The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
Private Pension Plans and Manpower Policy Bulletin No. 1359 UNITED STATES DEPARTMENT OF LABOR W. Willard Wirtz, Secretary BUREAU OF LABOR STATISTICS Ewan Clague, Commissioner Private Pension Plans and Manpower Policy Bulletin No. 1359 May 1963 UNITED STATES DEPARTMENT OF LABOR W. Willard Wirtz, Secretary BUREAU OF LABOR STATISTICS Ewan Clague, Commissioner For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington 25, D.C. Price 25 cents Preface In March 1962, President John F. Kennedy established an interagency Committee on Corporate Pension Funds and other Private Retirement and Welfare Programs, under the chairmanship of the Secretary of Labor, “ to review legislative and administrative practices relating to these programs.” One of the issues to which the Committee's attention was specifically directed was how these programs “ may contribute more effectively to efficient man power utilization and mobility.” The responsibility for compiling the basic information needed for the Committee's consideration of this issue fell to the Department of Labor. In consultation with the Bureau of Labor Statistics, the Department invited Dr. Hugh Folk, of the University of California, to prepare a report analyzing the interaction of private pension plans and manpower utilization and mobility. Dr. Folk's report was prepared during the summer of 1962. Because of its long-range interest in this subject, the Bureau of Labor Statistics gladly accepted the opportunity to bring Dr. Folk's comprehensive report to the attention of a larger audience. Detailed studies of the patterns of labor mobility and features of pension plans bearing upon mobility, now in progress in the Bureau, promise to add substantially to the data resources available for manpower policy research. m Contents Chapter I. Summary of findings______________________________________________________________________ Labor mobility____________________________________________________________________________________ Employment opportunities of older workers____________________________________________________ Retirement and withdrawal from the labor force_______________________________________________ Chapter II. Labor mobility___________________________________________________________________________ Summary__________________________________________________________________________________________ Introduction_______________________________________________________________________________________ Patterns of labor mobility________________________________________________________________________ Trends in labor mobility_________________________________________________________________________ Effects of pensions on mobility__________________________________________________________________ Pension trends affecting mobility________________________________________________________________ Chapter III. Employment opportunities of older workers__________________________________________ Summary__________________________________________________________________________________________ Introduction_______________________________________________________________________________________ Unemployment of older workers_________________________________________________________________ Characteristics of unemployed older workers-----------------------------------------------------------------------------Discrimination against older workers-----------------------------------------------------------------------------------------Pension and insurance costs and hiring practices_______________________________________________ Effect of pension plans on hires and separations of older workers_____________________________ Chapter IV . Retirement and withdrawal from the labor force_____________________________________ Summary__________________________________________________________________________________________ Aged workers in the labor force---------------------------------------------------------------------------------------------------Relation of pensions to retirement_______________________________________________________________ Involuntary retirement___________________________________________________________________________ Tables: 2.1. 2.2. 2.3. 2.4. 2.5. 2.6. 2.7. 2.8. 2.9. 2.10. 2.11. 2.12. 2.13. 2.14. 2.15. 3.1. 4.1. 4.2. 4.3. 4.4. 4.5. 