not be able to do the work in the sequence o f which you are starting up the plant. This does not make much sense. This is probably exaggerated, but there was one company that told me that according to their seniority they had to pull back a high seniority man for a job that was open and it was a clerical job, and the fellow could not write. So the answer of the union was put him on the job and train him to write so he can do this job. I think that is an abuse of seniority. These practices get in. The real question that is coming up—sure, there are these peripheral uses and extraneous and obsolete uses—the question is whether collective bargaining has handled this question so inadequately that now we remove it from collective bargaining, and this becomes a matter of management prerogative. This I cannot see as a way of solving the question. We get some questions o f human relations that you cannot solve by a gimmick or simple formula, but people have to get down to work and do their collective bargaining responsibilities in these regards. It would be wonderful if you could get some gimmick that would solve these industrial relations. I don’t think you can. I think we are dependent on the way unions and management live together in particular plants. This is why I suggested, Senator, maybe the benchmarks would facilitate this process. I f management could say, in an industry, “We will cut out all these seniority rules,” they would have to invent new ones. This is true with a lot of these work rules. Senator B u s h . I think you are right. Mr. T a y lo r . The question is, Whose judgment is to be deemed the reasonable judgment when you develop these work rules? I think we are committed to doing it by collective bargaining. Senator B u s h . I think you are right. Mr. T a y lo r . It is disturbing that collective bargaining does not function better in some of these areas. I f I might just add one point. I mentioned our problem in Penn sylvania. This disturbs me no end as a loyal Pennsylvanian and Philadelphian. Some companies have accumulated a lot of barnacles by way of these work rules. The union officials know it. The com E EM PLOYMENT, GROWTH, AND PRICE LEVELS 2619 pany knows it. They say, “ Let us get to work and convince these people that we need some changes, and we will work out ways of ameliorating the impact on them.” Finally a company will say, “ Oh, boy, even with union cooperation this will take a long time to do. These people are set in demanding these rules. So instead of putting our new plant in Pennsylvania as an addition to the old, we will go somewhere else where we can start over brandnew on the work rules.” This is an important problem in some of these older areas. The mere process of getting rid of the barnacles can have costly economic implications. I don’t think you can solve this by some gimmick or a general rule. I talked to some of these companies and said, “ Maybe you are making a mistake by putting your plant in some new terri tory.” I hope I am not impinging on your interests in getting in dustries to move to other States, which wTe do not want in Pennsyl vania. I say to them, “ I think you are making a mistake because you might get better working conditions in a new spot, it will be so much harder to refurbish your old plant.” Problems of this sort intervene. I do think there are these adverse economic consequences of the slowness o f our process in a democratic country in making these adjustments. I don’t think anyone would say that we want to drop the democratic process to make it faster. I think these human factors are very important. The only other thing I can see is that we facilitate the bargaining process so that it deals with the obsolete and peripherial problems more quickly. Senator B u s h . I think you could draw a distinction between the use of seniority for tenure against layoffs and its use for promotion. Mr. T a y lo r . Y o u can indeed, Senator. Senator B u s h . There I think it is much more questionable. On tenure I would say it ought to be given the benefit of the doubt every time. Mr. K a h n . Here again, though, wTe must remember that we are talking about promotions within a bargaining unit. Characteristi cally, m mass production, we are talking about promotions to jobs which are only slightly higher than the jobs the incumbents are hold ing, and which in the typical case almost any man could perform. Senator B ush . That would not be my definition of promotion. That is not what I had in mind. Mr. K a h n . I also want to add, sir, as a generalization, that sen iority rules rarely restrict promotion out of the bargaining unit into managerial jobs, which are much more critical forms of promotion than the kind typically occurring within the uni4. In terms of what Professor Taylor was saying, I would like to focus attention on this kind o f subtle problem. Certainly it is unreasonable to say that we must teach the man to write because he is the next man in line. That is unreasonable. On the other hand, let us take the other extreme view. Supposing a company says we don’t want to give the senior man first crack at this job unless he can perform it perfectly and immediately. Or take an intermediate position— 1 hour’s training, 1 day of training. What is “ reasonable” depends on all kinds of things: the imperatives of the company’s production problem, the costs of training, the seniority status of affected employees, what the company can afford to do. 2620 EM PLOYMENT, GROW TH, AND PRICE LEVELS This is something where you just cannot apply a stereotyped notion of “ ability to perform.” The parties really have to bargain it out. Senator B u s h . I would like to change over to another subject. Going back to the question of featherbedding, Mr. Stieber talked about it at some length. In this present steel strike, I observe that one of the contentions of the management is that they have to have some revisions in the work rules in respect to the matter of the feather bedding. I haven’t seen anything in the paper to outline their posi tion or justify their position or explain it. I wondered whether any o f the panel are in a position to give us a little light on that. It is a very interesting current situation. Does anybody know what the featherbedding rules are that management wants to change, and why they want to change them ? Mr. S t i e b e r . I do not know the specific illustrations, although I am sure that each side could adduce specific illustrations to support its point of view. I think one thing that is important in the steel in dustry is this: The argument over local working conditions and work ing rules in the steel industry is not over whether or not the company is free to introduce any new equipment or make technological changes which then may serve as the basis for a change in working rules. Specifically, the common provision in steel agreements is that where a working rule or working condition or local working practice, what ever you may want to call it, or local working practice, whatever you may want to call it, exists—and that may be either an oral or a written practice—the company may change that condition; provided, however, that it changes the underlying basis for the existence of that provision. In other words, if a company puts in a new piece of equipment, and it can demonstrate—before an arbitrator if necessary—that it has changed the underlying conditions which have served as the basis for a particular working condition, that company lias no problem. In such cases there is no argument between the parties that it is per fectly legitimate to change the working condition. However, dis putes do occur when the company does not change any underlying conditions, such as the mechanical aspects of the job, new equipment, automation, but wants to change the work process in order to intro duce a more efficient way of doing things. Then, there are difficulties because the underlying basis for the existence of the condition has not changed. Therefore, in arbitration, management would have a diffi cult time demonstrating that it had a right to change the working condition involved under the contract. I just want to comment on some points that have been made earlier by various people. First, on the matter of studies which I think Congressman Reuss raised. A t the risk of being charged with making a little work for professors, I would suggest that this committee or another appropriate committee might very wTell consider a series of studies in specific industries—not when there is a collective bar gaining agreement being negotiated as in the railroad industry but when things are relatively quiet—with a view toward evaluating the degree, if any, to which these various practices we have talked about today exist, and the degree to which they may be impeding or perhaps even assisting in productivity. This has been done with considerable benefit in some industries. For example, Harold Levinson, one of the committee’s technical assistants, has examined so-called restrictive EMPLOYMENT, GROWTH, AND PRICE LEVELS 2621 p ra ctic es in th e b u ild in g tra d e s. T h e r e has a lso been so m e w o r k do n e in th e r a ilr o a d in d u str y o n w o r k in g ru les. T h e r e h as n o t been a n y w o r k d o n e to m y k n o w le d g e in th e steel in d u s tr y or in a n y o th e r m ass p r o d u c tio n in d u strie s. I th in k th is c o m m itte e m i g h t v e r y w e ll c o n sid e r h a v in g su ch stu d ies m a d e o ver a p e rio d o f y e a r s, so t h a t w h en th e tim e com es th a t th ese w o r k ru les b ecom e a n issu e th e y m a y h a v e s o m e th in g to d r a w u p o n . Secondly, on the matter of how quickly changes can be made. I do not think that you can change a working condition that has existed for a long perior of time overnight, simply because you think that it is wrong. It is very difficult to do it, to say the least. This is not new. For example, before the 1949 pension agreements in steel were nego tiated, there were study commissions established to study pensions and whether or not they should be introduced. Before the supple mentary unemployment benefits were introduced in various indus tries, there were studies made of the guaranteed annual wage, which was a demand that unions had been making ever since the early 1940’s. I think that before changing en masse, as Professor Taylor puts it, so-called restrictive working practices, one might very well consider having some studies made to see whether or not these prac tices which have grown up over the years have some reason for being in existence and whether they should be changed and how rapidly. Finally, I want to comment on this question of relocating plants to get out from under restrictive working conditions and seniority problems. This will not long, perhaps, be a way of getting away from those problems, because even now there are seniority provisions which go beyond individual plants and cover all of the plants in a par ticular area or within a company. So when a company opens up a new plant, the provision requires that it give an opportunity to the workers who are being displaced in an old plant to move to that new plant. This is already being done under some agreements. There is also ■an argument whether workers should receive expenses for moving to new plants. Many seniority provisions are being broadened, thus cre ating new problems, although I think one should recognize that there are often very good reasons for broadening these provisions. Senator B u s h . The railroad industry is in a terrible mess finan cially. It is claimed there that the featherbedding rules are a very serious handicap to economic operation. They are faced with the most severe competition that they have ever been faced with right now and some of them are on the order of bankruptcy. I have one in mind in New England that runs through my home town, and it is a source of great apprehension to all of us in that area, A lot of our people use it every day. Does anyone want to comment on the question of featherbedding in this industry, particularly as distinguished from other industries ? I don’t know that the Congress is in a very good position to do much about it, but is there anyone who would care to express an opinion about it for the benefit of the wTorld at large? I f not, I won’t press the question. Mr. T a y l o r . I happen to have served on a presidential emergency board in the railroad industry when the demand was for an additional fireman and engineer beyond the current load. I had this much con 2622 EM PLOYMENT, GROWTH, AND PRICE LEVELS tact with the railroad industry and a great deal of evidence was given. It seems to me from a public standpoint you cannot single out par ticular instances without relating to the whole schedule. I think it is fair to say that while the hundred million rule and some of these other rules require looking at, there are other working conditions that are not favorable to employees in this industry. I think our tendency is to point out what are construed as the horrible examples, without looking at the structure as a whole, so you come out with a balanced problem of working conditions. I would make this observation on very limited experience, Mr. K a h n . For example, sir, the union representatives, or spokes men, recently have been pointing out that favorable working condi tions of the types which became standard in other industries, includ ing penalty rates for working at inconvenient times or to deal with the fact that railroads have to run 7 days a week, including holidays— factors like this are not adequately built into the structure of wage payment. This whole business becomes very complex. I f you look at the total earnings—I really should say the average earnings—o f various categories of railroad employees, I think you would be hard put to conclude that they are completely exorbitant and out of reason. I f you start to pick apart the particular formulas on the basis of which those earnings are arrived at for particular people, I think you would readily conclude that they don’t all make sense in today’s world. Senator B u s h . I thank you very much. Those are interesting ob servations on a very difficult subject. I have no further questions. The C h a i r m a n . I understand that Congressman Reuss has a com ment on the seniority system. Representative R e u s s . I am not sure it is a comment I wanted to make on the record. I did suggest to the chairman that it was inter esting. I think it was Mr. Taylor who said that the field where senior ity really worked was where the job was not very difficult. Senator B u s h . I do not agree to that. Mr. T a y l o r . I was speaking solely about industry, Senator. Representative C u r t i s . The only question to which I want to direct a little attention is this problem of classification and reclassification. It seems to me to be underlying this problem as much as anything. I was interested, Mr. Stieber, in your suggestion of studies being made of certain industries in this labor-management field. I thought one that we should be studying on our own is the post office, where we actually do handle the labor-management, or one aspect of it. To me it was very interesting that, in our recent 3 or 4 years, the problems that come up have not been so much around the increase in wages or the pay as much as in this area of classification. It seems as automation continues and we advance in economic growth, the prob lem of classification and reclassification is a constant one and an ex tremely difficult one. Perhaps if we directed more attention into this area we might come up with better solutions. Yesterday in the panel there was a point made that over a period of years, and I think the table showed from 1900, the percentage of wage costs in the price was increasing. I suggest possibly that might be due to the change in the mix of employees, using the word “ mix” in relation to skilled, semiskilled, unskilled, technical, and profes EM PLOYMENT, GROWTH, AND PRICE LEVELS 2623 sional. And whether we were not, as wTe continued in automation, getting more and more into the skilled and technical area and more away from the unskilled. I f that is true, then this problem of classification and reclassifica tion is going to be more acute. I f anyone would care to comment, I would appreciate. Mr. T a y l o r . I was going to put it this w ay: This is true, it seems to me, we are upgrading. In most manufacturing operations you don’t have common labor any more. Even a sweeper has got a machine which even does the corners good now, and things of this sort. It is hard to fail to recognize this upgrading. I think it is in the process of the upgrading that we are getting the big problem in industry. For instance, you make a technological change and you take away half of job A and half of job B. You should create a job C with one of the people. There is coming into industrial operation a craft conscious ness, if you will, where each one says that part of my job which re mains is mine to do even though it is not a full day’s work. There was one case quite recently where a very famous company in this country dealt with an independent union and sought to bring this about. You just got an uproar among rank and file. They invited in the Teamsters to represent them. They were not going to permit this reclassification to take place. It gets back to something else we are talking about. I think for job A and job B to be combined to job C in this upgrading process requires some attention to what you are going to do with this fellow who gets displaced. This is a very difficult thing to do. What you get in these cases, you get automation and technological change, it raises high hopes among the people. Now we ought to share in this increasing productivity. So you get wage-increase demands very naturally. It also raises deep fears about this person w7ho gets laid off. He thinks the added productivity ought to be used for protecting him from the impact. So the demands get rather high. I think one of our big problems in this country underlying a lot of these labor disputes, if we could come to a better understanding than we are about how you share in the fruits of increasing productivity. This might seem a little afar, Senator, but I think it is related to this prob lem of upgrading and reclassification. You get into very difficult human problems. Mr. S t i e b e r . Mr. Curtis’ mention of classification problems leads me to mention something that I think should be pointed out. The steel industry is coming in for a great deal of criticism, union-management relations in the steel industry are being regarded with a very critical view. The question is often asked, W hy can’t these people ever get together and settle things without a strike ? I think perhaps it might be well to point out that this industry and his union have developed a joint job classification or job evaluation system start ing in 1945-47 through joint negotiation, which both sides consider to be working extremely well. It has helped to take care of many of the problems in this area of classification and reclassification. I would also like to point out that while not an architect of that system one o f the men who had a great deal to do with the eventual negotiation o f that agreement is Dr. Taylor, who sat on the War 2624 EMPLOYMENT, GROWTH, AND PRICE LEVELS Labor Board in 1944 in the dispute over inequities in the steel in dustry. I might add that a definitive study o f this point job evalu ation program will be published in a few months. Modesty forbids me to mention the author of this work, but the Harvard University Press will publish the book, which will point up one area in which the industry and the union have been able to work out their problems through free collective bargaining. Mr. T a y l o r . May I comment just 1 minute. I think this is one o f the finest collective-bargaining jobs ever done in this country, when the Steelworkers Union and the steel companies worked out this classification system. They agreed on a description of the jobs and then they evaluated the jobs and put them in the proper classi fication. It was a tremendous collective-bargaining achievement. I think it should be added that this applied only to the hourly rated jobs. They never were able to get over some of their incentive prob lems, which is very urgent in this industry. But this one was a magnificant job. Representative C u r t i s . I will be very interested in reading that study. Mr. S t i e b e r . Let me add one more thing which has some relevance to our discussion. That joint job evaluation program was not nego tiated without cost. It was an extremely costly program. It has been estimated that it cost the industry something like $100 million to put the program into effect. But nobody deplores that cost. The companies consider the money well spent. Now we might ask whether in a situation where you are trying to do something equally as radical, where you are trying to change working conditions which have been in effect for many years, whether this too might require some cost. This is why I mentioned in my statement the Armour settlement and the west coast longshoremen’s settlement. These companies are paying money to accomplish some thing that they probably think they have a right to do without it costing them a penny. They probably feel that they are trying to exercise their prerogative as management to manage efficiently. Yet they seem to have recognized that this is a difficult decision for the union and perhaps they might be well advised to contribute something to ease the process. The job evaluation program in steel was negotiated at considerable cost, but over the years it has proved worth while. Perhaps the same approach might work with regard to current problems in the steel industry. Representative C u r t i s . I am happy to have you make that em phasis on cost. I believe that a great many of these advances in other areas do cost. I am willing to pay the cost. I don’t like people to assume that we can do things without it costing something. Did you have a comment ? Mr. K a h n . Yes, sir. I wanted to observe that while we have all been arguing for care and caution and recognition of diversity and o f human relations problems, and so on, it is also desirable to recog nize that whatever economic costs are involved in somewhat obsolete or inefficient or make-work working rules, should be appraised from a broader perspective. The costs of undesirable practices in terms of productivity are insignificant compared to our national growth in EM PLOYMENT, GROWTH, AND PRICE LEVELS 2625 productivity over periods of time. This does not mean we should not try to deal with the shortrun in wdiich we live, but we should also recognize that over time our output and our real income increases are of much greater importance. The railroad industry, which has a very difficult competitive problem at this time, is also an industry which has been doing very well in its technological development and in its improvement in man-hour output. Over a period of decades this is far more significant than the im mediate cost of a particular working rule. I would like, if I may, to focus attention on a point which Mr. Stieber brought up in his written statement and which has been ne glected. This is with reference to the fact that economic growth is not only a matter of man-hour output but of how many men are working how many hours, and relates to the shorter workweek and its consequences. I bring it up because I want to make a point in minor rebuttal to my colleague from Michigan State. At least I am not as certain about one point he has made as perhaps he is. This is with reference to his comment on page 16 : “ The shorter workweek can be had but only with some sacrifice of additional income.” My point is this: there is a good deal of evidence which suggests that as we continue to reduce the standard working week, which will un doubtedly happen, that we may not decrease and may perhaps even increase the effective labor supply in terms of man-hours of people who want to work. I say this first because many people who are on the margin of decision as to whether or not they should become regular members of the labor force will decide to become members of the labor force when the regular workweek is less. Housewives who feel they could not meet their home responsibilities with a 40-hour week may find it easier to do it with a shorter workday, for example, if they can then get home in time to make dinner. Handicapped people or older per sons who can’t stand the strain of a longer workweek might find it possible to work regularly on a shorter basis; and so on. Another factor is that where people are dissatisfied with the income which accrues to them from their shorter workweek and want to do something about it, they will have a much better chance to engage in multiple job holding or moonlighting, or whatever word you want to attach to it. I suspect the portion of people who hold second part-time jobs will undoubtedly go up as the regular workweek declines. Another factor will be the increased relative importance of overtime work. For example, if you have a standard 4-day week and in a pinch the employer asks his workers to put in 5 or 6 days, the pro portionate increase in man-hours involved is very substantial and provides a new dimension of flexibility for short-term needs. Finally, I think we should not neglect the fact that there will be an increase in the economic importance of do-it-yourself activities, which, incidentally, are a form of real income not readily subject to progressive income taxation. Perhaps there will be a great deal of additional part-time selfemployment o f various types. The man who wants to write a novel may find it more possible to do this and eat while he holds his shorter workweek job, and therefore need not apply for some kind of fellow 2626 EMPLOYMENT, GROWTH, AND PRICE LEVELS ship in order to accomplish his goal. What I am suggesting here, since the committee is concerned with economic growth, is that some careful study into the impact on labor force participation of a shorter standard workweek would be very desirable. We must not leap to the conclusion that from the point" of view of the economy as a whole our economic growth in material terms would be seriously retarded if in our society there emerged a general trend toward a 32-hour week or something like that. Senator B u s h . Your contention is that it would not be retarded at all ? Mr. K a h n . Individuals who are satisfied with the income provided by a shorter workweek job might indeed work fewer man-hours. The basic question concerns our values with respect to income as against output. Assuming a generally prosperous environment, I think a shorter workweek will mean that many more persons will hold more than one job in one form on another. This will be an offset and per haps an increase in the labor supply available to the country. Others will enter the labor force. I think all observers are agreed that in this country, for better or worse, we still have a very, very high pre ference for additional income over additional leisure. Senator B u s h . Y o u are saying that a shorter workweek would promote economic growth ? Mr. K a i i n . I th in k it w o u ld p r o m o te m o r e fle x ib ilit y in th e d e c isio n s b y in d iv id u a ls as to th e ir r e la tiv e ch oices b etw een in c o m e a n d le isu r e a n d f r o m th e p o in t o f v ie w o f th e c o u n tr y as a w h o le I a m s u g g e s t in g , as a b a sis f o r s tu d y , at le a st, the h y p o th e s is t h a t it m a y even in cre a se th e m a n -h o u r s o f la b o r a v a ila b le to e m p lo y e r s in th is c o u n tr y . Senator B u s h . And therefore promote economic growth? Mr. K a h n . And therefore promote economic growth. Mr. S t i e b l e r . Let m e comment briefly on these four points. I would certainly agree that the effect of a shorter workweek on the labor supply is one that should be and in fact has been examined as much as one can examine these things before the fact. It will un doubtedly have some ramifications in this area. A greater proportion o f women may seek work and so forth. This certainly ought to b e considered. The point that I was making, and I think it is still the most im portant point, is that the one aspect of a shorter workweek that has existed for the last 50 years that does not now and has not existed, you might say, ever since we reduced the workweek from about 48 to 40 hours, is that by and large in most industries we cannot anticipate a greater increase in output per man-hour as a result of the reduction in hours itself. Automation and other factors of the shorter workweek will still affect productivity. But this one aspect that has been so important in the past—the effect of a reduction in the workweek in terms of reducing fatigue, wastage, absenteeism and so on is less likely to be important in the future. Labor supply considerations are impor tant and should be taken into consideration. However as for the two-job-holding aspect, I would say that this should be considered but it is hardly a desirable way of increasing labor supply. EM PLOYMENT, GROWTH, AND PRICE LEVELS 2627 The C h a i r m a n . I am told that Akron has the shortest workweek in the country and that a very high percentage, no one quite knows how large, of the workers have a second job, is that true ? Mr. S t i e b e r . This apparently is true in Akron where they have a 6-hour day and there are two jobs being held by a sizable proportion of workers. The C h a i r m a n . I am also told that it is very hard to get people interested in civic matters because of this very thing. Mr. S t i e b e r . I would not be at all surprised. Overtime work may also increase as a result of a shorter workweek. But again I would hardly consider this a desirable consequence be cause overtime work increases cost of production. I f you reduce the workweek to a very low level workweek you might get considerable overtime work. As for the trend toward “ do it yourself” that is a matter of per sonal preference. The question is whether you want increased leisure, whatever that leisure may be used for, or whether you want increased real income. Finally, and this has to do with increasing labor supply, this con sideration is important only if we can maintain full employment. This, of course, is what these hearings are all about. Mr. T a y l o r . I would like to add a footnote to what Mr. Stieber said. I think we have had a steadily decreasing workweek in the last 10 years, probably as rapid a decline in the workweek as any period in history. Only we are taking the shorter hours in a different way. We used to go from a 44-hour week down to a 40-hour week and main tain the pay. Now we get 2 weeks with pay, which is equivalent economically in terms of reduction of the workweek. I think this is a very interesting development that has taken place in this hours movement. The holidays with pay. You get a lot of cases now where you get 4 weeks pay after 25 years of service. This is a new way of reducing hours of work in this country. It may be that it has great advantages over the former approach that we took, really to spread the employment. That was the big drive in earlier years. It does permit a better participation in these other activities and all the rest of it. Forty hours evidently does not seem to be an arduous workweek for most people. In these terms I think the hours move ment has gone very steadily in this direction. Mr. K a h n . I f I may add just a short footnote to Professor Taylor’s comment, I also think we need to learn more about what is happening to nonworktime than we now know. I recently participated in a brief inquiry into the question of leisure time and recreation time, particularly with respect to finding out what we knew about a metro politan area like Detroit. The answers are not simple. For ex ample, we have no idea about the aggregates. We don’t know how many man-weeks of vacations are taken in the Detroit area. For another thing, many workers, including all of the production workers in the automobile industry, do not get "paid vacations. They get pay in lieu o f vacation and are not permitted to take vacations if they are needed on the job. They take their vacations when they are laid off, if at all. ’ I suspect that there is a great difference in actual behavior with regard to vacation practices between workers who are told by their 2628 EM PLOYMENT, GROWTH, AND PRICE LEVELS employer, “ You will get your regular paycheck and we don’t expect you to come in for 2 weeks,” which is characteristic o f salaried workers, and production workers who are simply given a cash bonus because they are not getting a vacation. In addition, we don’t really know enough yet, and the data are not available, about what people do with time which the employer thinks may be leisure time. Do they employ this to paint a house because they cannot afford a painter ? Do they sit at home because they can not afford to travel ? We need a lot of information here. I mention this to emphasize that where people have a strong desire for more income and more leisure, they will find ways, if they can, to increase their income and not take advantage of the leisure which even their employer may believe he is giving them. The C h a i r m a n . Gentlemen, this has been a very interesting session. Thank you very much for coming. We will meet this afternoon at 2 :30. (Thereupon at 12:10 p.m., a recess was taken until 2:30 p.m., the same day.) AFTERNOON SESSION Representative R e u s s . The committee will be in order. The subject matter of this afternoon’s panel is market power of unions: Facts and policies. I notice that each of you three gentlemen has submitted a formal paper. Each will be printed just as you have submitted it in the committee’s record. We would like to have you proceed in your own way, either to read or paraphrase or summarize the papers or to gro beyond them in any way you wish. Mr. Killingsworth, would you care to start out ? STATEMENT OF CHARLES C. KILLINGSWORTH, MICHIGAN STATE UNIVERSITY Mr. K i l l i n g s w o r t h . Thank you, Mr. Chairman. My name is Charles C. Killingsworth. I have been a member o f the economics faculty at Michigan State University since 1947. I have also had experience as an arbitrator of labor-management dis putes, and as an administrator of the Government’s wage stabilization program during World War I I and the Korean war. The assignment of this afternoon’s panel is to consider “market power of unions”—the nature, sources, and limitations of such powery and ends and means of public policy. Many people, including some prominent economists, would consider this a simple assignment. They would say that unions can shut off the labor supply of plants, firms, or even entire industries; therefore, they are monopolies and should be brought under the antitrust laws. I believe that this diagnosis and this prescription rest on a common error in logic—the fallacy o f the inexact analogy. Even if it were justifiable to regard unions as the equivalent of business monopolies, or quasi-monopolies, the ap plication of the antitrust laws to unions would not achieve the same results in the labor market that the antitrust policy is supposed to achieve in commodity markets. Furthermore, the market power of most unions is subject to certain limitations and restraints which are EMPLOYMENT, GROWTH, AND PRICE LEVELS 2629 inapplicable to the typical business monopolist ; so I believe that the term “ monopoly,” as it is commonly understood, cannot justifiably be applied to unions generally. The basic assumption of the antitrust laws is that free competition is the best regulator of economic activity. An economic theorist can readily demonstrate that what he calls “ perfect competition” will bring about the most efficient possible distribution of productive re sources in relation to consumer desires, with a minimum of government intervention. In the interests of economic efficiency, the antitrust laws undertake to prohibit business combinations and business practices which hinder the operation of the forces of competition. Whether or not the anti trust lawTs have actually been effective in the business field is a question which we need not consider at this time. What we must consider is the fact that, even if there were no unions at all, the labor market would be far different from the economists’ model of perfect competition. Several conditions are essential for even reasonably effective com petition. One is that sellers are able and willing to seek the highest price for their product, whatever it is. This condition implies knowl edge of alternatives offered by the market and ability and willingness to take advantage of the best opportunities, Empirical studies of the labor market have shown that workers, as sellers of labor, simply do not behave in these ways. They generally have quite poor knowl edge about alternative opportunities, and most of them have, from the narrow economic standpoint, “ irrational” attachments to localitiees, occupations, and particular jobs. Because they are human be ings, they don’t behave like that calculating machine known as “ economic man.” Another necessary condition of competition is that there must be no collusion among buyers. All buyers must compete actively with each other for the available supply of a particular com modity, bidding on the basis of the value to them of added units of the commodity. Many research studies have shown that, under normal circumstancest, most employers regard wage competition as highly improper behavior. “ Labor pirating” is the term usually applied to such com petitive bidding for workers, O f course, there are a few pirates left, but their behavior is not the norm of the labor market. The norm is the “ gentlemen’s agreement” not to upset the going rate. Indeed, hiring policy is sometimes so gentlemanly that one firm will not ac cept an application from a worker who is currently employed by another firm in the community. Competition on the buyer’s side of the labor market can hardly be called vigorous. I have considerably oversimplified in the interest of brevity; but I think that most stu dents of the subject would agree that, even in the absence of unions, the forces of competition operate but feebly on both sides of the typical labor market. The U.S. Supreme Court has said that literal application of the antitrust laws to unions would outlaw almost all means presently used by such organizations to achieve their objectives. In other words, collective bargaining would be made impossible, And even this dras tic measure would not achieve even a reasonably competitive labor market. To be sure, most of the advocates of regulation of unions 2630 EMPLOYMENT, GROWTH, AND PRICE LEVELS through the antitrust laws do not propose such literal application of them; in fact, it is impossible to determine precisely what most of these advocates favor. They want competition “ restored” in the labor market, and they want to achieve this end through some “ adap tation” of the antitrust laws to union practices. I do not believe that it is possible, in our kind of society, to legislate a competitive labor market. I believe that attempts to apply the antitrust principle in the labor field while preserving collective bargaining involves an in escapable contradition. Many years ago the Federal Government launched upon a policy of protecting and encouraging collective bargaining. In my opinion, experience has abundantly demonstrated the wisdom of this basic policy. Our objective should be to improve the operation of collec tive bargaining rather than to substitute for it a fundamentally differ ent system of industrial government. The remainder of this state ment will consider some of the results of collective bargaining rather than the merits of an alternative to it. How much market power do unions exercise under our system of collective bargaining ? No answer to this question can be both simple and accurate. Undoubtedly some unions have sufficient market power to fix an arbitrary price for the services of their members and to force the payment of that price. Other unions have so little market power that their influence on wage rates is negligible. In this field, which is as complex and diverse as the American economy itself, generaliza tion is exceedingly difficult. Nevertheless, I will venture some broad observations to provide a basis for discussion. In the first place, I suggest that the market power of unions is usually a derivative of the market power of the employers with whom they deal. In an industry like garment manufacturing, where com petition between employees is vigorous, what unions can get in terms o f wage concessions is severely limited by that competition. In an industry like steel, where competition is much more limited and em ployers have great discretionary power over prices, the unions are likely to have some discretionary power over wages. I f we could really achieve the basic objective of the antitrust laws, which is e f fective competition in most product markets, we would thereby con siderably reduce the market power of most unions that have such power. Second, the market power o f unions generally is subject to one restraint that is inapplicable to the classical monopolist. Typically,, the monopolist is assumed to set his price at a level which will result in the maximum net return; the number of units that he sells is of subordinate importance to him. With most unions, the situation is reversed. With few exceptions, unions must be concerned with the employment effects of wage bargains, insofar as they can be deter mined. I f a particular wage rate would force the employer to go out of business or would demonstrably reduce the number of jobs he could offer, the typical union will temper its demands. To be sure, the employment effects of wage bargains are often obscured by cyclical and other factors. But for most unions, the fact that certain wage policies will result in unemployment for some of their members frequently acts as a brake on the exercise of whatever market power they may possess. EM P L O Y M E N T , G R O W T H , AND PRICE LEVELS 2631 Third— and closely related to the foregoing—market pressures on employers are frequently transmitted to the unions with which they deal. The automobile industry is usually considered to be one in Avhich administered prices are prevalent. But in recent years this industry has been subjected to increasing competition from other consumer goods industries. This was one factor which prompted the industry to insist on a modest wage settlement in 1958. Even though the U A W is one of the largest and most aggressive unions in the country, it had to bow to these market pressures. It is notable that automobile prices are not being increased this year, even though the industry is paying slightly higher wages. Finally, there is no convincing evidence that unions have had any substantial effect on the general level of wages over the long run. I f unions generally had a substantial amount of market power and consistently used it to push the wages of their member higher than would be justified by the economic forces at work, surely it would be possible to find statistical evidence of such an exercise of market power. Some students of this matter believe that they have found such evidence; but usually the use of a different base year, or slightly different time periods, or slightly different statistical measures, will yield contradictory results. We cannot say with confidence that unions have not affected the general wage level; but neither can we say the opposite. It does seem safe to venture the guess that, if there has been any effect, it has not been a spectacular one. I certainly would not argue that no union has ever possessed sub stantial market power. Neither would I argue that such power has never been abused. Some unions have barred goods made by fellow unionists from certain markets solely for the purpose of creating a local monopoly which can be exploited by the union and the employers with which it deals. Other examples could be cited. I, for one, would support legislation narrowly drawn to prohibit such clear abuses—and, of course, we already have some legislation which achieves that purpose. But my broad conclusion is that unions gen erally do not possess sufficient independent, unchecked market power to justify Draconian measures which would imperil the great bene fits to our economy flowing from our present system of collective bargaining. That concludes my prepared statement. I have a few additional observations. After I agreed to appear here today I started thinking about it and I realized that I did not know precisely what the term “ market power” means. I asked one of my colleagues on the faculty who ought to know what it means and he said he did not know, either. So I have assumed for the purposes of today’s discussion that per haps the committee is suggesting market power as a substitute for the term “ monopoly.” Representative R e u s s . I think the committee had no more in mind by “market power” than strength power, or whatever. So I am sure your paper will be relevant to the subject matter. Mr. K i l l i n g s w o k t h . Thank you. I hope so. I f market power should come to be substituted for the term “ monopoly,” I think this would be a real contribution to clarifying discussion in this difficult 2632 EMPLOYMENT, GROWTH, AND PRICE LEVELS area. It is my feeling that in the past 20 or 25 years this term “ monop oly” has come to lead a rather dangerous double life. In the political sphere, monopoly is something that is evil, some thing that conjures up the picture of oppression and grasping greed, and the exclusive control o f some commodity. I think that econo mists, on the other hand, have developed a quite different concept of monopoly over the last 20 or 25 years. Starting back in the 1930’s there has developed the idea that monop oly elements are ubiquitous in our economy. Some economists would even argue that the corner druggist in a sense is a monopolist. He may sell the same kind of aspirin as every other drugstore, but there is no other drugstore that is located exactly where he is. So, for the people in the community in his neighborhood this par ticular druggist’s location gives him an element o f monopoly power. In that sense it is quite correct, of course, to say that labor unions have an element of monopoly power. But this is not the sense in which the general public uses the term. I think that the result has been considerable semantic confusion in public discussion of the problem of “ labor monopoly.” The commit tee’s use of the less emotional term, “ market power,” may be a con tribution to understanding the problems. Thank you. Representative R e u s s . Thank y o u , Mr. Pierson? STATEMENT OF FRANK C. PIERSON, PROFESSOR OF ECONOMICS, SWARTHMORE COLLEGE Mr. P i e r s o n . Mr. Chairman, I shall start, too, with a brief con sideration of the definition of the term that we are dealing with here. I agree with Professor Killingsworth that the term “ monopoly” has so many connotations that perhaps it is not appropriate in this con text. I would suggest the following definition: Union market power is the capacity of labor organizations to with stand or exceed employment terms that would have obtained in the absence of collective action on the part of workers. The existence of such power on a continuing basis is the result of some advantageous circumstance—a strategically located craft for which no substitute exists, an insistent demand for a particular product or service, a localized labor or product market largely cut off from other areas— but its implementation requires collective action through a union organization. All unions, taken together, may also possess market power, but again the power rests on some advantageous circumstance, such as high national employment or a protective governmental policy, which affects most unions simultaneously. The exercise of union market power entails hardship for one or more groups, workers who may be excluded from jobs, employers who may be cut off from using more efficient methods of production or from reaching certain customers, consumers who are limited to a narrower range of choice in their purchases than would otherwise be the case. The issue which these situations always pose is whether the benefits which flow to the union groups can be said to outweigh the hardships imposed on other groups. Put even more broadly, the issue becomes one of deciding whether the present system of unregu- EM PLOYMENT, GROWTH, AND PRICE LEVELS 2633 lated bargaining leads to excesses which could be effectively controlled without undermining widely accepted ways of handling employeremployee relations. The problem assumes its most ominous propor tions when many unions possess market power and when conditions in the national economy favor the exercise of such power on a con tinuing countrywide basis. Under present public policy, employers and unions are free, within wide limits, to seek any changes in their bargaining relationships they want and can agree upon. Some legal limitations exist on the methods each can use, but there are hardly any restrictions on the objectives of their bargaining. This laissez faire policy toward the bargaining process acccords well with the country’s social and politi cal traditions, but it leads to very questionable results when power gravitates to one side of the bargaining table, or both parties join forces to exploit the consuming public. There is one notable exception to the Government’s laissez faire policy toward the results of collective bargaining. Union-employer agreements to fix product prices or to exclude outside firms from selling within a certain area are prohibited. (See the Supreme Court decisions in the Hutcheson and Allen Bradly cases.) The specific question which this committee is considering at this time is whether certain other bargaining objectives of unions should be outlawed and whether there are any other steps which the Government might take in this area. Impartial investigators, who have looked into these problems inten sively, tend to question the advisability of imposing direct curbs on union market power for the following reasons. First, the exercise of such power is hard to isolate and identify. A few extreme cases can be readily singled out but most are closely related to some change in the economic situation—an expansion in sales, a mounting scarcity of a certain type of labor and the like—which suggests that the change in employment standards would have occurred in one form or another anyway. Often, too, union W'orking rules are inbedded in longstanding prac tices in a particular company or industry, collective bargaining agree ments merely formalizing what w^ould have been done in the absence of union organization. These practices become part and parcel of the folklore of industrial life and their significance to the worker groups involved cannot be measured in economic terms alone. To try to cut them out, or to eliminate comparative norms which have come to be widely accepted, will be viewed as an attack on highly cherished traditions to be resisted at all costs. Second, it is most difficult to determine in a given case whether the exercise of union market power is or is not socially desirable. If, for example, a union prohibits employers from speeding up the pace of work to a point where the health of their employees would be seriously impaired, few w^ould question its action; but if a union forces a slowdown in the pace of work to a virtual standstill, its action would generally be considered antisocial. The rub comes-—and this is the typical case—where elements of socially desirable and undesir able behavior are intermixed. It follows that no single rule or guide to public policy such as a blanket prohibition of featherbedding, or 38563— 59— pt. 8-------13 2634 EMPLOYMENT, GROWTH, AND PRICE LEVELS of industrywide bargaining, or of monopolistic wage setting can be applicable in all cases. Consider the present impasse in the steel industry from this point o f view. I f the final settlement is for a 15-cent-per-hour increase, presumably it would be taken as an exercise of union market power, but would an increase of 10 or 5 cents per hour ? Where does union market power fade off and empolyer market power enter in ? These are disturbing questions because if union power is to be controlled, will not employer power in labor relations need to be controlled too; and, if so, under what circumstances and how ? Third, most systematic reviews of the available evidence indicate that verified abuses of union market power are relatively few and far between. For example, there is no clear evidence that in the 15 years since World War I I unions have been able to secure substantially greater wage increases than nonunion groups. Moreover, in some strongly unionized fields like bituminous coal mining, spectacular in creases in productivity have occurred over this same period. It is true that excessive wage increases and extreme limitations on produc tion have been secured by certain unions in locally sheltered markets where labor organizations deal with many small firms and where they control vitally necessary craft labor, as in the building construc tion field. However, the most recent comprehensive study of union working rules in this industry by Haber and Levinson, “ Labor Rela tions and Productivity in the Building Trades,” even raises doubt whether serious abuses exist in this field. Union market power is likeliest to pose difficult problems when eco nomic conditions deteriorate or change in ways that are unexpected. A rule about train crews or train mileage that is quite appropriate at one time may prove quite inappropriate at another. Incentive wage standards that are applicable under one set of conditions may be wholly inapplicable under another. Wage levels or wage increases geared to one pattern of consumer demand may not mesh with pat terns that develop at a later time. The attempts on the part of unions to maintain the status quo or to block forces of change may well be ill-advised but they are generally symptomatic of underlying diffi culties which need to be recognized and dealth with. It is not enough merely to castigate a particular union, say, in the entertainment or printing field, for insisting on make-work rules which violate every canon o f economic logic; nor is it enough simply to try to prohibit such rules in the future. The underlying difficulties which led to the adoption o f the rules in the first place would still remain, and some remedial action of a positive nature would therefore seem indicated. Enough has been said to indicate why it is so difficult to identify union market power or to decide exactly where or when serious abuses of such power occur. Based on present knowledge, corrective action by legislative means does not seem called for. Further investigation, however, might show otherwise. There is need for an intensive ex amination of these problems in key industries—building construction, railroads, trucking and steel, to name but a few. The services of employers, unions, and government bodies should be enlisted in carrying out such an examination if the results are to receive the