The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
Authorized for public release by the FOMC Secretariat on 2/25/2020 REC BOARD OF GOVERNORS OF THE N RECORDS SECTION SEP 1 71957 FEDERAL RESERVE SYSTE WASHINGTON 25, D. C. ADDRESS DF'FICIAL CORRESPONDENCE TD THE BOARD September 12, 1957, Dear Sir: As requ.ested at the open market meeting on September 10, I have prepared a summary of my comments on the arguments presented in the hearings before the Senate Finance Committee by critics of the System's policies. Several copies are being sent to you for your use. Sincerely yours, • I / , Winfield W. Riefler, Secretary, Federal Open Market Committee. Enclosure TO THE PRESIDENTS OF ALL FEDERAL RESERVE BANKS. \ Authorized for public release by the FOMC Secretariat on 2/25/2020 INFLATION CAN BE STOPPED: Print More Money and Reduce Interest Rates The above caption may seem bizarre to specialists in the behaviour of money, but it epitomizes, not too unfairly, the case presented to the Senate Finance Committee by an important group of critics of the Federal Reserve System. It is crucially important that all elements in the Federal Reserve System understand the reasoning by which these conclusions are reached. Essentially they rest on five assertions, only one of which (and not the crucial one) is easily demolished. The remaining four are highly plausible, so plausible that many people, unless they thought the problem through, might well find themselves assenting. The tactic used is to get assent to each one of the five assertions individually. Then, the seemingly ines- capable conclusion is drawn, namely, that easy money, including more spending and less saving, is the only way to stop the rise in the cost of living. The first, and most easily refuted proposition, is that since in- terest rates are costs, higher interest rates lead to higher costs. There- fore, higher interest rates account for an important part of the inflation of prices. This proposition is still being asserted, but less confidently asserted than when the Hearings began. Statistics presented at the Hearings, showing that interest costs are in fact a very small proportion of total business expenses, have made this assertion a pretty farfetched one. When it comes to the remaining four principal assertions, refutation is not so easy a matter. Hence, they are still being pressed by Federal Reserve critics with vigor and confidence. They run as follows: Authorized for public release by the FOMC Secretariat on 2/25/2020 -2Assertion I--The American economy today is not characterized by a shortage of manpower, since unemployment is one-third higher than in 1952 when prices were stable. Assertion II---The American economy today is not confronted with a shortage of physical capacity to produce since new capacity has been and still is being greatly enlarged. Assertion III--Consumer disposable income, in terms of real purchasing power, has not grown during the past year. Assertion IV--It is generally recognized that more produc- tion is the best cure for inflation. Given assent to these four assertions, the conclusion follows that measures to stimulate consumer spending, rather than saving, would so increase the output of goods and services, for which both manpower and capacity are available, as to cure inflation. This conclusion really asserts that the creation of more money, by increasing the demand for output, would curb inflation. The logical validity of any conclusion can be tested by stating it in reverse. In this case, the reverse proposition would be that the sure way to cure a deflation would be to raise interest rates and force contraction of the money supply. But such logical refutation of the main conclusion does not meet the need of refuting each of the assertions separately. done. This must also be Accordingly, the discussion below is devoted to their respective pit- falls. I. What About Excess Manpower? It is true that the percentage of unemployment today is higher than in 1951, '52 and '53, a period of price stability, but this does not mean that the current rate of employment is not pressing on our manpower resources. In his opening statement the Chairman of the Board stated, Authorized for public release by the FOMC Secretariat on 2/25/2020 -3"Despite the existence in some lines of reduced employment and slack demand, many employers face rising costs when they seek to expand activity by adding appreciably to the number employed. bid away from other employers." Often the manpower required has to be In other words, under current conditions of very high employment, further efforts to stimulate output on the scale suggested would soon spark a further rise in the inflation spiral. costs which would accelerate Before the Hearings adjourned, the Chairman placed in the record the appended technical appraisal of the unemployment figures cited in the Hearings. It throws considerable light on the statistical prob- lem of measuring unemployment and should be studied