Speech

For Release on Delivery Thursday, April 13, 1972 12:30 p.m., C.S.T. (1:30 p.m., E.S.T.) A NEW AMERICAN DILEMMA The Task of Reconciling Growth in Productivity and Employment Remarks By Andrew F. Brimmer Member Board of Governors of the Federal Reserve System At a Luncheon Sponsored by the Joint Boards of Directors of the Federal Reserve Bank of St. Louis Upon the Dedication of the Memphis Branch Building Memphis, Tennessee April 13, 1972 Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis A NEW AMERICAN DILEMMA The Task of Reconciling Growth in Productivity and Employment by Andrew F. Brimmer At this juncture in our history, we are confronted with a vital question involving multiple choices: how can we increase the rate of real growth in the economy, reduce the persistently high rate of unemployment, and—simultaneously--press on with the campaign to check inflation? That we find an answer to the question is imperative. In its Annual Report for 1972, the Council of Economic Advisers anticipates that--although real output is expected to increase by about 6 per cent--the unemployment rate probably will still be in the neighborhood of 5 per cent as the year draws to a close. At the same time, the target set for the price stabilization program visualizes that the rate of inflation will decline to a range of 2-3 per cent by the end of 1972—compared with a rise in prices of 4.3 per cent in 1971.— * Member, Board of Governors of the Federal Reserve System. I am grateful to several members of the Board's staff for assistance in the preparation of these remarks. Mr. Lawrence Slifman helped with the analysis of trends in output, productivity, and employment; his contribution included an econometric analysis of the response of productivity to changes in real output. Mrs. Diane Sower provided material on the performance of the Public Employment Program. Mr. David Wyss did the computer-based simulations of the national economy to estimate the differential effects of pursuing alternative approaches to reduce unemployment. Needless to say, the views expressed here are my own and should f not be attributed to the Boards staff—nor to my colleagues on the Board. 1/ This is the annual rise in the Consumer Price Index. The ~~ Wholesale Price Index advanced by 3.2 per cent, and the GNP implicit deflator rose by 4.7 per cent. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis -2- Thus, achievement of the goal set with respect to the rate of increase in prices would represent a significant accomplishment in the fight against inflation. So too would the reduction in the unemployment rate represent genuine progress. In March, those without jobs butt looking for work totaled 5.2 million, and unemployment has lingered in the neighborhood of 5 million for more than a year. And if the jobless rate should still be around 5 per cent at the end of the current year, it would mean that perhaps more than 4-1/2 million people would be unemployed. Why might we be facing that prospect? In searching for an answer to that question, I reached the following conclusion: the growth in employment this year may be greatly dampened by the rise in output per manhour (productivity) as the economy expands. Since the civilian labor force may also expand more rapidly, the level of unemployment can be expected to decline only moderately. These conclusions were reached on the basis of a fairly detailed analysis of the interrelations among the growth in output, productivity, and employment during the last two decades. The highlights of the examination can be summarized here: --Productivity in the private nonfarm economy rose at an annual average rate of 2.6 per cent during the period 1948-1971. However, while the rate of increase was just over 1.0 per cent in the years 1967-70, it jumped to 3.4 per cent last year. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis - 3 - --The dampening effect of reduced growth and declines in output on wage increases and unit labor costs during the recession phase of post World War II business cycles has been noticeable. But this impact was far less evident in the case of the 1970-71 cycle. Thus, employers have been counting heavily on increased productivity in the current recovery to help restrain upward pressure on labor costs. --Consequently, last year, when the economy was achieving some expansion in output, a large share of the rise in production was accounted for by higher output per manhour, and only minor benefits accrued in the form of higher employment. --The outlook is for a further large rise in productivity in 1972--perhaps in the neighborhood of 3-1/2-4 per cent. The net results might be an increase in employment only moderately more than the growth in the labor force--and thus only a modest reduction in unemployment. Since the nation definitely cannot afford to call a halt to improvements in productivity, we must search for alternative ways to stimulate employment--if we are to cut substantially into the persistently high level of unemployment. Among several alternatives which might be considered, an expansion of the Public Employment Program appears to be especially promising. The major points summarized above are explored further in the rest of these remarks. Long-Run Trends in Productivity During the last two decades, productivity in the private nonfarm sector