attention they deserve. The investigation should not be EMPLOYMENT, GROWTH, AND PRICE LEVELS 2635 aimed at cutting out as many union working rules as possible or at, achieving any specific partisan objectives. Rather, it should be de signed to find out what are the principal barriers to increasing productivity in particular industries and how employers and unions might work together for their removal. Approached in this ligM, all parties to the inquiry—employers, unions, and government—could be expected to contribute and all could be expected to benefit. The ultimate answer to abuses of union market power lies in more efficient use of our physical and human resources and in extending the benefits of increasing productivity as widely as possible. Atten tion to the part that unions and workers play in these endeavors should not begin and end at the bargaining table. A cooperative approach is needed in which all parties at interest can participate. Studies of particular industries, if undertaken in this spirit, would help lay the basis for a constructive attack on this important problem. Thank you Mr. Chairman. Representative R e u s s . Thank you, Mr. Pierson. Mr. Segal? STATEMENT OF MARTIN SEGAL, DARTMOUTH COLLEGE Mr. S e g a l . Mr. Chairman, I will read a slightly different version o f the statement that has been mimeographed. I simply added a few sentences to points which appear unclear in the mimeographed statement. I Virtually every aspect of collective bargaining is related to the problem of union market power. Hence a brief statement must choose among the multitude of issues. In the present statement I concentrate on a few aspects of union market power as it exists in a particular environment—in an economy characterized by moderately full employment, absence of unusual pressures on its resources, and by optimistic expectations of prosperity. I am also concentrating on a shortrun period, ignoring longrun changes in technology or basic shifts in demand for the products of particular industries. Two points deserve emphasis at the outset of the statement. First, we should recognize that the possession of some degree of market power—that is of some degree of freedom from competitive restraints emanating outside of the particular labor-management unit—is in herent in the very institution o f collective bargaining. Without elimi nating at least some degree of competition in the labor market the unions could not function effectively as negotiators for the sale of labor services. Second, and this is clearly a related point, freedom from the re straints of competition in the labor market is not necessarily achieved by union control of the work force of a particular employer, of a particular area, or even of a whole industry. Competition in the labor market exerts its impact also through competition in the prod uct market among different firms, different areas, or even different industries which produce close substitutes. These rather obvious statements point toward the two chief de terminants o f union market power. The first one of those is the degree of unionization of the particular industry either on a national 2636 EM PLOYMENT, GROW TH, AND PRICE LEVELS scale or in a locality, depending on whether the firms compete in national or local markets. The second factor is the nature of product competition facing the firms in a given industry. These two factors are generally highly interrelated. A third important determinant o f union market power in a particular industry is the ratio of labor cost to total operating cost. The importance of competitive conditions is most obvious in two cases: (a) In the case of foreign competition; and (&) in a situation where products of two different industries are close substitutes for each other. Both foreign and interindustry competition can impose rigid limits on union market power even in a completely unionized industry. And in the case of interindustry competition the third factor—the relative importance of labor cost in the total cost struc ture—may play a significant role. For example, a union in an in dustry in which labor costs constitute large share of total costs may not be able to negotiate wages which are as high as those gotten by workers in the competing nonunion industry in which labor costs constitute a relatively small portion of total costs. In considering the impact of competitive conditions within a given industry it is fruitful to draw a distinction between two groups of industries. The first group consists of industries which because o f transport cost and other considerations compete primarily in local markets. Fluid milk or retailing provide an example. The second group consists o f industries with national or broadly regional product markets. In a well-unionized locality the power of unions in the local market industries, those sheltered from outside competition by transport or other costs, is likely to be quite considerable. To be sure, even here the competitive conditions of the industry play a role in influencing the extent of union market power. In highly competitive industries (with easy entry and many small firms) it is more difficult to have complete unionization even in a given locality, or to avoid “ chiselers” in negotiated standards, than in local industries in which there are only a few firms in the area, such as beer. What is more, the evasion of union control is likely to be more widespread, even in a limited geographical area under the jurisdiction of one local, in an industry in which labor costs are an important part of total costs. And finally there may be always the problem of interindustry competition, or com petition with various do-it-yourself endeavors. But with a strong unionization o f the area, and where there are no close substitutes for the product o f the organized industry, the labor organizations can possess a relatively high degree of freedom from competitive re straints, and a considerable ability to push labor costs upward. The role of competition within a given industry, or of the lack of it, is much more crucial in the industries selling in national markets. In highly competitive industries the entry of new firms is generally easy; moreover, it is frequently relatively easy for an established firm to change location or open a plant in a new area. These features, combined with a high degree of price competition, have some obvious effects on the degree of market power held by unions organizing such industries. For one thing, wTith easy entry and high geographic mobility of plants, it is difficult, if not impossible, to organize completely a par EMPLOYMENT, GROWTH, AND PRICE LEVELS 2637 ticular industry. It follows that unions in organized plants must be constantly aware of competition from nonunion employers, many of whom may be located in areas which are generally characterized by relatively low wage levels. Secondly, it is very difficult for the labor organization to impose uniform standards in the unionized plants. The resistance o f em ployers located in generally low-wage areas is considerable; and they can always threaten to transfer their plants to other locations. More over, the local union leaders in relatively low-wage areas, concerned about employment effects and under no strong pressure from their members, need not be very anxious to equalize their wages with those of higher-wage establishments. Even when regional or national bar gaining exists, there is a tendency to chisel in enforcing such agree ments at the local level. A ll these factors constantly expose individ ual unionized plants to outside competition o f lower costs, and thus limit the market power o f local unions. As in the previously de scribed case o f a local market industry, the difficulties of the union are likely to be greater the larger the share o f labor costs in the cost structure o f the industry. The situation in the “ oligopolistic” industries—those with a few sellers, high barriers to entry, heavy investment in the existing plants, limited or nonexistent price competition—offers a sharp contrast. Most of these industries are strongly unionized ; no strong interre gional or interfirm competition imposes limitations on negotiating relatively high-wage advances or enforcing national patterns in indi vidual firms. Indeed uniform wage changes throughout the industry are probably viewed as another way of maintaining “ stability” in the product market structure. Under the conditions of general prosper ity and optimistic expectations, labor-cost increases can generally be passed on to the consumers in the form of higher prices. Here then is a bargaining situation in which labor organizations share with the employers a very considerable degree of market power—that is of freedom from the coercive restraints of competition. The two cases—one of a very highly competitive industry and the other one o f a “ tight” oligopoly—present the two extremes o f a spec trum of possible competitive situations in the product market of indi vidual industries. In between these two cases there is a wide variety of competitive environments, and hence a wide range of degrees of union market power. II For several years labor costs in many industries have been rising because wage advances exceeded productivity increases. The debate centers on the issue whether these wage changes were a result of a pull of demand or of exercise of market power of unions. I do not think that the problem of wage changes can be answered by singling out any one factor. In a complex economy one is more likely to en counter interactions rather than one-way causal relations. In the years between the last two recessions there were, I think, some relatively “ tight” labor-market situations in some of the lowwage services, clerical work and low-wage manufacturing of large urban centers; perhaps also some shortages o f selected skilled work ers. Moreover, we should also emphasize the generally optimistic 2838 EMPLOYMENT, GROW TH, AND PRICE LEVELS expectations o f the business community, based, I think, partly on the mildness of the 1953 downturn, the change in the national adminis tration, and a belief in a continuous prosperity, more or less assured by a potential intervention of the Government in the case of further recessions. But what I said above about the existence of high degree of union market power in particular industries is also highly relevant to the explanation of the upward trend of wages. The ‘ demand pull” argu ment alone does not provide an adequate explanation. An adequate explanation must assign, I think, an important role to collective bar gaining in situations of high degree of union market power—both in the oligopolistic industries and in some of the local market, well unionized, activities. Virtually any method of attempting to identify the special impact o f collective bargaining on the wage trend of recent years is open to some questions. But, apart from the deductions of the analysis of various types of market structures, there is, I believe, some evidence o f the influence o f unionism. For one thing, the annual rate of wage increases in manufacturing in 1953-58 was four times as large as in the roughly comparable, but generally nonunion, period of the twen ties. And secondly, additional evidence is provided by the behavior o f the interindustry wage structure. The period of the Second W orld War showed that an economy char acterized by very tight labor markets and significant excess demand witnesses a considerable compression of the interindustry wage differ ences (in percentage terms). As employment expands in the highwage-paying industries, the low-wage activities in nondurable manu facturing and services experience labor shortages and advance wages more rapidly than the industries which, to begin with, paid relatively high wages for unskilled or semiskilled labor. .No such compression appears to have taken place in the recent years. In fact, a comparison of the earnings trends since 1953 in some of the high-wage industrial groups, where unions have high degree of market power (e.g., petroleum, rubber, chemicals, primary metal, machinery) with those in the highly competitive and low-paying in dustrial groups (leather, lumber, furniture, textiles, apparel), and with some low-paying services (laundries, cleaning plants), suggests that we have probably experienced a widening of the interindustry wage differentials, with high-wage industries experiencing larger per" cent age increases. This behavior of wages seems to be a pretty good evidence of the effectiveness of union market powder in the oligopolistic and well-or ganized industries. The essential point is that under the conditions of mild labor shortages, the high-paying industries need not raise their wages very much, if at all, in order to expand employment; the mere presence of high-paying jobs is probably sufficient to cause an influx o f workers from low-paying activities and agriculture. For example, the steel industry, whose employment expanded by a smaller per centage in 1955-57 than the total non-agricultural-labor force, could undoubtedly attract new wTorkers without the large increases in its already relatively high wages at any level of skills. The fact that wages in the high-paying industries rose as much or more than in the EM PLOYMENT, GROWTH, AND PRICE LEVELS 2639 other sectors must be attributed, in a large measure, to the action of unions. Accordingly, I view the upward trend of wages in recent years as a joint result of some selective labor shortages and of an effective exer cise of union market power—both factors interacting in a favorable environment of optimistic expectations, some increases in the supply of money, and increasing income velocity of money. P e r c e n t a g e in c r e a s e s i n a v e r a g e h o u r ly e a r n in g s , 1 9 5 3 -5 8 LOW-WAGE AND LOW-CONCENTRATION INDUSTRIES Average hourly earn ings, 1953 1.62 1.54 1.33 1.37 1.14 .98 1.40 Lumber______ Furniture_____ Apparel______ Leather_______ Cleaning plants Laundries_____ Retail stores___ Percent increase, 1953-58 17 16 13 15 16 15 21 HIGH-WAGE AND HIGH-CONCENTRATION INDUSTRIES 24 22 26 29 Petroleum__________ Rubber____________ Chemicals__________ Primary metal______ Machinery_________ Transport equipment Electrical machinery. 21 21 22 III Having exhausted my allotted time, I shall comment only briefly on the problem of public policy toward union market power. (a) I must reject the idea of application of antitrust laws to labor organizations. The application of these laws cannot be effective in reducing union pressures on wages. And the present laws—including the recently enacted amendments to the Taft-Hartley Act—provide remedies for situations in which union activity may affect adversely the degree of competition in the product market, i.e., labor-management collusion, secondary boycotts, contracts restricting doing busi ness with nonunion employers. ( b ) Insofar as the exercise of union market power leads via higher wages to higher prices, such inflationary pressures can probably be reduced by restrictive monetary and fiscal policies. But the result of such policies is periodic unemployment and a general reduction in the rate of economic growth. I do not think wTe can afford to pay such a heavy price for reducing a relatively mild inflationary trend. (c) In the immediate future the most fruitful policy seems to be that of attempting to increase the employer resistance to wage de mands in the strategic, pattern-setting sectors of industry—sectors in which labor and management have a great degree of market power. Thorough investigations of price increases in the oligopolistic indus tries by such bodies as the Senate Antitrust Subcommittee or this committee put these industries in a position somewhat analogous to that of public utilities which must justify rate increases. Unless the 2640 EM PLOYMENT, GROWTH, AND PRICE LEVELS managements would be willing to accept in the future a lower rate of return on investment, the publicity attending such investigations is likely to increase their resistance to union wage demands. (d) The more fundamental and longrun policies would be those designed to increase competition in the oligopolistic industries and to increase the extent of interregional labor mobility. But these objec tives probably cannot be accomplished without some fairly drastic public measures. (e) An equally fundamental policy is that of reducing barriers to foreign competition. To be sure, at the present time foreign competi tion appears to exert its greatest impact on industries in which union market power is very limited. This, however, may change and some of our concentrated mass production industries may be exposed in the future to the pressures of lower cost competition from abroad. Thank you very much. Representative R e u s s . Thank you, Mr. Segal. Representative Curtis ? Representative C u r t i s . I might say I tend to agree with the panel’s general agreement that our antitrust laws do not lend themselves well to doing anything in this area, as a generality. I was wondering whether, in light of what is said in some of the papers, you do not believe there are some special areas that perhaps something through the antitrust laws would help. Let me give an example. Some of your building trade unions, take the bricklayers in St. Louis, not only control who can be a bricklayer, for example, but actually limit the number of bricks any bricklayer can lay. I think Mr. Killingsworth, in his paper, indicated that this kind of thing would benefit the employer. This kind of thing cer tainly does not benefit the contractors at all. How do you handle a situation like that? Our present antitrust laws won’t do anything. I do not believe that the amendment to the Taft-Hartley Act will do anything. I f anything, with their recog nition— and I happened to be in favor of this— of the hiring hall principle in the building trades, it is aggravated. Do we not need to do something in that area ? Mr. S e g a l . Sir, actually I think if one were to enforce the present law which says that no union shop or other type of union security would be given unless the employer would be assured that workers can enter the particular union— at least the membership of the union is offered to the workers— would solve the problem. Furthermore, I would like to point out in some of these cases you probably will not have interstate commerce anyhow because it would be highly local activities. I would say that you probably could in this particular case to apply if you wanted this provision of the unamended Taft-Hartley law, saying that the employer could say, since the membership is not given to people on an equal basis, he would not have any union shop. Representative C u r t i s . He could have a strike. That is the point. In fact, you do have bricklayer strikes. This does not apply just to bricklayers. O f course, there is an economic pressure that keeps them from going too far, which means that they just use substitute materials. EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2641 In our St. Louis area there has been a decline in brick houses, a very noticeable decline, as a result of this. Mr. S e g a l . Also, I think some of the bricklayers who are not per mitted to work on union contracts would probably work on nonunion contracts. I think the union would be rather anxious to keep away from this nonunion competition. Eepresentative C u r t i s . There is not any. That is the point. They use their economic power here over the contractor. I am using this really as an example o f whether or not there are not some special areas, although I agree with the general proposition, or I think I do, that our antitrust laws would not be well applied in this particular area. I have one other special point that I might throw in, although it opens up a big area. Let me limit it to the bricklayers’ example. Did you want to comment ? Mr. P i e r s o n . I just wanted to raise the question whether this would not involve a degree of Government meddling in matters which are generally thought to be private between employers and employees. Whether the suggestion of trying to control that degree of detailed relationship would not involve a very serious matter of Government intervention that most people would question. Representative C u r t i s . It would only be where you had a monopoly, where one union controlled all the bricklayers in an area and then used that power. Mr. P i e r s o n . Y o u would attempt to control, I take it, just for the purpose of argument, all the terms of the agreement between such a union ? Representative C u r t i s . N o . As I interpret the antitrust law, you break up the trust. You just create a situation where there is not the concentration of power. That would be the extent, I would think, again using this argument. No, I would not want the Government in this area to get into labormanagement. But simply if you had a concentration, as we do in the business sector, of economic power you proliferate-----Mr. S eg a l. First o f all, obviously, i f we a pply antitrust laws the same way as we do to business combinations, any union would be foun d right away an illegal organization in this sense. We would have a rule per se that certain organizations restricting competition are illegal. Representative C u r t i s . That is why I made my preliminary re mark saying that I essentially agreed with your presentation. I was raising the question if we did not have some special areas that we might want to write additional antitrust. I am merely speculating. The other area which opens up a bigger field is this: I am won dering what happens in the field of public utilities or railroads, where you have a concentration of economic power in the hands of manage ment. Recognizably, it is a legal trust regulated by a Federal agency or a State agency. What happens to union power ? In some States, like in Missouri, we have a law that prohibits strikes in the area of public utilities. In the morning panel I raised the question about railroads. The point there is that, say, a union simply struck one railroad. 2642 EM PLOYMENT, GROWTH, AND PRICE LEVELS The fact that it is part of a network, the economic power in the hands of that union is such as to strike in such a way that so much more of the economy is affected. In a utility you have complete regulation of the management end of the thing. That might be a special case where we would amend our thinking as to whether antitrust legislation would help. Mr. P i e r s o n . I take it this wTould not be so much antitrust legisla tion as control of a public utility. Representative C u r t i s . It would be the other way. Let us face it. In dealing with trusts we do two things: One, we proliferate where we think we can compete; the other where we think we have a natural monopoly, which is uneconomic to have duplication, we would regulate. You raise a proper point. I would answer to that, that w^e would treat the labor end just like wTe would anything else. Even though it might be in the public utility field, if you could proliferate, you would accomplish the same thing which I think I would regard as being a lot better than regulation. Regulation is second best- to using the marketplace. Mr. S e g a l . I would like to elaborate in this wra y : It seems to me that because of what I said it is very clear that a union is interested in the degree of competition in the industry which it organizes. In the past there have been cases in which the union has attempted to limit the degree of competition. It seems to me that in most cases that I can think of the present laws provide the remedy. For ex ample, for labor management collusion, then the kind of situation where a union pickets independent businessmen who are competitors. I would think under the amendment to the Taft-Hartley law this type o f picketing would be illegal today. I would say the same thing would be true of limiting the area ini which a particular businessman can sell his products because this would come under a contract which is invalid because it compels the employer to cease doing business with somebody. This is still a mat ter of speculation because we do not know how these amendments will be interpreted. I would think that what was meant perhaps originally as the “ hot cargo” clause has also an application to a limi tation of an area in which an employer can do business. It seems to me this again is another aspect. Any price control it seems to me must involve an employer organization. By that I mean some sort of association. There are some problems, but I think this is generally considered, according to the Supreme Court interpretation, as being a violation of antitrust laws. I would say in virtually every case where union activity has in fringed somehow the degree of competition or limited the degree of competition in the product market, virtually every type o f activity is proscribed by the existing legislation. On the other hand, the problem of application o f antitrust laws as it is frequently mentioned is to somehow split up the unions, to limit the area of collective bargaining. This, I think, is highly un realistic for two reasons: First of all, in the case of local industries it is slightly impossible to visualize a situation in which any other type but multiemployer bargaining would take place. In New York it would be impossible to write a separate contract with every apart- EM PLOYMENT, GROW TH, AND PRICE LEVELS 2643 merit house owner for the building service union, the local 32(b) of the building service employees. Moreover, it is by no means obvious that an individual owner would have more bargaining power if he were bargaining with a union which has 20,000 members or more and need not care about the employment of this particular j anitor in Brooklyn. In a local industry like that it is impractical and unrealistic to expect that any splitting up of multiemployer bargaining would have any effect on the pressure of costs in so far as such pressures exist. And secondly in the highly concentrated industries, as I already indicated, the pattern of some sort of wage leadership really precedes union organization. Moreover, even if individual bargains were struck with an individual company or even with the plant of an individual company we would expect that this pattern would be followed by other firms. Representative C u r t i s . My time has expired. I am going to come back with this question which I will pose now. Take a situation like the Teamsters where their economic power is such over a proliferated management that they can completely con trol transportation. I pose the question because the subject matter is the market power of the union. I do think we have a situation where you have proliferated management and have one union dealing with it, and the union derives some power that goes way beyond an industry, as in transportation. True, they do not have a public utility in that sense in a big field like trucking, although some truckers are subject to it. The Teamsters Union has control over all kinds of trucking, private as well as the common carrier. I don’t see that these amend ments to the Taft-Hartley Act that we just passed are really going to solve that problem. That is interstate for sure. Is the antitrust field the area in which to approach this? I f so, would you make an exception in your recommendation that we do not do anything about it ? Senator B u s h . Would you accept an addition to that question, too. I f it is not, what is the way to come to grips with the enormous eco nomic power of a union like the Teamsters Union, which has got so much power that it could really halt economic life in this country. Mr. S e g a l . I think you are dealing here with a problem of an emer gency dispute. It seems to me we are bringing this problem of market power with respect to one particular point, that is the danger o f an emergency dispute that will somehow put in jeopardy the health and safety of the Nation. It seems to me that the way to attack it is somehow to improve the provisions of the Taft-Hartley law which deal with this special prob lem rather than use the antitrust law^s. Obviously unless you would want to have a special application of antitrust laws for the Teamsters, it seems to me we would run into problems. Representative C u r t i s . It would not be for the Teamsters. It would be anything that fits the circumstances. O f course, transportation is transportation, so it would be an area, I happen to believe quite strongly that a strong, independent labor movement is essential to a properly operated private enterprise system. So I am anxious for it to have the power of bargaining. Couldn’t your Teamsters’ group have sufficiently adequate bargaining power 2644 EMPLOYMENT, GROWTH, AND PRICE LEVELS for these things that we seek to solve in their own real labor manage ment relations, and at the same time through proper laws not give them this tremendous power that they have over the entire society. I think possibly you could, just as we did in breaking up management through our antitrust laws. I think maybe something like that might be possible. You seem to answer me by presenting it as an extreme, that I would take away their bargaining power which I do not want to do. That is, bargaining power within their own union and based on the economic power that they have. Mr. P i e r s o n . Mr. Curtis, I w^ant to agree with you first that I think the limits on the power of a union such as the Teamsters are very, very wide under present law. The Teamsters Union is in a position where it can exert very great market power if it chooses to do so. The second point, I think, is important. That is, public policy is moving in the direction of internal checks within these unions and an attempt to control racketeering elements and other elements of arro gant power in labor organizations, and that should be the first line of attack, namely, to make the leadership at least more legitimate and in some sense more responsive to the more usual considerations of social welfare as well as their own membership welfare, The third point is that I agree that there is no guarantee that any effort to clean up a union like the Teamsters will necessarily result in satisfactory contract settlements in that industry. It is very pos sible that we have uncovered here an area of very serious abuse and further steps must be taken. I come back to my first proposition, that the Government should move very slowly in breaking up any established institution or in interfering in any way direct or indirect with the detailed terms of collective bargaining agreements. Mr. K i l l i n g s w o r t h . May I comment 011 that, sir ? I believe if you examine carefully the sources of the Teamsters’ power you will find a good deal of their force in these matters has been derived from various secondary boycott techniques, including among other things the so-called hot cargo clauses in their contracts. The recent amend ments to the Taft-Hartley Act have considerably stiffened the pro hibitions. Representative C u r t i s . May I stop you there just for clarification? I think the hot cargo clauses have been used more to assist other unions in their problems than it has to do anything for a direct Teamsters’ strike, because a direct Teamsters’ strike can certainly shut down the transportation on which any plant is counting, because they will be truckdrivers of the Teamsters. I don’t see where hot cargo enters into that. We are talking about the primary strike of the Teamsters. They are the ones who are going to strike. The hot cargo situation has been where they have been trying to help somebody else. Mr. K i l l i n g s w o r t i i . I believe it has been quite effective also in organizing efforts of the Teamsters themselves. For example, you have a nonunion truckline running from St. Louis to Chicago and freight then has to be transferred in Chicago if it is going farther east. Under the hot cargo clause the union can say this §t. Louis firm is unfair, that is hot cargo, and no Teamsters member in Chicago EM PLOYM ENT, G R O W T H , AND PRICE LEVELS 2645 will touch the stuff. This puts tremendous pressure on the St. Louis trucker who is completely unorganized. Representative C u rtis. I see your point. One point has been raised here. The firms can hire nonunion labor in place of union labor. From the standpoint of skill you don't have too much of a problem. Mr. K i ll in g s w o r t i i. I would go back to the contention that to a very large extent it is secondary boycott techniques which have been used by the Teamsters to organize their industry, and further to put pressure on recalcitrant em ployers in that industry. The Teamsters also possess very great power in the affairs of other unions by virtue o f the fact that if the Teamsters refuse to cross a picket line, then the employer involved in the particular dispute is under tremendous pressure. In certain circumstances he is effectively blockaded by the picket line. That might be regarded as another type of secondary boycott. The driver of the truck simply refuses to make a delivery because the employees of this employer are on strike. Representative C u rtis. I keep thinking there is a tremendous power for one big trucking firm, or for a Teamster Union striking a big trucking firm, to be brought into line through economic power. I f the strike is just against one trucker and not against the others, that competitive element that exists in management plays a part. Rut if all the Teamsters in the whole area strike and shut down all truck transportation, that is a power I don’t see that they need in order to accomplish collective bargaining. Mr. K i l l i n g s w o r th . I f I may continue on that particular point, you have something of a dilemma here. I f one employer is struck wdiile all his competitors continue to operate, then they could in some instances get almost anything they want from that one employer and the other employers can be picked off one at a time. I f all of the employers band together and no one settles on terms that are not acceptable to everybody else, then the employers have a sort of union of their own. They have a combined power to match the combined power of the union. As a matter of fact, a lot of the association bargaining is initiated by employers rather than unions. Representative C u rtis. That exists because the employers have found out that it is a process of picking one off, and it is a pattern that is in the control of one group, the Teamsters. The Teamsters can decide to pick off. I f actually you didn’t have that concentration of power, you wouldn’t have the picking off. You would have the strike going against the one firm. That really proves my point, That is what occurred before the employers have banded together. They have banded together, as you say, but it is as a result of meeting this business of picking one off. They went, as a matter of fact, right down the line on the thing. Mr. K i l l i n g s w o r th . May I add that here is one of these situations where we simply cannot find a perfect answer to the problem. In an industry which is characterized by multiple unions instead of the monolithic kind of organization that you have with the Teamsters—■ for example, the west coast maritime industry—you get all kinds of fancy whipsawing^ and leapfrogging resulting from the fact that the employees o f the industry are organized into many different unions. 2646 EMPLOYMENT, GROWTH, AND PRICE LEVELS These unions compete with each other. They can manipulate various things. For example, one will negotiate a wage concession one year. A n other rival union will negotiate an overtime concession passing up the wage concession. The following year each union will get what the other one got. So if you were to apply the antitrust laws in such a way as to break up the Teamsters in the St. Louis locality into 10 separate independent competing unions, you would have a much worse situation in my judgment than with centralized control on the Teamsters’ side and corresponding organization of the employers on the other side. Representative R e u s s . Senator Bush. Senator B u s h . Mr. Curtis just about exhausted the questions I have in mind very ably, and the gentlemen have answered them very well. I take it, then, that you gentlemen all more or less agree that these new amendments to the labor-management legislation are sig nificant in respect o f unions with as much power and size as the Teamsters U nion: That the secondary boycott and hot cargo amendments to the law are sufficient to substantially check the power of the Teamsters Union to bring this economy to a halt as has been threatened from time to time by leaders of that union. So you should not put in time trying to figure out some amendment to the antitrust laws until we have seen whether in fact these new elements of law, these new additions to the law, don’t have the hoped for result or the expected result in limiting power. Is that about it ? Mr. S e g a l . I would say generally I would agree with what you said, sir. I think there are some problems in some of the provisions. I think the picketing provision has some problems of this sort, I think in a strongly unionized area this is an eminently reasonable provision. On the other hand the trouble is that the very same pro vision applies also to some town in Mississippi which never had a union, in which the weight of the public opinion and the local service and business organizations exert all kinds of pressures on the workers not to vote for the union, in which the whole tradition is against any unionizing drive. It seems to me that in such a situation perhaps the restrictions on picketing—organizational picketing—which have a great deal of validity in New York or in Detroit are not as justifiable in this little town in Mississippi. This is the problem of trying to write a law which will deal with some situations and at the time may not be equally justified in some other situations in a rather complex economy. Senator B u s h . That is a very interesting point. Mr. P i e r s o n . Senator Bush, I think the broad experience under Taft-Hartley would indicate that regulations of powerful unions with respect to the closed shop and other such areas has had little or no effect. Therefore, I would be very hesitant in saying that the new amendments would actually deter the Teamsters Union from exercis ing as much effective power as they have in the recent past. The problem I keep coming back to is that if your union leadership is arrrogant and ruthless in its tactics and objectives there is not much you can do other than some very extreme measures, as suggested by EMPLOYMENT, GROWTH, AND PRICE LEVELS 2647 Eepresentative Curtis, to break up the union or to impose very tough sanctions. But this notion of creating a climate or a general environment in which union leaders feel some sense of social responsibility is not just naive thinking or theorizing. There is a real possibility that the union leaders will come to recognize some of their obligations through the force of public opinion. This approach seems to me much prefer able to the use of direct sanctions. Senator Bush. I am glad to hear you say that. It gives us all hope. Mr. K i l l i n g s w o r t h . I would agree and emphasize the points that have been made. That our experience with legislative restrictions on really powerful unions in the past have not been particularly fortunate for a variety of reasons. There were some laws passed during the wartime period that were specifically aimed at the United Mineworkers and John L. Lewis. Some of the provisions of the Taft-Hartley Act in 1947 were aimed at some of the more powerful unions. I presume that the Teamsters helped to pass, in effect, the restrictions that were recently added to Taft-Hartley. Our experi ence has been that the most powerful unions with very able legal counsel, with great resources, can often fight a lengthy delaying battle against the application of such restrictions as these, whereas it is the weaker unions that you didn’t have in mind at all that are strongly affected and sometimes almost fatally affected by this kind of restriction. I think this is something that it is important to keep in mind when we are thinking about additional restrictions. They almost of neces sity have an extremely uneven impact. Senator B u s h . I am afraid I didn’t quite get the implication of the weaker union having so much more strength than the bigger union. Would you phrase that a little differently for me? Mr. K i l l i n g s w o r t h . I d in ’t m ea n to sa y t h a t th e w e a k e r u n io n s h a v e g r e a te r s tr e n g th . I sa y t h a t a le g is la tiv e r e str ic tio n lik e th e lim ita tio n o n se c o n d a ry b o y c o tts o r o r g a n iz a tio n a l p ic k e tin g h a s a m u c h h e a v ie r im p a c t on th e w e a k u n io n s th a n on a s t r o n g u n io n . For example, the closed-shop prohibition of Taft-Hartley in 1947 is virtually ineffective in the building construction industry where the unions are old, powerful, well established. It has had consider able effect in the South where unions are weak and struggling to get a foothold. That is my point. Senator B u s h . I see. Thank you, Mr. Chairman. Eepresentative E e u s s . I wonder if there is not a difference of opin ion among the members of the panel on quite a fundamental question involving the market power of unions. As I understand the testi mony of Mr. Killingsworth and Mr. Pierson, you were saying that you found no really convincing evidence that unions have had any perceptible effect in raising the general level o f wages. Mr. Segal, on the other hand, laid it on the line that particularly in the admin istered price, concentrated, pace-setting industries, and alro in strongly organized local trades, there did seem to be a body of evi dence that union collective bargaining had advanced wages over what they otherwise would be. My first question is, am I right in detecting discord ? 2648 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS Mr. P i e r s o n . I think yon caught a minor difference. I think the main point, Mr. Reuss, is that it would be dangerous to base public policy on some recent shifting around of the data for the period from 1953 to 1957. We have to be very careful when we analyze an in volved, fast moving economic situation. Mr. Segal made the comment that employment has not risen very much in steel. On the other hand, capital expenditures in steel have risen sharply during the 1950’s. This indicates a level of profits and a demand from that influence that might offset his reference to employment. I mention this just to indicate that it is a very complicated matter and it would be dangerous to change labor law for the next 20 years on the basis of a 4-year statistical showing. Representative R e u s s . I remarked, too, that despite some difference in observation on the part of you panel members as to whether or not collective bargaining produced a substantial increase in wages over what otherwise might have been the case, your recommendations for policy do not seem to differ very much, if at all. You are all against wooden use of the antitrust power. Mr. Segal comes forth with a little more specific line of action, it seems to me, on toning down wage increases by stiffening management’s back in the concentrated industries. Mr. S e g a l . That is right. Representative R e u s s . I was impressed, Messrs. Killingsworth and Pierson, by Mr. Segal’s comparison with the 1920’s. I f I am not mistaken, Mr. Segal, you said that wage increases in the twenties when unions by and large were pretty weak—I think you used the years 1923-29—were very much less percentagewise than in postWorld War II. Mr. S e g a l . Between the two recessions. I refer to the years 1953-58. Representative R e u s s . That may be one of those short little periods that your colleagues objected to. How about this, gentlemen? Mr. S e g a l . I think the very same thing applies if you take 1947. But the union influence in early postwar years is more doubtful. Still a recent study by Ozanne appears to show some influence of unions on wTages between 1947 and 1957. I would like to comment very briefly on Professor Pierson’s remark about the profitability of the steel industry. It seems to me that this profitability is consistent with my argument because obviously profitability is one of the things that unions look at when they ask for wage increases. Indeed, profit ability, I think, is one of the incentives that make these unions where they have market power to press for wage increases. Representative R e u s s , What about that point, Mr. Killingsworth, and Mr. Pierson, that the 1920’s and the post-World W ar I I period, are broad periods. Mr. P i e r s o n . I think that is a very illuminating comparison. I think Mr. Segal's comments are essentially correct, as I understood the two periods. However, the source of the increase in wages in the latter period is open to considerable controversy. For one thing, unemployment was a good deal less in the 1950’s than it was in the twenties, as far as available evidence would indicate. The rise in wages in service industries and among white collar and salaried groups was a good deal greater than in the twenties. This meant EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2649 more pressures on wages generally, that is, in order to keep a wage bal ance, wages in the heavy and unionized industries had to move along in order to maintain that balance. Representative R e u s s . May I stop }^ou right there because I am a little surprised? I would have thought on this business o f the service industries bringing pressure to bear in the last few years, that hardly anyone would have given up being a steelworker to go to barber college let us say. Barbers service charges have risen largely because steelworkers were making a lot of money and barbers felt they could get, and did get, some part of that. Do you really think it constituted a pressure on steel wages ? Mr. P i e r s o n . I think the word “ pressure” would be too strong. In the long period of narrowing of wage differentials, say since 1940, there is something of a reverse process now underway. The narrow ing process occurred during the forties when wages for unskilled labor, white-collar workers, and, in service industries were pulled up more in line with the high-wage industries. Now, you are getting a widening back again toward what might be called a more normal relationship, although Mr. Segal apparently interprets it as undue union pressure. It is debatable. Representative R e u s s . I gathered from what has been said that this difference in analysis is emphatically one of degree, and nobody is in violent disagreement. Mr. Segal simply sees a somewhat more drama tic effect o f collective bargaining on wages than you gentlemen do. Mr. S e g a l . That is right. I would like to add this comment. The fact that certain industries have always been high-wage industries is very well known. In fact, way back in 1900 when the Immigration Commission of the U.S. Senate examined the wages of the immigrants they noticed this. I remember this particular phrase: “ The very same Lithuanian gets a much higher wage if he gets a job in the steel in dustry than in the textile industry.” On the other hand, I would like to say this : It seems to me that while the preservation of the rela tively high-wage positions certainly is one of the wage policies of the well-entrenched oligopolistic firms, it seems to me unlikely they would do it where they were at the same time permanently raising the level of costs and increasing the wages beyond the productivity. They would be stuck with these high costs. They would more likely be doing it in a period when there would not be so much pres sure on the cost. I would think that in this particular period in w^hich the wages in the concentrated industries exceeded productivity, this must be to some extent attributed to the activity of the unions. I also emphasize the interaction. There were some shortages. I was at that time, in 1956, studying the labor market of the New York metropolitan region and there were very obvious tight market situa tions in such industries as textile dyeing and finishing in Passaic, or in some of the lower paid clerical occupations. You had then an inter action which together put the wages above productivity increases. Representative R e u s s . Mr. Curtis. Representative C u r t i s . I have one comment I wanted to make first. Last week, when we were holding hearings in the area of antitrust legislation from management’s standpoint, I could not help but feel impressed, or I was impressed, with most of the recommendations that we needed to do something further in the 38563— 59— pt. 8-------14 2650 EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS antitrust area there is a selective way. It is not to do it generally but in selective areas. It seemed to me that a similar type of se lectivity might be indicated here. I don’t know. I do appreciate the panel’s papers in clarifying this issue and bringing it out where we might see it. There is one other general subject that I would like to pose. First, I want to make another comment which I liked about the papers. I think all of them more or less agreed that the market power o f the union was effective—in fact I think you said it—where the union was located. I was trying to get the panel this morning to bring this out. The other thing that I tried to bring out in previous panels in this question of the labor mix between unskilled, semiskilled, and skilled on up to technicians and professionals. Doesn’t the mar ket power of the union depend considerably on whether or not the labor it is selling is highly skilled, or even in the building trades they have the day laborers organized? Isn’t the market power af fected considerably by what kind of labor they are trying to sell in relation to skilled, semiskilled, and technical skills ? Mr. K i l l i n g s w o r t h . Offhand I think I would say that is less true today than it was 30 years ago. Representative C u r t i s . I s that because we have more skills today and there is less market for unskilled labor ? Mr. K i l l i n g s w o r t h . N o . My reason for saying that, and this is purely an offhand reaction, is that it is no longer fashionable to under take strikebreaking activities. The reason why a skilled craftsman would have greater bargaining power would be the greater diffi culty o f replacing him, because you don’t go out on the street and pick up skilled men easily. Representative C u r t i s . Maybe I can pose it in this light. My hypothesis could be wrong and, if it is, correct that. I understand that in the U A W there has been a recent development in the past few years where the more highly skilled members actually broke loose from the U A W because they felt they were being used as the primary force to bring about wage increases for the semiskilled or not as skilled workers down the line. Actually there was a compres sion of the differentials between the wage rates of the skills. It would suggest to me that the marketing power o f the skilled group was a great deal more organized than that of the semiskilled. Mr. K i l l i n g s w o r t h . On this U A W problem I don’t believe that the skilled group actually broke off. Representative C u r t i s . I think they did. Mr. K i l l i n g s w o r t h . I think there was some talk of the formation of a society o f skilled trades and it ultimately came to very little. The skilled tradesmen were given more autonomy within the U AW . I think this development probably did indicate some degree of dis satisfaction within the skilled trade groups within the U A W , and the dissatisfaction probably stemmed from the narrowing of differ entials. The interesting thing there is that this narrowing of differentials appears to be a worldwide phenomenon. We have had perhaps less of the narrowing of differentials in this country than in a great many other highly industrialized countries where the U A W doesn’t operate. Obviously, it was not merely a matter of U A W policy that was in EM PLOYM ENT, G R O W T H , AND PRICE LEV E LS 2651 volved here. You can look elsewhere in American industry. For example, the locomotive engineers have been extremely unhappy in recent years that their percentage differentials over the lower skilled groups in the industry have been squeezed and squeezed even though they have their own independent union. There are a great many factors that tend to account for this narrowing. Representative C u r t i s . Y o u still don't feel that the marketing power, which is the basic question, is affected greatly by wrhether or not the union is representing high skills or semiskills. One way of attacking the same problem is this: Does a vertical union have a d if ferent power in the market than a horizontal union ? Mr, K i l l i n g s w o r t h . I think I would want to answer that by say ing that the skill factor may be one component in bargaining power. As someone was saying earlier, driving most kinds of trucks does not take a tremendous amount of skill. Yet the Teamsters Union is extremely strong and has great bargaining powers in most situations. Representative C u r t i s . That is because of the other factor we dis cussed, the kind of industry. The marketing power is affected by the marketing power of the industry. I was throwing this other ingredient in as to whether or not it was not really an active and important ingredient. Mr. S e g a l . I think I would agree with Professor Killingsworth. In other wTords, in 1870 or so when the mule spinners in Fall River went out on strike they had considerable market power because the French Canadians who were imported to break strikes could not take their place. I think today, since you normally cannot really organize such a large strikebreaking activity, the situation w^ould be quite different. Also, we have many examples besides the Teamsters. I think of industries in which relatively unskilled people, or a union of unskilled people, has very considerable market power; for example, in some local industries such as breweries. There are many people in the breweries who are not highly skilled. In New York they start with very high wages even though they have never been inside a brewery. This is a result of a very tight organization combined with a very advantageous economic position, in which the percentage of the labor cost is very low in the total cost. You have only a few breweries which control a very large share of the sales in the metropolitan area. It is a local industry, protected by transportation costs from extreme outside competition. More over, with respect to this example of the U AW , you can find examples where the unskilled people have threatened with breaking away and led to a change in union policies and in union personnel. Representative C u r t i s . But the key would be: Was that power they had as strong a power as that which the skilled would have? I doubt it. I think you have posed a very proper and good argument on strikebreaking. Take, for example, the strike in the telephone company; there, through the retention of very skilled people—in fact, management—they were able, without strikebreakers but in effect having their skilled people do jobs that normally they would not do, to keep the thing