carefully. There is always a labor supply at a price outside the current labor force (housewives, students, retired workers, etc.). that the rate of unemployment in The facts are this inflationary period has been very low and at frictional levels; there has not been a margin of manpower available at current wages with which to raise total output above levels actually realized. When such a margin obtains, the curve of unemployment takes a definite U-shape, being quite high at the younger and older age groups as well as higher for all age groups. When the demand for manpower is unemployment stays relatively high at lower age levels, active, though lower than when demand is less active, but declines sharply to middle-aged groups, from which it tapers off without much rise in the oldest age group. This kind of pattern has obtained consistently since early 1955. II. Elimination of Specific Shortages Does Not Necessarily Expand Overall Capacity. The Chairman's statement pointed out that specific bottlenecks in capacity that impeded growth in production a year ago have been largely Authorized for public release by the FOMC Secretariat on 2/25/2020 -4relieved and that individual bottlenecks are no longer the cause of bidding up of prices of individual commodities because of limited availability. This was seized on by critics as an affirmation that we are suffering from underutilization of resources. The problem, of course, is much more complicated. Employment figures show that consumer demand has shifted somewhat away from goods and toward services, private and government. Under such circumstances, particular segments of the economy can have unutilized plants without the economy overall being in a position to expand greatly its total output. So- called excess capacity, furthermore, does not mean that materially higher rates of utilization would not entail rising costs. not. It might or it might That would depend upon the technical efficiency of the reserve capacity as well as many other considerations. III. It Is Inflation Not Underemployment That Has Impinged on Real Earnings. The assertion that consumer disposable income has not increased during the past year disregards the inflation, which is our problem, by resort to statement in terms of ex-inflation dollars. Consumer personal income available for spending has grown appreciably both in absolute amount and on a per capita basis during the past year. It is the largest single component, by far, of the total spending stream that has sustained the continued rise both in wholesale prices and in the cost of living. If consumers had saved a larger proportion of this income, it would have been available for the financing of schools, highways, and capital plant without contributing further to inflation and the reduction in the value of their spending dollars. As it is, the inflation that has actually occurred has offset in large part the buying power of the increase in consumer disposable income. Authorized for public release by the FOMC Secretariat on 2/25/2020 This sort of development is flation. in For example, manufacturing stability. practically all not exceptional during a period of inof the gains in industries since World War II real weekly earnings have come in periods of price From mid-1946 to mid-1948, a period of sharp inflation, both consumer prices and average weekly earnings rose by close to 25 per cent and there was little gain in real wages. there was little change in prices, From mid-1948 to mid-1950, but a weekly earnings of about 10 per cent. gain in however, the purchasing power of From mid-1950 to the sping of [sic] 1952, sharply rising prices again offset rising weekly earnings. During the en- suing long period of stability in the cost of living, lasting until early 1956, the rise in money earnings was reflected in comparably large gains in real wages. Since that time further large wage increases have been largely nullified by the inflation. Thus during the whole period since the war, the appearance of inflation has coincided with a leveling off of real wages. IV. Increased Production Per Se Does Not Cure Inflation. Money income is generated in part of the spending stream. the process of production and becomes As has frequently been pointed out, one man's expense in general is duction in themselves add to the stream of spending as well as to the stream of goods. Increased output is another man's income. Consequently, increases in pro- desirable to the full extent permitted by our capabilities, provided that it is the right production and is financed in such a way as to promote continued prosperity. However, if there is excess money demand present in the economy at a time when resources are actively employed, that excess will cause a rise in prices. Increased production under these circumstances will add to the spending stream as well as to the stream of goods and services. It will not, therefore, eliminate the Authorized for public release by the FOMC Secretariat