4.6. Workers who changed jobs as percent of all persons who worked, 1955 and 1961____ Number of jobs left per 100 persons who worked during 1961, by reason forleaving job, age, and sex________________________________________________________________________ Annual turnover rates, by age and sex, seven areas, 1955-------------------------------------------Separations of workers by length of service and age, seven areas, 1955----------------------Separation rates, by age and length of service---------------------------------------------------------------Percent of job shifts in 1961 involving change of major occupation group, by occupation group of job left___________________________________________________________ Percent of job shifts in 1961 involving change of industry group, by industry of job left_______________________________________________________________________________________ Turnover rates in manufacturing, annual averages, 19 4 8 -6 1 __________________________ Number of voluntary and involuntary job shifts per 100 persons who worked, 1955 and 1961__________________________________________________________________________ Reason for leaving job, by occupation: Distribution of jobs left in 1961 and 1955, by major occupation group____________________________________________________________ Hires and separation rates, by industry, age, and pension coverage, 1955, six areas_____________________________________________________________________________________ Annual separation and hire rates, by size of establishment, age, and pension coverage, 1955, six areas_______________________________________________________________ Annual quit rates, by industry, age, and pension coverage, 1955, six areas___________ Annual quit rates, by size of establishment, age, and pension coverage, 1955, six areas_____________________________________________________________________________________ Likelihood of workers of given age and length of service having vested pension rights, 124 large company plans-----------------------------------------------------------------------------------Unemployment rates of men, by age, 19 48 -6 1---------------------------------------------------------------Percent of population in the labor force, by sex and age, 1940 -6 1____________________ Employment by occupation groups of aged men and all men, and changes in employment, 1952 and 1961______________________________________________________________ Employment of aged men, by major occupation groups, 1952 and 1961______________ Percent of aged men with part-time and full-time jobs, 1 9 50 -6 0---------------------------------Normal, compulsory, and automatic retirement ages in 300 pension plans under collective bargaining, late 1958-----------------------------------------------------------------------------------------Normal and compulsory retirement ages in coventional plans, 1956-59 and 1 9 53 -5 5— _ Page 1 1 2 3 4 4 5 6 10 12 16 19 19 19 21 23 24 25 27 29 29 30 32 33 7 7 8 9 9 10 10 11 11 12 13 14 15 15 17 20 31 32 33 33 34 35 y Private Pension Plans and Manpower Policy Chapter I. Summary of Findings This paper reviews the effects of corporate pen sions on various aspects of manpower policy, in cluding labor mobility, the employment opportuni ties of older workers, and retirement and with drawal from the labor force.1 Manpower policy, broadly defined, includes Government programs affecting the size and quality of the labor supply and its allocation among competing demands for labor. Measures which increase or decrease the number of people seeking work, or which introduce or remove ob stacles to the movement of workers between jobs may be considered aspects of the Nation's man power policy. Among the specific parts of the Nation's manpower policy are the operation of the employment service, apprenticeship and training programs, and the requirement for a nondiscrimi nation policy in performing work on Government contracts. The goals of manpower programs in the United States are to aid persons who are willing and able to work in finding jobs, to assist employers in meeting their manpower requirements, to facilitate the free movement of workers between jobs, and to assist workers in gaining the education and skills needed in the modern labor market. Manpower programs thereby contribute to the attainment of high levels of employment and to the adjustment of the labor force to the changing patterns of demand for labor. However, these manpower programs cannot by themselves assure a properly functioning labor market and a high employment economy. Ob viously, they cannot create jobs where the demand for labor is insufficient. Thus, the use of other economic instruments, such as monetary and fiscal policy, may be needed to provide an economic environment in which consumers, investors, and Government create adequate demand for labor. Labor Mobility A growing economy requires changes in the occupational, industrial, and geographical distri bution of labor. In a period of rapid change such as the present when new processes are being intro duced, new centers of industry growing up, and other industries declining, the tasks for manpower policy are great. The remarkable increases in labor productivity in industries for whose products demand is limited, such as farming and coal mining, imply sharp reductions in demand for labor in these industries. At the same time, growing industries such as education and health services require an increasing work force. To meet these changing demands, a certain degree of job shifting is needed, but there is little agreement as to how much constitutes the “ optimum" or “ necessary" level of labor mobility. In recent years, the extent of turnover among factory workers has declined, and some observers believe that present levels are too low. Similarly, the continuing high unemployment rates of workers in some areas and industries