going. That is another reason it would seem to me that the skilled people would have more marketing power. I don’t know. I am posing this for discussion and I appreciate your answers. Did you have any comments to make % 2652 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS Mr. P i e r s o n . No ; I don’t think so. Mr. F r u c h t . I w a n te d to ask you whether the marketing power is affected by the cycle of the economy. In other words, boom and bust. Mr. P i e r s o n . I would say very definitely that the market power of unions, especially as it becomes a nationwide phenomenon, is very closely associated with the level of employment, the volume of spend ing, the prospects for prices, and, more particularly, the past move ment of prices where the cost of living is rising. It is in that environ ment that you get examples of union market power. The difficulty, however, is that in that environment it is unclear what is the moving force and whether the wages would not have risen in this inflationary environment anyway. Mr. F r u c h t . I am sorry, my point is with reference to the skilled and unskilled even perhaps within a union. I am wondering whether or not the skilled part of a labor force will have greater power rela tive to the unskilled part of that same labor force in a depression than it does in boom. I f you have an answer here, I would like you to relate the whole question of union power in a boom. Mr. P i e r s o n . There, again, I think the general pattern that is usually accepted is that in a period of very full employment the greatest relative scarcities are likely to occur in the unskilled field simply because of the change from the previous phase of a cycle when unskilled labor was likely to be in much less demand. Such a shift, as in the W orld War I I period, is a very striking cyclical phenom enon and you get the narrowing that Mr. Curtis referred to between skilled and unskilled. Mr. F r u c h t . Would y o u relate this to the salaries of instructors, assistant professors, and full professors? It seems to me that it is very relevant. Mr. P i e r s o n . It is not only relevant, but painfully so. Mr. K i l l i n g s w o r t h . May I comment on one thing, Mr. Chair man, that seems to me to be a recurrent theme. I would like to ad vance the thesis that perhaps we have tended to make unions scape goats in a sense. We have tended to impute to them certain things which would have happened anyway. I would like to go back for just a moment to this question of the bricklayer who lays 400 bricks a day, let us say. In my travels around the country, I have visited quite a large number of factories, and it is simply astounding the extent to which you find pegged production on occupation after oc cupation in plant after plant. The vast majority of these limitations on production are not union imposed. As a matter of fact, a very interesting study was made a number of years ago called “ Limitation of Output by Unorganized Workers.” That author reached the conclusion that unorganized workers probably tended to limit their output more severely than the organized workers. Representative C u rtis. T o keep the jo b g o in g more or less, Mr. K i l l i n g s w o r t h . To keep the job going. They felt more in secure. They don’t have the same kind of job protection that the unionized workers have. In other words, what I am suggesting is that here we have what is apparently a fairly universal human tend ency not to do too much. We see the thing in the classroom where the majority of students will mutter about the “ curve buster,” this fellow who goes and bones up and gets a very high score and makes EM PLOYM ENT, G R O W T H , AND PRICE LEVELS 2653 everybody else look bad. This is a fairly universal problem. When we say that unions have imposed limitations on output, we are at tributing to them powers and designs and policies that they do not really have. I have some skepticism, too, about attributing the in crease in wages in this postwar period to unions. I know that the automobile industry, for example, and a number of other industries, were spending a great deal of money in very intensive recruiting efforts, indicating rather pressing labor shortages during that period. My guess is that wages might have gone up just about as much without unions during that period as with them. That is a guess. I do not think I can prove it any more than Professor Segal can prove his interpretation of the thing. To reiterate, I think there is some tendency to attribute to unions things and policies that existed long before there were unions and that would have come about without unions. Eepresentative C u r t i s . I appreciate your comment. In fact, I agree with that. I think it is very basic. The unions will tend to reflect not their own decisions but really the economic forces. The only comment I would make is that what we are trying to dig into is what differences does it make when there is an organization dealing with these f orces as opposed to unorganized. In the bricklaying thing, it is true that you do have the same tend ency in unorganized people, but occasionally you find some character that will go out and want to lay more brick. But when you get a union, all those characters are just knocked out. You do not have any of that. That is a difference, although you are dealing with a similar thing and you are dealing with human nature or economic forces. The organization does tend to produce conformity, which, I might add, sometimes—in many instances—is good economics. I am not at all sure that we do not do better by knowing that you can get 400 bricks laid by whomever you hire. Eepresentative E e u s s . Mr. Pierson has a comment and then, per haps, we should bring the hearing to a close. Sir. P i e r s o n . I would like to throw out one suggestion that grows out of the great dearth of authoritative data in this field. While we have grounds perhaps for some interesting speculations we are not on a firm footing such as we are dealing with the market power of employers in the antitrust field. Given this lack of basic data in individual industries, I think some authoritative investigations should be made. I f this committee were to ask the Federal Trade Commission, for example, to dig into some of these alleged practices in particular industries and particular cities, the results could be very helpful. Eepresentative E e u s s . Thank you, gentlemen. We are very grate ful to you. The hearing is now adjourned until 10 o’clock tomorrow morning at which time we will meet in this chamber. (Whereupon, at 4 :15 p.m., the hearing in the above-entitled matter was recessed, to be resumed at 10 a.m., Thursday, October 1, 1959.) EMPLOYMENT, GROWTH, AND PRICE LEVELS THURSDAY, OCTOBER 1, 1959 C o n g r ess o f t h e U n i t e d S t a t e s , J o in t E c o n o m ic C o m m it t e e , Washington, D.C. The committee met, pursuant to recess, at 10 a.m., in room P-63, the Capitol, Hon. Richard Bolling presiding. Present: Senator Bush; Representatives Bolling, Reuss, and Curtis. Representative B o l l i n g . The committee w ill be in order. The first witness will be Mr. Robert Ferber, on the subject of “ The Service Sector: Longrun Trends and Relationships to Inflation and Growth.” Mr. Ferber is from the bureau of economics and business research o f the University of Illinois. Mr. Ferber, you may proceed as you wish. STATEMENT OF ROBERT FERBER, BTJ&EATJ OF ECONOMIC AND BUSINESS RESEARCH, UNIVERSITY OF ILLINOIS Mr. F e r b e r . I have submitted a brief statement which I think would take more than 8 minutes by itself to read. I have selected various abstracts. Representative B o l l i n g . I f it is not too long, you can read the whole statement. Mr. F e r b e r . Thank you. Service activities have come to occupy a position of increasing im portance in the American economy. Thus, during the 1950’s the average annual rate of increase in consumer service expenditures after deflation for price increase has been a substantial 4 percent; the average annual rate of increase in consumer goods expenditures during this period has been about 2.3 percent. A t the same time, indications are that services will continue to grow in importance with the rising standard o f living. PAST INCREASES SUBSTANTIAL The growth in service activities has been accompanied by sub stantial price increases. A general indication of the extent to which these prices have risen during the past two decades is provided in exhibit I. As is evident from this exhibit, prices of services generally have increased at an average annual rate of about 3.6 percent between 1936 and 1950 and at an annual average rate of about 3 percent during the 1950’s, It is also clear from this exhibit that prices of different services have not increased uniformly. Not only have prices for the same type o f services increased at different rates during the two periods 2655 2656 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS shown, but disparities are evident in the extent of increase of different prices. Prices which increased most rapidly between 1936 and 1950 include domestic service, personal care, private education, foreign travel, recreation, and religious and welfare activities. A ll of these prices increased at a rate of approximately 4 percent per year or more. During the 1950’s only three sets of service prices increased at a corresponding annual rate; namely, purchased transportation, per sonal care, and medical care. Note that there is a little duplication between the two categories. In fact, some tendency is apparent for prices which increased at a rapid rate during the earlier period to increase at a lesser rate during the fifties, whereas prices which had increased relatively little during the earlier years rose at a much more rapid rate in recent years. Particularly noteworthy in this con nection is the decline in the rate of price increase of domestic service, o f foreign travel and of private education, while at the same time an acceleration of the price increase has been evident in utilities, in public transportation, and in automobile operation. Price increases have accounted for large portions of the rise in total service expenditures in these two decades—exhibit II. In some in stances all of the increase in expenditures is attributable to price increases. This is the case for public transportation and for cloth ing and jewelry services—laundry, cleaning, watch repairs, etc.— during the 1950’s as well as for domestic service during the entire period since 1936. Indicative further of the extent of the effect of prices during the past tw^o decades is that in the great majority of the comparisons shown in exhibit II more than one-third of the rise in expenditures that has taken place is accounted for by price in creases. In interpreting these price data, it should be noted that improve ment in quality of service or goods often cannot be incorporated in a price index. For example, a price index of air travel can hardly include an allowance for the greater comfort and speed at which air travel is now possible. In addition, a price index may overstate the true amount of increase because of new products or services. A new good or service, such as a new type of medical treatment, would only be included in the regular price indexes after it has gained fairly wide use, and, as a result, after its price has come down markedly. This is not the place to discuss the various conceptual problems in volved in the preparation and interpretation of a price index. Suffice it to say, however, in many areas that available price indexes may be overstating the extent of true price increase in recent year. DEMAND FACTORS D O M IN A N T The price increases that have occurred during both of these periods would seem to be explained primarily in terms of the dynamic inter action between a growing demand and a relatively stable supply rather than in terms of cost factors, with some exceptions. Thus, in areas like domestic service, medical care, and education, it is apparent that the available facilities have been insufficient to meet the demand. For various reasons the supply of the service could not be expanded quickly enough to meet the demand even over the period of two decades. As a result, the shortage of facilities has EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2657 brought about continuing upward pressures on prices, which have been manifested to the extent shown in the accompanying exhibits. In a similar w^ay, it would seem that the rise in prices of housing, of clothing and jewelry services, of personal care, automobile operation, foreign travel, and personal business have been brought about by rapid increases in the demand for these services relative to available supplies. This increase can be traced to the rapid rise in incomes and in standards of living coupled with the emphasis during past years on the manufacture of goods for war and defense purposes. Only in relatively recent years does the supply of services seem to be coming again into line with the demand for them. Cost factors would seem to be a primary determinant of the rise in prices of utilities and of purchased transportation. Because of the regulated nature of these activities, the usual demand and supply relationships are not operative. Assuring certain minimum rates o f return necessarily requires that costs become a major factor in price determination, at least on the upward side. Admittedly in both these sectors, the extent of price increase that did take place was probably less than what would have occurred had there been no Government regulation. Yet in the case o f purchased transportation it is clear that price increases were frequently brought about not because of in creased demand but rather as a result of reduced demand for the serv ice and a willingness on the part of the regulatory body to offset the reduced revenues by higher prices. All this is not to imply that costs have no influence in determining the price of services not under regulatory bodies. Costs are influ ential in determining the extent to which prices are increased in the light of price or cost increases in other areas. Thus, the rising wages of manufacturing employees helped bring about higher pay for do mestic help because of the continuing drain on the supply of domestics. In a similar fashion, increased construction costs of new buildings have entered into higher educational fees. The fact remains, however, that for most services, costs are more important on the downward side than on the upward side. In this connection, it is worth noting that costs, as applied to services, repre sent almost entirely labor costs. To be sure, depreciation and certain rental and overhead costs are incurred in the operation of some serv ices, but these are primarly in the nature of fixed costs. Because of the predominant role of labor costs, a rise in the price of a service once it has taken place is likely to be rescinded only under extreme cir cumstances. This is true not only of services which involve the pay ment o f wages but also of services which involve professional care such as medical care and education. For this reason, costs in the service area appear to act essentially as a floor to prices, helping to raise price levels and serving as a barrier to price reductions. PRICES AND GROWTH Price increases are likely to promote growth only if the supply of services, or goods, can be readily expanded as a result of these in creases. In the case of services, this seems to be more the exception than the rule. Thus, the 56 percent increase in the price of purchased travel during the 195G’s has been associated with a sharp decline in the extent o f such travel. Here, it is not unlikely that price exerted 2658 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS a negative influence, leading people to abandon public transportation in favor of the automobile. In medical care, price increases have had little effect on the supply of physicians. In addition, in the areas where service activities have expanded most rapidly during the past decade—household utilities, foreign travel, and personal business (exhibit I I I ) —the extent of price increase has been relatively little. On the other hand, in some areas it is evident that the price in creases may have acted to maintain growth by preventing the transfer of resources to alternative uses which had experienced prior sharp increases in price. Domestic service is one such example, as noted previously. A similar statement might be made for education: had teachers’ salaries not been raised to some extent, then the teacher shortage would be even greater than it is now. Nevertheless, the studies that I have made on the subject would seem to indicate that, within the present range of observation, with a few exceptions the growth of services is influenced primarily by the rise in the standard of living, by the distinctive factors associated with that standard and by changes in taste. For example, fluctuations in expenditures on recreation, adjusted for price variations, have been found to be related closely and inversely to the number of hours of work, rising over time as the number of hours of work per week de clined. In general, where prices do influence growth, the effect is most likely to be of an indirect nature, a price increase acting to pre vent transfer of resources to other activities where price rises have already taken place. In evaluating the growth of a particular service, it is also important to consider the extent to which other services or goods may compete with it for the same function. For example, there has been very little increase in consumer expenditures for laundry services during the past decade. This is not because people are no longer interested in wearing clean clothes, but rather it is the result of the increasing shift o f the laundry function from outside establishments to home washers and dryers. In a similar manner, there is little doubt that expendi tures for spectator fees and for movies would be much higher than they are currently were it not for television, which in turn has brought about a tremendous increase in the use of repair services. Both o f these approaches have their distinctive uses. The functional approach is perhaps more pertinent from an overall point of view, whereas the individual item approach is primarily applicable from a business point of view. OUTLOOK FOR TH E FUTURE The average anual rates o f growth in different service activities dur ing the past two decades after adjustment for price increases are shown in exhibit III. A ll things considered, the rates shown for the 1950’s are likely to continue at least for the next decade as well, with the follow ing modifications: Housing activity is likely to slacken in the next few years and in crease again sharply thereafter, in line with the expected fluctuations in family formation. The decline in public transportation expenditures will probably come to an end, primarily as a result of increased business o f airlines and local transit services. EM PLOYM ENT, G RO W T H , AND PRICE LEVELS 2659 The rate o f growth in foreign travel in all likelihood will diminish and approximate more closely to the 1.8 percent annual growth rate of the previous decade than the 9.4 percent o f the current decade. The rate o f growth in pricate education expenditures and in personal business activity will probably increase further, the former because of the increased interest and emphasis on higher education, and the latter because a growing standard of living tends to bring about a more than proportionate increase in the use of consumer credit and of other busi ness services. A t the present time, further substantial increases in price appear likely in the areas of medical care, public transportation, and educa tion. In each of these areas, the supply of the service promises to remain relatively stable (with the primary exception of air trans portation) in the face of a rapidly expanding demand. Whether at tempts should be made to prevent such price rises from taking place would seem to be in part a matter of longrun national policy. In each instance, a balance has to be sought between the benefits of price stability as against possible restriction of supply because of price stability—the latter being especially serious if other prices should be moving up too. Government action would seem most desirable in the ease of services, or goods, where price increases are brought about by artificial restriction o f supply. Such action might be in the direction of either promoting price stability or increasing available supplies. On the whole, with the exception of a few scattered areas, there seems to be relatively little danger of substantial price increases in consumer services during the 1960’s—barring any major catastrophes. A t the same time, in view o f institutional rigidities that exist in so many services, price declines are not at all likely either. The outlook for the next decade or two in the service area, therefore, appears to be for a gradually rising price level. (Additional comments of Robert Ferber:) I should like to add the following brief comments to the discussion at the hearing on the services sector on October 1. These are comments which I did not have a chance to make at that time. 1. Question might be raised whether the presumed shortage of applicants for medical schools in recent years might not be attributable in part to the restric tive practices of many of these schools in limiting numbers of students from certain areas or population groups. 2. Professor Ginzberg’s remark that the growth of the services sector may be relatively less in the near future because of the concentration of the popula tion increase among younger people is not clear to me. For one thing, substantial increases in the population of older people is in prospect too. For another thing, more than half of all service expenditures relate to housing and household op eration. The growth of such expenditures is hardly likely to diminish because of any relative increase in the number of younger people. The same is true for education, recreation, transportation, both public and private, and personal care, which constitute other major portions of the services sector. 3. I am not convinced that the presence of a large number of self-employed people contributes to greater flexibility of the economic system. If a few taxi drivers experience bad business, undoubtedly they can increase their income somewhat by working longer hours. It is an economic fallacy, however, to generalize this by saying that if all taxi drivers experience poor business all can increase their income by working longer hours. The same reasoning would apply to most other self-employed groups. 4. The implication through most of the discussion was that medical resources are most desirable where they are most likely to contribute to increased overall productivity, and hence that emphasis on medical care for younger people would be relatively more desirable. This would seem to be an extremely narrow cri 2660 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS terion to use in allocating medical resources. From the point of view of economic welfare, the health and well-being of all the people should surely be as important as the extent to which medicine can be used to increase, say, manufacturing output. 5. It should be recognized that the discussion of medical care was focused pri marily on conditions in New York and in the large eastern cities. These cities clearly have medical facilities which are far superior to those possessed by the rest of the country, particularly those of smaller towns and rural areas. It is doubtful if everyone in these areas receives all the medical attention or surgery needed at the proper time. Of course, people that do not receive such care are not likely to be around to talk about it. 6. For the same reason, the fact that municipal hospitals may no longer be necessary in this day and age seems to me rather questionable. In 1958, the last year for which figures are available, roughly one-third of American families were earning less than $8,000 per year, most of them undoubtedly concentrated in the rural areas. In New York City, it might be noted that there has been sub stantial immigration in recent years, mostly by people with virtually no resources from areas like Puerto Rico. 7. It might be noted that in Champaign-Urbana and in other parts of the country, public school students do pay for books and supplies. There is no evi dence, however, that this policy contributes to the desire of the parents or the students to get more out of the educational system. About all that can be said for the practice is that it helps to reduce school costs somewhat, to the extent that some parents may insist that the students take better care of the books. At the college level, it seems to me that students are generally more interested if they have to pay for their education, particularly if they do the paying and not the parents. For this reason, among the many unwritten laws that I would like to see passed is one which would require that all college students earn at least a certain portion of their educational costs by working for it themselves. This would go a long way toward eliminating students that go to college just for the joyride. In any event, it must be recognized that the ultimate solution for having students get more out of education, whether it is at the secondary level or at the college level is in raising standards. Such a policy is not inconsistent with free education at the public school level and would certainly do a lot to assure better use of college and university facilities. (The exhibits referred to follow :) E x h i b i t I . — E s t im a t e d c h a n g e i n p r ic e o f c o n s u m e r s e r v ic e s , b y c a t e g o r y , 1 9 3 6 - 5 8 [Percent] Relative increase in prices Average annual rate of increase Expenditure category 1936-50 H o u s in g ____ _ __ _____ _________________ _________ Household utilities_____ _ ________ _____ ________ Domestic s e r v ic e ____________ _____ __________________ Clothing and jewelry se rv ice s _______ __________ . Personal care __ ___________________ __ _ Medical care ________ _________________ ________ __ User-operated transportation________ _______ _________ Purchased transportation ____________ _ ________ __ . _ Foreign travel ____ _______________ Recreation. _________________________ _________ _____ Private education________________ _____ ______ ____ Personal business ________________________ ________ Religious and welfare _____ _______ _________________ All services - _ _______________ ________________ All eroods____ __ __ ________________ ______ ____ Goods and services_____________________ ________ 36 9 184 69 116 54 44 49 89 74 104 48 117 i 63 i 94 i 83 1950-58 20 14 33 31 46 39 34 56 4 34 23 29 32 i 27 i 15 i 19 1936-50 2. 2 .6 7. 7 3.8 5. 7 3.1 2.6 2.9 4. 7 4.0 5. 2 2.8 5. 7 3.6 4.8 4.4 1950-58 2.3 1.6 3.6 3.4 4.8 4.2 3. 7 5. 7 .5 3.7 2.6 3.2 3.5 3.0 1.8 2.2 1 Implicit increases based on U.S. Department of Commerce data; not comparable with estimates for individual service categories. Source: Estimates derived b y methods described in Ferber, Robert, “ A Statistical Study of Factors, Influencing Temporal Variations in Aggregate Service Expenditures,” app. A , in Clark, Lincoln, edition, “ Research on Consumer Reactions.” E M P L O Y M E N T , G R O W T H , AND PRICE E x h ib it 2661 LEV ELS I I .— In f lu e n c e o f p r ic e i n in c r e a s e i n p e r s o n a l s e r v ic e e x p e n d it u r e s , b y c a t e g o r y , 1 9 3 6 -5 8 Expenditure category Dollar out lay, 1958 (in billions) Percent increase 1936-50 1950-58 Proportion of increase attributable to price rises (percent) 1936-50 1950-58 Housing------ --------------------------------------------------Household utilities____ _______ _ ____ ____ Domestic service______ _ _ _ _ _ _ ___ _ _ Clothing and jewelry services-------------------------Personal c a r e _____ ____ ____ ____ _____ _ Medical care- _ _______ ___ _ ----------- --User-ooerated transportation.__ _ _ _ _ _ ------Purchased transportation_____ . _ ___ Foreign travel ---------------- ------- --------_ -Recreation _____ ___ _ _ ------- ------------Private education------- ---------------- --- __ Personal business_____ _ _ _ _ _ _ _ _ ___ Religious and welfare------------------ ------------------ 38.0 13.4 3.5 3.4 1.9 13.5 5.9 3.2 1.9 5.5 3.4 16.5 3.9 170 183 163 238 126 216 303 159 142 177 228 140 173 77 99 30 20 82 83 111 4 113 39 75 102 60 20 9 109 43 95 38 28 42 71 54 55 44 76 32 19 109 147 63 55 39 106 5 88 37 36 59 All services-------------------------------------------A ll goods_____________________________________ 114.0 179.6 178 230 75 39 48 55 43 44 Goods and services___________ _______ _ 293.6 210 51 53 42 Source of dollar outlays: Survey of Current Business, July 1959, p. 17. 150-151. “ National Income Supplement, 1954,” pp. 206-207. E x h ib it “ U.S. Income and O utput," pp. III. — C h a n g e i n d e fla te d p e r s o n a l s e r v ic e e x p e n d it u r e s , b y c a t e g o r y , 1 9 3 6 -5 8 [In percent] Average annual rate of increase Relative change Expenditure category 1936-50 1950-58 1950-58 Housing _______ __ ______ _ Household utilities __ _ ______ Domestic service ____ Clothing and jewelry services_____ _______ __ __ _ Personal care-. _ ____________ ___ _ __ _ __ ____ ____ _ Medical care _ _ User-operated transportation.___ __ __ _ _ Purchased transportation_______ _______ __ _ Foreign travel___________________ _ . . Recreation. _________ ______ _____ __ ______ _ Private education___ _____ _____ _________ ___ _ Personal b u sin ess______ ___ _____ _______ ____ Religious and welfare _ __ _ _ _ _ _ ___ _____ 98 159 —8 101 4 105 180 74 29 59 76 63 26 -2 —8 24 31 58 -3 3 105 4 42 57 21 5.0 7.0 -.6 5. l .3 5. 2 7. 6 4.0 1. 8 3.4 4.1 3.6 1. 7 5.0 7.2 -.2 —1.0 2. 7 3. 4 5.9 -3 .6 9.4 .5 4.5 5.8 2. 4 All services. _____ _________ . _____ __ _ _ All g o o d s ______________ __ ____________ . __ __ 70 70 37 20 3.9 3.9 4.0 2.3 Goods and services_____________ ______ _ 70 27 3.9 3.0 Source: Exhibits ___ 48 1936-50 I and II. Kepresentative B o l l i n g . Thank you, Mr. Ferber. Our next wit ness is Mr. Eli Ginzberg of Columbia. I understand Mr. Ginzberg mailed copies of his statement in good time, but for some reason they have not arrived. STATEMENT OP ELI GINZBERG, COLUMBIA UNIVERSITY Mr. G in z b e r g . I have one copy, Mr. Chairman. I have entitled my remarks “ Manpower Aspects of the Service Sector of the Economy.” I want to go back a little bit and look at the personnel and manpower factors that loom so large in the service sector and pay less attention 2662 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS to the indexes of wages and prices, which I don’t study very care fully. My learned colleagues right and left know much more about them. I want to begin first by saying that I am not quite sure that the spectacular relative rate of growth of the service sector, the nongoods producing sector which has characterized the economy in the past 15 or 20 years, is necessarily going to be maintained during the coming decade. The reason I believe we will again see some further relative increase in the goods producing sector, manufacturing and construc tion, is that we are in for not only a continuation of a very heavy population boom, but that population increase is going to be con centrated in persons reaching marriageable age— that is between 20 and 30 when they are likely to purchase a large amount of durable commodities. You set up a household. You buy a lot of furniture. You buy a lot of durables. The current issue of the Chase Manhattan Bank Bulletin deals with this aspect of the changing nature of the consumption pattern. So I would suspect that we may see some shift back to a somewhat speedier growth of manufacturing, construc tion and transportation, which has lagged a little over the last years. I just state that by way of preliminary background. The next thing that strikes me is that from the point of view of the committee’s interest in stabilization a substantial service sector, which includes on the one hand Government and on the other hand finance, insurance, and trade implies that a large number of people have been shifted from hourly work to a salary base. I think the employment and the income of these people is likely to be more stable over the swings of the cycle, and may be a positive factor for the reduction in the cycle itself. Just consider expectations. I f a person has a regular job and is on salary, the likelihood that he will not be too frightened if the economy starts to go off and will maintain his level of consumption is much greater than if he is subject to layoffs as an hourly employee. I would suspect that the high proportion of persons in the service sector who are on salary is on the whole a contribution toward sta bilization of the economy. The next point that strikes me is that you have in the service sector a large number of self-employed persons which gives you another kind of flexibility. I do a lot of talking to taxidrivers in New York. Business gets bad. They can maintain their income by stretching their hours. An employee of a manufacturing plant can’t do that. A professional man also has more leeway in transferring hours into income than a typical employee of a manufacturing plant. So I would argue again that the service sector has, by virtue o f the large number of self-employed people, another kind of positive flexibility from the point of view of the economy. Then there is the question of the part-time worker in the service sec tor. The National Manpower Council’s recent report on womanpower shows that a high proportion of women are more loosely attached to the labor force than the typical male between the ages of 18 and 64— They tend to come into the labor force when business is good and then move out when it is bad. The impact on family earnings and on the economy of the whole is somewhat cushioned by this swing group. I would say that there is another way in which service industries contribute to the flexibility of the economy. Many offer seasonal em EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2663* ployment and can draw in people who do not have to work and in turn if they are let go, they are not solely dependent on their job. The next thing that strikes one is that many people in the service sector of the economy work for nonprofit organizations. That leads me to suggest that some of these price increases that we have seen over the last years are simply a lag in the adjustments of these wagestructures. That is, if you take a look again at the Chase Manhattan Bank Bulletin for September-October, the service index moved very slightly between 1939 and 1947, while the commodity index in general moved twice as fast or three times as fast, and the food part of the commodity index moved twice again as fast. What we have since 1947 is really a belated adjustment in the wage structure of the service industries which lagged in particular because the nonprofit sector of the economy was slow to adjust to inflation. As you know, the Government does not move very fast in raising the salaries of civil service employees and when you come to hospital and educational institutions, the lags have been even very serious. That means I agree with my colleague Mr. Ferber that I do not anticipate as large an increase in the years to come, because much of the lag has been made up in the immediate past. It is not a popular thing to say, but I would argue on the whole that women, who are schoolteachers, especially in the larger metropolitan centers, are now getting quite a good competitive wage. That may even be true o f small towns in the South. But if you take the larger metropolitan centers, there has been an adjustment so that a schoolteacher’s salary o f $4,500 to $5,000 a year for a 9-month year is not too bad. It may not be competitive for males, but it is competitive for females. The next thing that strikes me about the service sector is that partly because we have so many nonprofit institutions which tend to be capital poor, productivity tend to be low. Being a professor, I am impressed with the fact that only at a place like a university would you ask a man whom you pay a modest wage of ten or twelve thous and dollars to type his own letters. The kinds of supporting per sonnel that even a broken-down business organization would have on the payroll to economize the use of the more expensive personnel are scarce in nonprofit institutions. Being capital poor, these institutions squeeze their dollars and try to make them go as far as they can. From a productivity point of view, I think you have a substantial underinvestment in capital, with corresponding underutilization of personnel which on balance gives you a bad result. The next thing that strikes me, and Professor Ferber mentioned this, is that the whole concept of productivity in the service sector is a very complicated one to appraise. It is relatively easy to measure changes in productivity in goods production but I am impressed with the difficulties, let us say, in the medical field. The advances in surgery resulting from improvements in anesthesiology, the use of whole blood, and so on may mean the difference between survival and death on the operating table. I don’t know how you measure this in terms o f hospital price increases. My father underwent a prostatectomy in the 1930’s and even with the best medical skill in New York, it was a life and death operation. He pulled through after 3 months. Today this is a routine operation which takes about 4 or 5 days in the absence of special complications. 2664 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS I would like to stress that concern with price changes without close attention to what is being produced in the service sector really throws you way off base. The next thing that strikes me is that the very rapid growth of the service sector, especially in trade, sets the stage for almost in evitable technological breakthroughs which will result in economy in the use o f personnel. We have the classic case of the A.T. & T. which, had it not moved to dial telephones, would have required all the fe male labor supply in the United States today just to keep the tele phones operating. You can reach a point in the service sector where if you want to have a further expansion of service you simply can’t get it by adding people. I think the movements toward self-service stores, the pur chase of goods through machines, the use of the telephone for ordering, I think would seem more on the trade side to suggest that in the next decade there may be no comparable increase in employment in trade that we have seen in the past. There is a basic logic to the economy. After you reach a point where you can’t easily add more people, you are forced to get more economical. I never thought I wxmld see this trend around universi ties, but there is now the beginning of a glimmer of hope that the utilization of teachers will improve. For a long time it seemed hope less. I have been talking with my colleagues for 25 years, ever since I started to teach, that it doesn’t make much sense to have classes indiscriminately limited to 20 students or so. It makes very good sense to have some teachers who know how to lecture Avell to have a class of 150 students. I find it much more stimulating to have a large lecture class. It is true that for seminar work I want a smaller num ber of students. A pattern of work may give way very slowly, espe cially where you have strong professional groups, but it does give way. That is the important point. It finally gives way. Currently there is a considerable interest in experiment in the use o f teaching personnel. So I think we will see some technological breakthroughs in the use of personnel in the service sector. The next thing that strikes me about the service sector is the great difficulty of establishing standards to measure either the effectiveness of the personnel or the effectiveness of the service. Dr. Henderson, the famous biochemist at Harvard, used to say that as late as 1910 it was a matter of chance whether the average patient coming into contact with the average doctor would on the average profit from the exposure. I think that is true also about the average student coming into contact with the average teacher. He may still have on the average only a 50-50 chance of net profit from the exposure. There are subtle questions as to what constitute gains to the individ ual in service areas. It is obvious that children who are exposed to certain kinds of television shows may lose rather than gain. There are many more subtleties in the whole service sector than in the manu facturing end of the economy. One of the serious problems in the service sector is the development of sensible standards of output or of the effectiveness of the people who are contributing to the output. The public appears to want many services that it simply is unwill ing and unable to pay for, either in terms of personnel or dollars. EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2665 Among my current duties I am serving as Chairman o f the Commit tee on Studies for the White House Conference on Children and Youth that is coming off next spring. I have been reading the re ports of the 400 national organizations. The thing that strikes me as I read these reports is that everybody wants more people. They need more personnel. But we have still a finite population in the United States. I f you add up all the people that apparently are needed so that we can do better in preventing delinquency, expanding medical services, supporting people in old age, et cetera, you get to a figure greater than the total adult population of the United States. The thing that strikes me is that in the service area we have to re think what part of the services that we want are perhaps obtainable in part through the use of volunteers in their leisure time, after all we are releasing more and more the total work time of the American population. I was first struck with this when I went at the end of W orld War I I on an Army mission to Japan and went through some Japanese hospitals, I found that the family moved into the corner of the hospital room, set up a stove, remained to nurse the patient, and all the hospital provided was the surgical and medical specialist skill. But the family really continued to provide the essential services for the care of that patient. That seemed very strange until my wife was delivered in New York City, and wTe were unable to get any nurses to stay with her overnight. So I wTas the person who stayed on duty. This made perfectly good sense. I think that with a 40-hour week or less average work many people don’t know what to do with their free time. Many of the personnel demands for the service sector should be able to be met by new com munity organizations relying heavily on volunteers. That is hap pening all over the country, but we will have to do more. There has been, in my opinion, a spurious development toward excessive professionalization in some service fields. There is no reason why volunteers in mental hospitals cannot take an ambulatory patient to a ball game, and sometimes make a sizable contribution to his recovery, sometimes as great as skilled medical personnel. My friend, Jack Ewalt, who is professor of psychiatry at Harvard, and used to be the commissioner of mental hospitals in the State of Massachu setts, organized volunteers very successfully. There are simply not enough people in the labor force, nor can we possibly raise all the taxes for getting all the kinds of services w© would like to get. It makes good sense to me to think in terms of making more use of community resources than we have in many of these fields involving services to children and older people. The next point suggested by Professor Ferber, emphasizes that some problems in the service fields are due to the strength of the organized professional groups which frequently stand in the way of more economic organization of the pattern of work. I mentioned earlier the difficulty of changing teachers. It is very hard to get a faculty to do anything different from what it has been doing. Or to take a case from the field of medicine, just contrast what the armed services were able to do when they had medical personnel under their own control during World War II in the structuring of the flow of work and what you can do in terms of a private practice in which 38563— 59— pt. 8^— 15 2666 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS each man is his own entrepreneur. It is not easy to get a profes sion to reorganize itself along more economical patterns since it has interests different from that of the consumer. Nevertheless, given enough time something happens. The thing that has impressed me about American medicine has been the fact that since 1910, when Dr. Abraham Flexner’s famous report was issued, we have not made our great advances in medical care through the multiplication of doctors. W e have made it through the tremendous expansion in auxiliary personnel of improved quality in the education of physicians. That has been a very economical and sensible way to proceed. So despite the fact that you have here the most powerful group organized for the protection of its own position, nevertheless, there has been logic to the expansion of medical care which has meant that you now have a medical industry of about 2 million people, of whom the doctors represent, at the outside, less than 200,000. Senator B u s h . May I ask the gentleman just one question? Do we have a shortage of doctors, in your opinion ? Mr. G in z b e r g . I represent even a more—from one point of view— cautious position than the A M A on this point. The A M A now says we do have a shortage of doctors and is willing to contemplate the expansion of medical schools. Incidentally, there is a considerable expansion underway. A large number of the States that did not have medical schools are developing medical schools. A fair number o f the 2-year medical schools have become 4-year medical schools. Some o f the smaller medical schools have been expanded. So there has been an increase underway in the postwar period resulting in raising the annual output from about 6,000 to over 8,000 graduates per year. That is not small. Rep resent ative C u r t i s . May I interject another point? The 2year medical school, which at one time people thought was of no value,, has now been given the stamp of approval because of the fact that in your last 2 years you have a small class, and they want this new blood coming from 2-year schools. I say that knowing about the Dart mouth Medical School. Representative R e u s s . So that the interjections may be across the board, let me say I am with you 98 percent of the way. But when the time comes for questions, I do serve notice now that I am going to ask you to substantiate your feeling that we are doing all right on the number of physicians in this country in view of the fact that people are living longer, they are making more money and demanding more medical care, and medical care, as you have cogently observed, is so much better that we want it and need it. Mr. G in z b e r g . I will come back to that. The next point I want to make is that there are a large number of community resources and individual resources going into the training of the professionals that operate primarily in the service area. But even there we have begun to see some light on aiming at better utilization. Johns Hopkins is leading the way in the reduction of the length of medical education from 9 down to 7 years. They have decided that instead of 4 years o f college, 4 years of medical school, and 1 year of internship, they can condense this to 7 years. EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2667 This is now being copied by other institutions. O f course 9 years is an understatement; it is necessary to add 3 to 5 years of residency and 2 years in the armed services so that a man has a gray beard be fore he starts to earn a living. I think there is always a need to restudy the kinds of ways in which professions seek to establish standards which are often methods of controlling the supply. I think in the State of Wisconsin, the embalmers managed to get a licensing measure across in which you had to be a college graduate before you could become an embalmer. I have been arguing very unsuccessfully until recently with the schools of social work in and around New York that there must be a differ ence between a young woman of 20 or 21, a college graduate, who knows little about life, who goes on to social work, and a woman who comes back to school at 35 after she has brought up three kids. The older person asks to be given some kind of constructive credit for hav ing survived with a husband and three kids for 15 years. Professions tend to develop standards in the first instance, because they want to raise the quality of their service. But then they become the victims o f the standards which they do establish. I think there is a constant need for reassessment of our training and licensing programs. I am impressed from my management studies that we have become entirely too committed to the question of a man’s educational back ground, and assume that the more education he has, the better he will do, even where there is no objective evidence that additional educa tion will help him in his work. Obviously there are a lot of fields where the additional education is relevant. I f you go back to the mental-hospital problems, which require a large number of personnel, the really important point in the treatment of mental patients in terms o f day to day, hour to hour, is the quality and humaneness o f the hospital attendant. That is very much more a question of per sonality than it is a question of education. It is true that he ought to have a minimum IQ and have some capacity for further training. But the character, logical and emotional determinants are much more important than his educational background. I use this only as an extreme example to emphasize that American management generally is overimpressed with degrees irrespective of the work that somebody is to do. Macy’s made this mistake in the depression of 1933. It had an easy market and decided that it wTould take only Ph. D.’s to sell silk stockings. They were worse than a young girl out of high school or somebody who had not even finished high school. The presumption is that the automobile industry walked into a strong automobile union in Detroit by taking on people in the early thirties who were really overeducated, and who became much more frustrated with the work than less well educated persons. My final point simply is that the problem of supervision is a very difficult one in the service area. O f course, in the absence of good supervision the effectiveness of personnel is apt to be quite low and the real costs will be very high. We have been just through, as von, know, a hospital strike in New York. I am around hospitals enough , though I am not a physician, to know that the people who were being paid $35 a week were in many instances giving $15 worth of work a week. This is again a situation where the capital poor nature of 2668 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS nonprofit institutions catches them in a vicious circle. They have not been able to pay a living wage, and they have gotten only half of what they paid for in my opinion. It is a very serious problem. But how do you establish a system of effective supervision ? This is a very difficult question. It is difficult in any kind of large organiza tion like the military which I know considerable about. My general impression is that about 25 percent of the people do 75 percent of the work in most organizations. However, in a straight manufacturing operation work standards are easier to establish. In the service area it is hard to pin down the requirements to a specific person and check in terms of time and out put whether he has met them. I think, viewing the economy as a whole, that major gains can result from more intelligent management: more realistic requirements for entrance; more use of volunteers; better supervision; new and more imaginative patterns of work. One more word about hospital costs since you have a particular interest in that area. I am currently finishing a year’s study of the 10 Federation of Jewish Philanthropy hospitals in New York. The pressure on hospital costs has been in part a long delayed and neces sary adjustment in what a very low wage scale—the $35 weekly that I mentioned earlier. Representative R e u s s . For what ? Who gets $35 a week ? Mr. G in z b e r g . A fair number of people at the lowest level of gen eral service work; that is, the people who help bring the meals up to patients, the people who wheel patients around, cleaning women. Un fortunately a hospital needs many hands. They have attracted, in gen eral, the less capable members of the labor force, the most recent ar rivals to a large city. They have been so squeezed financially, or so they thought, that they could seldom attract anybody else. I think they made a basic mistake in management. They thought this is the best way to keep costs down. Senator B u s h . This is what the big strike was about ? Mr. G in z b e r g . That is right. The wage structure was ridiculous. Earlier, the hour structure was also ridiculous. Think of the nursing situation some years ago. You used to have nurses on 12-hour or even longer split-shift operations. What happened was that when the change came you not only had to raise the salaries of nurses to some thing like a living wage, but you also had to absorb the changes due to the reduction in hours. Representative C u r t is . I wonder if you will comment on the impact o f a lot o f volunteers in this whole thing on the labor force in hospitals ? Mr. G in z b e r g . Interestingly enough, I would say, for the general hospital, at least in New York— I have not looked at other hospitals recently—my impression is that the volunteers are most conspicuous with respect to certain types of specialized hospitals. The general hospital in New York, except in connection with social work—that is— the bulk of the work, the day-to-day routine work on patients— is not being done by volunteers. Only in the W orld War I I emergency did F ifth Avenue ladies handle bedpans. Representative C u r t is . Y o u do get a lot of that in your religious hospitals. EM PLOYM ENT, G R O W T H , AND PRICE LE V ELS 2669 Mr. G in s b e r g . I estimate that in the Catholic hospitals in New York City the uncompensated labor contribution of the nuns runs to over a million dollars a year. That is a very special group. I f you take the average voluntary hospital non-Catholic, what is really hap pening is that the volunteers are adding something to the quality of the service. They come and visit the youngsters when the parents can’t come. I f the patient has a special dietary problem they may help to feed them. It is really peripheral services. The bulk of the work in the general hospitals is performed by paid people. Representative C u r t i s . The reason I asked the question, in St. Louis the religious hospitals, and not just Catholic, Lutheran, Baptist, Jew ish, and so on, having talked to a lot of them, state that one of their problems seems to be the impact o f the volunteer, which evidently is a great deal more than the one you described in New York. Mr. G in z b e r g . I think New York, Philadelphia, Chicago, that is correct. I would say the eastern seaboard hospitals which I know pretty well do not generally rely on the volunteer in the general hospital. The next point I wanted to make is an interesting one, and has to do with hospital costs that come about from the fact that the hospital is not only providing medical care. The larger hospitals of the country have become teaching institutions and the training of all medical specialists in the United States is now hospital-based. After you have finished medical school and have a year’s internship, all the residency programs are being carried by hospitals. They have become educa tional institutions and have been absorbing a large number of costs which have nothing to do with their basic mission. I don’t mean that it is not better for a hospital to have a teaching program, but it has become a big educational institution. That is a very important factor raising hospital costs. This is another way of saying that as a country we have not faced up to how we are to pay for the costs o f medical education across the board, and the least logical w^ay to pay for them is to have the sick person carry those educational costs. The arguments with the various Blue Cross systems around the country have finally focused attention on what is a sensible and logical part o f this cost which ought to be put on the insurance system. You also have had some interesting demographic and social changes. In the old days no man was permitted to become a hospital intern if he was engaged, no less married. This used to be the first question o f an examining board. Today the average house physician, by 5 :30, or quarter to six, wants to go home and help his wife with the youngsters. That means the hospital has a problem in scheduling coverage from 6 p.m., to 8 or 9 a.m. You have had a complete change by virtue of the fact that the hospitals used to get a tremendous amount of free labor. An intern or resident used to be lucky to get 2 nights out a week. I think the next point about medical care is that it is inherently ex pensive. In a certain sense doctors are always working themselves out of the easy part of their work. As medical science advances, and you bring more and more diseases under control and a single shot or some pills, prevent disease. That means that doctors are working on the residual problems. They are the difficult ones. Take older peo ple : they require more hospital care. You prolong their lives. This 2670 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS means that costs will tend to rise. The minute you really succeed in one area, as in the control of polio, attention is shifted to something new. You start to do cardiac surgery. You have to face up to the fact that as a nation you can put as much or as little money as you want into making sick people, especially older people, more comfort able, but you are coming, and this is very important from an economic point of view, into the area of less and less production return. When you save children, you have their whole working life ahead of them, and you get a big return. When you save people in middle age, they make a social return in terms of employment, taxes, and so on. But as to older people, that is something else again. I remember very well a visit to a New York State mental institution. I started on on the top floor and walked down. I saw not only very ill mental patients, but physically dilapidated patients who were of no use to anybody, including themselves, who were going to linger on. With each improvement in medicine they were going to live longer. In the old days before the new drugs, they would get a cold, pneumonia, and die. W ith improvements in medicine they live longer. But we are coming increasingly into the nonproductive areas of medicine. This has not been well understood. It raises some very serious problems from the point of view o f the country. I have seen a large number of middle-cl ass families almost impov erished providing terminal cancer care for one of their loved ones. This becomes a very serious problem. It is increasingly a nonproduc tive expenditure. We have saved the children. We are increasingly saving the middle-aged people. We are moving up against old people. What we are really trading is a kind of easing of life in the terminal years o f life, and really trading one kind of illness for another. So that there are difficult problems ahead. I would expect medical costs to go up. It is quite clear that the question of standards in this area has much to do with the total amount o f money the country has to spend on medical care and other worth while purposes. Does it want to devote more to older or to younger people ? I think the question will get more and more serious because the numbers o f older people will increase. There is no promise yet in medicine that anything spectacular will prevent the slow decline of older people. I think the big need is to be as imaginative and as economical as possible. I think that is the challenge to try to develop as many sensible sys tems o f support for older people so as to prevent their institutionaliza tion where it can be prevented; that is to use home care programs in stead o f prolonged hospital stay; to enable them to visit clinics for examination and treatment purposes and then for them to return home. Unless we really design reasonably sensible and economical patterns, we will find our costs going very high. These are some of the points that I thought would be o f interest to you. Representative B o l l i n g . Thank you very much. Next is Werner Z. Hirsch, of Washington University. EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2671 STATEMENT OF WERNER, Z. HIRSCH, WASHINGTON UNIVERSITY Mr. Hirsch’s prepared statement follows : T h e P u b l ic S c h o o l s S ector : L o n g r u n T r e n d s a n d R e l a t io n s h ip s to I n f l a t io n an d G ro w th 1 (By Werner Z. Hirsch, professor of economics, Washington University, and Resources for the Future) in t r o d u c t io n Public primary and secondary schools are employing today more than 2 million men and women, who serve about 35 million pupils. This sector in 1958 expended more than 30 percent of all the funds spent by State and local governments; it accounted for about 3.3 percent of national income and 3 percent of gross national product. Its pivotal position and importance for the future of this Nation is all too clear and needs no elaboration. a look at the record Total current expenditure (plus debt service) for public primary and secondary education has advanced from $193.9 million in 1900 to $11.01 billion in 1958— about a 56-fold increase in 58 years.2 How consistent was this advance? A glance at table 1 and chart 1 points to the fact that during the first 30 years of this century, the advance was quite steady and persistent. Depression and World War II greatly reduced the rate of increase; but since the end of World War II the steady increase has been resumed. factors a f f e c t in g e x p e n d it u r e level While, no doubt, a large variety of factors can be expected to determine expen diture levels, the following are likely to be of particular significance: 1. Number of pupils in average daily attendance. 