on 2/25/2020 -6excessive money demand that is the cause of rising prices. For inflation to be curbed, excess money demand must be absorbed from the spending stream. This may come about by the development of a budget surplus, by increased planned savings, by curtailed borrowing from banks, the growth of the money supply or in its turnover. or by a slowing down in It does not result automatically from increased production. No one would maintain that a cessation of production--the reverse of this proposition--would stop a deflation. Likewise, an increase in pro- duction does not in and of itself stop an inflation. of France today is a standing example of this fact. The unhappy condition It sharply increased its production as well as its productivity, but it failed to take measures adequate to reduce the excess money demand that was necessary to avert an acute inflation crisis. Authorized for public release by the FOMC Secretariat on 2/25/2020 TECHNICAL APPRAISAL OF FACTORS ACCOUNTING FOR DIFFERENCES IN UNEMPLOYMENT; 1951 - mid-1953 and 1957 Current unemployment First, in terms of perspective, current levels of unemployment. 3.0 million, or it is worthwhile examining In July 1957 unemployment totaled 4.3 per cent, of the civilian labor force based on new definitions which were adopted starting January 1957. If old definitions were used (and data on the old basis are the only data comparable with earlier periods), unemployment in July would have been reported as 2.7 million, or 3.8 per cent, of the labor force. The summer months tend to be the high months in the year in respect to unemployment because of a large influx of students and graduates looking for summer jobs. As students leave the labor force in September and as the usual fall expansion in industrial activity gets under way, unemployment drops rather sharply. Between July and October unemployment can usually be expected to decline by 700,000 to 800,000. Thus, if only seasonal factors affect unemploy- ment between now and fall, the number of workers seeking jobs in October of this year will be only about 2.2 million (new definition). Since early 1955 seasonally adjusted unemployment has remained virtually unchanged with the unemployment rate moving within a one-half per cent range and with no consistent trend in either direction. During this period over 3 million workers were added Authorized for public release by the FOMC Secretariat on 2/25/2020 -2- to the labor force, a much larger increase in the labor force than would have been expected on the basis of growth of the population of working age, reflecting the continuing strong demands for workers. This fact also indicates the frictional nature of current unemployment in that it has been necessary to go outside of the labor force to meet demands for additional employees. Other indications of the current low level of unemployment are about two-thirds of the unemployed have been looking for work less than 6 weeks and that only 500,000, or less than 1 per cent of the labor force, were reported as having been unemployed for 15 weeks or more in July. Except for the very young age groups who are just starting their work careers or looking for summer work, unemployment rates among adult workers are very low. In each age group 25 years and over the unemployment rate was substantially below the average. For married males with wife present, the unemployment rate in July was only 2.3 per cent. While there are a number of areas which re- port substantial labor surpluses (unemployment rates of 6 per cent or more) they consist mainly of textile towns and mining areas, in which the age, sex, past work experience, and geographical location have in large part prevented the absorption of these persons into gainful employment in a period of expanding demands for workers. In contrast, there are still reported shortages for engineers, teachers, and other professionals along with some kinds of skilled workers. Authorized for public release by the FOMC Secretariat on 2/25/2020 -3Comparison of current unemployment with 1951 to mid-1953 1. In the first half of 1957 average unemployment was about 800,000 to 1 million more than in comparable months in 1952 and in 1953. (Old definition used in both instances for purposes of comparability.) Since 1952 some 6 million people have been added to the labor force. If the data is standardized to take account of increases in the labor force and differences in age and sex distribution in the two periods, unemployment would have increased by 200,000. 2. The major differences in unemployment in primarily result from the Korean hostilities. the two periods Between mid-1950 and the defense peak the armed forces increased by 2.3 million. This resulted in a sharp reduction in the number of unemployed males under 25 years of age. Since mid-1953, however, the armed forces have been reduced by 800,000 men, from 3.6 million to 2.8 million, and this to some extent accounts for a slightly higher unemployment rate among younger men in 1957 than in the earlier period. 