of chronic labor surplus are seen by some as evidence of a reluctance of workers to change location or occupation, thus obstructing the achievement of high levels of employment and rapid economic growth. Despite the apparent low mobility of workers in some depressed areas and industries, there is probably no widespread reluctance of workers to change jobs where good jobs are available. The proportion of workers who change jobs during the year falls off with increasing age and with in creasing length of service, but even among workers in their fifties and among workers with 10 years or more of service with a single firm, there is a sizable number of voluntary job shifts. The reasons for i Excerpts from this bulletin appeared in the Monthly Labor Review, March 1963, pp. 285-288. 1 the reluctance of older workers or long-service workers to leave their jobs are obvious. Long familiarity with the work, seniority and job security, and the privileges which come with long service hold workers willingly to their jobs. More over, the older worker usually can expect some difficulty in finding another job. The need for an older worker to decide to leave a job seldom arises because the opportunity to change to another job is seldom presented. It is difficult to determine the effect of pensions as such on mobility. Studies have demonstrated that firms with pensions have lower separation and quit rates than firms without pensions, and this dif ference is often cited as evidence that pensions reduce labor mobility. However, firms with pen sion plans frequently have other characteristics which tend to hold older workers on their jobs. The same firms are also likely to be unionized, to pay higher wages and salaries, and to have more effective job-security provisions based on length of service. These other factors probably account for much of the relatively lower mobility in pension firms. No study of mobility has been conducted in such depth as to isolate the effects of these individ ual factors. The effect of pensions may, however, be significant for certain occupational groups (such as senior white-collar workers and executives), and may also reduce the mobility of older manual workers in some depressed industries and areas. For older workers with greater seniority the effect of pensions on voluntary job changing de pends partly on the extent to which the individ uals pension rights are vested. For longer service workers the existence of nonvested pension rights may be, in many cases, a deterrent to voluntary job changing. To the extent that the individuals pension rights are vested (or some degree of portability is provided in other ways) the immobi lizing effects of pensions are counteracted. The trend in recent years toward the adoption of vesting provisions and the increasing prevalence of multiemployer plans has acted to limit any adverse effects of pensions. In summary, the effects of private pensions on mobility are significant but limited and selective. Effects seem to be most important with regard to senior white-collar workers and older workers in some depressed industries and areas. However, 2 there are other more important factors than pen sions that lie behind the long-term decline of voluntary job changing. While mobility appears to have decreased in recent years, there is still a substantial amount of job changing, and it does not appear that inadequate job changing is now a general problem in terms of the needs of the economy. Employment Opportunities of Older Workers Older workers have long been faced with a lack of job opportunities. While younger workers usually have higher unemployment rates, older workers who become unemployed tend to remain without a job for longer periods of time. It is this fact that gives the older worker problem its special importance. Firms with private pensions often refuse to hire workers who are older than a certain age. The reasons given by employers for not hiring older workers often relate to the presence of a pension plan. Thus employers appear to be reluctant to hire older workers who are too close to retirement age or whose pension costs are higher than the pension costs of younger workers. Under most definite-benefit pension formulas the pension costs of older workers are likely to be somewhat higher than those of younger workers. While the additional costs in many instances will not be large enough to be a significant factor in the hiring decision, they may be given undue weight by employers. The reluctance of pension firms to hire older workers is usually not a direct result of the pension plan itself. Rather, it is likely to reflect the same factors that led to establishment of a pension program—namely, a desire to keep a balanced age distribution of the work force, to promote from within the organization, and to train as new em ployees individuals who have a relatively long future working life. While the difficulties older workers experience in finding jobs are undoubtedly complicated by hiring practices associated with pensions, it cannot be concluded that pension plans are primarily responsible for the older workers’ lack of