2. Sociological characteristics of population: (a) Age structure, e.g., primary versus secondary school population. (b) Geographical distribution, e.g., urbanization of school population. 3. Economic characteristics: (a) Price level of goods and services bought by schools. (b) Income level, i.e., ability to afford services. 4. Physical characteristics : (a) Productivity of school system. 5. Government characteristics. 6. Variety, scope, and quality of education services: (a) Addition or deletion of services, variety. ( b) More or less of a given service, scope. (c) Superior or inferior services, quality. 1 To be presented at the Oct. 1, 1959, hearings before the Joint Economic Committee of the Congress of the United States in connection with its study of employment, growth, and price levels. a There exist numerous expenditure and cost concepts that can be used. But since one of the important issues under consideration is how the cost of a given bundle of educational services has changed over time and what forces generated these changes, it appears appro priate to use total current expenditure plus a reflection of the cost of providing physical facilities. The latter can be approximated by a debt service figure. Thus, the unit of measurement throughout will be total current expenditure (plus debt service). 2672 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS T o t a l c u r r e n t e x p e n d it u r e ( p lu s deb t s e r v ic e ) f o r p u b lic p r im a r y a n d s e c o n d a r y e d u c a t io n , s e le c te d y e a r s , 1 9 0 0 -1 9 5 8 M illio n s Year \ 1900_______________________________________________________________ 1902_______________________________________________________________ 1910_______________________________________________________________ 1913_______________________________________________________________ 1920_______________________________________________________________ 1922_______________________________________________________________ 1930_______________________________________________________________ 1932_______________________________________________________________ 1940_______________________________________________________________ 1942_______________________________________________________________ 1946_______________________________________________________________ 1948_______________________________________________________________ 1950_______________________________________________________________ 1952________________________________________________________________ 1954________________________________________________________________ 1956_______________________________________________________________ 1958_______________________________________________________________ $193. 9 214.2 384. 8 473. 0 938. 5 1, 332. 0 2, 095. 0 2,147. 9 2, 192. 0 2, 378. 6 2, 995.1 4, 092. 4 5, 090. 7 6, 219. 8 7, 458. 3 9, 054. 6 111, 020.0 1 Estimate. Source : Based on data on worksheets, made available by the U.S. Department of Health, Education, and Welfare. BILLIONS OF DOLLARS E M P L O Y M E N T , G R O W T H , AND PRICE LEV ELS 2673 2674 EM PLOYM ENT, G R O W T H , AND PRIC E LEVELS The qualification of variables 1 to 3 is reasonably easy. However, distinct difficulties are encountered in relation to the other three items, of which Gov ernment characteristics are likely to be the least important. To start with, the three elements of item 6— variety, scope, and quality— might be considered. Over the years, the variety of services offered by schools has greatly expanded. For example, more and more services are offered which most likely do not directly affect education achievements. Among them are transportation of children in schoolbuses, feeding of children in school cafe terias, health services, etc. No one will claim that these services are not im portant ; and yet they are merely auxiliary in nature. Their expenditures have skyrocketed from zero in 1900 to about 1.5 billion in 1958, i.e., about 16 percent of total current expenditure (plus debt service). Insofar as the scope of educational services is concerned, it appears reason able to assume that the longer the school term, the more can be learned by the pupil. Major changes have taken place in the length of the school term, which expanded from an average of 144 days in 1900 to 179 in 1948— a lengthening of 24 percent. In order to determine the quality of education, it might be best to identify and then measure major ingredients of good education. But there exist some perplexing measuring and weighting problems, both of which are further aggravated by a general paucity of data. With full recognition of its limita tions, the number of principals, superintendents, and consultants per 1,000 pupils in average daily attendance is used in this study to represent quality changes.3 These are the educational experts who evaluate the curriculum and introduce changes, provided guidance, supervise teachers, etc., and their number is considered to be as reliable an indication of the quality of education as can be obtained at present— superior to any other simple measure, as for instance, teacher-pupil ratio. About the latter, Halph Cor diner of the General Electric Co. is reported to have made the controversial statement: “There are some educators who are actually proud of the declining ratio of students to faculty * * * To my knowledge, there is no other field in human endeavor which actually prides itself on declining productivity.” 4 The number of principals, superintendents, and consultants per 1,000 pupils in average daily attendance has about doubled since the turn of the century. At the moment, no simple way can be found to measure the productivity of the school system directly. But there is an indirect measurement which appears appropriate and which will be attempted below. A N E M P IR IC A L TEST In order to shed light on such questions as, Why has the public education sector had such large expenditure increases? and, What are the underlying relationships for the near term ? Two wrorking hypotheses have been enunciated and empirically tested. The first hypothesis is as follows : X la= daily total current expenditure (plus debt service) for public primary and secondary education per pupil in average daily attendance,5 is a function of Xs^percent of public high school enrollment relative to total public school enrollment.6 3 In a cose study of the St. Louis City-County area it was possible to use a more complex quality measure. Quality was measured in terms of the following six equally weighted subindexes: A— number of teachers per 100 pupils in average daily attendance. B— number of college hours of average teacher. C— average teacher salary. D— percent of teachers with more than 10 years of experience. E— number of high school credit units. F— percent of high school seniors entering college. For detailed discussion see Werner Z. Hirsch, “ Measuring Factors Affecting Expenditure Levels of Local Government Services” (St. Louis: Metropolitan St. Louis survey, 1957), 94 pages. * Quoted by J. A. Livingston in “ Get With It, Professor, Your Show’s Slipping,” the Washington Post and Times Herald, Nov. 27, 1957. 5 It would have been possible to have the number of pupils in average daily attendance, as well as the length of the school term, as independent variables, except that such a step would have further reduced the already very small number of degrees of freedom available with but 17 years. 6 This information is not available on the basis of pupils in average daily attendance, but the pupils in average daily attendance and the enrollment series appear closely correlated). E M P L O Y M E N T , G R O W T H , AND PRICE LE V ELS 2675 X s=per cent of pupils in average daily attendance living in urban areas. X 4— average annual salary for member of instruction staff. X 6— number of superintendents, principals, and consultants per 1,000 pupils in average daily attendance. In brief— X la= / ( X 2,X 3,X 4,X«). The following multiple regression equation was obtained for 17 selected years during 1900-1958: ■ .0Q2067X2 .005288X3 , .000441X4 , .022391X6 „ C0257) _ = " (.1069)“ 09895) (.02171) -(L1) The coefficient of multiple determination, adjusted for degrees of freedom lost— R 21 .2346— is 0.998 and is statistically highly significant. Thus, about 99.8 percent of the variation of daily total current expenditure (plus debt service) per pupil in average daily attendance is explainable in terms of these four variables of which salary level is statistically significant. The general picture changes little if expenditures for auxiliary services are deducted from other expenditures. If X lb— daily total current expenditures (plus debt service) minus expenditures for auxiliary services for public primary and secondary education per pupil in average daily attendance— is used as the dependent variable, the following equation is obtained for 1900-1958: nn.non .000603X2 ,0Q0336X3 , .000372X4 , .013847Xf) „ X u —.006929— ( 0030)- -(.0 0 0 5 8 ) + '(.9889) ^ 7.01110', Once more i22i.2346:=.998. The foregoing analysis leads to the conclusion that the cost of an education unit, so standardized that its variety and scope are held reasonably constant, and expressed in per pupil in average daily attendance terms, is overwhelmingly determined by the prevailing salary level of the instruction staff and perhaps the prices of all goods and services bought by schools. However, they, in turn, are determined by the general demand and supply situation and price level of the rest of the economy, as well as the particular demand for and supply of teachers.8 The first consideration apparently by far overshadows the latter. Or, to put it differently, a very large part of the increases in public school expenditures have been fed into this sector via the price level changes of the rest of the economy. Population growth and lengthening of the school term are assumed to be additional factors. 7 The figures in parentheses are partial correlation coefficients. Since there are 17—5 or 12 degrees of freedom, coefficients are statistically significant at an oo of 0.05, in case they are larger than 0.532. Statistically significant regression coefficients are underlined. 8 According to some calculations made by Roger A. Freeman, for instance, during 1929-57 teachers’ salaries advanced more rapidly than the earnings of all workers— 94 versus 82 percent. This indicates that the demand for and supply of teachers cannot be neglected. (See Roger A. Freeman, “ School Needs in the Decade Ahead” (Washington, D.C., Institute for Social Science Research, 1958), p. 1931. > 2676 E M P L O Y M E N T , G R O W T H , AND PRICE LEVELS The importance of salary level changes can also be illustrated by comparing 1900-1958 total public school expenditure (minus auxiliary expenditures) in current terms with those in constant (1954) dollars. This can be done by re ferring to chart 2, In current dollars the 1900-1958 increase was 48-fold, while in constant dollars the increase was merely 2.5-fold. If further adjustments are made for the number of pupils in ADA and length of school term, expendi tures in real terms are approximated. They too are represented in chart 2. Expenditures in real terms exhibit amazing stability during 1900-1958. For the years for which data are available, 1922 was the low year wTith $1.37 daily expenditure per pupil, and 1913 was the high year with $1.60. Over these 58 years, an overall decline of about 3 percent was registered. To the extent that the standardization of the educational unit has been successful and appropriate data in general were used, the analysis can reflect on the productivity variable as a residual. Apparently, productivity in the public schools has changed very little, if at all. This finding appears to bear out the claim made by Burek EM PLOYM ENT, G R O W T H , AND PRICE 2677 LEV ELS and Parker some 4 years ago that “The productivity of schools and teach ers * * * cannot and properly should not increase very much.” 9 IN C O M E E L A S T IC IT Y OF PU BLIC ED UCATION For projection and policy purposes, estimates of the income elasticity of pub lic education are truly relevant aids. Measuring income elasticity requires that a net regression coefficient relating expenditures to personal income— X 5— be computed and then multiplied by the mean personal income per mean expendi ture.10 This formulation assumes that “amount” of education and expenditures for education are highly correlated. To the extent that X s succeeds in measuring educational quality, it should be excluded from an equation designed to estimate income elasticity. Income changes must be permitted to find their expression in quality changes, which, in turn, can then lead to expenditure changes. With these considerations in mind, a new hypothesis was formulated and a multiple regression and correlation analysis made. The following multiple regression equation was derived : xn _ 1 , -004274 X 2 (.0024) .005200 X 3 , .000924 Z 5 (.0022) + (.8715) ' { ON } As before, the statistically significant net regression coefficient is underlined. R*21.235 is. 969 and highly significant at an X of 0.05. Income elasticity of public education was estimated to have averaged -f 1.09 during 1900-1958. Thus, during 1900-1958 a 1 percent increase in per capita personal income was on the average associated with a 1.09 percent increase in daily total current expenditures for public primary and secondary education per pupil in ADA. It assumes that the effect of other factors included in the analysis is held constant. It is only slightly above unit elastic and apparently lower than that of some other public services, not to speak of such consumer amenities as air conditioning, automobiles, golf, speedboats, etc. Such low elasticity is of concern to those who are convinced that improvements in public education are essential if the United States is to remain a leading world power, and that therefore an increasing portion of the American people’s income must be channeled into public education. Certainly, people’s attitude toward educa tion and our general tax system are mainly responsible for the prevailing in come elasticity of education. If public education is to be improved in the United States and more funds to finance education are to be found, serious con sideration must be given to changing both, including further shifts in the re sponsibility of financing education from local to the State and possibly to the Federal Government. T H E N E A R -TE R M PROSPECTS Making projections is at best a hazardous undertaking, particularly in a case where the phenomenon awaiting projection so greatly depends upon exogen ous forces, such as personal income and teacher salary levels. How hazardous predictions in the field of education are is well demonstrated by recalling James B. Conant’s 1938 prediction: “By 1960 or thereabouts we shall have a stationary population. The expensive pressure on our schools will soon be gone.” 11 In spite of serious misgivings, at the request of this committee 1960 and 1965 projections will be made with the aid of two of the equations given above. The fact that predictive equations have been derived offers flexibility and makes possible later projections by others who might want to change the assumptions. By and large the following projections are based on the same general assump tions that were made by Otto Eckstein in his “Trends in Public Expenditures in the Next Decade.” 0 Gilbert Burck and Sam Ford Parker, “Productivity: The Great Age of 3 Percent,” Fortune, November 1955, pp. 249-250. 10 Using the same notations as before, income elasticity of public education is defined as yXi y X _5 VXs X i 11 James B. Conant, “The Future of Our Higher Education,” Harper’s magazine, May 1938. 2678 E M P L O Y M E N T , G R O W T H , AND PRIC E LEVELS 1. The degree of international tension is not expected to change much. 2. The political attitudes toward expenditures will not undergo a revolution. 3. Prices are assumed constant at 1958 levels. 4. The economy is to grow at a 3 percent annual rate. 5. The division of functions between Federal and State and local govern ments is assumed to persist unchanged.12 Three basic sets of projections will be developed. The first or “low” projec tion assumes that by and large 1958 conditions will persist, except that the number of pupils in average daily attendance will in all three cases increase as projected by the National Education Association. The “medium” projections assume a cumulative annual increase of 3 percent in the magnitude of per capita personal income, teachers’ salaries and length of school term. The other independent variables are assumed to increase very little.13 Finally, the third projections are high, in that they make rather optimistic assumptions about the future, i.e., cumulative annual increases of 6 percent in per capita personal income, teachers’ salaries and length of school term. On the basis of these assumptions daily per pupil expenditures can be pro jected with the aid of equations 1.1 and 1.3. They wil be converted into annual total current expenditures (plus debt service) estimates by assuming that the number of pupils in average daily attendance will advance from 29,859,000 in 1958 to 32 million in 1960. and 37,200,000 in 1965.14 Under these assumptions the following sets of projections were derived, which can be compared with 1958 expenditures: [In billions of dollars] 1958 L ow projection__ _____________________________________ ____ ____________ M edium projection__________ _______________________________ __________ High projection. _________ ________ ___________________________________ 11.0 1960 11.8 12.7 13.8 1965 13.8 17.4 22.8 Equation 1.1 emphasizes teachers’ salary level and, in certain respects, it mainly reflects expenditure conditions. Using equation 1.3, which emphasizes the income side, projections are produced which more nearly reflect the reve nue situation. As is to be expected, projections by those two equations show differences. Thus, with the aid of equation 1.3 the following three sets of projections are derived and can be compared with 1958 expenditures: [In billions of dollars] L ow projection________________ ____________________ _________ ________ M edium projection______ ______ _______________________ ____ ____ _ H igh projection__ _ _____ ___________________________________ 1958 1960 11.0 10.4 11.1 12.1 1965 12.1 15.3 20.0 It is interesting to compare these sets of projections with 1958 expenditures, which amounted to $11 billion. 1960 and 1965 projections (in terms of 1958 dollars) emphasizing the expenditure side are 16 and 58 percent, respectively, above the 1958 figure. The corresponding projections emphasizing the income revenue side are 1 and 39 percent, respectively, above 1958. 12 Otto Eckstein, “Trends in Public Expenditures in the Next Decade” (New Y ork : Committee for Economic Development, 1959), p. 4. 13 Xo-percent of public high school enrollment relative to total public enrollment is as sumed” to be 23 in 1960 and 1965, compared to 22.4 in 1958, and X 3-percent of pupils in average daily attendance living in urban areas is assumed to be 52 in 1960 and 1965, compared to 51.5 in 1958. u Enrollment projections were taken from “ Status and Trends: Vital Statistics, Educa tion and Finance” (Washington, D.C., National Education Association, 1959), p. 9, and with the aid of an enrollment— pupils in average daily attendance regression, pupils in average daily attendance figures were estimated. EM PLOYM ENT, G R O W T H , AND PRIC E LE V ELS 2679 SUM M ARY 1. The huge increases in current expenditure (plus debt service) for public primary and secondary education since the turn of the century are mainly due to exogenous forces, such as factors raising salary levels and per capita personal income. 2. It appears that there have been no significant productivity changes in the education sector. 3. Since the turn of the century, the percentage increases in educational expenditures— to be more specific, daily per pupil current expenditure (plus debt service)— were on the average about the same as the percentage increases in per capita personal income. The former exceeded the latter by a very small amount. 4. A “medium” projection suggests that if annual teachers’ salary increases of 3 percent are granted— and funds to finance them can be found— total cur rent expenditure (plus debt service) for public primary and secondary edu cation will have advanced in terms of 1958 dollars to about $17.4 billion by 1965, i.e., a 58-percent increase in 7 years. However, if people’s attitude toward public education and the way of financing it remain unchanged, a 3 percent annual increase in per capita personal income may produce merely $15.3 billion in revenue in 1965. While this would be an increase of about 39 percent over the 1958 figure, it would constitute about a $2.1 billion deficit, measured in 1958 dollars. Mr. H i r s c h . Mr. Chairman, my learned colleagues discussed the whole service sector. I would like to address myself to one element of this service sector, namely, the public school sector; in many respects it can serve as an example for other State and local government sectors. Public primary and secondary schools are employing today more than 2 million men and women, wTho serve about 36 million pupils. This sector in 1958 expended more than 30 percent of all the funds spent by State and local governments; it accounted for about 3.3 percent of the national income and 3 percent of gross national prod uct. Its pivotal position and importance for the future of this Nation is all too clear and needs no elaboration. Representative C u r t i s . What is the public primary school sector in relation to the total ? What is in the private sector ? Mr. H i r s c h . The statistics o n the private sector are very poor. Representative C u r t i s . I mean just the bulk. Two million men and women serving 35 million pupils. How many pupils in the private sector ? Mr. H i r s c h . A s I say, the information is rather poor, both in terms of expenditures and number of pupils. I do not believe it would be much more than 10 percent in addition. I am also in a sense carving out a smaller part than all public education in that I do not address myself to the problems of higher education. We might want to briefly look at the record. I have an exhibit which indicates the changes in total current expenditures, plus debt service, for public primary and secondary schools. It indicates an increase from less than $200 million at the turn of the century to more than $11 billion in 1958, the last year for which we have information. You might say a 56-fold increase in 58 years. Yet a look at this chart indicates that this increase was not a steady one. I am rather in trigued by the twro tails of this line. What are the factors that are responsible for this tremendous in crease in expenditures for public primary and secondary education ? I would like to submit a rather general hypothesis and later briefly 2680 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS test it. I think there are six major forces that have been leading to the changes in expenditures for education. The most obvious one is that the number of pupils in average daily attendance has increased. The second involves changes in the sociological characteristics of the population: age structure changes which affect the primary-sec ondary school population ratio. Supposedly, a pupil in high school is more expensive to educate than a pupil in primary school, both because of smaller classes and more expensive facilities. Also the geographical distribution of our school population has changed. In this context urbanization is important. A third element is economic in nature; that is, the price level of goods and services bought by schools and people’s ability to afford services; namely, their income. The physical characteristics are primarily related to productivity changes in the school system. Government characteristics center around the issue of whether school systems are primarily controlled by a single specialized governmental unit or whether they are part o f a much broader complex. Finally there is a factor which I would like to call the variety, scope, and quality of education services. Let me briefly indicate by way of an example what I mean by va riety in the education field. At the turn of the century, not many school districts used school buses to transport children to school. They had neither medical facilities nor cafeterias, and so on. Today, out o f $11 billion, about a billion and a half a year is spent for tliese services. I would be the last to suggest that these are undesirable or unnecessary services; and yet I think we should recognize them as expenditures which do not necessarily directly add to the quality of education. Senator B u s h . I would like to interpose a question about teachers in these six points you make. Are the teachers concealed in one of those points ? Isn’t the question o f supply of teachers one of the major points? Mr. H l r s c i i . Teachers and their salaries would be reflected, as I hope to show later, in the price level of goods and services bought by schools. Teacher salary is the main element. Indeed, about 80 percent of the current expenditures for public schools take the form of salaries. Senator B u s h . Thank you. Mr. H i r s c h . The second part, under the heading of “ Variety, Scope, and Quality,” that is, scope might be measured in terms of length of the school term, which has greatly changed over time. Today the average school term lasts about 179 days compared to merely 144 days at the turn of the century. By the way, I am not trying to imply here that it would be proper to talk entirely in terms of the number of days that a pupil is in school and say that this is indicative o f the amount of education he gets. Representative C u r t i s . On that point, may I ask: Is this average attendance of a single pupil ? Mr. H i r s c h . That is right. Representative C u r t i s . In other words, the fact that there is a lot of summer school makes no difference ? EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2681 Mr. H i r s c i l N o ; the summer school would not be included here. This is a relatively recent development. Summer terms usually are not financed exclusively by the school system. Instead, the student often pays for the summer term. Thus, I am talking about the very same school term for which the regular teacher is hired. Representative C u r t i s . That is what I wanted to be sure of. Thank you. Mr. H i r s c h . Measurement of the third element, that is, quality, in the narrow sense, is the most perplexing one. I f one works on the na tional level, empirical problems are almost insurmountable. In my prepared statement I refer to a case study that we were able to carry out 2 or 3 years ago under the financing of the Ford Foundation. In it we were able to develop a quality index composed of six subindexes. In talking with numerous people in the field of education, we tenta tively concluded that perhaps the very best single measure to reflect the quality of education in this narrow sense would be the number of specialists, namely, of full-time equivalent consultants, superin tendents, and principals per 1,000 pupils in average daily attendance. We assume that the school that has relatively many experts per pupil makes similarly major efforts to provide good education and succeeds in its efforts. I am fully aware that many educators have been emphasizing the pupil-teacher ratio as an indication of the quality of education. Per sonally, I have always been somewhat embarrassed by this concept, and I feel not very differently than the president of General Electric did when he commented some time ago: There are some educators who are actually proud of the declining ratio of students to faculty. To my knowledge there is no other field in human endeavor which actually prides itself on declining productivity. For this very reason, I would not want to suggest this as a good measure. Without going into details o f the empirical test of this hypothesis, we found that the cost of an education unit, so standardized that its variety and scope are held constant and expressed in per pupil in A D A terms, is overwehlmingly determined by the prevailing salary level of the instruction staff and perhaps the prices of all goods and services bought by schools. Moreover, they in turn are determined by the general demand and supply situation and the price level of the rest of the economy, as well as the particular demand for and supply of teachers. It appears that the first consideration apparently by far overshadows the latter. To put it differently, a very large part of the increases in public school expenditures have been fed into the sector by means of price level changes in the rest of the economy. Population growth and length of the school term are assumed to be additional factors. Salary changes in education are very important. The equation devel oped by us can explain roughly 09.8 percent of the changes that have taken place in per pupil expenditures during this 58-year period. I f we now take education expenditures, subtract expenditures for auxil iary services, and deflate this series by salaries, the initial 48-fold increase is reduced to one of about 250 percent. (See chart 2). 3856B— 59v—tpt. 8------ 16 2682 EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS I am using 1954 here as a base, because the work the Office of Busi ness Economics has been doing in relation to the implicit price de flators for State and local governments expenditures is also using 1954 as a base. Thus, while the current dollar increase for 1900-1958 is 48-fold, the constant dollar one is merely 2 ^ -f old. Or to put it differently, during the last 10 years annual per pupil expenditures in current dollars increased an average of 6 percent, while the increase in 1954 dollars was less than 2 percent. Senator B u s h . What is the population increase since the turn of the century ? What was it at that time, approximately ? Mr. H i r s c h . It might be best to answer this question by comparing the number of pupils in average daily attendance. I believe I have these data here. In 1900 we had 10.6 million pupils in average daily attendance, and in 1958 there were just a shade below 30 million. Senator B u s h . So that has increased b y threefold as against the dollar increase of 2y2 times. Mr. H i r s c h . We can use this analysis to shed light on the problem o f what has been happening to productivity in our public, primary, and secondary school sector. We do this by looking upon productivity as a residual. I f we take expenditure data and adjust for price level changes, and standardize, to the best of our ability, to account for scope, quality, and variety changes, we have a series which is quite stable for 1900-1958. (See lower line in chart 2 of the exhibit.) W e find that expenditures in constant terms have hardly changed at all over this 58-year period. Apparently, productivity has remained unchanged. Unless this line goes down, we have no increase in productivity, we have been produc ing over about a 58-year period, in constant terms, roughly the same quantity and quality for about the same dollar cost (expressed in 1954 dollars). S e n a t o r B u s h . D o y o u m in d m y a sk in g w h a t y o u m e a n b y “ pro d u c t i v i t y ” in t h is case ? Mr. H i r s c h . It would be output per unit of input, recognizing fully that it is very difficult to define here the output, namely, the amount of education offered. We tried here to standardize this out put by trying to hold quality constant. Senator B u s h . I think that would b e awfully hard t o a r r iv e at. I congratulate you for the effort. Mr. H i r s c i i . Next, I would like to raise a second point of some con cern, I believe. It will shed some light on a very important policy issue, and I would hope, it would make possible projecting the rev enues that will be produced for the public education sector. In my earlier comments, I indicated that one might want to hypoth esize that also income has a distinct bearing on expenditures. In come is a reflection of people’s ability to afford education, or any other service or product. I f we again rely on statistical multiple correla tion and regression techniques and use the very same per pupil per day expenditure figures we had before, we find that a very substantial amount of the expenditure changes can be explained in terms of in come changes. Specifically, roughly 97 percent of the changes that took place during these 58 years in per pupil expenditures can on the average be explained in terms of income changes, plus some other factors that I had mentioned in my earlier hypothesis. Now, we can EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2683 attempt to answer the question: What happens to per pupil expendi tures if personal income goes up ? For this purpose we estimated the average income elasticity of edu cation. We find it to be slightly above one, that is, +1.09. It means that holding the effect of the other factors constant during these 58 years a 1-percent increase in per capita personal income was on the average associated with about a 1.09-percent increase in daily total current expenditures for public primary and secondary education per pupil in average daily attendance. This figure is only slightly above unit elastic and apparently much lower than that for some other pub lic services, not to speak of such items as speedboats, golf, air-condi tioning, and so forth. This low-income elasticity of education is of con cern to me since I am convinced that improvements in public education are essential if the United States is to remain a leading world power. I f this is to be accomplished, an increasing proportion of the American people’s income must be channeled into public education. Income elasticities of education have been calculated also on a cross section basis, namely, State by State or city by city. We find that in this case income elasticity is much smaller. Three studies are avail able, and they find an income elasticity varying from 0.6 to 0.8. What do these results mean ? This means the follow ing: As we go from one State to the next, or one city or school district to the next, we find that those who have a higher per capita personal income do not spend proportionately more on education. W hy? Because, we so heavily depend on property taxes. I think it is very important to realize the difference that exists in the provision of services by the public and private sector. Insofar as the private sector is concerned, we can decide to purchase a service, and we pay for it if we want it badly enough. But there is a great difference with regard to public sector services. I f we want better education or better health services from the public sector, or better police, we must work ourselves through a very long line of very com plicated channels that exist between our desire for services and our ability to convince our fellow voters to help us get them. We have to get a consensus of the people to favor these additional services and we have to vote higher taxes. As long as public education is so heavily financed by property taxes (which are proportional in nature), it is most difficult to channel substantially larger amounts of our incomes into this particular endeavor, I submit this fact is of great concern to many of us. I f public education is to be improved in the United States and more funds are to be had to finance education, serious con sideration must be given to changing the tax structure and peoples’ attitude toward education. In order to change our tax structure, one of the only hopes I can see is that we shift more of the financing from the local to the State and the Federal level. A t the moment there is no other way that will help persuade local units to change over from the very heavy reliance on the proportionate property tax to the more progressive income tax. Senator B u s h . May I ask a question there, Mr. Chairman ? Representative B o l l i n g . Certainly. Senator B u s h . In higher education in our State universities they do charge a very modest tuition relative to what is given out in edu cation and very modest in relation to what the private universities 2684 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS charge. In your thinking about this need for additional financing have you given any consideration at all to the possibi 1 i tv of a modest tuition for students in the primary and secondary schools? It is i foim of taxation, but only to be paid by those who use the facilities. M Pn sen. Senator, I believe this issue <s beyond the realm of (< » I believe your question touches on one of the most basic issues 111 a person’s philosophy; namely, how important universal free education *■io this country. Sena' *,y. ]»i w. You have gone into the whole subject pretty deeply; I thought maybe you would have some ideas about that. Mr. H irscji. They would be very subjective. I feel that one o f the mo- * i;i‘ }v>i\ant elements that will assure a dynamic, progressive,, and free society is that good education is offered virtually regardless of a person's abil itv to afford it, S^?n*or B rsn , 1 ani not tak< ;--ue with you on that. a! Ive (V n n ^ Or to ab-orb it. P e ^aid ?s1> wy to afford it. i \' i I' i»<* w evt-.-o ihat to an abi*;ty to a Vorb it. Senai o. IU sir, I am n< : sujize^* '«no; that th>s a*.. 'fi■m»ive is a satis factory one at all. 1 \va> wondering whether you had considered it along with the m;ny considerations that you have spoken of here, because of the need for additional funds. I agree with you wTe need a lot more. The problem really is where we get it. Mr. H ir s c h . There is, I believe, an interesting experiment taking place at this moment. Most of the 6 or 8 weeks of summer education,, are either partly foundation financed, or partly privately financed. Once they catch on, I personally would want to see the question raised whether this is really not an element that will work against those who are not fortunate enough to afford this enrichment program. What is the near-term prospect? I take the liberty to quote a man whose name has been very closely associated with the problems of public education; namely, James B. Conant, who in 1939 predicted: By I960 or thereabouts we shall have a stationary population. pressure on our schools will soon begin. The expansive The making of projections is a hazardous undertaking. I f we make the very same assumptions, that the very able technical director o f this study has made in one of his pubications for CED, three basic sets of projections for 1960 and 1965—high, medium, and low— can be produced. I f we use equation (1.1), which emphasizes teacher salaries, we can project expenditures. On the other hand, if we use equation (1.3), which emphasizes income; namely, people’s willingness and ability to afford education, we can project revenue. I f we do so, we find a distinct gap between expenditures and revenue in 1965. This gap will not necessarily exist, However, if we cannot change our tax system and if we cannot change people’s attitudes to education, I submit that the projection that uses the income equation will tend to be approached. On the other hand, if we change our attitudes and tax system, the conditions underlying the equation that emphasizes the expenditure side will prevail; we then will have a much higher figure. In brief, my tentative medium projection, using the first equation (emphasizing the expenditure side), suggests that the 1958 expenditures of $11 billion—and these projections are in 1958 dollars— will increase to about $12.7 billion in 1960 and $17.4 EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2685 billion in 1965. On the other hand, if we emphasize the revenue side, we will have a very small increase in the next 2 years, from $11 bil lion to $11.1 billion and to about $15.3 billion in 1965. To put it differently, in terms of percentages, the 1965 figure em phasizing the expenditure side will be on the order of magnitude of about 58 percent in 1958 dollars, while if we cannot change our atti tude and tax system, the increase will be merely 39 percent, or the gap in between will be slightly in excess of $2 billion. Representative R e u s s . (presiding). Thank you very much. Senator B u s h . I might say this has been a most interesting morning indeed. What are we going to do about that $2 billion gap, Mr. Hirsch? Mr. H i r s c h . From a public finance viewpoint, it seems to me we have two main opportunities. One approach would involve matching funds in order that we can bring to bear the progressive element of the Federal income tax on this problem and in this manner increase income elasticity of education. Senator B u s h . Y o u are talking about grants-in-aid of the Federal Government to the States in the form of matching funds without restriction as to use ? Mr. H i r s c h . I think there should be restrictions that are designed to assure the proper use of funds. They should be designed to lift the quality o f education. I do not believe that they should be of such a nature that local school districts would feel that interference from the Federal Government makes their operation difficult. Senator B u s h . This deficit here is dissociated with school building requirements. This is in the service sector; is it not ? Mr. H i r s c h . This is current expenditures plus debt service. Senator B u s h . It does not have anything to do with school construction ? Mr. H i r s c h . No, sir. Senator B u s h . S o your feeling is that the solution lies in Federal grants to the State to be dispensed by their departments of education on a matching basis ? Mr. H i r s c h . That is one approach. The other one, which I under stand is being considered by Dr. Conant, involves a much more com plex proposal, one that would allow a given percent, let us say up to 20 percent of the Federal income tax as credit to States, in case they shift from sales and property taxes to income taxes. Senator B u s h . In other words, he proposes to take a 20-percent cut out of the income tax for each State as it contributes, return that to the S ta te for these purposes. Mr. H i r s c h . In principle this would be the objective. Senator B u s h . I have not heard that before. Mr. H i r s c h . I think the main issue is to realize the very low-income elasticity. I f we have to channel more income into education, what techniques do we have ? I suggest that we have to have a change of people’s attitude toward education, but we also must shift from a proportional tax as the major revenue source to a progressive tax. Senator B u s h . I think we have to be very much concerned with this problem. I think it is probably one of the most serious problems that we face. I am very much impressed with Admiral Rickover’s statements about our educational processes in this country. Also 2686 EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS Senator Fulbright from time to time has made some impressive com ments. Only recently in August he made a statement on this subject which I thought was excellent, the burden being that we have to for get some of the extravagances and recognize more of our necessities in which he rates education, I think very properly, right at the top. I f we can work out some way to do away with some of the subsidies where they are really not needed, if my friend will indulge me for a moment, like the farm surpluses-----Representative R e u s s . I will not only indulge you, but I will agree with you. Senator B u s h . Even in another field where we have some $9 bil lion we have accumulated in recent years. This is a shocking waste of the taxpayers’ money, when we have these needs that you talk about. Even in the field of housing, where we now find ourselves the proud owners of some $5 billion worth of mortgages which was never in tended that we get into that position. There is $14 billion in those two items alone. Representative C u r t i s . $26.5 billion in excess military surpluses. Senator B u s h . There you are. I didn’t realize that was the figure. That is an amazing figure. Representative R e u s s . And an extra figure this year in service on the national debt. Representative C u r t i s . N o w w e are getting controversial. Senator B u s h . But in order to take care of the necessary increases of the national debt and to take care of the problems that Mr. Hirsch is talking about, a very serious one, I would like to see some of our economists come up with a program that w^ould come to grips with the overall problem and make us some realistic proposals about where we can cut back so that we can put this money into these areas wThere it is so greatly needed. Public education I have in mind as almost No. 1. I don’t know how’ we are ever going to get that kind o f recom mendation out of our group of economists, but I do believe it would be one of the most constructive things we could possibly have, that is, recommendations of that kind from the economists of this country. I am very much afraid as a practical political matter that you are going to have a hard time doing what Dr. Conant wants to do, or otherwise increasing taxes at the Federal level. I believe from what I have learned that the town and city level is going to have it very hard to make any substantial increases as far as the rates are con cerned. Don’t you think so ? Mr. H i r s c h . Obviously I cannot speak for all economists. I would say that if the economists would be asked by this very distinguished committee to submit a proposal, and given due time, I believe it would be done. On the specific issue you raise, I think we would want to distinguish between the population in core cities versus that in suburban communi ties. When we talk about the metropolitan population and do not make this distinction, we forget that today the wealth of the nation is concentrated in suburbia. Core cities are at best stable. The subur ban communities are increasing in population and wealth. I am not quite sure whether one would not want to work out a system that wrould induce metropolitan areas to tax their suburbs more heavily. This could be done, in my opinion, realizing full well that EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2687 the young family that moves into suburbia needs most of the services and cannot afford them. Yet the adjacent community that needed the services 10 years ago, and whose families are now matured and are in a better earning position has the funds. We don’t have at this moment a system or inducement to tax those who have the means, except through the Federal income tax. I would hope that a system could be worked out on the local level that would siphon off some of these rather substantial tax bases to support education locally. Senator B u s h . This raises the problem we have in the New York area where the State of New York does tax nonresidents wTho live in suburbia, whether it be Pennsylvania, New Jersey, and so forth. While that is very unpopular in suburbia, I think on the whole it is not an unreasonable proposition. After all, if you are going to use all the facilities of the great city of New York to earn your living, it is not unreasonable to call on you to help support some of the serv ices which are at your disposal every d a y: protection, streets, educa tion, everything. I think in line with what you are saying that this may be one of the ways in which these centers of population, the large cities are losing their higher income groups into suburbia, may have to turn to that as a source of income to take care of the deficit that is being left there by the removal of these families who can afford to move, so to speak. Mr. H i r s c h . May I emphasize the point that even if we shift the responsibility from the local level to the State level, we will not necessarily succeed in raising