of the American economy rose at an average annual rate of 2.6 per cent. (See Table 1, attached.) However, the growth of productivity has been far from steady. For example, in the Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis - 4 - first half of the 1950's, output per manhour in private nonfarm activity grew at a rate equal to the long-run average. But in the closing years of the same decade, the rise in productivity was substantially below the postwar trend, averaging only 1.8 per cent per year from 1955 to 1960. Growth rates considerably above the long-run trend were achieved during the 1960's. From 1960 to 1966, private nonfarm output per manhour climbed at an average annual rate of 3«4 per cent. In the succeeding five years, the performance of productivity slackened noticeably. Over the years 1967-71, output per manhour rose at an average annual rate of 1.7 per cent per year. In fact, nonfarm productivity actually declined through 1969 and in the first quarter of 1970. In the manufacturing sector of the economy, productivity has risen more rapidly than in the private nonfarm segment as a whole. During the last 20 years, the average annual increase was 2.8 per cent. But the pattern of growth has been roughly parallel to that of the nonfarm sector taken as a unit. Large, above-trend gains were posted from 1962 through 1965, and below-trend growth f rates were recorded in the last half of the 1960s. Cyclical Behavior of Productivity The behavior of productivity is greatly influenced by the phase of the business cycle through which the economy might be passing at a particular time. The close linkage between the cyclical Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis - 5 - behavior of total output arid the growth of productivity can be seen in Chart I (attached). In the chart, one can trace the two measures over the years 1947-71, and peaks and troughs of the five post-World War II business cycles are also identified. It will be noted that, in the contraction phase of the cycle, growth in productivity typically slowed down--and sometimes there was a decline. Several factors help to explain this pattern: these include the intentional and accidental hoarding of labor (especially white-collar labor), the time that lapses after the cyclical peak is reached in the economy and individual employers recognize this fact in their own businesses, and shifts in the industrial mix of output from high productivity sectors (which tend to be sensitive to cyclical variations) toward the more stable but relatively low productivity sectors. Employers need to maintain a minimum labor force to meet future production needs. Consequently, they tend to maintain employment even after the growth of output begins to moderate, and they persist in doing so until they are convinced that the economic slowdown is not temporary. Moreover, even after it is evident that a contraction is in process, employers have often been reluctant to reduce overhead labor--such as administrative and professional staff. In the early recovery phase of a business cycle, productivity growth tends to accelerate as overhead and other relatively fixed labor costs decline per unit of output. Having adopted cost-cutting Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis -6- programs to trim excess personnel during the recession, employers hesitate to add non-production employees when new orders and output begin to pick up, and there is little need for additional overhead personnel. This pattern of behavior is clearly evident in the statistical record. For example, in the 1958 recovery, employment of production workers in manufacturing rose by 8.8 per cent during the first four quarters after the cyclical trough was reached while employment of non-production workers increased by only 2.9 per cent. Similarly, in the recovery from the most recent recession, employment of factory production workers grew by 0.4 per cent between the fourth quarter of 1970 and the fourth quarter of 1971. Yet, over the same period, jobs held by non-production workers declined by a further 2.7 per cent. The extent to which increases in productivity can be achieved during the recovery from a recession depends significantly on the rate of growth of real output. This linkage can be seen in Table 2, showing the behavior of output and productivity during the recovery phase of business cycles. Chart I presents the same evidence in a different form. In general, the sharper the upturn in output during a recovery the larger the jump in output per manhour. However, the relationship has varied somewhat in each postwar recovery. For example, during the first year of recovery from the 1953-54 recession, gross national product (GNP) in real terms (i.e., corrected for price changes) rose by 8.6 per cent; productivity Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis - 7 - increased by 4.3 per cent, and manhours climbed by 5.6 per cent. In the 1958 recovery, the rise in output in the first year following the trough was especially large--9.2 per cent. However, while there was a substantial increase in productivity (4.6 per cent), manhours worked also rose appreciably (5.6 per cent). In contrast, the strong advance in output in the first year of recovery from the 1960-61 recession (7.6 per cent) generated an exceptionally strong increase in productivity (6.0 per cent), while the rise in manhours was rather moderate (2.2 per cent). The magnitude of productivity increases achieved during the sluggish 1970-71 recovery was not typical of that recorded in other recovery periods. During the four quarters following the cyclical trough (which occurred in the last quarter of 1970), real output rose by only 3.5 per cent (substantially less than in earlier periods). Productivity rose by 4.1 per cent, and manhours rose by 1.2 per cent. Moreover, the growth of output and productivity during this period (1971) was stimulated by the recovery from the General Motors strike. After correcting the data to allow for this factor, the slower growth of productivity than in previous recoveries is more evident. Another aspect of the cyclical behavior of productivity (shown in Table 2) should be noted. During the second year of recovery, the rate of growth of output, productivity, and manhours Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis -8- slowed down appreciably in the three earlier periods. However, the typical second-year behavior may be different as the current recovery progresses: rather than a moderation in the rate of growth of productivity, we may witness an acceleration in output per manhour. This may come about because of the anticipated faster growth in real output this year--in contrast to the slowing in growth in the second year in each of the other periods. The implications of such an outcome for employment will be discussed at a later point in these remarks. Productivity and Unit Labor Costs Productivity growth is a key element in checking inflation. This is true because the relationship of the increase in output per manhour to compensation per manhour determines unit labor costs. The record of changes in productivity and unit labor costs during the years 1948-71 is presented in Table 3. The cyclical behavior of unit labor costs is sketched in Chart II. The typical cyclical pattern has been for unit labor cost pressures to ease during the recession phase of the cycle as slack labor markets dampen the rise in compensation per manhour. During the recovery phase of the cycle, it has been the above-average productivity gains which have tended to offset rises in compensation and to limit unit labor cost increases. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis - 9 - However, the experience during the 1970-71 recession and recovery so far has not conformed to that typical pattern. Compensation per manhour continued to rise at a rapid pace throughout 1970 and into early 1971. This occurred even as unemployment rose. In addition, productivity showed little change in 1970, and it did not grow nearly as fast as is usually the case during the 1971 recovery period. As a result, unit labor costs continued to advance fairly rapidly prior to the imposition of the Wage-Price Freeze in August of last year. Productivity and Employment As indicated earlier, the drive to improve productivity poses a dilemma: while productivity growth helps to dampen cost pressures, it also has a retarding impact on employment gains. The source of this dilemma is easily understood. For a given amount of expansion in production, the larger the increase in labor productivity the smaller the increase in manhours required. In a recovery period, strong productivity growth and increases in the work week tend to limit employment gains. The interrelations among output, productivity and employment during the recovery phase of business cycles are shown in Table 4. It will be noted that, in recovery periods, private nonfarm output tends to grow rapidly in the first few quarters after the cyclical trough is reached: but so does output per manhour. During the first half-year following the low point in output, productivity gains and increased hours tend to account for much of Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis -10- the output growth. But for a variety of reasons which need not be catalogued here, there is a limit to the increased production that can be obtained through higher productivity and overtime work. At some point, as the demand for output continues to expand, additional workers must be added to the payroll. The data in Table 4 allow one to trace this interplay of changes in output, productivity and employment. Essentially, the figures show the ways in which the economy was able to achieve the growth in output recorded in the full year of recovery following the trough of the last four business cycles. In general, one can visualize changes in productivity, hours of work, and employment as three sources of inputs required to generate the observed growth in output. The specific share of each in the increased output varied substantially, but several clear tendencies are observable. Among the three, increased productivity appears to take the lead, accounting for half or more of the output gain in the first six months of recovery. Increased employment was of somewhat less importance, and longer working hours accounted for the smaller proportion. As the recovery progresses, the share of output growth attributable to productivity and longer work weeks tends to diminish and the share contributed by increased employment climbs. Again, however, the experience during the 1971 recovery varied significantly from the typical pattern sketched