job opportunities. This is supported by studies showing that in spite of age limitations, in practice firms with pensions appear to hire about as many older workers as they separate and have about the same proportion of older employees as do firms without pensions. Retirement and Withdrawal From the Labor Force Private pensions, as a major supplement to in come from public retirement programs, have served to induce voluntary retirement and with drawal from the labor force. At the same time, it has been asserted that through the practice of involuntary retirement, private pensions have also had the effect of forcing into retirement many who still wish to work and who remain capable of contributing effectively to the national output. The attitude of workers toward involuntary retirement is difficult to determine. On the whole, there appears to be little dissatisfaction with re tirement among those involuntarily retired, or among those older workers for whom involuntary retirement lies in the near future. From many points of view, however, “ automatic” systems of involuntary retirement (which apply without ex ception to all workers reaching a specified age) are less desirable than flexibly administered “ com pulsory” systems which permit exceptions. From the point of view of manpower policy, the desirability of retirement and the resulting re duction in the labor force depends chiefly on the level of demand for labor. In wartime when man power is severely limited, the appropriate man power policy is to induce older workers to remain active workers. With unemployment at more normal levels, retirement serves the interest of manpower policy by permitting many aged workers who are partially disabled or for whom work is especially burdensome to withdraw from the labor force, providing retirement income is sufficient, thereby opening greater opportunities for other employed and unemployed workers. In occu pations and industries in which unemployment is especially severe, retirement reduces the level of unemployment and facilitates the employment adjustments essential in a rapidly changing economy. Changes in retirement programs, both public and private, will have manpower effects by making retirement more or less desirable and thus in fluencing the incentive for aged workers to remain in the labor force. One example of such a change is the recent adoption of early retirement pro visions in both public and private programs. It is yet too soon to evaluate the effect of the recent change permitting early retirement at age 62 under the old-age, survivors, and disability insurance program. The decentralized nature of private pension systems provides an important element of flexi bility in meeting manpower problems. The OASDI system does provide flexibility in the re gard to the age at which an individual may retire, but it does not include any special provisions applying to the needs of particular industries. These would be difficult to justify in the public system, in which uniformity of retirement age and benefit schedules for all covered workers is a well established principle. On the other hand, private pensions are well suited to meet such problems, since the content of the plans may be varied to meet the special needs of an industry or firm. Illustrative of the flexibility inherent in pensions negotiated through decentralized collective bar gaining are provisions for early retirement at em ployer’s request included in certain recent bar gaining agreements which provide twice the normal private pension from the time the employee re tires until he becomes eligible for OASDI benefits. 3 673173— 63------- 2 Chapter II. Labor Mobility Summary Labor mobility in the United States appears to have decreased in recent years, although it is still considerable. The causes of the decline are by no means certain, and it is impossible to say whether the forces leading to the decline are growing in importance. It is true, however, that some countries exhibit higher mobility and others have lower mobility than the United States. The Soviet Union, for instance, has been plagued by high turnover, and Soviet labor law places heavy penalties on workers who quit without adequate notice. Other countries, such as Japan, the Federal Republic of Germany, and Sweden, typi cally exhibit low turnover. In Japan, the “life time job ” is a well-established principle, and workers are very closely bound to the employer. In Sweden and Germany efforts have been made to increase labor mobility. Yet all of these countries have recently exhibited high rates of economic growth, whether faced with high or low turnover. This suggests that labor mobility is only one of the factors contributing to growth and development but not the primary factor, and that other ad justments can adequately offset either high or low labor mobility. Movement of workers from one job to another is but one of the adjustments which are needed in a growing economy. Judging by the evidence for factory workers, mobility in the United States has decreased, but it does not now appear that in adequate mobility is a general problem of the economy, although