more funds. The elasticity will change, however, if we shift from a proportional to a progressive tax. One of our great problems is not only to shift taxes from the local level to a metropolitan or State level but also to raise the income elasticity of education. Mr. G in z b e r g . A quite liberal economist far to the left of the A M A is Seymour Harris at Harvard, who, after taking a look at the edu cational problem in the State of Massachusetts, sees no special rea son why you can’t begin now to rethink the whole sales tax structure and other taxes in relationship to services. I wanted to come back to a question Senator Bush asked earlier about payment and services in terms of what is happening to the income distribution of the United States, I think if you begin to look back how the Govern ment got into the service business, we have to realize it was an en tirely different kind of distribution of income which brought it into the hospital field. Take the big municipal hospital plant of New York City. This was established at the time when we were getting a tremendous number of immigrants into New York who literally had nothing on their backs but the shirts they were wearing. Obvi ously if they were to get medical care, the city had to do something about it. That is quite a different story from where one stands today. I think there is a very serious need to emancipate ourselves—I am talking about the economic profession as well as the public at large—■ of a large number of, attitudes and points of view that were applica ble to an entirely different level of both national income and personal income and look again at the services we want and how to pay for them. It is quite obvious that the State of New York, with respect to certain new services, has deliberately moved toward at least partial payment by the user. The whole recreational development of Long 2688 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS Island-J ones Beach operation, the development of certain tunnels and bridges and the rest. The question as to just where it is appropriate and where it is not appropriate to reassess some part or the whole part or a considerable part of the services on the user thereof ought to be carefully evalu ated. I think you could quite well say that while it was reasonable for New York City in an earlier day to develop a single municipal col lege which was very valuable in the sense that it provided poor people with access to a higher education which they might otherwise not have had, this is a quite different matter from saying that in the year 1960, not to mention the 1970’s, New York City ought to continue in discriminately, without any question of ability to pay, to provide mu nicipal higher education for middle-class families. I have a view about secondary education which is a little ambiva lent. I would like, for incentive reasons and not for income reasons, to try to get youngsters to buy their books if they can afford them. I think we have a very bad attitude in this country: What you don’t pay for is unimportant. I would like to have them buy $25 or so of books a year which they would keep at home, and I think the whole educational system would begin to have more meaning. I f parents had to put out a token amount of money per month for some part of the educational costs, they might be more interested in what is going on. I don’t want to submit that I am now in favor of payment for secondary education across the board. Viewing the United States— and I am not an expert in educational finance as is my colleague, Dr. Hirsch—I don’t think this ought to be handled as a homogeneous problem. I would say that the State of New York can raise the money it needs to have the kinds of schools it wants. I don’t think the Federal Government has to assist the State of New York to raise money. I would say that the same is true of Connecticut, Massachusetts, Pennsylvania, Illinois, and so on. The essential problem arises in my opinion with respect to the low-income States who produce a large number of children who are making a more than average effort. I believe that the legitimate role of the Federal Government is to view children as a national asset and to say that it will not stand by and because of the accident of birth deprive children and fail to help bring the weaker States up to some kind of a minimum. I felt that Senator T aft’s general proposition was very sound, and would prevent getting more and more Federal machinery which one doesn’t need. There are too many problems that derive from ana chronistic government forms, State, local, and surban. They have to be handled anyhow because each separate problem will be impossible o f sensible solution unless you get a reorganization of local govern mental structure. We have a 1900 structure for the 1960 world. I don’t think education should be the excuse for preventing a large number of changes in that structure. I do believe that the Federal aid to education should be thought of very specifically with respect to States like Mississippi and the Carolinas and Arkansas and Ten nessee, where they are making a more than average effort and simply can’t find the money because they have too many youngsters. I wanted to raise another point about the flow of funds into educa tion. To put more money into a structure which does not change EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2689 makes little sense to me in terms of trying to improve the structure. The way I want to argue the point is, why do you assume that the women who are now getting $5,200 a year— as teachers in large cities in the North. I have not checked this figure but it cannot be far off— would be better teachers if you pay them $6,000 ? There are many serious problems where money is blocking the reorganization of American education, but you will not get the results you are seeking by an indiscriminate flow of funds unless the educational leadership will make changes. In high schools it is perfectly clear that it is very serious not to have more male teachers teaching boys. I don’t believe it is possible to do an effective job with adolescent boys with mostly women teachers. We have need to get a higher quality of male to stay in the school system for a career. You cannot do that unless you move toward some kind of merit system. There is not enough money in the United States, because each thousand dollars of average salary increase will cost about a billion and a half dollars a year. What you need to do is to get some widening of the range of recompense for teachers of different capacities and abilities and so on. You have to go back and say, “ Why are some of the teachers as weak as they are ? They are weak because the normal schools or the liberal arts college is not what it should be.” I f you have weak teachers, you will have weak students, I would like to think that the problem of American education is not an indiscriminate flow of funds into an existing structure but the identification of a series of priority weaknesses, and then trying to see what, money and other things can do to remedy them. I think we are hopelessly naive when we assume that just multiplying the dollars in the same system will give us much of a return. It just doesn’t make sense to me. Senator B u s h . While I have the floor, may I ask one more question ? Representative B olling. Yes, Senator. Senator B u s h . On the subject of doctors, you very briefly— I don’t believe you really answered my question, and you said you would come to it— would you come to it ? Give us your views about this question about the supply of doctors. Mr. G in z b e r g . I have an ambivalent position. I am not really as conservative as I sounded. There are some areas of the country which do not have enough of any good things, including “good” doctors. They do not have enough good teachers, good roads, good hospitals. They are obviously the more impoverished rural areas o f the United States which are short of a lot of things, including doctors. There are classes who, because of low incomes and other reasons, do not have access to all the good things in the United States in the amounts they should have and, therefore, you can identify sectors of the population which are not having proper access to medical atten tion. There is no question about that, in my opinion. Then, there are particular shortages in certain fields within the medical profession. I f you take psychiatry, which is a field I know more about than some of the others, I suppose you could say that there will always be a shortage of psychiatrists if you believe that everybody is a little nutty and that everybody could profit from some association with a doctor. 2690 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS In the case of the President you have a standard of one doctor to one patient. That gives you a kind of upper limit. When the Presi dent gets sick he has more than one doctor to one patient. I think the problem has to be viewed a little bit differently. The question is whether you want to make a differential effort from a public point of view at the present time to increase the number of people entering medical schools beyond those which are now there. I would say it makes good sense for a State which has no medical school to see whether it can start one. There is a very important relationship between where doctors get trained and where they practice. There are some discussions going on in New York City now as to whether there ought to be another medical school in New York. I have a “ withholding” attitude to ward that because I know that by and large, if we have another medical school in New York, most of the doctors will be in New York City. We do not really have a shortage of doctors in New York City, if you take a look at any kind of ratios. We are in a very favorable position. Next to Boston, probably, and the city of Washington, the most favorable. Therefore, I am in favor of selective activities here as in the case of education. There are problems with respect to the numbers of doctors, but I am not impressed with a simple ratio approach. The U.S. Public Health Service, which has argued the question of so many doctors per so many population, is not on very solid ground in my opinion. I am not impressed with the fact that because we are going to have more children or more older people that is enough of an explanation of a demand factor that we need more doctors. The thing that impressed me about pediatrics is that it has become primarily a preventive field. Next, there is a question of whether you really ought to encourage the American public to use a lot of doctors. In testifying some years ago down here in a different con nection I presented a very unusual statistic, and I remember Repre sentative Hoffman of Michigan almost chewed off both my ears. I said that I was impressed with the fact that during W orld War I I about 40 percent of the active doctor manpower of the United States was withdrawn from civilian use and by all the objective in dexes o f health that I knew anything about, the health o f the Amer ican public had improved. I really was not arguing, “ the fewer the doctors, the better the health.” But there comes a point at which health is not a function solely or primarily of doctors. The question of how much income the population has, how it eats, where it lives, what kind of educa tion it has, how it learns to take care of itself, are all very important aspects o f health. The simple equation that doctors equal health is a very foolish and incorrect equation. I would like to see certain things done on the health front which I think are much more economical and would pay off much better than any broadside approach aimed at the rapid de velopment o f new medical schools. Senator B u s h . Generally speaking, the income level attainable by doctors is satisfactory enough to attract a good quality of persons into that most important profession, is it not ? EM PLOYM ENT, G R O W T H , AND PRICE LE V ELS 2691 Mr. G in z b e r g . The deans have been complaining but I think they have misunderstood the manpower problem. Over the last 10 years, some of the leading deans have been complaining about the applicants they have been getting. There has been a misunderstanding o f why they have been “ suffer ing.” They did not always understand, first, that this was a thin age group, during the last 10 to 15 years. We had a small age group of people becoming 22. So the number o f applicants declined from pre vious levels. Secondly, there has been more excitement in some new scientific fields. I think there are young men today wTho prefer to be atomic physicists than to go into doctoring. Thirdly, this inordinate extension of medical education which Johns Hopkins has begun to reverse had something to do with this. Many smart people did not want to start earning a living at 40. On the whole, I think medicine is getting a reasonably good cut of the trained college graduates and we do not have to worry about the fact that there are not enough people interested in entering medicine. Representative B o l l i n g . Mr. Reuss, do you have any questions ? Representative R e u s s . I will take up where Senator Bush left off on this most interesting subject. I take it that you feel that the problem of medical care is mainly one o f certain geographical areas— southern Appalachians, people are poor, generally; hence, they do not get enough medical care, and in certain social nongeographical groups. I take it, although you have not said it, that among those groups are older people whose income is not sufficient to pay the necessary costs o f adequate medical care. Once you get beyond those geographical and social groups there is no particular problem of the adequacies of the number of physicians. I want to pursue that a bit. I f I am not mistaken, the number of physicians being turned out by our medical schools today is, in abso lute numbers, very little if at all greater than it has been at times in the past when our population was very much smaller. Mr. G in z b e r g . That is not quite right. What is looming ahead is because of the rapid increase in population, some decline in the ratio o f doctors per hundred thousand of the population. We have had a rather sizable postwar increase in the number of graduates but it is also a time when the population lias been increasing very much. This is the refinement on your point. Representative R e u s s . It is a necessary refinement. Mr. G in z b e r g . The number of graduates has been going up quite rapidly. Representative R e u s s . In terms of population, it is about the same percentage. Mr. G in z b e r g . W ith some likelihood of a decline in the next decade. Representative R e u s s . The likelihood of the decline is because our present foreseeable plans of medical schools for expansion are not enough to take care of maintaining the same ratio. Mr. G in z b e r g . That is right. Representative R e u s s . This is what concerns me. We do have the fact that people are living longer as a result, incidentally of better medicine. Mr. G in z b e r g . I am going to comment on that eventually. 2692 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS Representative R e u s s . Let me name the three or four factors that I have in mind and then you may comment. We have the factor of population increase which has been mentioned. We have the factor, which I think is very important, that medical care is dramatically better than it was in 1940 when you reported that with 40 percent of the physicians around, people seemed to be about as healthy as they were with 100 percent of physicians. Finally, we have the fact that not only is medicine more efficacious, but demand becomes effective because of a higher national individual income. Taking into account all these factors and looking at the next 10 years, I wish you would give me a little more reassurance than, frankly, you have so far given me that all is well with our future medical picture, and conversely, we do not need to do some things other than what we are now’ doing to see that our existing medical schools are enabled to expand somewhat their enrollment more than the presently envisaged expansion. We will not talk about whether this is to be done by private fund raising drives, Government help, or whatever. Mr. G in z b e r g . I would have several comments to make. I would say that the increase in population per se really does not tell us too much about the needs for medical care because one of the things that happened in medicine, unlike education, has been important changes in work patterns. Productivity changes. I f you take the fact that the practice of pediatrics—kids used to get pneumonia and mothers used to have to sit up with them for many days and nights through the crisis and the doctor used to come twice a day—you have had the situation where the very success of medicine has made for very substantial reduction in the use of doctor services for many sectors of the population. The way that I put it frequently is, that if you are lucky and born healthy, you see a doctor for X number of preventive services in your early childhood and you really do not begin to see a doctor again bar ring an appendix, which takes a few days, or being hit by a truck, until a time in your life when he cannot do too much for you. The doctors are working themselves out of the payoff areas by the very advances of medicine from the places they used to make their spectacular therapeutic contributions. Now you get more and more of the problems connected with older people. I f you say, what does decent medical care mean for an older person, this gets to be a very, very tricky problem. I am convinced that the use of the doctor is in part for emotional support of older people who frequently do not have anybody to talk to. I do not say that is bad. This is a kind of emotional support for an older person that might be quite good. But we must understand the nature of the relationship. I would say that the thing that impresses me about the American public is that for the first time in the history of the world a sub stantial part of the population is eating itself to a premature death. This has nothing to do with doctors. We have reached a level of income where we are shortening our lives by too high a level of consumption. This is just a statement of fact that most physicians will generally substantiate. EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2693 Representative R e u s s . Everybody but the manufacturers of women’s bathing suits, who are coming out with a bikini program, had the opposite view. Mr. G i n z b e r g . I think in a certain sense you can argue that a civilized and humane society, if it wants to use doctors to support older people or people in trouble—where in the old days ministers were more frequently used—you can double your medical profession. I do not think this is a very good way to move, I think it is much more sensible to ask ? How can we accomplish what we want or need less expensively ? I wTill give you an illustration of what I mean. Take the mental hospitals. I f you look at their budgeted positions, they do not have nearly enough doctors. I am not talking about a theo retical ideal, but in my opinion there never will he enough because relatively few who have gone through psychiatric training, no matter how many people are in the pipeline, will decide to live in an isolated mental institution. There are a whole series of personal reasons why psychiatrists do not want to do it, I would go a step further. I would like to raise very serious ques tions such as how important the psychiatrist is for the care of chronic mental patients. Admittedly they are very important in the active section in which you have an active therapeutic program for 6 to 9 months while you try to reverse the disease. After that, the key factors in a mental institution may be the quality of the ward attendant, the amount of money you have for food, be cause many mental patients are suffering from malnutrition. The whole relationship of the doctor to the totality of the require ments for the maintenance of health and the recovery of people really warrants restudy. I am convinced that a lot of patients would never go back to mental institutions if you could get them a job when they come out. But all the doctors in the world will not keep them well if they do not have jobs. You have subtle relationships here between the social factors that contribute to health, the economic factors and the more specifically medical factors that contribute to health. When I did the New York State Hospital study for the Governor of New York in 1949,1 was satisfied that in the State of New York— and I held hearings and all over the State listened to all complaints— that there was nobody who needed surgery that was not getting it in the State o f New York. Medical as distinct from surgical care is something else again. You can put a medical patient into the hospital and that relieves the family by doing some tests on him, on older patients in particular, and then you let them out. My father went through the last 7 years of life with a chronic illness. We had available the best medical talent in the world for him, but, doctors could not help. This is true, regrettably, with a large number of chronic diseases. I would say that I would want much more specification, much more pinpointing of just what are the current medical needs before con cluding that we have a gross shortage of doctors. I remember an ar gument I had with Aneurin Bevan, the former Health Minister in England. The poor people in the city of New York through our volunteer hospitals, public assistance, and municipal hospitals, were getting 2894 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS better medical care in my opinion than the British health scheme could provide. That does not mean that everybody in New York gets ideal service by a long shot. There are a series o f factors having to do with the nature of medical care where I think we should move toward the establishment of minimums. I do not think you can talk about adequate care because it is too loose a concept. But minimum, essential care, I think you can move against. I think that is what is important for different communities to worry about. I would say my hunch is that the problems have more to do with the organization of medical services than with the numbers of doctors around. Representative R e u s s . This does state the issue, and where I re main skeptical is that I am not at all convinced that the Park Avenue physicians who are being nice to nice ladies and your comb-out of redundant psychiatrists at mental institutions, who are doing work that a social worker ought to do or a Gray Lady ought to do, whether you are going to find enough scientific bodies by doing this and whether you are going to involve acceptable institutions for com bining them out so as to meet the problem. But certainly, one can agree with you, as I do, that there is much waste and inefficiency in our system without being entirely convinced of your feeling that really the problem will work itself out because increasing productivity of the medical profession will take care of the lag in production of physicians at current ratios of physicians to total population. Mr. G in z b e r g . I do not think it will work itself out. I would like to look at this in terms of other social priorities. I would rather see some more money going into schools intelligently, for younger children, for people at the beginning of their productive life, than I would for indiscriminate increase in medical resources where I think the payoff will be much less. I think the service sector of the economy is really the central issue that underlies all of these discussions. That is, you cannot do every thing you would like to do across the board in the service sector. I just have a feeling that in general medical care is “ in better relative shape” than are some other essent ial services that we need. As I look at New York City, the problems of the poor, and we have a lot of poor families in New York, I would not put their problems with respect to medical service anywhere near the top priority. Representative R e u s s . Along that line, would you agree, as I think you would from what you have said, that the possibilities of dramatic increase in productivity in the teaching profession are; in the nature o f the animal, much less than in the medical profession. One physician armed with modern education and drugs and equip ment can apply medicine to an increasingly large number of people. But it is much less true, is it not, that one elementary schoolteacher can teach 90 kids as well as he can 30. Mr. G in z b e r g . I will turn that on its head. I am going to say that I really think maybe one of the great problems in the medical field comes from the educational side and on the question of discipline and the kind of human being one is. One is always one’s own first doctor. I f one neglects a cold or does not, whether one drinks too much or does not, whether one overeats or does not, whether one does EM PLOYM ENT, G R O W T H , AND PRICE LEVELS 2695 all of these things is always in the first instance dependent on the person himself. The doctor is only useful on the preventive or repair end of the story. The thing that impresses me is that one of the big neglected areas, and we have done very poorly, is in the field of health edu cation. The great contribution that was made in the United States to health education came through the armed services. I have had a lot to do with the medical departments of the armed services. The armed services since 1940, have made an outstanding contribution to the health of the American population. This has come about because they have taught people how to eat, they have introduced them to prophylaxis. I f you look at the rates o f venereal disease toda}^ and in 1940, there has been a spectacular de cline. I had occasion some time ago to review the rejections at an induction station in Tampa. I knew what they were in World War II. I said, “ Where are the venereal disease cases?” A druggist told me o f the tremendous increase in the sale of condoms. There had been a whole new attitude toward prophylaxis and a very sizable contribution has been made to the health of the younger generation as a result of what they learned in the armed services, I do not want to leave you with the impression that I think every thing is good on the health front. I would like to pick a series of targets and especially expand the number o f public health nurses, be cause I think they are the biggest educators and much less expensive to train and give a much bigger payoff in many areas of the United States than physicians. I think there are a series of objectives that we ought to set up for health as for other fields. We ought to put a price tag next to each objective and figure how do we get the greatest social gains for least dollars. I know getting doctors is a very expensive way to proceed. Senator B u s h . Metropolitan Life Insurance Co.’s “ Good Hint for Good Health,” has helped some. Representative C u r t i s . Mr. Chairman, I first want to extend my appreciation to this panel for what I regard as very splendid papers and a very splendid discussion, and then to make this comment: Had I been setting up the studies of this committee in this area of price stability, employment, and economic growth, I think I would have made this subject matter, the service sector, probably a base, and then proliferate, because there are so many questions that are raised here that go right on into the very issues that we are supposed to be studying. As has been pointed out, this is the area where we have seen the primary cost index increases, and there is an area that needs attention. Furthermore, it is in this area, too, that I think we can learn some thing about the relationships of the various levels of Government as to how they relate in here, particularly the medical and educational field, plus its relationship to the private sector of the economy. So, in a sense, I am distressed that we are having one morning on the very area that I think we should have been spending and should spend much more time. I hope that possibly next year we may take this area itself and devote several weeks to digging into it. There are some comments and possibly questions that I would like to make for possible future observations. 2696 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS There is the question of financing schools and the idea of possibly coming to the Federal Government as a source of capital formation to solve it; after all the Federal Government is just another area of capi tal formation. Maybe in the end it has its peculiarities. Sometimes we do need to utilize that area for capital formation for various reasons. I think it should be based upon the fact that there are some good reasons, and not because it looks like it is an easy way. In other words, because you just have to persuade one body, a Congress, to possibly do it. The big problem our communities face right now, and our school boards rest primarily on the fact, is that the Federal Government did not perform its main function, which was to preserve the stability of the dollar. In other words, through inflation the whole real estate tax struc ture has just been damaged. Real estate tax, which is the primary source of school funds and sewers and all the community facilities, is a primary tax that is used for those services. Real estate tax, in turn, is based upon appraisals that go on the books over a period of years. When, through inflation you do not affect those previous appraisals that are on the books, but you do increase the cost of the services, you finally end up with what this United States is faced with— a very difficult technical job and a very difficult political job of reappraising all the real estate on the assessor’s books. Some communities have faced up to it by conducting a complete re appraisal. I might say a further thing on that subject. There is no easy way out, as some people said, by just raising the real estate tax rates. The new homes that come on the real estate books come on at the inflated value, and they are of the small-income groups. I f you just raise the rates you would be hitting that class the most. I am disturbed when people suggest the way you solve community problems is to come to the Federal Government for the capital forma tion, when the Federal Government, by improperly handling the one thing which it certainly has responsibility for, has created a good bit o f the damage. There is one other general statement in this area of the service field, and I w^ould add to it the distributive field. It is a strange parallel that that is the field where strong business—using it in its real sense, the smallest units—are actually engaged and have the area pretty much to themselves. At least they predominate in this area. It is also the area where you have the noncorporate form of doing business. So many people talk about helping small business and then they think in terms of helping small corporations. Eighty percent o f small business is not in the cor porate form o f doing business. Most of our economic statistics—the reason why I am developing why I think this area needs to be gone into— relate to big business and the bigger units and corporate statistics. Very little has been devoted to this big area of service and distribu tive categories. Ie seems to me that the cost increase in this area in many respects is the traditional one of too much money chasing too few goods. Certainly, when you try to get a plumber to come home and fix your bathtub or something, because there are not enough plumbers, frankly, you have difficulty. E M P L O Y M E N T , G R O W T H , AND PRICE LEVELS 2697 It is also indicated in the do-it-yourself trend. It is shown in the precooked food field, which is a form of do-it-yourself, and transporta tion, where the big factor has been the use of the private automobile instead o f the bus and streetcar. In this area of doctors, Professor Ginzberg, wouldn’t you say that there has been a tremendous increase in productivity in the individual doctor ? I was listing four factors: Automobiles, the fact you can call up the doctor by the phone, the hospital where you concentrate them, and then the use of technicians which you mentioned. I would think if that is true, that there has been a tremendous in crease of productivity on the part of a single doctor, there would be or should be a decline in the ratio of doctors per thousand population, and at the same time of a tremendous increase in health facilities. On the school thing, I wanted to ask you a question, Dr. Hirsch. I am wondering whether there is this great need you are talking about, just on your aggregate figures. There is one thing I was interested in* You think there was a threefold increase in the number of pupils. In the 1900’s you said 10.6 million, average daily attendance, against today of 30 million. Mr. H i r s c h . Yes. Representative C u r t i s . I think the population o f the United States, and this is a guess, was about 76 million in 1900 and today it is about 177 million, as a rough figure, which shows a little more than a doubl ing of the population. Yet, the average daily attendance has almost tripled. The figure I would like to know is this: In 1900 how many potential pupils there were that were not going to school and what that ratio now is, because, obviously, it has not de clined considerably. Whether or not a factor that should be thrown in is that we have probably come close to solving the problem of quan tity education in the United States. I f we have, our problem then comes to something wrhich I think is quite exciting. It would be, I think, the first time a nation or society has ever gotten to the point where they could really talk intelligently about quality education. To get quality you have to first sift quantity. We are now in a postition where we can sift quantity. I f that is so, then these aggregate figures do not have the implications that you suggest. Maybe we have reached the point where we have quantity education. I do not know. I would raise the question as to whether we do. Mr. H i r s c h . These projections took into consideration expected changes in the ratio between school-age population and average daily attendance. It is true that years ago we started out with a much smaller ratio between school-age population and average daily attend ance, but this ratio has stabilized in recent years. I would not fore see any major change over the next 5 or 10 years. ^Average daily attendance in relation to enrollment has changed over time because we have become more urbanized. Let me add that I was surprised to find in working on this particular subject, that the percent o f pupils in urban areas has not been in creasing over the last 5 or 6 years. Representative C u r t i s . A s a matter of fact, it is decreasing. Take Missouri, where our rural population has decreased Incidentally, on 38563— 59—pt. 8------ 17 2698 E M P LO Y M E N T, G R O W T H , AND PRICE LEVELS the schoolroom shortage—I happened to have voted for school con struction, which is a little out of character for me— I was convinced the aggregate of school classrooms did not show the shortage that was claimed, but a lot of those classrooms in rural areas were abandoned due to a wise consolidation. Then if you took the area where you really had a problem, which is suburban, you had a large problem. I regard that was caused by W orld War II and the Federal Govern ment, I felt, had a temporary responsibility in the area. I think this suburbanization factor is an exceedingly important one in this area. Mr. H i r s c h . May I just add one point? Representative C u r t i s . Yes. Mr. H i r s c h . I am the last one to suggest that the State and local governments should turn to the Federal Government for aid as such. I think it would be very important, and this was more or less implied in my earlier comments, to convince our suburban population that un less they pay higher and more progressive local and State taxes, voters sooner or later will force the Federal Government to assume the financ ing of many functions. Since the Federal Government would heavily rely on an income tax it would cost the higher income groups more than it would if they were willing to pay higher and more progressive local taxes now. This issue, to my knowledge has not been made clear to the local electorates and surely not in a very vocal fashion. The other point is that I do not suggest Federal aid for its own sake but only as an inducement which has carefully designed strings attached. Representative C u r t i s . Matching-fund idea ? Mr. H i r s c h . Yes; merely as an inducement. I would like to sug gest a criterion. One would be a test of tax effort. On the one hand, I would want to see every State spending a given percent of personal income for schools before that State is eligible for Federal aid. In the metropolitan areas, I would suggest that we would have a proviso that would insist that either entire metropoli tan areas or whole States pool their tax base to underwrite a certain floor, which is enough to provide for a minimum education. Tlius per public education expenditures could not fall below this floor. Representative C u r t i s . Let me throw out a suggestion. Possibly rather than a local incentive, maybe a family incentive would be better. I hate to mess with the income tax for that purpose. I am willing to do almost anything in this area to try to bring about a change. Possibly, give families a certain deduction on their in come tax for sending kids to school and the cost of it. Then you would have a pretty good one. I f they did not send them to school, they would not get anything. I f they did, it would be on a matching basis. I do think there is a possibility of increasing pro ductivity in schools, I am on the board of trustees at Dartmouth where we have started this idea of a trisemester proposition, with the idea of going to 12 months7 programs. Not that the students would go 12 months. We would use our physical facilities 12 months. Also, this is utilizing the professors, I think, in a more efficient manner. I have often theorized how did we ever get to the 9-month school system. I think it was because we were an agrarian society. EM PLOYM ENT, GROW TH, AND PRICE LEVELS 2699 That is not any longer a reason for it, We can now do some pretty good rethinking on this thing and utilize our physical plant 12 months a year, instead of 9 months. That in itself will make tremendous additional money available. Then, there is this question of whether or not T V and visual aids actually are good for education. That would be a quality decision. If, indeed, that is a technique, then wo can certainly increase produc tivity through that. Even more is the use— and I would like to get any comment on this—just as doctors have learned to use technicians, of technician instructors, in some teaching this could be done in schools. Even in graduate schools they are coming to that sort of thing. Is that being developed at all? Is that an area where there is some chance o f considerable increase in productivity ? Mr. H i r s c h . I had intended to refer to this issue in the paper I am preparing for the committee. I did not feel it fell in the narrow limits of this discussion. Representative C u r t i s . I can see that. I am asking the question, do you feel that there is a real potential or is this something that is of value but not o f any great significance ? It seems to me it is of great significance, Mr. H i r s c h . Teachers’ aids, as well as television, appear to be promising tools for raising productivity. Foundations have started to put up money for these two promising areas. You have to over come—and I think Professor Ginzberg mentioned this— a very hard core of resistance, parochial in nature. It will take more than im provements in technology; we will have to bring about changes in the entire education system. Mr. G in z b e r g . There are two really productive changes that should be easy to introduce and which are long overdue. W e have a crazy idea in this country that education is a direct function of the number of hours that a person is in direct contact with a piece of wood, with somebody talking up fron t: that you have to sit in class to be educated and the teacher has to be performing. In no civilized or uncivilized country in the wTorld do people put as much emphasis on formal classroom instruction as we do in the United States. I f you could ever get youngsters to learn to read— and you cannot educate them unless they learn to read—the better ones could move much more effectively with fewer hours of exposure to the teacher and more hours of exposure to books. This is a system in every European country which has been in terested in quality education. You have to vary the procedure ob viously. There are X number of subjects where you need the class room teacher more than other fields. The next point is that we have been wedded in this country fool ishly, on a mistaken psychiatric doctrine, that everybody has to move in a lockstep. That is, the chronological age progression is sine qua non for emotion statistically and so on. There is at least a 2-year variability among young pepole. There is no reason why bright children should not be moved through the system more expeditiously. I am not saying that an 11-year-old should go to Harvard. But a 2-year spread is reasonable. The sure way to get a smart youngster disinterested is to force him to stay with uninspired youngsters who are not moving at his clip. 2700 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS I would say that two of the really significant and very inexpensive productivity gains are simply to shift more work to the pupil and to move that pupil more expeditiously through the school. We have begun to do something about that but it is hard. In one of the finest public schools in New York City, the principal told me he recommended these procedures to his faculity. The faculty was worried, like any trade union, about the possible loss of work and loss o f jobs and said, “ Nothing doing.” Representative C u r t i s . Part of this experiment at Dartmouth, if you have followed it at all, has been more outside reading and less classroom. The figures for the first 2 years are very gratifying. We check through the use of the library. The graph has just gone up. This is an indication that you are not wrong in assuming that the students would devote more of their time to study. Mr. G in z b e r g . I f people would only begin to shift some of the expenditures from classroom costs to the purchase of books so they would have them at home and could begin to work with those books. I cannot understand how you can become educated without having a personal library. The gains are fantastic. I am talking about the upper half o f the class. Representative C u r t i s . I have one final question of the many that could be asked. When Mr. Reuss was mentioning the needs o f doc tors, he listed geographical groups. Then he mentioned social grounds, and he referred to old people. I was very much impressed with the testimony given before the Ways and Means Committee this year by representatives of the health groups. They said that their studies revealed that where a geographic group had adequate medical care and hospital care, no one in that group, whatever age bracket, lacked adequate or minimum health; but whenever you went to an area where there were inadequate health facilities, it did not matter whether you were old, young, or whatever, there was a lack in all age brackets. This indicated to me that probably it is a geographic problem rather than a social problem. I wonder if you would comment if you had reviewed that in any w^ay. Mr. G in z b e r g . I have what I would call an impression of some of these materials, but I cannot pull out of the back of my head data that bear directly on this. I suppose it is true. We know something about male and female use of doctors. Emily Hutchinson’s studies on the west coast in San Francisco indicate that on the whole women tend to use doctors more than males. One o f the reasons is that the male has got to get back to work. There are psychological factors as to how often you go to see a doctor. 1 thing it is true that with older people who are not very affluent there may be a tendency to stay away because “ He can’t do much for me.” I f the cost factor was not there at all, they would probably go more frequently, but older people are pretty reasonable and they have to live with their condition. I f money were no issue, they would tend to visit doctors more often. It is necessary, regrettably, for human beings to live with their chronic conditions. You want them to go to the doctor when he can help. E M P L O Y M E N T , G R O W T H , AND PRICE LEVELS 2701 I do not know what “ adequate” means for different groups. Old age is a very sad age for many people. I f they could be helped by seeing a doctor once a week, it would be nice if they could go. Maybe an alternative would be to have more social recreation where they could talk to each other, which would do just as well. Representative B o l l i n g . Gentlemen, we are very grateful to you for your contributions to this very interesting discussion. I f there are no further questions or comments, the committee will stand adjourned until tomorrow morning at 10 a.m. in this same room, wThen the subject will be structural unemployment, extent, causes, and remedies. (Whereupon, at 12:45 p.m., the hearing in the above-entitled matter was recessed, to be reconvened at 10 a.m., Friday, October 2, 1959.) EMPLOYMENT, GROWTH, AND PRICE LEVELS F R ID A Y , OCTOBER 2, 1959 C on gress of t h e U n it e d S t a t e s , J o in t E c o n o m ic C o m m it t e e , Washing ton, D .C . The committee met, pursuant to recess, at 10 a.m., in room P-63, the Capitol, Hon. Richard Bolling presiding. Present: Senator B u s h , Representing Bolling. Representative B o l l i n g . The committee will be in order. Today our subject is “ Structural Unemployment: Extent, Causes, Remedies.” Our first witness is Neil W. Chamberlain of Yale University. Mr. Chamberlain, you may proceed as you wish. STATEMENT 0E NEIL W. CHAMBERLAIN, PROFESSOR OP ECONOMICS, YALE UNIVERSITY Mr. C h a m b e r l a i n . Thank you, Mr. Chairman. I have a state ment here which I will read with your permission. Representative B o l l i n g . Please do. Mr. C h a m b e r l a i n . For analytical purposes, unemployment takes five forms. It may be seasonal, transitional— random and tempo rary—cyclical, technological, and structural. Whether these dis tinctions are relevant for policy purposes is a question to be examined later. Structural unemployment is easily differentiated from the sea sonal, transitional, and cyclical types. It and technological unem ployment are often closely related, however, and sometimes not easily distinguished. In general, structural unemployment attends the decline of an industry or region. As the demand for anthracite coal has fallen off, for example, some mines close down and others reduce their employment—or cheaper oversea competition may lead to an increasing loss of markets for a domestic industry, such as textiles, which may come to be viewed as a “ sick industry.” In other cases, even though an industry may continue healthy, a shift in its loca tion may leave behind a community of unemployed. I f a steel mill or a railroad yard is moved, perhaps because of a change in market orientation, numbers of its employees may not make the move with it, but may move instead into the ranks of the local unemployed. The relationship to technological unemployment lies in the fact that sometimes the decline of an industry is rooted in a change of technology: electric refrigerators displaced many men who held jobs in ice plants. Or a change in factory location may be attributable to new methods of manufacture which make it cheaper to build an 2703 2704 E M P L O Y M E N T , G R O W T H , AND PRICE LEVELS entirely new facility rather than remodel the old one: no longer dependent on the particular skills of its present labor force, and able in the new operation to utilize general skills such as are found in any labor market, a company is free to consider alternative loca tions more advantageous because of closer location to markets or suppliers. So technological change may be the agent o f structural unemployment. When structural unemployment emerges it is likely to be persist ent—hard core it is sometimes called. This persistence is attribu table to two factors. In the very nature of the case, numbers of men and women usually become unemployed within a short space of time. The area or industry becomes saturated with unemployed. These people do not readily find their way onto another payroll because (1) they are immobile, and will not leave the area of unem ployment density for other more promising areas; or (2) they possess either no particular skills or—even worse—a skill no longer wanted. Sometimes both factors are present. Structural unemployment poses a more intractable problem than other forms of joblessness. It is worth doing something about for three reasons: First, it represents a loss to our national productive effort, which is important even in our affluent society because of obli gations to the underdeveloped areas of the world, which can be expected to grow; second, because it lowers the standard o f living o f those affected, sometimes reducing them to the status of charges on relatives or the community; and third, because the accompanying sense of purposelessness and uselessness—unwantedness—represents a threat to their psychological health and stability. I make no estimate of the present magnitude of the problem. Unemployment figures do not allow a ready sorting out of those whose jobless status is due to something we can define as structural change. The Department of Labor has recently added Detroit to its list of centers having a “ chronic labor surplus,” for example, noting that increased mechanization and decentralization have led to a substantial loss of employment in that area. Yet we would not label either the automobile industry or the Detroit area as “ declin ing.” Regardless of this problem of estimating, however, and regardless o f the numbers actually affected, I would contend that it is important that newTpolicies be devised to provide greater protec tion to those affected. I f the numbers are great the need is evident. I f they are not great, a sense of added security can be provided to workers without substantial cost. What then can be done about the problem of structural unemploy ment? First, there can be a direct attack on the conditions giving rise to the immobility or skill deficiencies which root individuals to areas o f unemployment density. Second, where these efforts are unavail ing, there should be income support over a longer period than is now provided. Immobility o f the unemployed is due to a number of causes, among which are homeownership in an area of declining real estate values, dislike of breaking awTay from a circle of friends or relatives, lack of specific knowledge of jobs in other areas, and fear of the unknown. Some of these causes can be overcome. In particular, I should like to E M P L O Y M E N T , G R O W T H , AND PRICE LEV ELS 2705 see a stab made at the real estate problem. Where workers of X — perhaps 5 or 10 years—years of service in a declining community are made jobless, but are willing to pull stakes for another community if they can sell their homes at reasonable rather than distress prices, I should like to see some Federal agency ready to offer 90 percent of the fair market value prior to the labor market decline. Any loss involved in subsequent turnover of the property might well be less than costs of sustaining the worker’s family in unemployed status and would lessen the psychological costs mentioned above. Further effort to notify employment offices in other labor market areas of skills available in the surplus area is also likely to be helpful. It may not even be out of the question that transfers of numbers of families jointly from the distressed area to areas of labor shortage could be arranged, cushioning the shock of being uprooted from familiar associations. Probably more attention and effort has been put into programs for retraining displaced workers than into programs for increasing mobil ity, but here too the possibilities for improved performance might be reviewed. Underwriting of more extensive and expensive training programs in needed job areas, by those capable of assimilating the new instruction, would be highly desirable— an underwriting, that is, not only of the costs of instruction but living expenses where the training program runs longer than the period of the worker’s unemployment compensation. Such underwriting might take the form of a loan. The end result would be a worker reemployed in a skill area of real value to the Nation. Despite the most imaginative efforts of public and private agencies to avoid or shorten the duration of structural unemployment, however, we can be sure that some families will always be subject to its effects. In particular, it will always prove more difficult to retrain or relocate older workers. Youngsters caught by structural economic changes can be expected to make the adjustment on their own, with relatively little assistance, perhaps only a little guidance. But the older the worker the less adaptable is he likely to be to changes thrust on him. For those who are in the clutch of such circumstance, a longer period of transition is necessary—transition either to a new, probably less desirable, employment situation, or to new financial arrangements, probably involving a lower standard of living and a dependency rela tionship. These adjustments, even if they may be ultimately ines capable, should be cushioned by a longer period of preparation for them. Some form of income support of longer duration than unem ployment compensation programs usually provide is needed. I suggest the following simple formula as one possible approach. Recognizing that whatever is done in this area will have to be under taken by the Federal Government if it is to apply uniformly through out the Nation, we might consider Federal continuance of unemployment compensation— at State levels—past the date of their normal exhaustion. To take account of the age factor, I would sug gest additional benefits at the rate of 1 week for each year of unem ployment for those up to age 40; for those in the age bracket 40 to 50, iy 2 weeks of additional benefits for each year of employment; and, for those older than 50, 2 weeks of benefits for each year of employ ment. Thus, for example, assuming a benefit duration of 26 weeks 2706 EM PLOYM ENT, G RO W T H , AND PRICE LEVELS under a State unemployment-compensation plan, a man displaced at age 35 after having worked for 15 years would receive 15 weeks’ addi tional benefits or a total of 41 weeks. At age 45, after 26 years of employment, he would receive an additional 39 weeks of benefits or a total of 65 weeks of income support, And at age 55, after 35 years of work, he would be entitled to an additional 70 weeks or a total of 96. I f these totals sound high, may I remind you that they wTould be paid only if the individual remained unemployed. And I would argue the case that it is not excessively generous to allow a person who has had a work career of 25 to 35 years, after which his job has been shot out from under him, something less than i y 2 to 2 years to make a difficult adjustment to a less desirable economic status, I have made the above suggestion in connection with the problem o f structural unemployment, but having done so I would move on to say that I see no reason why the same approach should not be adopted for unemployment generally. The individual who is idled by transi tional, random, or cyclical circumstances and who finds difficulty in securing another job, perhaps for reason of age, faces the same hard ship as the person idled by structural causes. The principal danger that I can foresee in such a system lies in the possibility of its abuse by those who become unemployed and who simply make that event the occasion for their withdrawing from the labor force, which they may have planned to do in any case. Such abuse can be guarded against, however. It is likely to arise from people past 60 who might already have been planning retirement or from working wives who may already be contemplating withdrawal from the labor market, or from those in poor health who might find steady employment beyond their capabilities. Abuse from these groups could largely be avoided by providing that additional benefits could not be drawn by those (a) not meeting the eligibility standards for unemployment compensation under the relevant State system, o r (&) receiving a pension under either a public or private system, or (c) constituting part of a family unit where there is at least one em ployed member or where another member is receiving the additional benefits here proposed. How much of an added expenditure burden this system would place upon the Federal Government could be computed on a sampling basis. The costs, it seems to me, should be met out of general revenues rather than through any increased payroll tax. I f there is some doubt about the workability of this approach, it might be possible to try it out experimentally in one State before undertaking it generally. Senator B u s h . That would be out of general revenues of the State treasury ? Mr. C h a m b e r l a i n . Out of the Federal Treasury, Senator. In conclusion, it is worth stating the rationale for such a program— in reality, restating the philosophy underlying unemployment com pensation generally. Idleness due to economic circumstances is no fault of those who are affected. While they can be expected to exert themselves to find new employment, their future security should not depend on the problematical success of such personal efforts. While no one can ever be shielded from all economic risk, it is not a sign of softness for society to lessen personal risk wherever this can be done without detriment to the economic system as a whole. EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2707 The detriment feared by some is a lessening of individual incentive and self-reliance. It has yet to be proved, however, that the person without an assured future income is likely to take life easy if given temporary assistance at a level considerably less than his normal earnings. I might add at this point that by tying any additional supplement ary unemployment compensation benefits to past years of employment there is also a further safeguard here of simply supporting idle floaters. I f economic conditions are such as to give rise to prolonged unem ployment for substantial numbers of people, the weakness lies not in possible large expenditure from any benefit scheme but in the failure to generate a level of national output consonant with our capabilities. To summarize then, in the case of structural unemployment at least, governmental assistance should first take the form of stepped-up efforts to undo the strings tying people to areas of unemployment density, as through home-purchase arrangements, improved means of notification of worker availability and job openings, possible multiple family relocations, as well as further support for retraining. The possibilities of inducing new industries to move into declining areas I would leave to local initiative, partly to avoid Government entangle ment in intercommunity rivalries and partly because this line of attack seems so much less promising. Where unemployment cannot be re duced by these means, however, income support o f longer duration than present compensation systems allow should be provided by the Federal Government, on a basis which recognizes the greater difficul ties o f reemployment for older workers. Representative B o l l i n g . Thank y o u , Mr. Chamberlain. Next we have Philip Taft of Brown University. Mr. Taft, you may proceed. STATEMENTS OF PHILIP TAFT AND MERTON P. STOITZ, DEPARTMENT OF ECONOMICS, BROWN UNIVERSITY Mr. T a f t . I am making the statement on my own behalf, Mr. Chair man, and Prof. Merton Stoltz. Structural unemployment can be defined as involuntary idleness which arises from causes other than changes in aggregate demand. It is likely to be concentrated in an industry or a region or in both, and not widely distributed through the economy. It is a stubborn and per sistent type o f unemployment which can adversely affect thousands and even millions o f workers. Structural unemployment tends to con tinue even at the peak of the business cycle boom, and areas seriously suffering from it will generate a higher than average rate of unem ployment at all phases of the business cycle. It can arise out of a multitude of causes. Technological changes which enable an industry to produce the volume of demand for its product with fewer workers, the development of substitutes for existing finished products or raw materials, changes in tastes, the exhaustion of a new material—for example, coal or lumber— shifts in population, and changes in com parative advantages between regions may bring about structural unem ployment in a region, or in a specific community. 