3. During the period of the Korean conflict there was a well advertised manpower shortage. Public agencies and many employers conducted an active and extensive drive for workers. had a number of effects. This apparently It tended to reduce unemployment as well as the length of unemployment for those seeking work. On the other hand, it led to hoarding of workers and use of less efficient workers on the part of employers who feared that sufficient. manpower might Authorized for public release by the FOMC Secretariat on 2/25/2020 -1not be available in the future. On the whole in this period it appears that there was a good deal of under utilization of manpower and there was very little growth in productivity. It was not until after cessa- tion of hostilities that output per manhour started to rise again. 4. It seems likely, although it is difficult to prove, that during periods of hostilities (World War II and Korea) people interviewed in the Census household sample surveys may have been reluctant to admit to being unemployed -- on the assumption that unemployed persons were not contributing to the defense effort in view of stories of worker shortages delaying war efforts and other patriotic appeals. During World War II, reported unemployment fell to the very low figure of 500,000. 5. Since 1953 there has been a reduction in manpower require- ments in the railroad, mining, and textile industries which has resulted in some increase in longtime unemployment and is reflected in somewhat higher rates of unemployment among older workers now than in the 19511953 period. As mentioned earlier in this memorandum, this has resulted in what might be called some chronic unemployment; but the number of such persons appears to be small. 6. In 1956 and 1957 there have been a number of mixed trends in the employment situation resulting in some layoffs. In 1956 the re- duction of automobile production was definitely reflected in the unemployment totals but was offset by other gains. In 1957, while unemployment among automobile workers declined, reductions in residential construction, Authorized for public release by the FOMC Secretariat on 2/25/2020 -5lumber, electrical machinery and more recently aircraft employment have tended to keep the unemployment totals fairly constant, slightly higher than if 7. all but probably activities were currently rising. The unemployment series is based on a sample survey and has all the difficulties of such data including sampling error. A difficult factor to evaluate has been the improvement in unemployment data resulting from two changes in In 1957, month. the Census sample since 1952-1953. Census interviewed about 35,000 households in In 330 areas each 1952-1953 only 21,000 households were interviewed in 68 areas each month. Sampling error for unemployment in lated as 190,000. In 1957 the sampling error is 100,000 for unemployment. 1952 was calcu- The Census sample was increased from a 68-area sample to a 230-area sample in January 1954. The results of the new sample showed that for January 1954 unemployment exceeded the old sample figure by some 700,000, or 31 per cent, a considerably larger difference than could reasonably be attributed to sampling variability. An examination of the evidence by the Census Bureau's staff and a special technical committee led to the conclusion that the old sample figure was understated, partly because of inadequacies period of transition to the new sample. in interviewing during the On the basis of comparing the total unemployment statistics with the number of persons receiving unemployment compensation, it started in September 1953. was concluded that the understatement On this basis an adjustment was made by Authorized for public release by the FOMC Secretariat on 2/25/2020 -6arbitrarily graduating downward the percentage difference between the old and new sample estimates of unemployment from January 1954 to September 1953 and no change was made for prior months. 1956 the sample was again expanded, In April this time to 330 areas but the unemployment figure was reported as approximately the same for both samples. Authorized for public release by the FOMC Secretariat on 2/25/2020 UNEMPLOYMENT AND LABOR FORCE July of each year 1950 1951 1952 1953 August 16, 1957 Number unemployed (thousands) Unemployment rate Total all ages 14 to 19 years 20 to 24 years 3,213 5.0 1,856 .912 1,548 1957 1957 (Old definition) 2,686 (New definition) 3,007 3.8 4.3 10.9 11.8 5.5 3.4 2.3 2.5 2.7 3.0 6.2 3.8 2=8 2.7 3.0 3.3 7.0 2.9 8.1 3.6 25 to 34 years 4.5 2.2 2.3 2.4 6.6 3.2 1.9 35 to 44 years 45 to 54 years 55 to 64 years 3.3 3.6 3.8 3.6 2,0 2.3R 1.8 2.1 2.4 1.7 1.8 1.5 2.3 2,2 2.3 65,742 67,477 67,642 28,804 73,056 73,051 1,315 3,095 3, 466 3,590 2,823 2,823 64,382 64,1176 65,214 70,233 70,228 65 years and over Total labor force (thousands) Armed forces (thousands) Civilian labor force (thousands) 12.0 3.0 8.8 4.2 1.4
@FedFRASER