in certain occupations and industries more outward movement is desirable. Movement out of these industries and occupations is probably not impeded by unwillingness to change to other jobs, but rather by the shortage of other jobs to which workers might change and by the educational and training qualifications necessary for the jobs which are available. If mobility is generally adequate, then the prevalence of pension plans, which in some instances impede mobility, cannot be criticized from the point of view of economic policy for reducing mobility below de sirable levels. The frequency of job changing commonly is inversely related to age, although length of service, 4 rather than age, may be the dominant influence in the lower mobility of older workers. Between twothirds and three-fourths of the workers in all age groups who were separated from a sample of firms studied in 1955 had less than 1 year of service. Apparently most of the job changes are performed by the small proportion of workers who are highly mobile. The separation rates of workers in all age groups fall off with increasing length of service, and there are only small differences in the rates of workers all age groups younger then 65 years in the same length of service classes. Occupational attachment varies widely among occupations, and workers in those occupations which require formal education or lengthy training usually show a high degree of occupational attach ment although they may change jobs frequently or move to different industries or areas. Industry attachment appears to be somewhat less important than occupational attachment, and about threefifths of the job shifts in 1961 involved changes of industry. Data on mobility trends over time are limited. The quit rate in manufacturing has shown a secular decline on which cyclical variations attributable to fluctuations in the level of unemployment are superimposed. It has been asserted that the down trend is attributable to the growth of seniority rules and fringe benefits (especially pensions). Doubtless these factors have been important, but the stability of manufacturing employment and somewhat higher levels of unemployment are probably more important influences in recent years. The available evidence linking the presence of a pension plan with job mobility demonstrates that firms with pensions have lower separation rates and quit rates than firms without pensions. Data from a sample of firms in six major labor market areas give this result even when pension and non pension firms are grouped by major industry division or by size of firm. The difference in separation and quit rates between pension and nonpension firms is especially marked in the 55-64year age group. Lower separation rates in firms with pensions do not prove that pensions are the cause of the lower turnover. Pension coverage is more common in firms with higher than average wages and in unionized firms, and these firms are likely to have lower turnover regardless of pension status. Firms with pensions also have lower accession rates than firms without pensions. Since most of the workers in all age brackets who are separated have less than 1 year of service, the lower accession rates mean that proportionately fewer of the shortservice workers who exhibit high mobility are found in firms with pensions. Finally, separation rates and quit rates are generally lower for workers of all ages in firms with pensions than in firms without pensions. Since the effect of pensions on the mobility of young workers should be small, the lower separation and quit rates of these workers in firms with pensions suggests that factors other than pensions (and seniority) are important in reducing turnover. Three important trends have mitigated the effect of pensions in reducing mobility. First, vesting has become much more common both in unilateral and in negotiated pensions. Although 10 years or more of service is commonly required for vesting, vesting even on such restrictive terms can do much to reduce the immobilizing effects on the groups of workers for whom pensions are likely to have the most influence. A second important trend is the growing frequency of early retirement provisions. Early retirement serves as a limited kind of vesting for workers who leave their jobs a few years before retirement age. Finally, multi employer plans have become more common. Such plans provide portability of pension credits to workers who shift to other employers who are members of the plan. Despite the low level of industry attachment observed in national data on job mobility, industry attachment is probably much stronger in many of the industries covered by multiemployer plans. 2 The most comprehensive review of labor mobility studies is Herbert S. Pames, Research on Labor Mobility (Social Science Research Council, New York, 1954, Bulletin 65). For studies after this date, see Herbert S. Pames, “ The Labor Force and Labor Markets” in Herbert G. Heneman, Jr., and others (editors), Employment Relations Research (IRRA Publication 23), New York, Harper, 1960. 