2708 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS Unemployment is not limited to the industry feeling the initial im pact o f structural change. Ancillary industries which perform a wide variety of services for the firms directly affected will also suffer serious losses o f income and their workers will be forced to endure a higher than average amount of unemployment. In addition, a large volume o f primary structural unemployment affects the “ protected” indus tries, those industries which do not face competition from firms located in other regions. This secondary structural unemployment results from a reduction in the income produced in the community and from the possible loss of external economies if the base of an industrial com plex is destroyed by structural change. The volume of structural unemployment is not easily determined. Structural unemployment is not distinguishable from cyclical and seasonal unemployment, nor are the various types of unemployment independent. Structural unemployment may intensify seasonal or cyclical variations in employment and, conversely, structural unem ployment may be deepened by changes in aggregate demand. More over, the primary and secondary employment effects of structural change cannot be separated. Despite these difficulties it is conceptually possible to devise indexes of structural unemployment which will serve to indicate the dimensions of the problem. This is a task of some magnitude and cannot be undertaken here. Nevertheless, readily ob servable evidence indicates that structural unemployment is a problem o f major importance. Consider two examples from the New England area. Rhode Island experienced a loss of 50,000 textile jobs—production workers— over a 40-year period. O f these jobs 31,000 were lost in a single decade, 1947-57. The decline of the textile industry was the dominant factor in the long-term loss of 40,000 manufacturing jobs in a labor force numbering less than 300,000. Massachusetts experienced a similar decline in textile employment and, in addition, suffered severe losses in employment in boots and shoes. In one decade, 1919-29, boot and shoe job losses numbered 25,000. These examples do not indicate the extent of structural unem ployment at a given time but they do reflect both the magnitude and the duration of the adjustment problem. Many other examples can be cited. The direct effect of structural unemployment upon the work force depends upon the degree of regional concentration and the relative importance of the industry experiencing the decline. An industry which provides only a minor fraction of a region’s employment and is w7idely dispersed in location will not have the same serious effects upon a region’s economy as one highly concentrated in particular places. The incidence of structural unemployment in an industry which is geographically dispersed is likely to fall more heavily on the older worker. Younger workers are more readily able to move into other occupations and acquire new skills. The older worker is at a dis advantage. Not only will he lose the benefits which arise from formal and informal seniority systems and long tenure, but the experience which one inevitably gains from lengthy employment in an industry will be lost to him. In addition, the older worker is not likely to be as easily reemployed. It is generally assumed, and there is some E M P LO Y M E N T, G R O W T H , AND PRICE LEV ELS 2709 basis to the assumption, that he is slower in acquiring skills he does not possess. In addition, there are other considerations, such as pen sion costs, illness, and absenteeism which can only be mentioned here. We, therefore, conclude that the older worker will feel the impact of structural unemployment even in regions where declining industries will not generate a serious longrun employment problem. In many instances, the older worker will at least be forced to accept a serious downgrading in employment if he is to be reabsorbed into the active labor force. While the plight of the older worker ousted from a long-held job should not be minimized, structural unemployment in an industry highly concentrated in a particular region is even more serious for the economy. The older worker suffers just as cruelly and keenly as in the other instance, but in the case of a declining industry being heavily concentrated in a region, the effect is both more general and more profound. The impact is greater because the loss of jobs and income is likely to be much greater. The loss of employment, no matter how small, can seldom be greeted with equanimity, but the effect will not be limited to the initial unemployment. The secondary effect of a decline in a heavily concentrated industry may be that the region loses some of the indirect advantages it once held. For example, en gineering, chemical, and other industries suffered a reduction in out put and efficiency as a result of the migration of the textile industry from New- England. A region will also experience some outward migration of younger members of its labor force. The decline in employment in an industry highly concentrated in a region will induce younger members of the labor force to transfer out of the region. Older workers are less mobile geographically than younger members of the labor force and, perhaps for similar reasons, less adaptable. In addition, there are lifelong associations and family ties that act as deterrents. But aside from sentimental and family considerations, the same impedi ments which exist to the hiring of older workers in one region are likely to exist in the other. In addition, the older worker is unable to utilize his friends and acquaintances in the search for a job if he should transfer to another community. The movement of younger workers out of a region is a serious blow to its economy, for it means that the labor force tends to be made up of a larger than average number of older workers who are likely to be generally less vigorous members of the work force. It can mean that the region’s competitive position with respect to all indus try may be weakened as a dilution in the quality of the worker takes place. The existence of a large amount of structural unemployment is likely to lead to the creation over time of an older and economically less attractive labor force. The deterioration of the labor force places a region at a disadvan tage in attracting new industries, so that structural unemployment once it exists on a substantial scale tends to become part of the eco nomic life of a region. Whether we examine a region such as Rhode Island or West Virginia, w7e are impressed by the continuance of an amount of unemployment greater than the national average. Overall trends in the country affect the economies of these regions, but in good times they are not likely to do quite as well and in bad times 2710 EM P L O Y M E N T , G R O W T H , AND PRICE LEVELS they are usually in a much worse position than the country as a whole. Where structural unemployment is a significant regional problem a number of indirect consequences unfavorable to a local economy may ensue. Chronically depressed areas tend to develop an outlook toward the economic future different from that in the more buoyant regions. An area in which a declining industry generates a large measure o f structural unemployment is likely to affect investment adversely. In parts of New England, for example, the decline of the textile industry created the feeling that the economy was de pressed. In fact, most of the period of the greatest contraction of the New England textile industry has been one of unparalleled national economic growth. The experience of this industry, which historically was o f great importance, adversely affected investment decisions in other segments of New England industry. A large amount of structural unemployment may also attract in dustries and enterprises which provide lower per capita returns for the labor force. In addition, a large permanent pool of unemployed encourages undesirable practices in the use of the labor force. W ork ers are kept on short time. Staggered or alternate shifts can be ar ranged. Under these practices, workers are employed on alternate weeks, which means that a large measure of disguised unemployment exists. In addition, it encourages more workers to depend upon such employment than are necessary or desirable. Irrespective of its cause, a large volume of unemployment inevitably leads to losses of income by the community. Structural unemploy ment falls most heavily upon the older worker because the aged worker is more highly concentrated in declining industries. It is al most a corollary that declining industries are older industries which are likely to have a larger number of older workers in their employ. I f an industry in a region experiences a change in its favorable posi tion and the industry is still growing, the change will tend to show itself in less favorable growth rates than are taking place in other regions. Consequently, whenever an industry generates a considera ble amount o f structural unemployment, it is likely to be an employer o f a greater than average number of older workers. Structural un employment is extremely difficult to erase, for the very devices which are used to escape from its effects—migration— are likely to leave a residue o f difficult problems. One must recognize that once structural unemployment invades a region, it is likely to persist for a long time. A heavy dose of investment would obviously eliminate or at least reduce some of the idleness, but the conditions for attracting a large volume of new capital are not easily satisfied. There is a tendency for the more venturesome, the more vigorous, and innovating to leave the region because of its more restricted opportunities, and their departure makes the path of recovery more difficult to ascend. Experience has shown that labor surplus areas, once they have arisen, do not quickly absorb their surplus labor. It is clear that large-scale structural unemployment generates diffi cult social problems. The inability of distressed regions to generate a sufficiently large amount of investment to replace, lost jobs provides a justification for action by public authority. A variety of proposals have been made; a few of them will be listed here. It has been pro posed that the Government undertake either to invest directly in new EM PLOYM ENT, G R O W T H , AND PRICE LEVELS 2711 construction or to subsidize investment by guaranteeing to investors the security of their capital and a specified rate of return. Tax remission is another method that can and has in fact been used by local communities to attract or retain industries. Rapid amortiza tion allowances might be allowed to businessmen who establish new industries in an area suffering from permanent structural unemployment. Another suggestion that has frequently been offered is the use of some program of contract allocation. Such a plan has been urged upon the Federal Government which is the unit of government best able to effectuate such a program. In addition to these measures designed to promote investment, there are possibilities in plans for the retraining of the labor force and encouraging migration through subsidies. Many other schemes may be specified, but these measures, like those listed above, raise questions concerning equity, effectiveness, and sub sidiary effects. Most of the arguments for and against these proposals as specific means to attain a goal are well known. The fundamental question, however, concerns the rationale of the proposed policies. Is it more efficient, in the economic sense of the word ( 1 ) to attempt to maintain employment in a depressed area by investing social capital in new plant and equipment and in training labor or ( 2 ) to use social resources to shift workers out of depressed areas? We would stress that there is no easy or single answer to this ques tion. There may exist several conditions under which it may be more efficient to invest in a depressed area than it would be to subsidize the migration of labor. Government investment which resulted in the development of an industrial complex, for example, might yield a high rate of return whereas investment by a single firm may be un profitable; or the differences between the social and private costs of migration may be very large. Which mode of action should be undertaken will depend on a number of considerations: the marginal efficiency of investment in industries in different areas, the cost of transfer of workers, the rela tive costs of social services including the costs of social capital, and the magnitude of the gains in real income resulting from transfer are all factors that must be taken into account. Before answers to policy questions relating to structural unemployment can be given, careful and intensive study should be made of the problem elements which have been listed. Social welfare is not advanced by the application of uniform policy measures under all combinations of conditions that arise in practice. It would also appear that the objectives of policy measures may have to be divided into shortrun and longrun objectives. In the short run the purpose may be simply to alleviate distress b}^ bolster ing incomes. The longrun solution may be quite different. Whether this policy is one of rebuilding an industrial base or one of social in vestment in migration is not easily determined. If the problem pres ently has some of the characteristics of a dilemma, it is a dilemma which can be reduced by a systematic application of economic knowledge. Representative B o l l i n g . Thank you, Mr. Taft. Senator Bush, do you have some questions I 2712 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS Senator B u s h . Mr. Chairman, both of these statements have been interesting. In your statement, Mr. Taft, I would like to refer to where you say: In fact, most of the period of the greatest contraction of the New England textile industry has been one of unparalleled national economic growth. The experience of this industry, which historically was of great importance, adverse ly affected investment decisions in other segments of New England industry. I w^ould like to know just what you mean by that. Mr. Taft. Senator, we make some comparisons on the investment in New England, and in Rhode Island, as compared to the same indus tries in other parts of the country. In general they were below7. This is the period between 1947 and 1957. Investment in Rhode Island industry wras low, and investment in New England industry wras low, as compared to investment in the same industries in other parts of the country. Senator Bush. I am not so familiar with the transition in Rhode Island as I am in Connecticut, obviously. I think it has been more difficult. Mr. Taft. I think your experience has been somewhat different from Rhode Island. Connecticut’s experience approximates the na tional experience much more closely than other parts of New England, sir. Senator B u s h . Then this general observation does say in parts of New England. I am not on the defensive or trying to defend Con necticut, but just to find out. I think the record shows that we have made a pretty good recovery from the losses of textiles and so forth. Mr. T a ft. Yes, you have. Senator B u s h . We have been successful in bringing in new in dustries into Connecticut in this 10 -year period and even this year. It has been to a rather surprising degree. We have seen the town of Danbury change over from a town which had 87 percent, I think, of the industry dependent on the hat business, and that went down to 14 percent, and yet Danbury is in good condition from an employment standpoint and has been for a number of years. That has been a very remarkable transition. This has taken place in the eastern part of Connecticut where the demise of the textile business, which hurt very badly for 10 years, has had the slack taken up in the New London area by the chemicals and the submarine business and so forth. I didn’t mean to get so colloquial, Mr. Chairman. Representative B o l l i n g . What industry came into Danbury to re place the hat industry ? Senator B u s h . I think of three industries. One is pencils, which is our most recent one, and which is a pretty good one. Electronics has been the principal one, I would say. Also chemicals. I think in those three areas you find the bulk of the new development. Representative B o l l i n g . I don’t know about pencils, but certainly the other two are growth industries. Senator B u s h . That is right. I don’t know whether pencils is growing or not. Anyway, we were glad to get their plant there. When you say that once structural unemployment invades a region, it is likely to persist for a long time, I suppose that is true. Again, I imagine you have in mind Rhode Island and West Virginia ? EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2713 Mr. T a f t . I have in mind Ehode Island and West Virginia very largely, or some more restricted areas in Pennsylvania. Senator B u s h . Affected b y the coal industry. Mr. T a f t . That is correct, sir. Senator B u s h . I think when you have a case like minerals or coal where the whole thing is attracted by that one industry, and that gives out, or the product is outdated, I don’t think it is as easy to make a comeback as we in some parts of New England have, as it would in a community where skilled labor is more versatile and where it is not dependent on the natural resources of the community. Is that not so ? Mr. T a f t . Yes, sir. I think the point that was raised is that different methods might be used. For example, the Ehode Island labor force is trained in the industrial arts. It is esentially a labor force that a large part of makes its living in manufacturing. The basic skills are there. Even some of the occupations that are used in other industries, the machine tool industry, for example, or in the electronics industry. Some of the skills can be transferred. In in dustries such as mining and lumbering, you have an entirely different problem. There perhaps migration might be a better answer. Senator B u s h . Are you familiar with the bill we have had before the Congress two or three times known familiarly as the depressed areas bill ? Mr. T a f t . In a general sense, sir. Senator B u s h . Do you have any observations to make about that type of legislation at the Federal level or not? I am not pressing you for an answer if you are not familiar enough with it. Mr. T a f t . I wouldn’t want to say offhand without refreshing my mind on the details, sir, just what my position would be. Senator B u s h . In Ehode Island, particularly, which is your own State, I take it, has the State government been helpful or otherwise in connection with the severe problems that they have had there? Mr. T a f t . I think that the State government has tried to be help ful. It has a development corporation and it has encouraged the setting up of a group to finance industry. It has tried. Of course, the possibilities for a State doing very much, especially a State that is facing very severe welfare and unemployment burdens, may not be very great. I think the State has tried to do things, but I think the kind of things that a State can do are severely limited. They are much more limited, I would say, than the Federal Government, be cause its resources are much smaller and its area of operations is much more restricted. I think the State government has tried to help. Senator B u s h . I wanted to ask Mr. Chamberlain a question, Mr. Chairman. In your statement, what are the problems—you are deal ing here with the question of unemployment compensation and ex tending that, a suggestion I had never heard advanced before—you say it has yet to be proved, however, that the person without an in sured future income is likely to take life easy if given temporary as sistance at a level considerably less than his normal earnings. There has been considerable uneasiness in some communities about the abuses of unemployment compensation, and the uses of it for purposes that really were not intended. I am just wondering if a large extension of it along the lines that you are thinking of might not produce con 38563—<59— pt. 8------ 18 2714 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS siderably more difficulty of the type we have in mind, where the opportunities are abused or misused. Mr. C h a m b e r l a i n . I understand your fear. Senator Bush. For taking prolonged vacations at the expense o f the unemployment fund, for instance. That has been done, unfor tunately, to a regrettable extent. Mr. C h a m b e r l a i n . Yes. Senator Bush. Would you comment on that ? Mr. C h a m b e r l a i n . Yes, I would be glad to, Senator. The state ment that you quoted is perhaps a little too sweeping in its nature. I meant it as a general proposition. There have obviously been num bers of instances in which abuses have occurred. I think that what ever tightening in the administration of unemployment benefits can be undertaken to avoid these is so much the better. With respect to the specific proposal that I have made here, I have tried to suggest certain approaches that would help to obviate the abuses which obviously can be built into a system of this sort. I think by tying the additional benefits to years of previous employment, you avoid sweeping into the program those who may simply have been getting along, scrounging on other people’s earnings or income with out actually making any attempt themselves to provide for their own living. This is one safeguard that would be built into the system. But in addition, I think that the three specific measures that I sug gested to avoid including in the supplementary system people who might be withdrawing from the labor market altogether would also preclude some of the abuses which are to be found in the usual normal regular unemployment compensation plans today. For example, the third provision in which supplementary benefits would not be paid to a member in whose family another person was already employed or drawing such benefits would I think preclude a number of abuses that might creep in. Senator Bush. Then I will yield, Mr. Chairman, the possibilities of inducing new industries to move into declining areas, I would leave to local initiative. You mean by that, like the Connecticut Develop ment Commission. Senator C h a m b e r l a i n . Yes, that kind of thing. Senator Bush. Partly to avoid Government entanglement in inter community rivalries and partly because this line of attack seems to be so much less promising. I take it you are familiar with the depressed area bill ? Mr. C h a m b e r l a i n . I am not as familiar with it as I should be. Senator Bush. This statement would be in variance with the pur poses of the depressed-area bill. Mr. C h a m b e r l a i n . My thought in this statement is that while for any one community it might be possible to draw in new industries, incipient industries, the burgeoning industries, for any number of communities which might find themselves in a depressed state to do the same thing is really to hope for too much. They begin to get in competition with each other for the limited number of industries that are developing. It is true that the higher the level of economic ac tivity, perhaps the more possible may it be for depressed areas to in duce new industries to move into their locality. That is, for all of them to do this. But I think when you have the numbers of areas EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2715 which find themselves with substantial unemployment on their hands— chronic unemployment, to use the Labor Department’s term—whether or not this may be structural, it becomes very difficult for all of them to be able to use this route to secure the alleviation of their distressed condition. Senator B u s h . Mr. Chairman, I yield. Eepresentative B o l l i n g . I gather from both papers that it is very difficult to differentiate among the various types of unemployment. Would either of you care to attempt a very rough estimate, under stood as such, as to how much of the current unemployment is struc tural ? We will call it a guess. Mr. C h a m b e r l a i n . I would be glad t o defer to m y senior colleague on that. Mr. T a f t . I have an example. As long as my good friend Neil Chamberlain, has given me the opportunity, let me approach it in this way. In some areas—this is true in central Pennsylvania and in parts of New England and West Virginia, and parts of the Pacific coast—that have had a large amount of lumbering, you find that un employment is always above the national average by a substantial amount. That is, percentagewise it might be one-third as great. Areas such as Rhode Island, parts of Massachusetts and the coal areas, the lumberling areas, and perhaps others have developed some struc tural unemployment. It is not likely to be as serious in Michigan, as a result of the decentralization of the automobile industry, as in parts of New England and in West Virginia. I would not want to give a figure because a figure has the ring of accuracy which and off-the-cuff estimate certainly does not have. But it might amount to around several hundred thousand at this particular point; wouldn’t you think? Mr. C h a m b e r l a i n . I should think that would be a safe estimate. Perhaps one way of getting some indication of the amount of longrange unemployment, whether or not this is due to structural causes— and this is about the best one can do—might be gotten from a few figures here which are not current, but which suggest the general magnitude of the problem. Our absent colleague, Professor Haber, lias been undertaking a study on a sample basis of unemployment in the United States in the recession period. In the 1 2 -month period, October 1957 to October 1958, he estimates that aproximately 2,860,000 persons were unemployed for longer than 26 weeks. A good deal of this was cyclical and not structural. This is simply longduration unemployment within that period. Another basis for estimating would be to take the number of reeipents of temporary unemployment compensation under the Federal scheme, which in New York over the 10-month period from June 1957 to April 1958, approximated 300,000 in that State, This again in cludes the cyclical unemployed and not only structural. But it is a rough estimation of the number of people who were among the long term unemployed. I sh o u ld th in k o n a v e r y r o u g h b asis th a t p r o b a b ly o n e -fo u r th to o n e -fifth o f t h is m ig h t be co n sid e red to be due to stru ctu ra l u n e m p lo y m e n t. Mr. T a f t . A r e g io n m a y h a v e so m e stru c tu ra l u n e m p lo y m e n t even of a seriou s k in d , a n d it w ill n ot n ec e ssa rily m ea n th a t so m e o f the 2716 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS people that have been displaced by the shift of an industry out of the area, or as a result of some technological change, will not find em ployment sometime. I think the characteristics of a region that is suffering from structural unemployment is that there tends to be a higher residue of unemployment in that region than there is nation wide. This tends to persist in good times and in bad times. For example, right now we have depressed areas, and the economy cer tainly is operating on a very high level. That in itself is an indica tion, it w^ould appear, that you are dealing with some kind of unemployment that is persistent, that is of long duration, and that is extremely difficult to get rid. I think that is the characteristic, or those are the characteristics, of structural unemployment. It is recalcitrant. It is hard to do very much about it by the methods that are in use at the present time. Representative B o l l i n g . Thank you. A number of areas have been mentioned as having substantial amounts of structural unem ployment. Perhaps I don’t remember them all. New England in general, with Rhode Island and Massachusetts particularly, West Virginia obviously, portions of Pennsylvania, there w^as passing men tion of the west coast in connection with lumber. That would pre sumably include Oregon and Washington, particularly. What about the mining areas of Montana, perhaps Arizona ? Mr. T a f t . I don’t know about Arizona. I don’t know whether any of the areas there have encountered the problem. Representative B o l l i n g . Montana certainly has shown some signs of this; has it not ? Mr. T a f t . Yes, sir. Representative B o l l i n g . What are some of the other areas, if any, that you think of ? I recognize the peculiarities of the Detroit problem. Mr. ChamberLxMN. I am not sure that I could give a very good inventory or catalog of them. Another one that comes to mind is Gloversville, N.Y., where there has been serious and persistent un employment due to the decline of the glove industry. Representative B o l l i n g . That is an illustration of the point I was getting at. You may have structural unemployment that affects a State, or a small unit. Mr. C h a m b e r la in . Yes. Representative B o l l i n g . Its impact in human terms is very serious, regardless of the size of the unit. Its impact in economic terms varies perhaps with size of the unit, Mr. T a f t . Y o u get a community like Gloversville, where many of the subsidiary industries are dependent upon the spending of the workers employed in these glove factories. It stretches out. It af fects others. It affects the service industries. It affects the indus tries that provide engineering and other kinds of services for the industries, as well as the consumer industries. This is a very serious type of unemployment. But its effect is likely to be limited to the areas that are affected by it. Representative B o l l i n g . At this point, I wTould like to put in the record, without objection, the areas of substantial labor surplus as of September 1955, and they are pretty widely dispersed. In Indi ana, a major area is South Bend, Terre Haute. Massachusetts we E M P L O Y M E N T , G R O W T H , AND PRICE 2717 LEV ELS have already discussed—Minnesota with the Duluth-Superior area, I am not suggesting that because they have substantial labor surplus it all stems from structural unemployment at this particular time. Senator B u s h . What is the date of it ? Representative B o l l i n g . In 1959. In New Jersey, Atlantic City; in New York, Rome, Utica; a number in Pennsylvania, Puerto Rico; Tennessee, Chattanooga and Knoxville; Washington, Tacoma; West Virginia, Charleston; in smaller areas, a number in Alabama, one in Arkansas; a few in Connecticut; Georgia, Illinois a very substan tial number of small areas; Indiana, three more; and so on. It is fairly widely spread. Mr. C h a m b e r l a i n . If I may comment on that, Mr. Chairman, I think the list you have just read underscores one thing. It is a mix ture of areas that have been depressed in part because of structural unemployment, but also because of cyclical changes or industry changes. Your mention of South Bend, for instance; at this time it was in a state of chronic unemployment. A few years earlier it had a very low rate of unemployment. One of the things which I think is needed in this area is more and continuing detailed studies of local labor market areas which would allow the identification of those areas in which unemployment has persisted, where it has not been high one year, low the next, due to cyclical or industrial changes, but where you can identify those areas where the unemployment has persisted over a period of time, and thereby earmark those areas where some relief is more needed. Representative B o l l i n g . I find that I am in error. That was an old table. We will have a new one soon. I will correct the record if we are still in public session. A summary of the September 1955 and September 1959 classifications for labor market areas are shown below: Number of areas Labor supply group Major labor market areas: Total ___________ ____ _________________________________________ Group A____________________ ____ ____________________ __________ Group B . ______ ____ ____________ _ _______ ________ ______ Group C ______________________________________ ___ ___ _ - ___ Group D _____ ______________________________________ _ _ G roup E __ ___________________________________ _ ____ __________ _ . _ Group F _______________________ __________________________________ _ Smaller labor m arket areas in class of “ substantial labor surplus’’___________ N o te .—X indicates m ajor area of su bstantial labor surplus for date show n. September 1955 149 0 40 83 16 4 6 94 September 1959 149 0 27 87 24 6 5 124 2718 E M P LO Y M E N T, G R O W T H . AND PRICE LEVELS M a j o r areas o f substantial labor surpl us A labam a: B irm in gham ______ C onnecticut: N ew B rita in ___ _ _ __ Indian a: E vansville . . South B e n d _____________ T erre H a u t e _____ .. . K entucky: Louisville M assachusetts: Fall R iv er_______ _______ Law rence _ _ _ ______ L o w e ll____ ________ N e w 'B e d fo rd s. . ___ S p ringfield-H olvoke____ M ichigan: D etroit- ___ ____________ F lin t M innesota: D uluth-S uperior_ N ew Jersey: A tlantic C ity ._ . N ew Y ork: A lb a n y - S c h e n e c ta d y T ro y - Buffalo . ___________ U tica-R om e_____ . . . N o rth Carolina: A sheville. - ______ ______ D u rh a m ____ ____________ N o t e .— S eptem ber 1959 Septem ber 1955 State and area X X X X X X X X X X X X X X X X x X X X X X X X X X State and area X X X i Pennsylvania : A ltoona____ __ . ._ ____ Erie Johnstow n- ___ - .. P h ilad elp hia____________ P ittsb urgh S cranton____ ____ ____ W ilkes-B arre-H azleton.._ P u erto Rico: M ayaguez - - _______ Ponce. __________ _ ___ San Juan R hode Island: Providence___ Tennessee: C hattanooga K noxville___ _______ T e x a s : B e a u m o n t- P o r t A rth ur Virginia: Roanoke W ashington: T acom a__ ___ W est Virginia: C h arlesto n .__ . _______ H u n tin g to n -A sh la n d __ W heeling-Steubenville Septem Septem ber 1955 ber 1959 X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X Ix X in d ic a t e s m a jo r a rea o f s u b s t a n t ia l la b o r s u r p lu s for d a te s h o w n . Smaller areas of substantial labor surplus State and area Septem Septem ber 1955 ber 1959 A labam a: A lexander C ity _______ X A nniston __ ____ X D ecatur _________ X Florence-Sheffield IX G adsden X Jasper _____ Talladega A laska: Anchorage _ _ _____ A rkansas' F o rt Sm ith X C onnecticut: A nsonia ___ __ _____ Bristol _______________ X D anielson -_ X ____ ________ M eriden M iddletow n __ _ N orw ich ____ ____ Thom psonville ______ T orrington _____________ X W illim antie __ _____ Georgia: C edartow n-R ockm art___ x Cordele X Illinois: C entralia ____________ H arrisbu rg______________ X H e rr in - M u rp h y sb oroW est F ran kfort. _ _ X L itchfield_______________ X M o u n t C arm el-O lney___ X M ount V ernon ____ X Indiana: Connersville _ ___ __ M ichigan C ity -L aP o rte. . X M uncie ______ X V incennes X Iow a: B urlington. _ _________ x K ansas: C o ffe y v ille -In d e p e n d ence-Parsons _______ P ittsb u rg ____________ __ X N o t e .— X X X X X x x X X x X X X X X X 1 ! X X X X X X X X X State and area K entucky: C orbin __________ __ Frankfort __________ ____ H azard. ________ ______ H enderson- ____ __ _ _ H opkinsville____________ Lexington_____ __ _____ M adison ville________ ____ M iddlesboro-IIarlan_____ M orehead-G rayson___ _ O w ensboro________ ______ Paducah ____ __ _ __ P aintsville-P restonsburg. Pike v ille-W illiam so n___ Louisiana: A lexandria__ ___________ Opelousas___ __ _____ M aine: Biddeford-Sanford___ M aryland: C am bridge. _ __________ C um berland_____________ F rederick. _ ____________ H agerstow n___________ . M assachusetts: F itch b u rg ________ ______ M ilford__________________ N orth A dam s. _ _______ S o u th b rid g e-W eb ster.__ M ichigan: A drian________ . Bay C itv ______ ___ E scanaba________________ Iron M o un tain. ._ ___ M arquette ______ . .. M onroe- . . . . . . . . | P o rt H uron . __ . .. M ississippi: Biloxi-G ulfport_________ G reenville. __________ in d ic a t e s s m a lle r area s o f s u b s t a n t ia l la b o r s u r p lu s fo r d a t e s h o w n . Septem S eptem ber 1955 ber 1959 X X X X X X X x x X X X x X x X x x x X X X X X X X X X X X X X X X X X X X X X X X X X X X EM PLOYM ENT, G R O W TH , AND PRICE 2719 LEVELS Sm aller a r eas of s u b s t a n t i al labor s u r p l u s — C o n tin u e d | Septem - j Septem- j ! ber 1955 | ber 1959 State and area j M issouri: Cape G irardeau_________ W ash ingto n-. - - ___ M ontana: N ew .Tersev: B ridgeton__ - - . Long B ranch____________ N ew Y ork: A m sterdam _____________ A uburn E lm ira Glens Falls-H udsonF alls. H u d so n _____ _.. .. _. Jam estow n-D unkirk N ew b urgh -M idd leto w n Beacon _ . _ _ _ 0 lean-Salam anca. _ O n eid a.__ __ .. . ... P la ttsb u rg h _________ . . . W ellsville. ________ __ N orth Carolina: F ay etteville_____________ H endersonville _ _ K inston __ __ L um berton... . . M ount A irv - - ___ R ockington-H am let .. _. R ocky M o u n t. _ ___ Ru th e rf o r d t o n -F o re s t C ity. ___ _ _ . Shelby-K ings M o u n ta in . W aynesvilie _ . - W i l s o n - ____ _ -. . . . Ohio: A th e n s-L o g a n -N e lso n ville. _ . . . ___ _____ C am bridge______________ M arie tta ________________ P ortsm outh-C hilli cothe O klahom a: A rdm ore____ __________ _ M cA lester. .. . ___ M uskogee O km ulgee-H enryetta. _ Pennsylvania: Berw ick-Bloom sburg___ B radford_______________ B utler _. ________ Clearfield-D uBois. . . . __ In d ian a. __ . . . K ittanning-F ord C ity ___ N o t e .— X X X X X X X x x X X X X X X X X X X X X X X State and area Septem -1 September 1955 ; ber 1959 i j P enn sylv an ia—Con. 11 L ew istow n___ _______ ! x |J Lock H aven_______ .... _ i x M eadville _ ... ______ i X ;| N e w c a stle __________ 1 x H Oil C ity-Franklin-T itusj ville ____... . _ X X j Pottsville j St. M ary ’s ______________ X j Sayre- A thens-To w anda. _ | S un bu ry -S h am o k in -M t. I Carmel________________ X i U niontow n-C onnellsville. X X j W illiam sport . . . Rhode Island: N ew port. South Carolina: X M arion-D illon X W alterboro . . . ______ Tennessee: B risto l-Jo h n s o n C ity X K ingsnort L aF o llette-Jellico -T azeX w ell. . . __ . Texas: Laredo _ T exarkana __________ X V erm ont: i B urlington X Springfield X Virginia: Big Stone G ap-A ppalaX chia ... C ovington-C lifton Forge. X R adford-Pulaski _______ X R ichlands-B lucfield_____ X W ashington: A berdeen A nacortes. _ .... Bellingham .B rem erton . ___ P o rt Angeles W est Virginia: X Beckley _. X Bluefield _ __ _ _ _ C larksburg___________ .. X X F airm ont X Logan M artinsburg X M organtow n. _ _ Parkersburg _. ____ _____ X P o int Pleasant-G allipolis. X Roncevert-W h i t e S u 1X ph ur Springs W elch _______ X W isconsin • L a C-rossc I ix ix \ X X X X X x X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X I X IX !X jX iX ;X iX jX X 1X IX X 1 X X X ]X !!i x X X X X X X X X X X X X X X X X X in d ic a t e s m a jo r a rea o f s u b s t a n t ia l la b o r su r p lu s fo r d a t e s h o w n . E x p l a n a t io n of A rea C l a s s if ic a t io n s One of the six overall objectives of the Federal-State employment security program is “ to develop and disseminate employment, unemployment, and labor market information in order to assist in achieving economic stabilization and growth, and to meet the informational needs of labor, management and the public.” Among the major measures established to carry out this objective is the Bureau of Employment Security program of classifying areas according to relative adequacy of labor supply. These area classifications are intended to provide a quick, convenient tool to measure comparative differences in the availability of labor in the Nation’s major production and employment centers. These condensed, summary indicators of area labor market conditions have been, widely used by Government agencies and private organizations in the plan ning, administration, and evaluation of manpower programs and policies ever since the area classification program was first initiated in the early da vs of World W ar II. 2720 E M P L O Y M E N T , G R O W T H , AND PRICE LEVELS Area classifications represent a synthesis of a number of key elements which reflect the nature and the character of an area’s present labor market. The area classification for each area blends together pertinent data on the current level of unemployment in relation to the size of its labor force, on changes in employment and unemployment in comparison with several recent periods, on the area’s employment and unemployment outlook, as reflected by employer estimates of their manpower requirements, on the size of the area’s labor demand in comparison with available labor supply, and on the seasonal pattern of local employment and unemployment fluctuations, into a single symbol which char acterizes the status of that area’s labor market in comparison with those of other areas throughout the country. Area classifications thus permit general comparisons to be made between areas, comparisons which are not feasible through the use of any other single statistic. The present classification criteria, which have been in effect since May 1955 group the areas into six major labor supply categories. Classification group ings are designated by letters ranging from A to F, with group A reflecting the relatively tightest labor supply and group F the relatively greatest labor surplus. Areas classified in categories D, E, F are regarded as meeting the require ments for designation as “ areas of substantial labor surplus,” or “ areas of sub stantial unemployment” for the purposes of Defense Manpower Policy No. 4, the policy on accelerated tax amortization for labor surplus areas, and Executive Order 10582, implementing the Buy American Act. A summary of the criteria used for each of the individual classification groups is listed below. Classifications made under these criteria are not directly com parable with classification ratings assigned under system used prior to May 1955. Area classification criteria 1. C U R R E N T L A B O R S U P P L Y -D E M A N D S IT U A T IO N G roup A G roup B C u rren t critical labor shortage; expected to continue a t least through next 4 m onths. G roup C Job oppo rtu ni Job seekers ties for local slightly in excess of job w orkers slightly in openings; this situ a excess of job tion ex seekers; this pected to situation ex continue pected to over next 4 continue m onths. over next 4 m onths. G roup D Job seekers in excess of job open ings; this situation ex pected to continue over next 4 m onths. G roup E G roup F Job seekers Job seekers substantially considerably in excess of in excess of job openings; job openings; this situ a this situ a tion ex tion ex pected to pected to continue continue over next 4 over next 4 m onths. m onths. 2. R A T IO O F U N E M P L O Y M E N T TO T O T A L L A B O R F O R C E Less th a n 1.5 percent. 1.5 to 2.9 per cent. 3 to 5.9 p er cent. 6 to 8.9 per cent. 9 to 11.9 per cent. 12 percent or m ore. 3. N E T N O N A G R 1 C U L T U R A L L A B O R R E Q U IR E M E N T S F O R 2 A N D 4 M O N T H S H E N C E IN D IC A T E — Some increases in em ploy m ent. Sizable em ploy m ent gains. N o significant D eclining increases in em ploym ent levels or no em ploym ent. significant increase. D eclining D eclining em ploym ent em ploym ent levels or no levels or no significant significant labor require labor require m ents. m ents. 4. E F F E C T S O F SE A S O N A L O R T E M P O R A R Y F A C T O R S T he current and anticipated labor shortage not prim arily due to seasonal or tem porary factors. Reflect signifi Reflect signifi T he current or T he current or T he current or anticipated anticipated cant seasonal cant seasonal anticipated substantial labor surplus fluctuations fluctuations labor surplus labor surplus not due p ri in em ploy in em ploy not due p ri m arily to no t due p ri m ent and m ent and m arily to seasonal or seasonal or m arily to unem ploy unem ploy seasonal or tem porary tem porary m ent. m ent. tem porary factors. factors. factors. N o t e . -A reas m ay also shift betw een groups D , E , and F in response to significant seasonal changes in em ploym ent and unem ploym ent, b u t w ill no t be m oved in or out of group A or betw een groups C and D as a result of prim arily seasonal or tem porary fluctuations. E M P L O Y M E N T , G R O W T H , AND PRICE LEV ELS 2721 Area classifications are normally issued at bimonthly intervals (in oddnumbered months— January, March, May, etc.) by the Bureau of Employment Security of the Department of Labor. A total of 149 of the Nation’s major labor markets are regularly classified into the 6 labor supply groupings. The 149 major labor market areas regularly classified by the Bureau of Employment Security according to relative adequacy of labor supply account for about 34 million nonagricultural wage and salary workers. This represents nearly 70 percent of the Nation’s total. In addition to the 149 major areas, the Bureau of Employment Security also classifies smaller areas (ranging in size down to those with a labor force of 15,000) when they have relatively substantial unemployment. Such areas are designated as “ smaller areas of substantial labor surplus,” but are not placed in a specific classification category. From the standpoint of Government con tract awards under Defense Manpower Policy No. 4, it makes no difference whether an area is classified in group D, E, or F, or as a smaller area of sub stantial labor surplus. Each is regarded as a substantial labor surplus area and each receives equal preference under existing programs to assist labor surplus areas. The area classifications are assigned on a “labor market area” basis rather than to individual cities or communities. A labor market area consists of a cen tral city or cities and the surrounding territory within a reasonable commuting distance. It may be thought of as an economically and socially integrated, pri marily urban, geographical unit within which workers may readily change their jobs without changing their places of residence. A labor market area takes its name from the central city or cities, but may have many other com munities within its boundaries. Major labor market areas usually have at least one central city with a population of 50,000 or more, according to the 1950 census. In most instances, boundaries of major labor market areas coincide with those of standard metropolitan areas, as determined by a Federal interagency com mittee chaired by the Budget Bureau. Definitions of all classified areas are listed in a Bureau of Employment Security publication entitled “ Directory of Important Labor Market Areas.” This publication also lists all major communities located within the boundaries of the defined labor market areas. The area classifications are assigned according to uniformly applied criteria. They are based on labor market information—both narrative and statistical— submitted to the Bureau of Employment Security by affiliated State employment security agencies under a regular labor market reporting program. These reports are prepared locally, drawing on the vast amount of information avail able in local public employment offices, according to standard outlines, methods, and techniques. The usefulness of the area classifications is thus enhanced by their comparability and uniformity. The extent of unemployment in a particular area is, of course, a key factor in determining the appropriate area classification assigned to each locality. It is not the sole criterion used in classification, however. Consideration is also given to the area’s employment outlook, as reflected by local employer estimates of their manpower requirements, to the significance of essential activities, to the relationship between labor supply and demand, to the seasonal pattern of em ployment and unemployment fluctuations, and to several other factors. Senator Bush. What you want is a recent table. Representative B ollin g. That is right. We are having it sent up. What I wyould like to get at next is another area which may not fall into the category of structural unemployment. It may be a question of underemployment. That is the whole problem of rural unemploy ment and underemployment. Would either of you care to comment on that problem? I have a feeling, based on some studies that are fairly old by now, that there is a very substantial amount of very severe underemployment which almost amounts to structural unemployment. Mr. T a f t . I suppose the unemployment in rural areas is likely to be largely disguised, sir. In other words, what you are likely to have in a rural area is more people working at some jobs that are tech 2722 EM P LO Y M E N T, G R O W T H , AND PRICE LEVELS nically dispensable. This would be especially true in agriculture, on the family farm. You may have more people employed than are needed to fulfill the particular tasks efficiently. I suppose the answer to that kind of problem is an expanding and active industry. Because the flow of labor has been historically from the rural areas to the cities and to the industrial locations, this wrould appear to be one answer, that is, a prosperous and expanding economy is one of the needs to absorb this surplus labor. With increased mechanization, certainly agriculture does not offer any possibilities for reemployment. We have to think in terms of an expanding economy if we are going to take care of those people and provide them with jobs. That would appear to me to be the answer to the problem in general. Representative B o l l i n g . D o you have any comments to make? Mr. C h a m b e r l a i n . Yes, sir. I think by and large, although I have made no special study of this particular type of unemployment, the problem you mention is not as serious as the problem of structural unemployment in the settled urban areas. I would say that for this reason. Where the underemployment comes from the inadequate use of young people on the farm, the answer here is an improved educa tion, improved guidance to the youngsters, to send them into more prosperous and more promising forms of employment. Where the underemloyment is due to a lack of older people’s proper utilization on the farm, I think I w-ould consider this more in the line of the alle viation of private distress. As long as they are getting a satisfactory subsistence income from such operation, if they do face the problems of relocation attendant to older persons, I would be disposed to leave them on the farm where they can provide their own needs. I would emphasize the shifting of the younger people from the farm to the more promising areas. Representative B o l l i n g . As a sort of aside, What about the prob lem of undercapitalization, both in land and machinery ? Mr. C h a m b e r l a i n . You are thinking of the small farm. Representative B o l l i n g . I am thinking of the small farms. There are still a very great many of them. I think, based on a study by a subcommittee chaired by Senator Sparkman some years ago, it was demonstrated that a very substantial number of people, particularly in the Southeast, were existing on farms that were uneconomic per se. Mr. C h a m b e r l a i n . Yes. Representative B o l l i n g . Some of them might have been made eco nomic if there had been the possibility of a sufficient capital invest ment. You have the problem of more than the individual. When you combine that wTith the problem of a certain amount of structural unemployment indicating an underutilization of the manpower, you don’t have the situation whereby the manufacturing economy can absorb these people. That is the only reason I raise the point. It seems to me to link into the total problem from the national point of view. Mr. C h a m b e r l a i n . Yes. This would involve a matter of indus trial organization. These farms are organized in inefficient forms. If they could be brought together into larger producing units, per haps in some sort of cooperative arrangement. EM PLOYM ENT, GROW TH, AND PRICE LEV ELS 2723 Eepresentative B o l l i n g . A s a matter of fact, they supply a very small amount of the very large surpluses we have in certain areas. Mr. C h a m b e r l a i n . Yes. Eepresentative B o l l i n g . Taking as a beginning the last paragraph of Mr. Taft’s statement, it would also appear that the objectives and policy measures may have to be divided into shortrun and longrun objectives, and so forth. I gather the short run is in effect the prob lem of meeting the needs of individuals. Mr. T a f t . That is right. I would tend to go along—I have not given Neil Chamberlain’s proposals too serious thought—but it is in that direction which something can be done. That is, linking a sup plementary unemployment benefit to years of employment. I think that is the answer to the question raised by Senator Bush, which is a real problem in administering unemployment compensation—what do you do with malingerers, what do you do with people that are not attached to the labor market—his criteria of eligibility is along protracted attachment to the labor market. It seems to me as a shortrun proposal some scheme of this kind, supplemented by others, has a great deal of merit. On the other hand, the longrun objectives— and the point was raised here, too—in some of these distressed areas you have a population that is trained in the industrial arts. This gives them some advantages as far as being absorbed in other indus tries is concerned. In an old mining or old lumbering region you may have a completely different kind of problem, and you may need other devices to meet it. Consequently, the suggestion that was made about continuing study seems to me a desirable one. I think you made it, Mr. C h a m b e r l a i n . Yes. I think the continuing study might very well be urged by the Federal Government on the States, as I believe in fact our Department of Labor has done. If I am correct, the Department of Labor has even provided a generrl guide for the study ■of structural unemployment to the State labor offices. If more of this sort of thing can be done, then we can induce our State depart ments of labor and the State industrial commissioners to have a continuing weather eye on the development of these pockets of unemployment. I think Mr. Taft’s repeated insistence here about the difficulties of doing anything about an area that has been subject to structural unemployment for any period of time should act as a warning to such State agencies. If they can catch these situations at the time they are budding and do something about them promptly, it is much easier to remedy the situation than after it has been allowed to go on for a long enough time to give the area the name of a declining economic setting, which is not going to help either the attraction of new industry or the reemployment of those who are out of work. Eepresentative B o l l i n g . I certainly agree with that. If we could do a little more anticipating of the problems we would do a great deal better job of solving them. Mr. C h a m b e r l a i n . Yes. I think this can best be done at the State level. Eepresentative B o l l i n g . I think that is true. One of the difficulties is that the powers that be in any given community very much resent the thought that they are declining. 