above. In the first six months following the trough of the 1970 recession, Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis -11- the rise in output per manhour accounted for three-quarters of the advance in output, and the proportion rose further over the next half year. The share attributable to longer working hours fluctuated considerably, but on balance the net results were a positive contribution of about one-sixth of the output gain in the first six months and an offset of roughly the same magnitude in the next half year. Higher employment provided less than one-tenth of the net rise in output in the first six months and just over one-fifth in the succeeding half year. Thus, from an analysis of the data presented here, one overriding conclusion emerges: last year, when the economy made only moderate progress in recovering from the 1970-71 recession, a disproportionate share of the rise in output was accounted for by higher productivity, and only minor benefits accrued in the form of higher employment. In passing, I might note that the output-productivity- unemployment dilemma is not unique to the United States. In the United Kingdom, industrial productivity rose 4.9 per cent between 1970 and 1971--compared to an average increase of 3.8 per cent during f the middle 1960s. Productivity growth was also large and in part was reflected in a sharp increase in unemployment. For example, in March, 1972, the unemployment rate was 1-1/2 times greater than 2/ its late 1970 average.— Although productivity growth in the U.K. has 2/ British labor statistics are calculated on a somewhat different conceptual basis than U.S. data and are not strictly comparable. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis -12- had welcome implications for costs and prices, it has also been a major factor in creating their largest unemployment problem since the beginning of World War II. As far as the United States is concerned, this means that if employment is to expand sufficiently to reduce unemployment significantly in 1972—and if labor cost increases are to be moderated--large output gains will be required. Outlook for Productivity and Implications for 1972 Given these recent trends in productivity and employment, we ought to raise questions about the prospects for the future. I have raised some of these questions, and the answers I obtained can be summarized here. In undertaking this task, my purpose was not to make an independent assessment of the outlook for the economy as a whole. Rather, I took as a point of departure the general economic outlook contained in the Economic Report of the Council of Economic Advisers (CEA). That forecast is for an acceleration of real economic growth during 1972 to somewhere in the neighborhood of 6 per cent, compared to the 2.7 per cent rate of growth achieved in 1971. In particular, the expected strengthening of activity in the manufacturing sector (in which productivity growth is particularly high) is of special interest. Thus, the question to which I wish to address myself at this point is this: How will this outlook for the national economy as a whole affect productivity and employment? Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis - 13 - In seeking an answer to this question, I pursued several avenues. First, on the basis of informal soundings among businessmen, I have gotten a clear impression that many manufacturing firms are moving cautiously in expanding their payrolls. Among other factors, a strong desire to control costs by meeting increased output demands through higher productivity rather than higher employment appears to have moderated the pace at which factory jobs grew throughout 1971 and into the early months of the year. These qualitative impressions are reinforced by the statistics relating to employment. In March, 18.8 million people were employed in manufacturing industries. This was an increase of 310,000 from the low point reached in August, 1971. Yet, this was about the same number as a year earlier and still nearly 1.6 million below the peak set in the third quarter of 1969. For the economy as a whole, the outlook this year is for considerable improvement in the job s ituation. However, the civilian labor force is also expected to rise appreciably, and the net effect on unemployment is cloudy. In March, the civilian labor force (seasonally adjusted) totaled 86.3 million. This was an increase of 2.4 million above the level a year ago.—^ Total employment amounted to 81.2 million. This was also a year-to-year gain of 2.4 million in the number of people with jobs. The number of unemployed persons totaled 5.2 million in March, an increase of about 40,000 since March, 1971. 3/ This increase allows for an expansion due to the 1970 Census population control adjustment introduced into the household survey in January, 1972. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis - 14 - To look beyond the employment experiences during the early months of 1972 to the rest of the year is necessarily difficult. In their Annual Report, CEA noted that: "Our estimate implies an increase of 6 per cent in real GNP between 1971 and 1972. This is about the rate of increase achieved in the fourth quarter of 1971. There is no easy way to separate the forecast rate of real growth into employment and productivity gains. Clearly, it should yield substantial increases in both. The extent to which the employment gains will reduce the unemployment rate depends on the size of the increase in the labor force. It is estimated that the unemployment rate should decline from the 6 per cent level of December 1971 to the neighborhood of 5 per cent by the end of 1972. (Report, 1972, p. 108). I agree that it is difficult to unravel the effects of gains in output, employment,and productivity. However, as shown above, the behavior of productivity has a significant bearing on the amount of employment we should expect to result from a given increase in real GNP. Thus, if we are to form a reasonable idea of the prospects for employment in 1972, we must search for even rough clues to the probable trend of productivity in the current year. To aid in this search, I made use of computer-based econometric techniques which are becoming increasingly useful tools of economic analysis. The results of the analysis provide an insight into the productivity-employment contour which might be expected for 1972. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis -15- The statistical analysis took as a point of departure 4/ the CEA forecast of a 6 per cent increase in real GNP during 1972.— Past experience suggests a 3-1/2-4 per cent productivity growth consistent with a 6 per cent increase in real output. In using this estimate, I must emphasize that it is only a rough approximation. But it does seem to be broadly consistent with the growth of real output expected this year. Using the 4 per cent figure for productivity,-^ it was estimated that nonfarm payroll employment might rise by about 2 per cent in 1972. This would mean an increase in jobs of roughly 6 / 1.5 million.— Under these conditions, the growth of total employment would be greater than 1.5 million—because of increases in the number of self-employed workers, private household workers, and other workers not on nonfarm payrolls. In rough terms, the rise in these categories might amount to 0.5 million. Thus, the increase in total employment implied by a 6 per cent growth in real GNP in 1972 might be in the neighborhood of 2 million. What does this mean for the unemployment rate? Normal labor force growth of about 1-1/2 million per year is currently 4/ For those interested, the statistical method followed was to estimate a projection equation by multiple regression analysis. Using forecasted change in real GNP, the change in private nonfarm output per manhour was projected for the period from the fourth quarter of 1971 to the fourth quarter of 1972. 5/ It was also assumed that the average work week would change very little and that no change would occur in the distribution of GNP between the private and public sectors. 6/ In contrast, a 3.5 per cent increase in productivity implies a 1.8 million rise in payroll employment. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis -16- estimated as a result of population growth and other demographic and economic factors. But labor force participation rates were relatively depressed in 1971--especially among adult males—due to slack labor market conditions. Consequently, a rebound in participation rates would lead to a somewhat larger than normal labor force growth. In addition, the Defense Department estimates some further reductions in the armed forces which would add an additional 100,000 to the civilian labor f o r c e . In total, the civilian labor force could be expected to grow by about 1-3/4 million 8/ in 1972. Thus, an increase in total employment of 2 million would mean a decline in unemployment of about 250,000. a decline of this magnitude might still leave the unemployment rate somewhat above 5 per cent. Alternative projections through 1973 were also performed using three different growth rate assumptions for real GNP--5, 6, and 7 per cent. (Table 5.) As the table shows, an acceleration in the growth of real GNP by 2 percentage points (from 5 per cent to 7 per cent) is associated with an increase of over 1.0 percentage point in the growth rate of productivity. If there is no change in the work week, Table 5 suggests that payroll employment might increase by 1.2 million under a 5 per cent real GNP assumption and by 1.5 million under a 6 per cent assumption—or by one-quarter more due to the additional real GNP. The rise in employment associated f] In the U.S. budget, the Defense Department estimates armed "" forces strength of 2.4 million in fiscal year 1973. The armed forces totaled over 2.6 million in 1971:4. 8/ The civilian labor force in the first quarter of 1972 was about "" 2 million above its year earlier level. However, it is unlikely to continue to grow at this rate. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis - 17 - with a 7 per cent growth in real GNP might be approximately 1.8 million, a difference of 300,000 compared with a 6 per cent advance in real GNP. The conclusion I reach from the foregoing analysis can be stated succinctly: the substantial growth in output expected for 1972 should also lead to a sizable expansion in employment. However, the significant increase in productivity which is also expected will dampen the growth of jobs, and only a moderate impact can be expected on the level of unemployment. Alternative Approaches to Expanding Employment Given this prospect, I asked myself whether anything could be done to quicken the growth of jobs. In posing the question, I obviously had to keep in mind the conflict that might emerge between additional efforts to stimulate employment and the continuing campaign to reduce inflation. Moreover, the substantial changes that have occurred in the structure of the labor force in recent years (i.e., proportionately more women and teenagers in the labor force) make the task of generating more employment even more difficult. Nevertheless, I am personally convinced that we cannot simply take note of these difficulties; we should also be prepared to pursue alternative approaches which show promise of helping us reconcile our competing objectives. One such approach is to expand the public employment program. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis -18- The Emergency Employment Act of 1971 (EEA) established a two-year manpower program which is designed to provide transitional public service jobs in State or local governments for unemployed and underemployed persons. The Public Employment Program (PEP) is authorized by the EEA to make available up to 90 per cent Federal funding for a variety of temporary jobs to go into effect when the national unemployment rate equals or exceeds 4.5 per cent for a three-month period. A special program for local areas with jobless rates of 6 per cent or above for three consecutive months is also provided. All unemployed and underemployed persons are eligible for the public service employment with priority consideration accorded to Vietnam veterans and low-income groups. Jobs created under this program must be transitional--that is, they must lead to permanent jobs in public service or in private industry, and they must provide prospects for skill development and advancement. The program incorporates safeguards which attempt to insure that funds are used to create new jobs that would not have been otherwise financed by local revenues or existing Federal programs. The Federal funds are apportioned among the states by a formula which takes into account number of unemployed and the severity of unemployment in a given area. States and localities (acting as agents under the program) contribute 10 per cent matching funds. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis - 19 - Fundlng provides a total Federal outlay of $2.25 billion over a two-year period ending June 30, 1973, as follows: General funds are triggered by a national unemployment rate of 4.5 or above. A total of $1.75 billion has been appropriated: $750 million for fiscal 1972 and $1 billion during fiscal 1973. (2) Special funds of $250 million in each of the two years are available for areas with particularly severe unemployment. According to data supplied by the Manpower Administration in the Department of Labor, $981 million (of total Federal funds) 1 have been allocated. Agents shares account for an additional $105 million. In January of this year, there were a total of 134,374 funded job openings approved under the program. In the same month, current enrollment (people actually at work) under the program amounted to 102,858. There have been about 8,000 terminations to date. The Manpower Administration estimates that a total of 145,000 jobs will be created in fiscal 1972 at an average annual salary of $7,200. A study of the first 45,000 people hired under the PEP indicated that 30 per cent were Vietnam veterans, and about 30 per cent were disadvantaged—including 10 per cent who were welfare recipients. Altogether, about 30 per cent of the newly hired were minorities, and about 88 per cent of the jobs were filled by unemployed persons who had never worked for the local government hiring them. A summary of these characteristics is shown in Table 6. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis -20- The jobs created reflect a wide spectrum of public service employment, although State and local agents must give highest priority to creating jobs in areas where the need for additional service is the greatest. The first 101,000 jobs funded were distributed among the principal occupations shown in Table 7. In order to maximize the number of jobs created, 90 per cent of the allocated Federal funds must be used for wages and benefits, so funds for training and supportive services are limited. However, the Manpower Administration estimates that about 16,000 participants will receive some training under the program. In addition, many more may receive training and supportive services through linkages between the PEP and other manpower programs. Given this favorable experience with the PEP, I asked myself what impact on total employment might one expect if the amount of funds being spent on the program were increased significantly. As indicated above, the EEA is currently funded for $750 million in fiscal 1972 and at $1 billion in fiscal 1973--with an extra $500 million for exceptionally depressed areas. To assess the consequences of such an expansion, I again made use of the computer-based econometric techniques available 9 / to us. In this case, the behavior of the economy was simulated— 9/ The simulations were performed with the aid of a modified version of the Social Science Research Council-MIT-PENN quarterly econometric model. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis on the assumption that Federal outlays would be raised by $1 billion above those already anticipated in the budget. This would represent an approximate doubling of the amounts currently available under the EEA. The simulations were made on the assumption that $1.0 billion were added to the existing level of Federal outlays for each of the fiscal years 1972 and 1973. Thus, the task was to estimate the differential effects on the economy of using the funds in several alternative ways. To start the simulations, a Base Projection was made by applying the GNP assumptions in the CEA Report for 1972. The effects of alternative policies could then be calculated by comparing the results of each successive simulation with the Base Projection. Four alternatives were studied (1) a reduction in Federal personal income tax rates; (2) an increase in Federal procurement; (3) an increase in regular Federal grants-in-aid to State and local governments, and (4) an increase in amounts available for the public employment program. Several conclusions stand out in the results. The largest and most rapid, short-run impact on employment would be produced by the PEP. The number of additional jobs provided would approximate 200,000--twice the number generated by each of the other alternatives at the end of one year. The PEP would also have the largest and most rapid impact on unemployment. By following this route, the unemployment rate might be reduced by 0.2 percentage point after one year; an increase in Federal procurement or an Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis -22- expansion in regular grants-in-aid might reduce the unemployment rate by 0.1 percentage point—while a cut in personal income tax rates would probably have little if any effect on unemployment. However, the expansion in the PEP might also eventually exert relatively more upward pressure on prices. The fact that the unemployment rate declines earlier and falls farther under the PEP approach implies that the margin of unused capacity would be shaved earlier in the process. Consequently, as the process continues, available resources would eventually come under pressure. With a given availability of funds, interest rates would tend to rise—thus increasing the cost of investment. If permitted to continue long enough, the rate of growth in real GNP would be moderated. But in the near-term, given the substantial amounts of unused resources that actually exist, the main results of an expansion in the PEP would be an increase in employment and a decline in unemployment. An increase of $1.0 billion in Federal grants-in-aid to States and localities would also produce a sizable rise in GNP ($1.9 billion vs. $2.4 biliion for the PEP). However, the effects on employment would be much less (0.1 million vs. 0.2 million). The observed differences seem to be explained by the fact that-- under the regular Federal grants-in-aid programs—States and localities would probably use the additional funds for less labor-intensive projects than under PEP. As indicated above, the PEP has provisions Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis - 23 - designed to limit this possibility. In practice, however, these limitations might not be completely effective, and the actual outcome mightlie between the two alternatives involving larger Federal grants. Under all of the alternative policies, the deficit in the Federal budget would be increased—but in no case by as much as the rise in Federal outlays. The expansion in the deficit would be about the same ($0.7 billion) for a tax reduction and an increase in the PEP; it would also be roughly the same ($0.4 billion) for a rise in Federal procurement and an increase in regular Federal grants-in-aid. The size of the increase in the deficit related to the tax reduction can be traced to the fairly small rise produced in the GNP as well as to the lower tax rates on a given level of personal income. Under the PEP, the ultimate expenditures by State and local governments would represent primarily wages and salaries paid to lower-income earners. Among these, effective tax rates would probably be lower than those applicable to those who would receive payments under either the regular Federal grants- in-aid programs or through increased Federal procurement of goods and services. From the foregoing analysis, it is clear that an enlargement of the PEP would be a preferable route to travel--if the nation is searching for rapid progress in stimulating employment and making a further reduction in unemployment. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis - 24 - Concluding Observations As I stressed at the outset, we definitely need to foster a substantial increase in productivity. If we fail in that goal, we are unlikely to make much progress in checking inflation in the short-run and in improving our standard of living in the years ahead. On the other hand, the considerable rise in productivity expected this year may dampen the growth of employment. Thus, the major task immediately ahead of us is to assure—as a minimum— that we achieve the rate of real growth projected by CEA for the current year. But--if we want to make a larger dent in the persistently high level of unemployment—the PEP appears to offer a promising approach. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis TABLE 1 TRENDS IN OUTPUT AND PRODUCTIVITY 1948-1971 (Percentage Change) Gross National Productivity in Product Private Nonfarm Productivity in (1958 dollars) Sector Manufacturing 1948 4.5 3.0 5.8 1949 0.1 4.0 3.8 1950 9.6 6.3 6.8 1951 7.9 2.0 2.6 1952 3.1 0.9 0.3 1953 4.5 2.9 5.9 1954 -1.4 2.3 -0.7 1955 7.6 4.4 5.9 1956 1.8 -0.6 -1.1 1957 1.4 2.2 2.1 1958 -1.2 2.5 -0.1 1959 6.4 3.4 5.6 1960 2.5 1.2 1.8 1961 1.9 3.0 2.4 1962 6.6 4.6 5.9 1963 4.0 3.1 4.0 1964 5.5 3.7 4.9 1965 6.3 2.9 4.1 1966 6.5 3.5 1.5 1967 2.6 1.6 0.1 1968 4.7 2.9 4.7 1969 2.6 -0.1 1.3 1970 -0.6 0.7 1.5 1971 2.7 3.4 3.6 1947-50 U 4.7 4.4 5.5 1950-55 4.3 2.5 2.7 1955-60 2.2 1.8 1.6 1960-66 5.1 3.4 3.8 1966-71 2.4 1.7 2.2 1/ Average annual rate of growth (compounded). Sub-periods are inflection points for private nonfarm productivity. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis TABLE 2 BEHAVIOR OF OUTPUT AND PRODUCTIVITY DURING THE RECOVERY PHASE OF BUSINESS CYCLES (Percentage Change) Quarter of Cyclical Trough 1954:3 1958:2 1961:1 1970:4 Changes in GNP (constant 1958 dollars) 1st year after trough 8.6 9.2 7.6 3.5 2.0 2nd year after trough 0.5 4.2 Changes in Output-per-Manhour 1/ 6.0 4.6 4.1 1st year after trough 4.3 2nd year after trough -1.3 0.7 3.5 Changes in manhours 1/ 1st year after trough 5.6 5.6 2.2 1.2 2nd year after trough 1.9 1.0 1.0 1/ Private nonfarm sector. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis TABLE 3 PRODUCTIVITY AND LABOR COSTS IN THE PRIVATE NONFARM SECTOR, 1948-1971 (Percentage Change) Compensation Unit Per Labor Productivity Manhour Cos ts 1948 3.0 9.0 5.8 1949 4.0 2.9 -1.0 1950 6.3 5.5 -0.8 1951 2.0 8.7 6.6 1952 0.9 5.5 4.5 1953 2.9 5.6 2.6 1954 2.3 3.?. 0.9 1955 4.4 3.5 -0.9 1956 -0.6 5.8 6.4 1957 2.2 5.7 3.4 1958 2.5 3.8 1.3 1959 3.4 4.3 0.9 1960 1.2 4.1 2.8 1961 3.0 3.2 0.2 1962 4.6 4.0 -0.5 1963 3.1 3.6 0.5 1964 3.7 4.7 1.0 1965 2.9 3.7 0.8 1966 3.5 6.1 2.5 1967 1.6 5.7 4.0 1968 2.9 7.3 4.3 1969 -0.1 6.9 7.1 1970 0.7 7.0 6.3 1971 3.4 6.9 3.4 1947- 50^ 4.4 5.8 1.3 1950-55 2.5 5.3 2.7 1955-60 1.8 4.7 2.9 1960-66 3.4 4.2 0.7 1966-71 1.7 6.8 5.0 1/ Average annual growth rate (compounded). Sub-periods are inflection points for private non-farm productivity. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis TABLE 4 INTERRELATIONS AMONG OUTPUT, PRODUCTIVITY, AND EMPLOYMENT DURING THE RECOVERY PHASE OF BUSINESS CYCLES Percentage change Percentage change Percentage chanf in productivity in work week in employment Output Percentage change Percentage change Percentage change (percentage change)"" in output in output in output Trough: 1954:3 4.5 140.0 -20.0 17.8 1954:4 9.7 26.8 20.6 50.5 1955:1 15.8 54.4 12.0 29.1 1955:2 8.7 48.3 - 8.0 56.3 1955:3 6.6 28.8 0.0 68.2 Trough: 1958:2 2.7 298.3 -18.5 -166.7 1958:3 11.6 57.8 16.4 22.4 1958:4 10.9 46.8 16.5 3.-J.9 1959:1 7.0 34.3 11.4 52.8 1959:2 12.3 34.1 8.1 54.5 Trough: 1961:1 - 1.3 46.2 61.5 1961:2 10.1 110.9 - 2.0 - 8.9 1961:3 8.2 78.0 - 3.7 24.4 1961:4 8.5 43.5 4.7 50.6 1962:1 6.6 47.0 1.5 48.5 Trough: 1970:4 - 5.6 26.8 17.9 55.4 1971:1 8.8 75.0 1.1 22.7 1971:2 3.7 73.0 32.4 - 8.1 1971:3 1.8 127.8 -50.0 22.2 1971:4 7.5 65.3 10.7 22.7 1/ All percentage changes are from previous quarter at annual rates compounded. Columns 2, 3, and 4 may not add to 100 due to rounding and cross products. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis TABLE 5 ALTERNATIVE OUTPUT, PRODUCTIVITY AND EMPLOYMENT PROJECTIONS FOR 1972 Growth in Real GNP ($ 1958) 5 per cent 6 per cent 7 per cent Change in private nonfarm output per manhour (per cent) 3.3 3.9 4.4 Change in nonfarm payroll employment (millions) 1.2 1.5 1.8 Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis TABLE 6 CHARACTERISTICS OF PERSONS HIRED UNDER THE PUBLIC EMPLOYMENT PROGRAM Characteristics of First Percentage 45,000 Hired Distribution Sex Male 72.0 Female 28.0 21 or younger 11.0 22-44 72.0 45 or older 16.0 Group White 70.0 Nonwhite 30.0 Public assistance recipient 10.0 Education Less than high school 22.0 High school graduate 45.0 Some college 18.0 College graduate 16.0 Vietnam veterans 30.0 Previously employed by agent 12.0 Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis TABLE 7 OCCUPATIONAL DISTRIBUTION OF PERSONS HIRED UNDER THE PUBLIC EMPLOYMENT PROGRAM Occupation Percentage Distribution Public works and transportation 23.0 Education 18.0 Law enforcement 12.0 Health and hospital services 11.0 Parks and recreation 7.0 Social services 6.0 Environment 5.0 Fire protection 3.0 Administrative and miscellaneous 15.0 Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis CHART 1 OUTPUT and PRODUCTIVITY 19U7 19h9 1951 1953 1955 1957 1959 1961 1963 1965 1967 1969 1971 Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis CHART 2 CYCLICAL BEHAVIOR OF UNIT LABOR COSTS, PRIVATE NONFARM SECTOR Per Cent Change from year earlier 0 ^—aarters Before -After Quarters After h 3 2 1 " 1 2 1 2 3 k -5 Qgartif^a Bp,| q^apt^^B ter- -5 Quarters After 2 U~ Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis

Andrew F. Brimmer (1972, April 12). Speech. Speeches, Federal Reserve. https://whenthefedspeaks.com/doc/speech_19720413_brimmer

@misc{wtfs_speech_19720413_brimmer,
  author = {Andrew F. Brimmer},
  title = {Speech},
  year = {1972},
  month = {Apr},
  howpublished = {Speeches, Federal Reserve},
  url = {https://whenthefedspeaks.com/doc/speech_19720413_brimmer},
  note = {Retrieved via When the Fed Speaks corpus}
}