3 Labor mobility is a characteristic of workers. Hence, most mobility data is based on interviews with workers. Turnover is a characteristic of firms. Most of the available data bearing on the relation of pensions to job changing is establishment data. Nationwide estimates of labor mobility are derived from 1955 and 1961 studies of job mobility of workers conducted by the Census Bureau as part of the Current Population Survey. Most of the relevant turnover data is derived from unpublished tabulations from the Bureau of Employment Security’s study of older worker problems in 1955. Introduction The study of labor mobility has many aspects. In its broadest context, labor mobility may in clude entry and withdrawal from the labor force; social mobility, in the sense of intergeneration changes in social status; and job changing among industries, occupations, and geographical areas.2 The analysis of labor mobility here is limited to the analysis of job changing particularly with re spect to differences in prevalence of job changing among workers of different ages, industries, and occupations.3 This study is not concerned with the equities or inequities of unvested pensions although much concern has centered on this problem. Labor mobility has implications for personal freedom. The worker who is unable to get another job has less choice in directing his life. The worker who refrains from quitting his job because of accumula ted unvested pension benefits which he would for feit if he left in some sense may be deprived of liberty. Pension benefits are compensation and some observers feel that it is unfair to withhold benefits or make receipt of benefits conditional upon completion of lengthy service. On the other hand, voluntary movement is not prohibited by pensions and seniority. Viewed as one of a num ber of institutional practices which make job changing costly to workers, pensions do not appear to be unique in their effects. Labor mobility also has important implications for manpower policy and this is the principal con cern of this study. Changes in the structure of demand and in the techniques of production call for increased employment in some industries, oc cupations, and areas and reductions elsewhere. A highly immobile labor force does not adjust rapidly to changing demand for labor. Thus, in adequate mobility may impede the attainment of high levels of employment. Yet mobility may also be excessive. Too frequent job changes are likely to prevent the development of a steady, trained work force, and result in low labor productivity. Labor mobility is but one of the ways in which labor is allocated to changing demands. Capital mobility is important, and there is some tendency for firms to move into areas of high unemployment or low wages. Firms in declining industries may also shift to new products. These movements decrease the amount of labor mobility required. 5 Wage flexibility is another way in which the supply of labor is allocated among competing demands for labor. Firms with ample labor supply generally do not have to increase their wages as much as they increase overtime as do firms which must attract additional labor. Local unemployment problems may reflect inadequate capital mobility and price inflexibility as much as they do inade quate labor mobility. Since many workers are reluctant to move from the localities in which they are situated, capital mobility and wage flexibility may meet the wants of the unemployed far better than movement to a new area.4* The existence of areas of persistent unemploy ment or occupations in which unemployment is usually high does not necessarily imply that labor mobility is inadequate. Workers may be willing to change jobs, but the opportunity to change may be lacking. Opportunity depends on the avail ability of jobs, and availability depends in large part on aggregate demand. In recent years there have been no labor market areas classified as “ areas of tight labor supply,” in contrast to the periods of World War II and the Korean conflict when there was heavy demand for labor in most areas. Only if there were occupations and areas with persistent labor shortages while elsewhere there were persistent labor surpluses could it be concluded that labor mobility was inadequate. The occupations which show serious shortages (as shown in the public employment service job orders in interstate clearance) are typically jobs requiring a considerable formal education and training or lengthy periods of apprenticeship. Patterns of Labor Mobility In 1961, about 8.1 million workers changed jobs, and about 5 million of these were younger than 35 years old. Approximately 11 million job shifts were made by these job changers. An additional 2.3 million jobs were left by workers who did not start a new job during the year. Overall, about one-tenth of the workers who worked during the year changed jobs.6 Mobility commonly differs among industries and occupations. Farm laborers are usually quite mo bile, but farmers infrequently change jobs. Con struction workers frequently work for several employers during the course of the year, but usu ally retain considerable attachment to the indus 6 try over a period