2724 E M P LO Y M E N T, G R O W T H , AND PRICE LEVELS Mr. C h a m b e r l a i n . Yes. Representative B o l l i n g . Or facing decline. That causes a tend ency to refuse to face the set of facts. Mr. C h a m b e r l a i n . Yes. Representative B o l l i n g . I gather from these answers that you would both agree that on the first half, the shortrun portion of this division, enough is not now being done by all agencies of government combined to alleviate, whatever the right word is, suffering of the people subject to long-range unemployment. Mr. C h a m b e r l a i n . That w o u ld certainly be my opinion. Mr. T a f t . It would be my opinion, that more can be done. Representative B o l l i n g . Moving into the more difficult and less understood area—I suppose we simply don’t have enough facts yet— What kind of longrun solutions would we want? In your opinion— I would assume obviously from what you have said so far that we are not doing enough there, all the agencies of government com bined— are we even doing enough studying of the longrun solutions, all the agencies of government combined. Mr. T a f t . I don’t think so, sir, because we know as much about it today as we did 10 years ago. For example, one of the questions we should have answered about an area that has generated a higher than average amount of unemployment is the following: Is it possible by directed investment to create a new kind of industrial complex which would permanently reduce the level of unemployment to the national average? We don’t know that. If you “force”—I use the word in quotes—investment into an area that cannot maintain itself economically over the long run, then once you remove the props, it won’t be able to exist, and your problem would arise again. The question really is, Can you by some kind of aid create conditions which will enable the industry or the enterprise to compete successfully in its particular part of the economy with other industries operating in other parts of the country? Unless you can do that, then there is really no point in seeking to maintain old levels of employment. You are really prolonging the transition rather permanently curing the problem. That seems to me one of the reasons why we need per sistent and continual study of specific types of remedies rather than a general analysis. We have to find out what there is about an area that generates a higher than average amount of unemployment con tinually. Is there something that can be done ? Representative B o l l i n g . Thank you. I have one very specific question in an entirely new field. Would willingness to accept retraining or relocation be a condition of eligibility for supplementary unemployment benefits ? Mr. C h a m b e r l a i n . I have turned that over in my mind. I think it is quite possible that you would have to try this out experimentally before you could answer that question satisfactorily. It has an attractive ring, but I think it would have to be hedged about with safeguards. I would say that if a man could be offered a job that it was not a matter of his moving to an unknown situation where he did not know whether he would be able to locate a job or not, but if the man in the distressed area could be offered a job opportunity which was known and identified and where he would be able to fill it on the basis of his past training—then I would be inclined to say EM PLOYM ENT, G R O W T H , AND PRICE LEVELS 2725 that failure to accept that employment in another location should remove him from the right to obtain supplementary benefits. In effect, this is continuing the standards of eligibility under State unemployment compensation systems now except that it is on an interstate basis. At the present time we require a man to acccept reasonable employment as part of the requirements of our unemploy ment compensation program. Failure to accept a reasonable job opening removes him from the compensation benefit roster. The same thing could be done on an interstate basis if you had this kind of Federal supplementation. Failure of a man to accept a rea sonable job opening in another locality, even in a different State, would remove him from the right to receive supplementary unemployment benefits. I don’t think we should say that the offering of any kind of a job or the problematic finding of a job in some other area should remove him from the roster. Mr. T aft. If I may comment on that: This, it seems to me, is a reasonable test, because if there is a job and the individual refuses to take it, then he is making the decision— and there is no employ ment in his area—he would prefer to remain idle where he lives rather than move for good enough reasons. I don’t think there is any moral question involved here. Then you could set that up as a test of eligibility which would meet such a problem. I may also say that continuing unemployment calls for a relaxa tion of the standards that we use or a tightening up of the standards that we use to determine what constitutes an acceptable job. For example, a man is not required in most States to take a job except the one that he normally follows. I don’t think this would be allowed, for example, if his industry were completely eliminated or his skill was no longer significant in the labor market. He would be allowed to draw benefits for a few weeks perhaps, but then he might be asked to take a job, even one paying a lowyer amount. I think it has been the rule that this is a proper w^ork test in defining “eligibility.” I don’t think Mr. Chamberlain’s test would in any way be unreasonable. Mr. C h a m b e r l a i n . Mr. Chairman, if I may follow that up for a moment: Obviously, the important thing is to try to get these people into new job opportunities. This is much preferable to paying them any additional compensation. In this connection it strikes me that although we certainly do need more studies of the kind that Professor Taft has mentioned, we also need a little experimentation with ways of doing this. We have read all about the electronic mar vels of inventory taking, and in the airlines where they can find out if there is an opening on a specific plane at any office in the country and sell it to prospective customers. Railroads know which box cars are empty and available for assignment wherever they may be. I think we can do a better job of matching available men with avail able job openings if we had an improved system of finding out the requirements of each and matching them 011 a national basis. Representative B o l l i n g . This would be done through the State employment security offices ? Mr. C h a m b e r l a i n . It could be tied in with the State employment security offices, but it would have to be done on a national basis if 2726 EM P LO Y M E N T. GROW TH, AND PRICE LEVELS you are going to effect this interchange of people and jobs. W e don't have adequate ways of finding out these facts now, quickly, promptly, and certainly. If our commercial firms can make use of computers for this purpose, I see no reason why we cannot put them to use for this kind of an objective. Representative B o l l i n g . May I ask out of sheer unadulterated ignorance if we had any such system working during World War II ? Mr. C h a m b e r l a i n . Not to my knowledge. I am sure there were efforts at matching people but not on any kind of electronic basis. Mr. T a f t . There is somehow an effort to shift workers between areas through the employment service. I don’t think it is too efficient. Senator B u s h . Especially within a State, I would say. Mr. T a f t . Yes. Representative B o l l i n g . N o w I am going to shift entirely. Both of you have placed major emphasis 011 Government policies to deal with structural employment, and nothing has yet been said about the role, if any, of collective bargaining in the situation. In your opin ion, do collective bargaining provisions, pensions, seniority, and so on, tend to immobilize workers ? Are there any developments that you know of in the areas of obvious chronic structural unemploy ment that would give light on this question ? Mr. T a f t . Offhand, I would doubt it, sir. I say that because I know that the argument has been made by my colleagues in the field that this is the effect of seniority systems and pension systems, but it would seem to me a man of 45 who loses his job because a mill or some other kind of factory closes down, after he has been out of work about 6 or 7 months is not going to be deterred by his possible pension twenty-some years later. I think there are certain kinds of restraints upon mobility which are more important. That is, family ties, and the uncertainties that exist in most people’s minds about moving to another area. Of course, there is a great deal of movement in the United States. But what wre need in these depressed areas is a larger than normal movement. We always say that movement takes place at the margin and this is sufficient to bring about the needed adjustments. But what we need here is a greater than average movement, This is where you run into difficulties. I think of personal kinds of problems, of fi nancial problems. Movement is expensive. There is no certainty that when a man moves from one place to another that he is not going to face the same barriers if he is an older worker. If he moves from New England to the Midle West, he still has to search for a job, and he may not have friends or relatives to help him. If he is an older worker, he will run into the same problems of limitations on hiring. He may not have the skills. So I think movement is difficult. I think this is one of the reasons that Neil Chamberlain’s suggestion about having some central area that we could keep a continual inventory of avail able jobs may have some use. Mr. C h a m b e r l a i n . I would certainly agree with Professor Taft on that. When you raise the question whether seniority provisions and pension programs immobilized workers, I think we would have to break that down into two parts. One is: Have they tied workers to jobs? I think the answer here, from my point of view, would be: To EM PLOYM ENT, GROW TH, AND PRICE LEVELS 2727 a very minor extent. Our studies of labor markets have indicated that it is the younger person, who is lesss affected by pension plans or seniority, wrho does most of the movement. Most of our job changing comes from young people, say in the first 5 years of their employment. They are not going to be affected by pension plans or seniority systems as much as the older people. Representative B o l l i n g . I am not sure but what that argument is the other way. It seems to me that the people who are likely to be immobilized by a pension plan are the older workers. Mr. C h a m b e r l a i n . My point is that they would be immobile with out the pension plan or without the seniority system. We find by and large it is age itself that provides immobility. It is not the addition of collectively bargained pension plans or seniority systems that have brought that situation. Age itself creates a certain immobility factor. It seems to me that this element is not really germane to our present discussion because now we are talking about people who are already 011 jobs. But when you move into a situation wrhere they have been thrown out of wrork by structural unemployment, then it seems to me here you are faced with a different kind of situation. Studies of labor markets that have been made—I am thinking particularly of one that Professor Reynolds made of the New Haven market and Professors Myers and Shultz made of the labor market in a Massachusetts com munity—have indicated that people tend to stay in their given commu nity out of association with given people. This is a period of insecurity to them. They feel more secure living with friends and family situa tions, with always that lingering hope that a job will open up again after all. But this kind of situation tends to change if there is a specific job alternative offered to them. It is the absence of oppor tunities which gives them this feeling of insecurity and the desire to stay rooted where they are. When these same people are given oppor tunities of reemplovment in a different situation, they will move to them. Senator B u s h . This prompts me to go back to a question I wanted to ask you at the bottom of page 2 of your statement, where you speak of the immobility of the unemployed due to a number of causes, among which are homeownership in an area of declining real estate values. Then you say they dislike breaking away from their circle of friends. On that homeownership thing, would you expand that a little bit? Are you saying that because of the situation in the area that real estate values are declining and they are reluctant to sell at a loss ? Is that what you mean ? Mr. C h a m b e r l a i n . Yes. As you know we have prided ourselves on the proportion of our population which own their own homes. Homeownership goes right down into the working class. We can say this is good. We all like this. It is a nice aspect of our society. But it also means that they are creating an anchor to a community. If you are dealing with an area which has become depressed, where people are not moving in, and hence real estate values have declined, if you are trying to create mobility for them, you are really asking them to sell at a distressed situation, to recover from the sale of their homes, if they can sell it at all, only a fraction of what they put into it. I am not sure what the precise answer is to this. I think some real estate experts coul d hel p us on thi s. 2728 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS Senator B u s h . They are reluctant to dispose of their home at a price which would destroy a substantial part of their lifesavings. Mr. C h a m b e r l a i n . That is right. Senator B u s h . Of course, the sentimental ties are very important and very understandable. You have people whose fathers and grand fathers lived in the community and they are buried there, and they have been going to schools and churches there. Mr. C h a m b e r l a i n . These are the most difficult things to do anything about. Senator B u s h . This is one of the difficulties I have had in thinking about the depressed area bill and picking up and moving of communi ties from one part of the country to another and financing it by Fed eral Government grants. I have a lot of trouble with that from a sentimental angle. However, I can see where in areas like certain parts of West Virginia and Pennsylvania, where the natural resources of the community are exhausted, it is going to be awfully hard to bring that back to anything like it was. Mr. C h a m b e r l a i n . Yes. If you can keep the people together who have been associated with each other in the distressed area on such a move, I think this lessens the uprooting effect, the feeling of being cut off from friends. Senator B u s h . Let me say that I feel, and see if you agree with this, if the Federal Government would interest itself in this problem which is certainly a matter of concern to the Government, I would like to see our efforts more directed toward assisting communities to rehabili tate, to attract industries there, than I would to cooperate in the trans fer of people and industries from one part of the country to the other. Mr. T aft. Both plans present serious difficulties. The problem of rehabilitating the community is whether you can maintain employ ment at the old levels; whether competitively the industries can survive in those areas. It is certainly conceptually conceivable that some communities will not be able to maintain the level of employ ment that they had at some other time. We know historically com munities have gone down in population, and there is no reason to assume that this process is not a continuing one. I would say, Senator, one of your problems is to determine whether industry can be maintained after the initial push from the outside, on a competitive basis, or will it need continual injections of aid. It it does, then some other approach might really be not only the more desirable economically but actually from the point of view of the community. You would not build up a whole series of social services, for example, that cannot be maintained. We could help in a gradual transition so that there are no catastrophic secondary effects upon the community, so that its income does not decline too sharply. Mr. C h a m b e r l a i n . I think I have mixed feelings on this approach. I think the mixture could be separated out by saying that there are some kinds of situations to which your solution would be effective and others where I think it would be ineffective. This comes back, I believe, to the point that we were talking about a little while ago : How quickly you can catch a situation that is developing. If it has developed to the point, as I think Professor Taft was pointing out in his paper, where you have a persisting situation that has gone on for a period of years, it wTould be much more difficult to do something E M P L O Y M E N T , G R O W T H , AND PRICE LEVELS 2729 about that than if you were to catch it in its earlier stages when your proposal would have much more likelihood of achieving results. The other consideration that might be present also was suggested by Professor Taft. It may be insufficient to lure a single company or industry into a region or declining area, but if you can do some thing about a complex of activities—that is, give the area a completely different complex from what it had before—this may be satisfactory. I don’t know whether West Virginia has potentials for being a new tourist area. Conceivably it could be built into this, changing the complex of that area altogether. Senator B u s h . That is what they did in Danbury. That is a good illustration in our State. Mr. C h a m b e r l a i n . T o some extent, that is true. Representative B o l l i n g . Mr. Frucht. Mr. F r u c h t . I would like to make some preliminary remarks, and then ask some questions along the line that we have been discussing. What we are discussing, it seems to me, is part of the very general problem created by economic change. Economic changes— favorable economic changes, at least—benefit people in a very general way. The benefits are widely dispersed. Very often the costs of these changes are very narrowly concentrated. The incidence falls on a small group. There is a very big problem which I suspect we will be worrying about for some years to come as to what extent the people as a whole who benefit from the change might w^ant to assume some of the burdens of those changes from the shoulders of those who are particularly affected. When we come to the question of structural unemployment, we come to a problem which is beginning to become quite significant, I think, in terms of the thinking of American people, and in terms of the problems concerning us. There is the question of relief of human distress. There is also the question of obtaining maximum use of our resources. If we really w^ant to raise our economic growth level, one way to do it is to use resources that are unnecessarily idle. Coming to the question of Federal Government or State govern mental policies in regard to economic change, I think we have two traditional approaches. One is, let things go and let the victims make out as best they can. Of course, there is the hope, and very often justified, that the victims will adjust, take steps to protect themselves, and improve their positions. The other approach is to prevent the economic change from occurring. That is as traditional as the former one. Tariffs, subsidies, what have you, are as old as economics. I suspect that our goals are inconsistent. The goals that we as a people have are inconsistent with both of these approaches. On the one hand we are concerned about human distress. We are not quite prepared, I think, to let individuals suffer enormously for the sake of a few. So the approach of government that says ado nothing,” is encounter ing serious resistance from the American people. On the other hand, if we are interested in maximizing economic growth—that is, in governmental policies toward that purpose—the approach of stopping economic change is equally undesirable. I think our policies to date are pretty much suspended between those two approaches, and subject to question. 38563— 59-~pt. 8— 19 2730 E M P L O Y M E N T , G R O W T H , AND PRICE LEVELS There is another approach which might lie along the lines that Mr. Chamberlain has been moving. I was very much interested myself in this line of attack. One of the key assumptions that under lies a theory of competitive economy is that of knowledge, which is unattainable in the real world. But it would seem clear that any measures which would improve knowledge are going to be conducive toward the better operation and functioning of the system. So that your computer approach, I think, has considerable merit. I wTould like to ask you some questions with regard to the application of this general principle. You have spoken about an improvement in the provision by the State and Federal Governments of better information to the worker as to job opportunities. I am concerned as to whether this is enough, although clearly any improvement here per se is going to make a difference. Will a worker in New Bedford or Woonsocket, R.I., 45 years old, trained in loom tending or what have you, move to San Diego wThere there is a tight labor market because USES tells him there are jobs in San Diego? Mr. T a f t . If h e is 38, h e m i g h t m o v e . Mr. F r u c h t . He may move. Mr. T a f t . Actually there is a great deal of movement, as you know. The question, I think, that arises is whether we could, stimulate selec tive movement. In other words, here is a depressed area. If move ment is the answer—we are not sure—maybe rebuilding the industrial area is the answer because of the existence of all kinds of social serv ices. If you go to San Diego you may not be able to get your child in school. We have a lot of empty schoolrooms in Providence. There are questions of that kind which we have not discussed. A man of 45 may not want to go but someone who is 41 might, I think if you could induce some additional movement and concentrate your movement to an expanding area, if possible, from the areas that are depressed, I think you would have clone something to alleviate the problem, al though perhaps not solve it. I should have let you answer it. Mr. C h a m b e r l a i n . No. I think your emphasis on selective move ment is good, Professor Taft. Your point is well taken. I don’t think simply the provision of information as to the availability of a job will normally be enough to induce movement. It is one of several factors or one of a number of factors that might be involved. Maybe your worker owns his home. Then we are back in the real estate situ ation. Maybe it is a matter of his really not knowing enough about the job. Then we have to give him more information about the nature of the job opening, the company in which it is located, and perhaps the community. As Professor Taft was suggesting here, he may be concernecl about school systems. He may be concerned about religious opportunities and so on. It is interesting in this connection that a friend of mine in Connecticut has been talking with a number of companies up there about the development of some documentary films about local labor markets, the Stamford labor market, the Bridgeport labor market, which would run 20 or 30 minutes, and which companies in that area could use for recruiting purposes by showing them to workers in other communities in California, Michigan, wherever it might be, a picture EM PLOYM ENT, G R O W T H , AND PRICE LEVELS 2731 of what the town is like, and what it means to w7ork in these communi ties. The more information of this sort that we can provide—simple things of this kind—will add to a feeling of reassurance on the part of this worker whom you want to induce to move from somewhere on the east coast out to the Middle West or Far West. The more you can reassure him, and facilitate that movement, the more likely he will be to make it. Mr. F r u c h t . I wonder if you will comment on the feasibility or the problems that might be created if there were movement made by the employment services of the States, sparked by the Federal Gov ernment, perhaps, to go one step further and actually try to broker jobs, so that the man in New Bedford who is moving perhaps to San Diego is moving to an actual job. In connection with that, I wonder whether you feel it might be useful and we repay the cost for there to be testing of structural unemployed workers, an assessment and evaluation of their experience, their job records, their competence, and also their as yet unused talents which may be hidden, which would go along to bridge this gap. Mr. C h a m b e r l a i n . Yes. Mr. T a f t . The employment service does some of what you are sug gesting that it do, but it does it for selective jobs. I think one of the problems is that you are likely to direct people to areas where there are expanding opportunities rather than specific jobs. The employer, unless he is in an area wThere there is no available labor locally, or where he needs certain skills that are not easily obtainable, is likely to depend largely upon the local labor market, and the people from thf outside move into the local labor pool from which they are eventually fished out for jobs. I think it involves very serious problems that have to be thought out. That is, getting a man from one point to another, and the farther he is separated from his job, the more difficult the problem is. To find a job somewhere else is difficult. The man does not really know what kind of job will be available when he gets there, and you are asking him to spend several hundred dollars to go out into a new section of the country. This problem of migration, of shifting people to specific jobs, has some attractive features, but it raises some very difficult problems for the individual. Mr. F r u c h t . There are various approaches. One may take the approach of maximizing voluntary action on the part of the individ uals involved, with minimum interference with actual markets, in the direct sense of wage fixing or investment behavior. I am inclined myself to want to follow that approach until it can be demonstrated conclusively that this approach will not solve the problem. I am thinking, for instance, along the line that Mr. Chamberlain moved, that the Federal Government, or an agreement among the States, might act as a catalyst, as an information center, bringing to gether potential new employers, expanding employers, contractors in a tight labor market, local agencies such as county governments, school boards and so forth, and providing them with a flow of informa tion as to what the future seems to hold for workers moving in, what kind of tax base will develop, and so on, coordinating, and not forcing. ^Mr. C h a m b e r l a i n . I like this approach myself. I think it has addi tional benefits that we have not mentioned so far. For one thing, this is more likely to be successful the higher the level of your economic 2782 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS activity. To put it another way, we have two related objectives here, the level of production and the distribution of factors of production needed to achieve that level. We want to increase the level of our economic activity to the point where a high degree of mobility will allow the shifting of resources to go to those areas where most needed. The kind of arrangements that we are discussing now, of the improve ment of information flow, facilitates that movement, There is then this additional benefit. If you are in a stage of high level economic activity, to the extent that you can improve this flow of resources to where they are most needed, you are going to lessen inflationary pressures. This is an added benefit. Mr. F ru ch t. You get not only reduced price pressure, but you also get more economic growth. You get better use of your available labor force. Mr. C h a m b e r l a i n . Quite so. Representative B ollin g. I would like to correct the record where I was reading before what I thought was the current list of areas of substantial labor surplus. I was actually reading from the list of September 1955. We now have the list for September 1959. I think it is fair to say these are roughly comparable periods. 1955 was the first year of a recovery, and 1959 was the first year of a recovery. The fascinating thing about it is that 21 of the 35 major areas that are now areas of substantial labor surplus are also included in the list of September 1955. There have been some additions obviously, with a difference of 9 between the 35, 9 major areas today, and the 26, if I count correctly, in 1955. There have also been a few substractions. Specifically, if I have it correctly, Asheville is off, Philadelphia is off, and Knoxville, Tenn. is off. You have a fascinating comparison which certainly illustrates very dramatically that the large number of areas of substantial labor surplus have problems that are continuing, at least over a 4-year period. What I would like to have permission to do is to have the staff put in a comparison of these two tables at the earlier point where I incorrectly stated the 1955 list. Without objec tion that will be done. Are there further questions, Senator ? Are there any further com ments ? If not, gentlemen, we thank you very much for a very inter esting and stimulating discussion, and the committee will stand adjourned until this afternoon at 2:30 in the same room, when the subject will be the role of collective bargaining in the economy: An overall evaluation and emerging problems. The witness is John T. Dunlop of Harvard University. (Thereupon at 11: 50 a.m., a recess was taken until 2:30 p.m., the same day.) afternoon s e s s io n Representative B o l l i n g . This committee w ill come to order. Our subject this afternoon is the role of collective bargaining in the economy: an overall evaluation and emerging problems. Mr. John T. Dunlop from Harvard University. You may proceed, Mr. Dunlop. EM PLOYM ENT, G R O W T H , AND PRICE LEV ELS 2733 STATEMENT OF JOHN T. DUNLOP, HARVARD UNIVERSITY Mr. D u n l o p . Thank y o u , Mr. Chairman. Collective bargaining evokes a variety of images and caricatures. Some say it is like a poEer game. The players are seated about a table, often for long hours. The largest pots go to those who combine deception, bluff and luck, or an ability to come up with a winning hand on occasions when they are challenged. But the rules of the col lective bargaining game are not so well established, the players some times play quite different games, and victory is not so easy to measure. Collective bargaining is at other times likened to a debating society. There is the same flow of words, massing of arguments, rebuttals, more or less polite invective and appeals to the gallery. But unions and managements are engaged in more than an exchange of words, and the prize is more than a debate plaque or silverware. Collective bargaining is sometimes depicted as a rational process in which the parties are intellectually persuaded to alter original posi tions by fact and argument alone. Disagreement is identified and dissolved by careful investigation, scientific procedures, and rational discussion. The parties have no adamant positions. They may be portrayed, in somewhat idyllic terms, as coming to the bargaining table eager to listen and learn and with a fully open mind to the sug gestions of the other side. But the depth of feelings and the resort to strike and lockout suggest that logical persuasion is not the only ingredient in collective bargaining. Collective bargaining is finally pictured as a power struggle. The strong impose their terms upon the weak. Take it or leave it. The results of collective bargaining, including relative w7age rates, reflect the distribution of naked powrer. But power means so many different things— force, capacity to impede or to shut down, ability to restrict supply and affect price, highly placed friends—that little understand ing is conveyed by the analogy of collective bargaining to a struggle for power. Each of these cartoons of labor and management at the bargaining table, no doubt, has a counterpart in some actual experience. But analogy cannot provide a substitute for more systematic analysis. In the papers and hearings of the past week the committee has ex plored many features of collective bargaining. This final session and this brief introductory paper were designed in summary— as the as signed title indicates—to provide perspective and an overall view. The following discussion explores three broad groups of questions: 1 . What is genuinely distinctive about collective bargaining in the United States ? How does our collective bargaining system compare with the arrangements of other countries ? 2 . How do our collective bargaining arrangements in themselves affect the terms of the resulting agreements? Are they inherentlv inflationary ? 3. What are the major choices and problems that confront our col lective bargaining system in the decade ahead as it affects employment, growth, and wage and price levels in the economy ? A discussion of these basic questions in brief compass runs the danger of being superficial and dogmatic, but it has the merit to compel attention to fundamentals. 2734 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS T H E A M E R IC A N COLLECTIVE B A R G A IN IN G SYSTEM Every industrializing country develops an industrial relations sys tem just as it also fashions an economic system. Regardless of political and economic form, industrialization creates industrial workers and labor organizations, managers, and government agencies active in in dustrial relations. This is equally true in countries with such diverse economic and political systems as India, Brazil, Egypt, Spain, Yugo slavia, the Soviet Union, or in Western Europe. Industrial relations systems are universally tripartite, although the relative distribution of power, the allocation of functions and the relations among labor or ganizations, managers, and the state differ widely from country to country. Moreover, just as there is competition among ideologies and economic systems, so there is intense debate and competition today over the industrial relations arrangements to prevail in each newly industrializing country. Each industrial relations system confronts the same group of central questions. Although the institutions to make the decisions and the substantive answers may vary, each country evolves or consciously de signs arrangements among workers, managers and government agen cies to perform certain functions: to determine wages and compensa tion, to establish discipline and other rules, to settle disputes and to develop morale and cooperative attitudes. Countries differ widely in the way responsibility for these central questions is divided among labor organizations, managers and the state. Wage rates, left largely to labor and management in the United States, are set by governmental arbitration tribunals in Australia. Discipline imposed by managers in Yugoslavia is subject to review by a committee of the workers’ council elected by the workers and by the trade unions. The single grievance procedure in the United States with private arbitration is to be contrasted with the prolifera tion of grievance channels in France created by Government regula tion with multiple unions and plant committees. In many countries labor organizations are regarded primarily as instruments of national development, to increase productivity, to educate and to promote hard work and discipline on the part of the work force rather than to regu late management and redress grievances. The wide variety of in dustrial relations arrangements in other countries helps to identify the distinctive character our our collective-bargaining system. Four features stand out. 1. Degree of decentralization.—The industrial relations system of the United States is more decentralized than that of other industrial countries. While there are a number of national bargains as in air lines, railroads or basic steel, there is relatively much more wage setting at the workplace level. (a) Our local unions are relatively very much stronger and more active than those elsewhere. They have exclusive representation under our traditions of majority rule and do not confront rival workers’ councils at the workplace, (&) Our unions have been oriented more largely toward the work place and have had mode significant impact on managerial decisions. Collective-bargaining agreements applicable to the actual workplace are detailed, and there is effective machinery in the grievance pro EM PLOYM ENT, G R O W T H , AND PRICE LEVELS 2735 cedure to secure compliance. Indeed, the pressures of our unions on managements at the workplace have been a decisive factor in making our managements more efficient. (c) Our managements have been historically relatively unorganized in collective bargaining as compared to employers in Sweden and England. In recent years our employers have substantially increased the extent of their working together in collective bargaining, par ticularly in national product markets, just as employers in local markets have engaged in multiemployer bargaining for many years. (d) Wage setting is more widely distributed, but a wage decision is one operation. The bargained wage is almost always the actual wage. In peacetime the “wage drift” has no significant counterpart in our experience unlike Western Europe. It is not merely that we tolerate higher levels of unemployment, but wage rates are not set in two stages or operations, with a national agreement or regulation followed by plant negotiations or managerial wage changes to con form to the particular labor market. 2 . Industrial conflict.—The extent of industrial conflict, measured by man-days of strikes and lockouts as a ratio to days worked or union membership, is higher in the United States than in other ad vanced countries. The extent of violence has markedly declined in industrial disputes but not the resort to work stoppages. * * * the rate of industrial conflict in the United States is considerably greater than in Britain or Sweden * * * 3.1 in the United States, 1.8 in Sweden, and 0.4 in Britain. (Working days lost as a multiple of union membership in the period 1927-52.) Our higher rate no doubt reflects many features of our industrial re lations system: decentralized bargaining, greater interunion rivalry, less centralization in the labor movement, greater resources of indi vidual workers and unions, the traditional concern of our unions with economic objectives through direct bargaining rather than resort to the legislative method and, finally, our collective bargaining has not yet widely developed provisions dealing with steps to be followed in the negotiations of new agreements. 3. Bole of Government.—The role of Government in our industrial relations system is distinctive in two respects. It is difficult to find a country—possibly Australia—with so extensive and detailed Govern ment regulation of the bargaining process and internal procedures of the parties. Nowhere else in the world is there such minute Gov ernment prescription as to who shall bargain, about what they shall bargain, and what the parties may say or do in the bargaining process. At the same time, the results of the bargaining, the substantive pro visions which the parties may conclude, are less subject to Government suggestion, recommendation, or prescription than in almost any other country. (An exception should be made of the growing number of contract provisions subject to legislative prescription, such as union security, checkoff, and now “hot cargo” clauses.) This combination of regulation of procedural details and substan tive laissez-faire is strange in comparison with other countries. This asymmetry is reflected in and perpetuated by the sharp separation of labor relations and mediation in different Government agencies. It perhaps arises from the declared legislative quest for the fiction of “equality of bargaining power between employers and employees” 2736 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS established by one legal framework for parties in quite different markets and with different resources. The preoccupation with procedural regulation and the ideological avoidance of substantive regulation in collective bargaining makes it impossible for the Government to have a wage policy beyond the platitude that the lavel of wages should be related somehow to productivity. In our industrial relations system the Government, ex cept in wartime, can have no wage policy for it has no way directly to influence or even to make suggestions about wage decisions. Monetary and fiscal policies may affect output, employment and profits, and thereby indirectly wage rate decisions, but these mechanisms are quite different from a direct wage policy even if only by specific sug gestion or recommendation. No proposals for changing our system are necessarily implied, but it is to be observed that the role of Gov ernment in our industrial relations system shows a strange and unique asymmetry. 4. Conservative unions.—The labor organizations in our industrial relations system are strong supporters of the main economic and political institutions of our society. No stronger statements in favor of private enterprise have come from any segment of our community than from many labor leaders. Aside from the labor organizations in the eastern bloc, ours is the only labor movement not dedicated to fundamental transformation in the economy. Our unions may be thought to be radical in their wage demands or expectations from the economy, but they are alone in the Western World in being con servative to the Nation’s economic institutions. Our unions have become more businesslike; they are staffed by career officers and they are becoming more like the rest of the community. In our society, only a radical labor movement would be spartan or ascetic. In any comparison of industrial relations systems, ours is relatively unique. It is more decentralized; there is more industrial conflict in the form of strikes and lockouts; the Government plays a major role in bargaining procedures but has no part in the substantive bargains, and our labor organizations are conservative to the economic and political institutions of the society. Our system is so distinctive that it is probably ill suited for export. The attempt to impose major fea tures of our industrial relations patterns on Japan and Germany after the war were largely a failure. Our system is likely to have little appeal to the newly industrializing countries. Professor Slichter, whom you may recall passed away this past week, wisely concluded: Our system of industrial relations would probably not work anywhere else, but it gives the American worker better protection of bis day-to-day interests than is received by the workers anywhere else, it puts American employers under greater pressure than the employers of any other country to raise produc tivity, and, though it gives unions a wonderful opportunity to whipsaw em ployers, it gives employers a freedom to bargain which they like and for which they seem willing to pay a big price. Hence, we seem justified in being grateful that we have been favored by fortune and perhaps also in taking modest pride that we have pursued opportunist policies with considerable flexibility and good sense. T H E IM P A C T OF CO LLE C TIV E -B A R G A IN IN G A R R ANG EM EN TS The collective-bargaining process sets the price of labor services, the price per unit of time or per unit of output or other measure of performance. The process does not specify the amount of effort nor E M P L O Y M E N T , G R O W T H , AND PRICE LEV ELS 2737 does it typically specify the quality of labor service. The process does not ordinarily set the amount of labor employed. The wage bar gain does not determine income of the workers nor labor costs to the enterprise. Although the wage rate decisively affects both, many other decisions affect both costs and income. The question of the effect of our collective bargaining institutions upon wage rates may be approached as follows. It is clear that these institutions are not the only factors determining wages; a vari ety of factors such as labor supply, prospects for profit or the capital equipment used with the labor force affect wage rates through the collective bargaining institutions. The issue is the independent effect of our collective bargaining institutions on wage rates. In a sense the question in this form poses an insoluble problem for it asks what wages would be in the absence of our collective bargaining arrange ments, We can get as many different answers as the wage-setting arrangements that are postulated in the absence of the present col lective bargaining: the norm of perfect competition, various types of unorganized labor markets, government wage fixing, et cetra. Each of them would yield a different answer. Let us start with some simple ways in which the collective bargain ing institutions independently affect the wage-setting process. 1 . The frequency of wage rate charges is affected by the length of the contract and its provisions, 2. The time of the year, the period in a seasonal industry, at which wage rate changes are made may influence the resulting bargain. 3. The method of wage payment is subject to decision: time rates or various forms of payment by results. 4. The division of compensation between basic wage rates and various forms of fringe benefits and the number and forms of such supplemental pay practices is subject to the decisions of the bargainers. 5. The institutions tend to establish an explicit wage structure for an enterprise rather than a number of personalized rates for individ ual employees. 6 . The scope of duties within a job classification is directly affected by the bargaining mechanism. The job content to go with the wage rate is subject to decision. These points establish the proposition that the collective bargaining institutions affect the wage-setting process in a number of ways, apart from the question whether the resulting wage rates are higher or lower than they would otherwise be—under some other wage-setting arrangements. Wage setting under collective bargaining is a different process than it is in the absence of the institutions. Do the institutions of collective bargaining result in higher or lower wage rates on balance? The standard of comparison is very much a problem. Collective bargaining affects the supply of labor: it probably tends to increase the supply of highly skilled labor from what would prevail in the absence of collective bargaining; unskilled labor may be affected by the rationalization of labor markets and the attachment of workers to particular enterprises. Collective bar gaining affects the productivity of labor through its impact on morale, skill, and training and through the pressure it places on management. Collective bargaining affects the demand for the product through 2738 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS affecting the character of competition in the product markets and any effects on national income. By affecting the forms of compensa tion, such as health and welfare plans, and vacations with pay, it affects the demands for some products relative to others. The significant point is not whether wage rates are higher or lower under the institutions of collective bargaining—relative to some other way of setting wage rates—but that the institutions of collective bargaining alter the magnitudes of the factor which determine wage rates. Collective bargaining does not simply raise or lower wage rates and leave everything else the same; it changes the quantities of the wage-determining factors. Under these circumstances definite statistical work to measure the independent effect of collective bar gaining institutions is most difficult; the statistics are likely to remain highly debatable and only suggestive. Senator B u s h . What do you mean by “changing the quantities of the wage-determining factors” ? What factors ? Please explain that. Mr. D u n l o p . Yes, sir. We ordinarily say that the demand for the product would influence wages. If collective bargaining changes the demand for the product, it will in turn have an effect back upon the wage. The important point is that collective bargaining will change the quantities and the magnitude of the wage-determining forces. Supply of labor is a factor affecting the wage. Insofar as collective bargaining increases the number of skilled workers relative to what there might be if there were no collective bargaining, it would change the supply of skilled workers. The important effects of collective bargaining are to be found in the impact on the magnitude of factors which determine wages rather than in leaving everything else the same and simply raising or lowering wages. Senator B u s h . Thank y o u , sir. Mr. D u n l o p . The question arises as to the impact of the collective bargaining institutions upon the stability of the economy, quite apart from any impact on the level of wage rates. Are wage rates under collective bargaining more or less volitable? D o they rise slower or faster in response to changes in wage-setting factors under col lective bargaining? It is often said that wage rates have become more rigid; they are less likely to decrease in response to some factor tend ing to reduce wage rates. The available data do not support this conclusion. Wage rates have declined less frequently in recent years, but there have not been the periods of unemployment or price declines of former years. The degree of inflexibility in wage rates should be compared to movements in unemployment, prices, or profits. The secular change is not in wage rigidity but in the degree of unemploy ment for sustained periods and in the extent of price declines with a reduction in output. The impact of collective bargaining institutions on the inflation potential or upward thrust of the economy would appear to hinge on the policies actually adopted. There are instances in other coun tries, such as Great Britian, where the collective bargaining arrange ments were used to restrain wage rates below what they would have been otherwise for a period of several years. This happened in the United States in some industries in 1946-47 with the voluntary ex tension of wage stabilization. But in both illustrations the awage drift” became significantly larger. E M P L O Y M E N T , G R O W T H , AND PRICE LEVELS 2739 It is my judgment that our collective bargaining institutions con tribute, under conditions of high-level employment, an independent element making for a faster rise in wage rates. The basic reason for this judgment is that high levels of profits become the target of union wage policies under collective bargaining, and particularly under our decentralized bargaining, to a degree not likely in the absence of labor organizations. But my judgment is that this independent effect of collective bargaining is not large, and it is not even certain that it is larger than the independent effect of increased productivity arising from union pressures on management and on the skill and morale of the work force. While I am expressing judgments, it seems to me that the improvement factor and the escalation clauses do not have much independent effect. The escalation clauses may reduce the average lag a bit and thereby contribute to an independent rise in wages, but the longer term contracts with which these clauses are associated tend to have advantages in freeing management to concentrate on efficiency. M A JO R PROBLEMS A N D CHOICES I N T H E DECADE A H E A D The third and final group of questions concerns the major prob lems, the policy choices in our collective bargaining system, and who had the choices to make in the decade ahead as they affect economic growth and stability. It is conventional in most public discussion to depict these choices as black or white and all or nothing. But an economist, and certainly a mediator or arbitrator, is likely to see these issues in marginal terms, as a little more or a little less. But small differences may be significant, since a small change in direction can have large effects in the long run. 1. Union democracy and wage stability.— On the problem of the impact of internal union government on wage stability, the country has recently made choices which are certain to affect wage settle ments in the decade ahead. Large segments of union membership have been led to believe, wittingly or otherwise, that their leadership may have made “sweetheart” agreements with employers, and the leadership did not get all the wage increases there were to be had by tougher bargaining. Many of the devices of leadership control, such as trusteeship, have been weakened, and certainly the disposition of international union leaders has been materially weakened to take strong positions against locals or factional leaders bent on pressing for wage increases greater than thought wise by the top leadership. Contrary to the public sterotype, international union leaders have ordinarily been more restrained on wage increases than local leaders and active members. They have a longer experience and a broader view of the industry and economy. The experience of other countries also shows that leaders of national unions are more likely to accept policies of wage restraint. Already we are seeing employers who urged Congress to pass strong legislation affecting internal union government going to na tional union officers as of old seeking national union support to re strain the demands of locals and to make agreements. They are not likely to get as much cooperation; they could not be given as much. The country has chosen on the grounds of morality and democracy 2740 E M P L O Y M E N T , G R O W T H , AND PRICE LEVELS to make wage stability more difficult to achieve. The unforseen con sequences of labor legislation have often been the most significant, and the lessening restraints of national union leaders on wage in creases is likely to be a development of significance for the future. The country has been led to believe that rank-and-file participation would result in more wage stability; it will have the opposite effect, Senator Bush. Would you care to amplify that? Those last three or four sentences puzzle. “The country has chosen on the grounds of morality and democracy to make wage stability more difficult/ 5 You say it will have the opposite effect. How do you arrive at that con clusion ? . Mr. D u n l o p . In my^ judgment, up to this point in history the na tional union officers of most unions have been more conservative in wage demands, more willing to settle wage demands at lower figures than active rank-and-file members in local unions. In the course of the recent session the Congress for good reasons has chosen to pass legislation which will have the effect of substantially reducing the possibilities of control of national union officers over local unions. Therefore, in wage settlements in the future, the influence of the local union and the influence of union rank-and-file members will be stronger and therefore the wage settlements will be larger. Senator Bush. Let me ask your observation on this point: Before the steel strike, Samuel Lubell, who has been sort of a political ana lyst— I don’t know if we can call him an economist—made a survey of union membership in Pittsburgh and came up with the report that six out of seven of the steelworkers did not want a strike. They were afraid it would hurt them, that they would lose more from it than they gained. Nevertheless a strike was called by the top union leadership. That would seem to me to a point at variance with the conclusion which you draw. Mr. D u n l o p . Perhaps, although not necessarily. Of course the strike, as I recall it, was called in accordance with the union’s con stitution, so it involved ratification by duly constituted representatives all the way up. I had primarily in mind the situations where national union leaders are called in to advise, participate, and take some part in local bargaining rather than the steel type of situation where the bargaining is performed in the first instance by the national union. I am not sure that the effect would be very large in the case of national union bargaining where the national union people were doing the bargaining before the statute and will be doing it later. Their role in the bargaining process has not been changed by this, but the kind of situation which is much more typical of the 100,000 or more agreements wTe have in the country is where local parties are bargain ing and the national union is called in to assist these parties and to use its influence to settle the dispute. It is that situation that I feel very strongly will be affected by the new statute. Representative B o l l i n g . I think part of the point may be that sometimes we tend to forget than an international union is in itself a political entity and the officers of that union are their own sovereigns of their constituency and the easiest way in the world to run against anybody, whether it be in national politics or union politics or any other kind of politics, is to say “I will get you more.” And there are specific examples of this. There is a man not too far away from here in EM PLOYM ENT, G R O W T H , AND PRICE LEVELS 2741 one of the unions who is experiencing that particular kind of thing right today. People in a given area are saying we could get a good deal more for you, the members, than our international officers are doing. So you create^a kind of political competition within the institution which has the effect or might have the effect—I am not saying it would—but it might have the effect of having the leaders competing with each other as to the highest settlement they can get. How this will work out, we don’t know. But would you agree that is a valid point in this connection ? Mr. Dunlop. Yes, I would. Senator Bush. I think that is a pretty tough conclusion to agree with at the moment. We will find out in a few years. Mr. Dunlop. May I proceed ? Then perhaps at the end you may wish to pursue it a little further. 2. Adjustments to technological change.—The most important prob lem of collective bargaining in the next decade at the workplace is likely to be the adjustment to technological and related market changes. Enormous technical changes are in process and in prospect growing out of automation principles, data-processing machines, elec tronics, and the new era of organized industrial research. The intro duction of new machinery and processes always involves adjustments in a work force: the speeds of equipment, manning tables, workloads, job requirements, day or incentive rates, seniority rights, ladders of promotion, and layoffs. These are always tough problems; the human readjustments are severe irrespective whether there is a union at the workplace. Indeed, collective bargaining normally mitigates these problems by providing for orderly change, for acceptance of loss of jobs, new speeds and manning schedules, and for procedures for h a n d lin g the inevitable grievances in return for wage increases, severance pay, and other protections. Much public discussion has unwisely attributed the problems to the unions rather than to the underlying adjustment processes. In recent days these problems have received increased attention, as they are certain to in the next decade: the strikes in the airlines over the introduction of the jetplanes, the long stoppage in part of the flat-glass industry, and the advertisements concerning basic steel and the railroads. Two recent sets of negotiations have developed new and constructive ways for dealing with these problems, ^ Armour & Co. and other meatpacking companies with two international unions have agreed to establish a joint committee and a fund to deal with problems arising out of automation and to stimulate retraining, re location, and other programs of adjustment in addition to the existing arrangements for severance pay. In the west coast longshore industry the managements may place into effect under the new agreement new techniques for loading and unloading vessels, and a share of these savings is to be allocated to the workers by a joint committee of the parties. The details of these new agreements are not so important as their point of view. The country desperately needs to speed up the rate at which technical change is introduced and productivity increased. The inevitable and delicate readjustments can be handled best by a policy of “greasing the change” rather than “bucking it through.” The 2742 E M P LO Y M E N T, G R O W T H , AND PRICE LEVELS “get tough” policy on these adjustments does not pay in the long term since the full and efficient use of modern equipment requires full cooperation of the work force; practices of small work groups fearful of layoff or downgrading or speedup can offset the savings on expen sive new processes. The active assistance of unions is^required to overcome the natural reaction of workers and small work groups. Moreover, the increased participation of the rank and file in union processes and the greater timidity of national union officers make the “greasing the change’' policy all the more imperative. These vital problems of adjustments to technological changes at the workplace are largely outside the scope of Government, The problems are up to the parties and particularly to management to propose a philosophy of adjustment. The rate of economic growth and productivity will be very substantially determined by whether management seeks to “buck it through” or to “grease the change.” 3. The need for shill.—The next decade is likely to see critical short ages of skilled labor. Despite a rapidly rising population and work force, the age group 25 to 44 is not expected to show any increase in the number of men in the labor force each year. This results from the decline in birthrates which took place in the depression of the 1930’s. The rigidity in the supply of men in the age group 25 to 44 in the next decade will limit the number of skilled workers. At the same time the demand for skilled labor will grow much more rapidly than for semiskilled operations which will be increasingly subject to ma chine substitution. The demands for technical, professional, and managerial occupations will be much greater out of this age group. The pressure on skilled labor may be expected to have important wage rate consequences; larger increases at the skilled end of the wage scale can be expected to pull up the whole level of wage rates to a degree. The critical need for skilled labor in the next decade is likely to exert a strong inflationary influence. There is growing recognition of the need for training skilled labor, and the intensity of the problem can be reduced by a variety of meas ures. Industrial plants can introduce apprenticeship programs. The skilled trades can expand considerably their joint apprenticeship pro grams under collective bargaining. Moreover, significant results can be achieved by retraining existing jobholders in new developments and specialties. Encouraging beginnings have been made by the elec trical and plumbing and pipefitting groups. But much more is to be done with both apprenticeship for skilled trades and retraining or up grading of the existing work force. These choices are largely up to the parties in collective bargaining, although the Government has been playing an increasing role in stimulating programs. The influence of these programs on economic growth and inflation has not been recog nized sufficiently. 4. Bargaining arrangements.—Although the bargaining arrange ments in the United States may be expected to change in some signif icant respects in the next decade, they are likely to remain very much more decentralized than Western European countries. In national market industries wage leadership by a major producer may be ex pected to give way to association or group bargaining. The develop ments of the last automobile negotiations and the more formal ar EM PLOYM ENT, G R O W T H , AND PRICE LEVELS 2743 rangements among 12 companies in the current steel negotiations rep resent the tendency. Moreover, managements in these industries may be expected to develop techniques, such as the compact in the airlines negotiations, to reduce the possibilities of the unions splitting the companies where they have traditionally followed a common pattern of wage adjust ments. Bj sticking together the companies hope to secure smaller settlements. These developments should cause little concern provided they remain techniques of collective bargaining; they can serve the function of promoting some greater degree of stability. In Great Britain and some other Western countries there has been considerable discussion recently over whether the adoption of a more centralized bargaining mechanism, as in the Netherlands or Sweden, would contribute a greater degree of stability. The argument is that factional interests within the labor movement compete to make the best showing, and the greater centralization of bargaining would tend to eliminate this source of the wage push. It is argued that relative wage rates could then be adjusted more readily without having to raise all wage rates. Whatever the applicability of more centralized bargaining to Great Britain, and I have reservations as to its efficacy there to curb wage increases, it would clearly be alien to our industrial relations system. It is not a realistic choice here. Very few collective bargaining agreements in the United States contain provisions to be followed when the parties are deadlocked in the negotiations over a new agreement. Procedures in the railroad and airlines industries are specified by statute, and agreements in the transit industry and some branches of the newspaper industry provide for arbitration. Procedures might well be negotiated in agreements generally under which the parties follow steps they have previously agreed upon before resort to strike or lockout in disputes over the terms of a new agreement. Such steps might include resort to desig nated mediators, factfinding with or without recommendations or even arbitration. Provisions of this sort are general in agreements in Great Britain. The parties in the steel industry, for instance, should be encouraged for the future to specify in their agreements the steps they shall fol low after direct negotiations have failed to produce an agreement be fore resort to strike or lockout or resort to any public emergency dispute procedures. They might even be required to develop their own procedure within a specified period or confront the legislative mandate of a Government-appointed arbitration board in disputes that threaten the Nation’s health or safety. 5. Government ivage policy.—Professor Burns has well said that “Official appeals for restraint in wage-and-price adjustments may be salutary, but experience suggests that it would be unwise under ordi nary circumstances to expect a broad response to exhortation.” A command to halt wage-and-price increases spoken by the President has no more effect on the tides of inflation than the words of King Canute confronting the rising waters of the English Channel. Despite this hardheaded judgment, as developed before the last American Assembly, it seems to me that the full potential of the leadership of the Federal Government has never been used persistently and imaginatively to shape decisions by private parties on wages and 2744 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS prices or to influence tlie climate of ideas within which such decisions are made. If the Federal Government expects to influence directly the ideas of the parties to collective bargaining, it must leave repeti tive platitudes and generalities and meet with labor and management representatives regularly to discuss and to debate in free exchange and with detailed statistics the economic setting and outlook in which wage-and-price decisions are made. More specifically the following suggestion might be tried, which is a suggestion I also made at the American Assembly. After the Economic Report of the President has been transmitted to the Congress and there have been hearings before the Joint Eco nomic Committee each year, the Secretary of Labor might convene in the early spring a conference with leading representatives of labor and management. The Chairman of the Council of Economic Ad visers, the Secretary of the Treasury, and other Government officials should present their detailed analysis of the short term and longer term outlook. Representatives of management and labor should be given the opportunity to discuss these views of the administration and to present materials and judgments of their own. In addition to formal sessions there might well be informal off-the-record periods of free give and take. The purpose of these annual discussions should be to develop a concensus of opinion, insofar as possible, or to narrow the range of views concerning the major problems confronting the economy as a whole and the expectations of the short term and longer term business outlook by principal sectors. These discussions would not be nego tiations nor should they be designed to predetermine any particular contract negotiations. But the Government would help to sketch the economic problems and climate; the interchange would benefit all three groups in our industrial relations system. The fundamental point is that imaginative Government leadership should press be yond annual cautions and preachments to more direct exchange of ideas and information in a society of freemen and collective bargain ing, and particularly in a society in which the level of general educa tion and the detail and the quality of statistical and economic infor mation has been improving rapidly in the past generation. Representative B o l l i n g . Thank you, Mr. Dunlop. Senator Bush, do you have some questions ? Senator B u s h . I just have one or two matters, perhaps. You speak about the general subject of the need for more skills. You say: These choices are largely up to the parties in collective bargaining, although the Government has been playing an increasing role in stimulating programs. The influence of these programs on economic growth and inflation has not been rec ognized sufficiently. What programs do you have in mind there ? Mr. D u n l o p . I have in mind two kinds of programs. First, the programs for the training of apprentices under joint collective bar gaining arrangements which are stimulated by the Bureau of Appren ticeship and Training in the Department of Labor and, second, pro grams designed to improve the training of already existing journey men ; that is, to raise the average level of skills among people who are already qualified in the light of new changes in industry. E M P LO Y M E N T, G R O W T H , AND PRICE LEVELS 2745 For example, in a branch of the industry like the electrical branch, there are many new innovations dealing with electronics, with new circuits, and so forth, and men who have had their training many years ago need to have their skills improved and brought up to date. We are beginning to develop through the direct action of the parties and through Government influence a number of programs which need to be increased very substantially to upgrade and raise the level of skills and to diffuse new techniques to our existing work force. Senator B u s h . D o you think the Federal Government should take an increased interest in this from the standpoint of appropriations? Mr. D t j n l o p . I like to be frank. I am sure I do not know, in the sense that I have not looked at the financial appropriations to the relevant agencies. I am sure they could answer that much better than I. In general, I believe the programs have a major contribution to make, and they should be pushed. Senator Busii. I do not think I have any other questions and I will have to excuse myself, regretfully. Representative B o l l i n g . Mr. Dunlop, on this proposal of yours: “The Secretary of the Treasury and other Government officials should present their detailed analysis,” and so forth. In other words, the Sec retary of Labor should convene this early spring conference at which the Secretary of the Treasury and so on would do what you have out lined— does this require, as you see it, any legislation?" Is this not within the power of the executive at the present time ? Mr. D u n l o p . I ain not a lawyer, Mr. Chairman. Representative B o l l i n g . Nor am I. Mr. D u n l o p . I am not well qualified to answer it nor have I looked into it in detail, but it is my impression that legislation is not required. It is my idea that there is sufficient authority now to call the parties together and to experiment with this idea. Representative B o l l i n g . As you remember, there was very con siderable discussion at that American assembly which you mentioned as to the particular form this particular device would take, and I, as one of the participants, was violently opposed to any new legisla tion because I felt the authority existed and if we had new legislation we wTould be guilty of just gilding the lily. Mr. D u n l o p . I think that is right. It seems to me that if the Government is to develop a wage policy, it is going to have to begin to do it by some sort of more direct ex change of ideas with the parties. It seems to me this should be de veloped over the years gradually, and we need to experiment with the mechanisms by which to do this. I am interested in this proposal as a way of starting to narrow the range of differences between the par ties in periods of threatened inflation. In periods wThen our balance of payments problems were severe, this range of issue would be called to their attention. There is need for a forum in which these problems are directly discussed and thoroughly debated. In this way a con sensus of ideas, I think, can gradually develop rather than for the Federal Government to come out by pronouncement with a Govern ment policy in detail. Representative B o l l i n g . Perhaps in an attempt to get to the point where there is formal action on a conference called by the Secretary, do you think that it would be possible or appropriate for a modifica 38563— 59— pt. 8------ 20 2746 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS tion of that type of thing to be attempted by the Joint Economic Committee when it reviews the President's Economic Report? Ob viously, it could not do the same kind of thing because this goes much beyond anything the committee could formally do, but would you feel that it would be wise for the Joint Economic Committee in con sidering and preparing its report to the Congress to begin to go into more of this type of thing? It would not have the same purpose as the Secretary’s conference but it would have an effect perhaps. Mr. D u n l o p . I think that would be a good idea; yes. The confer ence idea involves not only prepared statements but the interchange of ideas. Representative B o l l i n g . This is where the joint committee could not undertake to function. Mr. D u n l o p . Some of it could be in formal sessions and some in informal sessions, but I do think that anything that begins to call the parties together in separate sessions to discuss with them the economic outlook, short term and long term, next year and the next decade, and what this means in terms of wage policy in the country is a helpful contribution to the long-term solution of this problem. Mr. E c k s t e i n . Y o u mention the rapid technological advance in your paper and the need to have an institution which facilitates the employment adjustments that have to be made. What do you feel are the needs in the way of either public programs or the develop ment of institution programs for improving this ? Mr. D u n l o p . I do think, as in many areas, we need a good deal more imagination and experimentation. I would like to throw out a couple of ideas to 3^011 in answer to your question, just more or less off the top of my head. I feel that there is a great deal of technological change in the wind, and the problem is how to make this smoothly enough so that you don’t fight each little machine change or each change in speed, each change in manning schedule as a major battle. It requires a great deal of consent on the part of people to see them lose their jobs or see them have traditional skills changed. This is a human problem of the first order. I think that a certain amount of this is illustrated by the collectivebargaining agreement which I cited in the meatpacking and in the west coast longshore case, but I think there are some things public policy can do. For example, our unemployment compensation system might be modified to provide that men who had 15 years’ seniority—I am not going to quarrel about the particular number of years—in a plant who are displaced for technological reasons shall be entitled to double benefits for a period twice as long, provided, however they shall make themselves available for retraining programs and retaining skills. This is a kind of severance-pay arrangement built into an unem ployment compensation system which includes retraining. Representative B o l l i n g . May I interrupt at that point because it is so pertinent to some of the things said this morning by our witnesses. Would that be the supplementary benefit which would be handled within the S t a t e unemployment compensation system or by a Federal supplement ? The suggestion this morning by one of the witnesses, EMPLOYMENT, GROWTH, AND PRICE LEVELS 2747 Chamberlain, was that that particular type of retraining would be over and above and outside the State system. Mr. D u n l o p . Mr. Chairman, as you know much better than I, that has been a great source of controversy. My preference about the mat ter would be that this additional idea would be a matter under the Federal program. I did not know, frankly, that Mr. Chamberlain made the proposal. Representative B o l l i n g . That is why I interrupted because I was fascinated to have three witnesses give the same statement. Mr. D u n l o p . This suggestion buttresses one of the points I was making under point 3 on the need for skill. I am advised that something like one out of two apprentices who starts does not finish. Many who start become unemployed part time, and as a result they take some other job to supplement their income, and that other job seems more attractive; they get married and have a couple of kids, and the country has lost an apprentice. It seems to me that this is an area in which the parties should be encouraged to provide some arrangements under which apprentices shall be guaranteed a minimum number of hours’ work if not full employment in order to cut down the wastage that arises because ap prentices leave when they encounter unemployment. This is a small matter but, Mr. Chairman, I believe that our large problems are made up of a lot of these kinds of small problems. This would be a second idea that occurs to me prompted by your question. I don’ know all of the things that might be considered to speed up technical change. I say it is primarily a problem for the parties be cause I do not think it is possible for Government agencies to deter mine what work loads ought to be, what speeds a machine ought to be, what manning schedules ought to be, what the rearrangements in seniority systems ought to be, and these are the problems that are particularly important at the work level in order to speed up the rate of technical change. Mr. E c k s t e i n . I have one other question, and I am sure some of our other people have some other things on their minds. If you look at the recent historical record, the rate of wage advance has been considerably higher than the rate of advance of productivity. Do you see any real prospect that these two things can be brought into line? Mr. D u n l o p . I can answer it by saying I see reasonable prospects that they may more nearly be brought into line and by movements on both of the series you mentioned. I believe that it is possible in the country by doing a number of things to raise the average rate of in crease in productivity more nearly to Sy2 to 4 percent. I see it possible to restrain the rate of wage increases more nearly to an average of 5 or 4 percent. The rearrangement of the bargaining on the part of the employers, I think, will have an effect in the future. However, as I say, in the decade ahead what really concerns me is that we are going to confront a new set of inflationary pressures in the skill problem, in the problem of the lack of authority of the national union and so forth. I can see a number of problems which we may solve but these new ones will make the problem greater. I can see the rate of increase being more modest in the next decade with wisdom all around. I don’t think it is a predetermined question. 2748 EM PLOYMENT, GROWTH, AND PRICE LEVELS I think it is within the capacity of private policies and Government policies to effect the result. I do not think it is predetermined. I think it is unlikely that there may not be some upward drift in prices,, but it is possible that the rate will be modest. Mr. F r u c h t . I am far from clear myself as to what possible effect might occur as a result of the latest legislation. I am also uncon vinced by the discussion and I am going to sharpen a little issue I have in mind. If national leaders are less optimistic of union strength and members and if their evaluations are correct as to the strength of the union, then stronger demands by members will weaken the union, presumably, over a period of time and presumably this would weaken our wage pressures. But on the other hand if the members are correct, they do not overestimate their own strength, then I wonder why you would feel then that the national leadership would either deliberately with hold obtaining wage increases that could be obtained without weak ening the union or else why they are so excessively pessimistic. Mr. D u n l o p . Let’s take an example and work this thing through. Let’s take a case of a local union which has been bargaining with employers. It asks for a wage increase which it thinks is justifiable but the employers do not accept it. Let’s say, they are asking for 30 cents an hour and the employers say they will settle for 20 cents. The union persists to a strike. In the past it has been customary to invite in the national unions and they have advised and tried to mediate it out and settle. In the past under some circumstances the national union leaders have said to the local fellows, we think 20 cents is right and you ought to take it, and they have gone further in cases I know and backed it up and said, We think it is not only right but by per sisting in this strike you are hurting the union in the long run. You are getting its employment out of line. You are keeping this industry shut down, and the international has moved in and put this local under receivership. But they are not going to do it to the same extent in the future. Mr. F r u c h t . The point is we would pay higher prices now for a lower price later? If the union has the power to raise wages now and we weaken that power, then presumably we will have less wage pressure in the future. I am not saying I agree that unions have power, but if they do, it would seem to me it would follow if the unions get overambitious and are weakened as you say, then in the future we will have less wage pressures. Mr. D u n l o p . I think the difficulty in our exchange is this: You have implied that the taking away of the power of trusteeship has weakened the power of the union to get increases. I did not say that. What has happened is that the trusteeship or other arrangements, a good deal of the atmosphere of the times, means that the national union will not exert their powers or authority to constrain the local people. Mr. F r u c h t . I am not clear. Either the exertion of this power is adverse to the members so far as wage increases are concerned or if that power is not used, the local unions will hurt themselves as you have already said, weaken themselves and then perhaps in the future E M P LO Y M E N T, G R O W T H , AND PRICE LEVELS 2749 the union will be so weak as to be unable to obtain the kind of wage increases it has been obtaining. Mr. D u n l o p . It does not necessarily follow that if they pursue a higher wage policy they are going to weaken themselves. Their level of employment could be smaller at a higher wage and they could be just as strong, but it might be an unwise policy from the standpoint of the international policy. It does not follow that if they pursue a high wage policy they will be weaker. Mr. F r u c h t . I guess I am not clear then about the way in which the interest of the international would differ from the interests of the members. Mr. D u n l o p . The international’s view is a longer run view in my experience. The interest of the international may be to keep a larger membership. If you push up the wages of a particular sector of industry, you may lead to the development of substitute products or substitute materials. Now, as far as the men working in that locality are concerned who remain in the union, they may be quite happy to have a smaller num ber of unionmen working at a higher wage. However, the inter national union’s position may be “We don’t want that kind of arrange ment. We would rather have a little lower w^age, a larger number of members and our branch of the industry or our product more fully em ployed,” and I regard that as the normal situation among internationa1 unions. Mr. F r u c h t . It is analagous to the Berle-Means thesis in regard to separation of ownership and control in the case of corporations. I would like to turn just very briefly to the points you made on page 8 in regard to the asymmetry between behavior in this country and other countries in regard to the willingness of our Government to set a framework but not to dictate the consequences of collective bargaining. This arises, you say, “From the declared legislative quest for the function of 'equality’ of bargaining power between em ployers and employees.” I would just like to raise the point that to me it seems that the reason is that this is our approach to economic policy or it has been our traditional approach in all phases of private economic activity, one, namely, of setting a legal framework, or say, rules of the game, and then letting the participants go to it and make their own decisions within that framework; that this is an old tradition that I think is being eroded in many ways, but I think this is the source. Mr. D u n l o p . I would not quarrel with that. Mr. F r u c h t . I would just question, not that I disagree or feel that the evidence is compelling the other way, your assertion that unions make managements more efficient. My own feeling is that it would be an awfully hard thing. Mr. D u n l o p . May I comment on that, Mr. Chairman ? My distinguished colleague, Professor Slichter, wrote a book on this subject in 1941. He spent something like 18 years in writing that book. The conclusion I expressed was his. He spent the last 3 years with two of my distinguished colleagues, Professor Healy and Professor Livernash, on a manuscript of 1,500 pages which has been completed. 2750 EM PLOYM ENT, G R O W T H , AND PRICE LEVELS That book will go to press at Brookings Institution very shortly and when it is out it will document the conclusions I have stated in detail. The old book’s title was “Union Policies and Industrial Management.” Mr. F r u c h t . I have just one more question to raise and it is highly ideological and it relates to your closing remarks. If, as you say, some unions do have the power to substantially af fect in an adverse way the health and safety of the country, I am not prepared offhand to agree with that myself, but if this is true, then I grant you that vour approach for dealing with this is consistent, and is one tenable approach. Is it not equally tenable to suggest that we might attempt to reduce the size of these bargaining units to make sure that they are not in the position of being large enough or powerful enough to affect the health and safety of the country ? Mr. D u n l o p . I have two comments. First, power to adversely affect the health or safety of the country in a collective-bargaining relation ship is always a joint one in the sense that if one side says yes and the other side says no, it is out of the disagreement that the problem arises, and I find great difficulty in saying of two parties who are bargaining about wages or conditions or something else that when they disagree blame is necessarily to rest with one rather than the other or that the consequences are inherently one rather than the other. Mr. F r u c i i t . I am prepared to agree that where either side or both sides have such power, there is a real question. This is not the issue I am bringing up. Mr. D u n l o p . I wanted to say that so we would be clear about it. The second comment I want to make is to inquire whether you are opening up for discussion the whole antitrust area about which this committee has had, I gather, a good deal of discussion. If so, this would require, I think, a few minutes because it is a very com plicated topic. I am prepared to do so if you want me to take 5 minutes to say what I think about it. Representative B o l l i n g . It is entirely all right with me, Mr. D u n l o p . Sometimes, when people talk about this problem they are thinking of antitrust in terms of getting at the emergency dispute problem. You seem to be in some ways talking about that. Other people have talked antitrust in terms of technological change, and other people have talked antitrust in terms of the kind of negotiations to be carried on. I think it is important to separate these problems and be clear about them in any elemental discussion. With respect to the concern with antitrust about what you might call the economic power of unions and not the emergency dispute question or the technological question, but the power to cause con tinually rising wages and prices, that power may be greater in many local markets than it is in national markets, in which case one would be confronted with the application of antitrust to the whole range of local markets, Mr. F r u c h t . This is analogous to business antitrusts. Mr. D u n l o p . The next point I would make is that it has never been clear what is meant by the further application of antitrust to labor, E M P LO Y M E N T, G R O W T H , AND PRICE LEVELS 2751 and I would be very interested if someone who advocates this point of view would come forward and tell us what he has specifically in mind. For the life of me I have not been able to think through and I know of no one who has been able to think through a systematic sketch of how this would be applied. For example, what about the problem of craft units ? Craft units split across a whole industry. Are you going to split those up? I don’t quite understand. Are you going to split up only industrywide bargaining but not craft bargaining that goes across industry or craft bargaining that goes across 10 or 12 industries? This is a problem that needs a good deal of thinking by those who are talking about it. I would hoM—and I think most of us who are arbitrators or media tors would hold—that the record of industries where there are two unions does not show any greater degree of stability than industries with only one union. Suppose we should take the aluminum indus try where there are two different unions. In fact, there are a number of different unions. Or suppose we take the can industry or the aircraft industry or the shipbuilding industry. These are all industries in which there are, if you like, rival unions, in the sense that major components of the industry are organized by different unions. I once did some study comparing these wage movements, whether there was any difference between industries where there was one union as compared to those where there were different union situations, and I could find no influence. I think it is quite doubtful that there would be any significant difference. The only point I would like to conclude on is this: If the problem you are worried about is the emergency dispute problem, I think there are much better ways of getting at the emergency dispute problem than applying any kind of antitrust action, and if you are talking about the inflation problem, I think there are a lot better ways of dealing with the inflation problem than with antitrust. Either way you put it, I do not think it deals with the problem. Mr. F r u c h t It seems to me that you asked something, let’s say, of the proposal for union antitrust that nobody really, I take it, would ask seriously of business antitrust. You say, specifically, What would you propose in a situation where there are so many different kinds of power and different exercises of power and different consequences and so on ? Under the Sherman Act, we have the very general kind of law which operates against monopoly. It is not even defined and the courts go ahead and argue that in this particular case market power exists for the following reasons and is manifested in such and such kinds of behavior and a remedy is put into force and the remedies differ. In one case it will be breaking up and in another case it will be a requirement that they do this or that or do not do something else. In other words, the remedies can vary with the circumstances as determined by the court. Yesterday, or the day before, your former student, Mr. Segal, testified on some of the bad effects of the pres ent law, arguing that while it might in many cases have good effects, beneficial effects, in other cases, such as in an organization drive in the South, it could well be a threat to the legitimate interests of the labor movement. I quite agree. 2752 E M P L O Y M E N T , G R O W T H , AND PRICE LEV ELS Tlie same proposition applies in the case of business antitrust. A general, across-the-board principle, can be a very dangerous thing. I just wanted to ask you then, in light of this, in light of the record of our courts in being very careful about interpreting the meaning of market power in the case of business firms, what objection would you have to a very general law such as the Sherman Act for the case of unions and then to that adding a process of precedents and developments and so on that we have had. Mr. D u n l o p . There are others who have taken this same point of view. I wish they would tell us what they have in mind. Here you are urging the courts to do something but give us some idea what you are talking about. Do you regard the steel industry as a situation in which you want to apply antitrust or not? Now let us take that case and talk about steel. You might split the union into as many unions as there are companies. Another way is to split it down to each plant. No one since the 1901 strike would seriously argue the point, because that would be a clear imbalance of bargaining power. So, the only proposal that makes any possible sense is that there shall be an international union or national union for United States Steel, a national union for Bethlehem, a national union for Republic, and so forth. If this is the proposal, let’s analyze it and look at it and see what it would do. If you are concerned about the emergency dispute, I would remind you there is a much simpler way of doing this which some people have advocated. Some people have proposed no more than one-half of the industry shall be shut down at one time and that has nothing whatever to do with antitrust. If you are interested in the wage problem, let’s look at it that way. I once made a study of how steel wages moved from the 1890’s up to the time the present union was introduced. While I do not have those figures in my memory at this time, what they showed was what one would expect, I think, namely that common labor rates in the steel industry moved very much together, indeed, moved the same day. Why? Well, it moved the same time in Pittsburgh and Chicago for the simple reason of the interdependencies that the market pro vided. You had wrage and industrial relations leadership develop long before there was any national union in the picture. It seems to me the splitting of the steel industry into separate industrial groups would not affect anything about the timing of wage changes. Now you may argue that splitting up bargaining will reduce the amount of these changes. Now I can’t sit here and tell you it will not have that effect, but I do not see any reason why it should, because it seems to me that just as there are interdependencies on the product market side, so there would be the same type of interdepend encies, indeed, more, on the labor side. One further point, if the application of the antitrust laws to the product market have not resulted in any splitup of the steel indus try, I cannot see that the simple application of the antitrust laws to the unions in the steel industry would change that situation. Mr. F r u c h t . I think in our hearings it was very well brought out by both sides, I think, of this table that there is some question as E M P LO Y M E N T, G R O W T H , AND PRICE LEVELS 2753 to the adequacy of the enforcement resources in the hands of the Justice Department and that this is not irrelevant and I would not suggest that we pass a law to deal with unions and give an agency $50,000 or funds of this nature in which to do it nor am I suggesting that I know the answer to the steel industry. The courts had a terrible time in coming up with a remedy in the Alcoa case. The Justice Department and the courts have had a rough time in dealing with the Du Pont and General Motors situa tion, but since it is not obvious as to what appropriate remedies are in dealing with Du Pont, would you, therefore, say that we should not have a law that would make us find one ? Mr. D u n l o p . I would say if somebody has a specific problem to deal with, let’s deal with it specifically. A general and vague act which puts everything under the sun under a cloud as to its status is precisely the kind of uncertainly in the collective bargaining area which we cannot tolerate. I would rather have you say to me, I don’t like this and this and this, and I would be prepared to discuss each problem. Mr. L e v i n s o n . I would like to go back, if I may, to a comment that you made in passing, while you were answering Mr. Frucht’s questions. You suggested that if what we were worried about was the inflation problem that you had a lot of better ways of dealing with it than through antitrust legislation. I should like to ask you then, What are your better remedies than antitrust legislation ? Mr. D u n l o p . That is a whole different world, a subject on which I suppose every economist could talk a full day. But let’s try to be fairly specific and brief. In general, I would think of two broad areas of approach. First, we need substantially to refine in the United States our instruments of fiscal and monetary control. We have only the most general controls. We are scared of inflation and so we do not take very many risks, We stay very far away from the brink. If we had more refined fiscal and monetary controls, our chances of pushing output and employment higher without the danger of inflation could be substantially increased. For example, and I am not proposing any one of these, but they are the types of instruments that need to be explored. We could perhaps make our social security taxes cyclically variable. We could perhaps make a number of our depreciation allowances variable. We could, for example, introduce various kinds of differential tax rates which we do not now’ have. We could develop more explicit monetary con trols over consumer credit. In other words, we have very general fiscal and monetary controls. If we expect to let the system reach higher output and employment levels without inflation, we have to develop a great many more detailed fiscal and monetary controls. There are men in this room who are perhaps more competent to deal with this than I am. I am answering your general question of policy. This is one area where as a country we are way off the beam. We have not really begun to experiment in this area as compared to what is possible. The second idea that I would emphasize is the point of view em phasized at the end of this paper, namely, I do think that it is 2754 EM PLOYM ENT, GROWTH, AND PRICE LEVELS possible in industry and in the economy generally to help the parties* in collective bargaining take a longer run view. May I make a remark off the record ? Representative B o l l i n g . Off the record. (Discussion off the record.) Mr. D u n l o p . The point I would put on the record is that I believe that it is possible for the parties to collective bargaining in industry after industry to establish forums in which they concern themselves with costs, productivity and other facets of the economies of their industries so that over the long run in sector by sector, we do some thing to slow down the inflationary rate, and this would supplement the annual kind of general conference where all parties take a look at the economy as a whole. Representative B o l l i n g . I have one question. In your statement, you talk about procedures being set up in an agreement so that, if they have trouble arriving at the new contract there is this procedure established. Then, your last sentence is that they might be required to develop their own procedure within a specified period or confront the legislative mandate of a Governmentappointed arbitration board in disputes that threaten the Nation’s health or safety. What is the way in which this is encouraged? Is the operative point here the threat of Government arbitration, or is the necessity in meeting emergency situations, strikes, or lockouts, or whatever they may be, to have on the books legislation that says if you do not do so and so then you come under this particular provision ? Mr. D u n l o p . That is an interesting question, Mr. Chairman, and I would like to make two observations. In the first place, as you know, the Congress has in at least one major area adopted exactly this device to encourage the private settlement of disputes, and I refer to the area of jurisdictional disputes where section 10 -K of the statute gives the parties the opportunity to settle their problems themselves or else the Government will. The government was sub stantially influential in encouraging the parties to set up their own procedures, and I think that device can be used with a number of other problems. In the second place, in the emergency dispute problem one ought to proceed only after a certain amount of preparation, a certain number of conferences, a certain amount of informal exploration. If in the last resort the parties do not make progress with respect to the long-run solution of some machinery, I would in the end say leg islatively to a major sector; either you develop your own procedures by which you spell out the steps that you yourself will take to settle future disputes, or we shall specify that if the dispute threatens the Nation’s health or safety you have to arbitrate. I think there is merit in saying that such an award should run for no longer than 6 months. You see, serious disputes must be settled by keeping at work on them. It is ultimate consent and acceptance by both parties that has to be had. By this device one would not only get over the immediate hump but you would keep working at the problem. There is an adage about keeping them talking and working, narrowing the differences, you get a recommendation, you narrow down the dispute in successive con ferences and proceedings. So I would want to limit the applicability of the award to a period of 6 months. E M P LO Y M E N T, G R O W T H , AND PRICE LEVELS 2755 Representative B o l l i n g . Thank you very much for a very stimu lating paper and a very stimul ating discussion. With that, I will announce the next set of hearings. They will be in this room on October 26, 27, 28, 29, and 30. The subject will be constructive suggestions for reconciling and simultaneously obtaining the three objectives of maximum employment, an adequate rate of growth, and substantial stability of the price level. With that the committee stands adjourned until October 26. (Whereupon, at 4:1*5 p.m. the committee was recessed, to reconvene at 10 a.m. October 26,1959.) X