of years, at least during the seasons in which construction work is at high levels. These obvious differences in mobility are sometimes concealed in studies of job changing over periods as brief as 1 year. Similarly, the number of industrial and occupational changes involved in the total number of job shifts is under stated by the use of broad occupational and industry classifications. The volume of job changing in a given year also reflects economic conditions. During periods of very high labor demand, few workers are laid off but many workers quit voluntarily for better opportunities elsewhere. Thus, the patterns of labor mobility to some extent reflect general economic conditions as well as propensities of workers with varying personal characteristics and in various industries and occupations to change jobs. Characteristics related to mobility such as sex, seniority provisions, and region are not treated separately in this section, but obviously they have a bearing on mobility in general. Age. Youth is the characteristic most highly associated with labor mobility. To many younger workers, job changing is a vital part of choosing a lifetime occupation. It has often been pointed out that workers find it difficult to judge in advance whether or not a job is worth taking. In many instances, the younger worker finds a job worth keeping after trying several jobs, and thereafter he does not move voluntarily. The younger worker loses little from frequent job changes. With little or no seniority or pension benefits to lose, the younger worker is not inhibited from changing jobs until he finds one that suits him well. After age 25, however, the frequency of job changing falls off rapidly. About one- <Implicit in such legislation as the Area Redevelopment Act of 1961 and the Manpower Training and Development Act of 1962 is the idea that in dustry should be attracted to areas with unemployment problems and that workers should be retrained to staff industries in which jobs are available, rather than that workers should move to areas where their skills are in de mand. The provision of adjustment assistance to firms injured by foreign competition in the Trade Expansion Act of 1962 is directed toward assisting these firms to remain in business producing other goods. The appropriate role of wage flexibility in the sense of slower rates of wage increase in in dustries with considerable unemployment is suggested in the Economic Report of the President Transmitted to the Congress, January l , 1962. Together With the Annual Report of the Council of Economic Advisers, Washington, 1962, p. 189. 5 Preliminary data from a forthcoming Bureau of Labor Statistics study of job mobility of workers in 1961. T a ble 2.1. W o r k e r s W ho C h a n g ed J obs of A ll P e r so n s W ho W o r k e d , 1955 P ercent 1961 i as and 1961 1955 Age M a le F e m a le M a le F e m a le ------------ 1 1 .0 8 .6 12.5 8 .7 14-17 y e a r s __________________________ 18-19 y e a r s ____________________ ___ 20 -2 4 y e a r s _____ _____________ 25-3 4 y e a r s ______ __ _______ 35-4 4 y e a r s ________ _ - ___ _ 45-54 y e a r s ____ _____ __________ 55-6 4 y e a r s ________________ _______ _______ _ __ 65 y e a r s a n d o v e r 8 .9 2 3 .5 2 4 .4 1 4 .9 1 0 .2 7 .1 4 .0 3 .4 5 .8 2 2 .2 16 .3 10 .6 7 .2 5 .2 4 .1 1 .9 12.9 2 7 .4 2 7 .8 1 6 .0 1 0 .2 1 0 .0 5 .6 3 .4 10 .8 2 0 .8 1 4 .9 9 .3 7 .8 5 .4 4 .3 1 .9 T o t a l , 14 y e a r s a n d o v e r * Includes workers who changed jobs during 1961 and who never held 2 jobs at the same time. Sourcf : Bureau of the Census, Current Population Reports, Labor Force Series P-50, No. 70, February 1957, and preliminary data from a forthcoming BLS study of job mobility of workers in 1961. fourth of the men 20 to 24 years old who worked during 1961 changed jobs, but only 1 out of 15 of the men 45 to 54 years old who worked changed jobs, and only one-thirtieth of the men 65 years and older who worked changed jobs during the year (table 2.1). The number of jobs left voluntarily by workers also decreases with increasing age. About onethird of the jobs left by men 18 to 24 years old were left voluntarily (improvement in status), while about one-fourth of jobs left were left volun tarily by men 45 to 64 years old. The total num ber of jobs left per 100 men who worked at some time during 1961 falls off rapidly with increasing age regardless of the reason for leaving (table 2.2). Nevertheless, the proportion of jobs left for in voluntary reasons is higher for older men. The Seven City Study conducted in 1955-56 by the Bureau of Employment Security in coop eration with State employment security agencies provides additional data on voluntary job leaving (table 2.3). Quit rates for each of the seven cities showed similar patterns, but the variation in quit rates between cities was substantial. Thus, it appears that local labor market charac teristics have major effects on quit rates. In the rapidly expanding labor markets of Los Angeles, Seattle, and Miami, quit rates of workers 45-64 years old were about twice the quit rates in the other cities. Hire and separation rates are approximately equal for workers 45 to 64 years old, hire rates for younger workers are higher than separation rates, and hire rates for workers 65 years and older are lower than separation rates. The quit rates decline with age much more than do the layoff rates. The separation rates for workers 65 and over are one and one-half times larger than the separation rates for workers 45 to 64, although quit rates and layoff rates for aged workers are less than one-third larger than the corresponding rates of workers 45 to 64. This difference is due to retirement and death. New hires are about four-fifths or five-sixths of the total hires for all age and sex groups, so that rehires cannot be more than one-half of the layoffs, even though the year studied was a period of recovery from recession. Length of Service. One of the most firmly estab lished findings in labor mobility research is that the probability of a worker leaving a job falls off sharply with increasing length of service. This is true both of voluntary and involuntary job leaving. Voluntary job leaving falls off with increasing length of service because the long-service employee has become well established in the firm, pre sumably is satisfied with his job, and has gained through accumulated seniority a greater degree of job security as well as other benefits, such as pension credits, longer vacations, and in many cases higher rates of pay. In some firms, workers with high seniority gain considerable prestige and status because of long service. This includes not only “ 25 Year Clubs/7 and service pins, but often larger bonuses, better parking facilities, and free dom from punching the timeclock. Voluntary mobility of long-service employees is impeded by T a b l e 2.2. N u m b e r of J obs L eft p e r 100 P e r s o n s W ho W o r k e d D u r in g 1961, b y R e a so n f o r L e a v in g J o b , A ge, an d Se x Reason for leaving job Age and sex Improve Termi ment in nation of Other tempo reasons status rary job Total Eco nomic Total, 14 years and over____ 16.4 5.3 5.3 2.1 3.7 Male, 14 years and over........ 14-17 years____________ 18-24 years..................... 25-44 years____________ 45-64 years____________ 65 years and over______ 18.1 15.2 42.8 19.6 8.9 6.7 6.8 2.8 13.0 7.6 4.6 2.8 6.1 3.0 14.0 7.9 2.1 .9 2.0 4.7 5.3 1.5 .9 1.4 3.2 4.7 10.4 2.6 1.3 1.7 Female, 14 years and over... 14-17 years____________ 18-24 years____________ 25-44 years_____ _______ 45-64 years____________ 65 years and over______ 13.7 9.9 30.8 13.3 7.2 3.9 2.8 1.4 5.5 2.8 2.1 .7 4.1 1.4 9.4 4.2 2.3 .1 2.3 3.2 5.4 1.8 1.3 2.2 4.4 3.9 10.6 4.5 1.6 .9 Source: Preliminary data from a forthcoming BLS study of job mobility of workers in 1961. 7 the worker’s recognition that these privileges will be lost and, in a new job, these accompani ments of long service will have to be earned again. The long-service employee is usually an older worker, so that his opportunity to find a new job is often restricted by maximum hiring ages. Thus the attractions of the old job in the form of security and privileges are made even more im portant by recognition of the difficulty of finding a new job and by the realization that once em ployed, the worker will have little security in the new job. It seems likely that many workers would not consider substantially higher pay as sufficient compensation for the security and privileges lost by changing jobs. Involuntary job mobility is also less likely among long-service employees. The protection against layoff provided by formal seniority rules T a b l e 2.3. A n n u a l T u r n o v e r R a t e s , b y A g e a n d S e x , S e v e n A r e a s , 19551 Number per 100 employed Men Women Age Separations Hires Separations Hires Total New Total Quits Total New Total Quits Total.......................... . Under 45 years___ 45-64 years............ 65 years and over.. 60 70 39 36 49 60 30 27 54 61 38 57 21 26 10 13 61 72 35 26 53 63 28 21 54 64 32 48 30 37 14 18 Detroit......................... Under 45 years___ 45-64 years............. 65 years and over.. 40 52 18 12 34 46 14 7 33 41 18 50 17 23 6 13 50 60 20 20 45 55 16 17 37 43 18 40 21 25 9 13 Los Angeles................. Under 45 years----45-64 years........... . 65 years and over.. 84 93 65 71 71 81 52 52 79 84 69 82 28 33 17 19 77 86 52 45 69 79 45 33 74 83 48 79 39 47 19 31 M ia m i...................... . Under 45 years___ 45-64 years_______ 65 years and over.. 112 117 86 91 76 82 50 51 108 114 82 102 43 48 25 16 123 127 96 88 87 92 63 51 137 118 87 66 62 70 38 16 Minneapolis-St. Paul. Under 45 years___ 45-64 years............. 65 years and o v e r- 54 59 30 27 47 52 25 23 47 50 28 47 19 23 7 10 52 63 26 27 46 56 20 14 46 54 21 39 30 36 11 15 Philadelphia............... Under 45 years___ 45-64 years........... 65 years and over.. 25 30 10 8 18 22 7 5 20 23 8 22 7 10 3 3 35 45 14 9 27 36 11 5 26 35 8 29 15 21 4 6 Seattle.......................... Under 45 years___ 45-64 years............ 65 years and over.. 98 101 86 79 74 75 65 60 89 89 81 106 28 34 15 15 82 91 46 57 60 68 30 24 76 84 43 68 43 52 20 12 Worcester..................... Under 45 years___ 45-64 y ears.......... 65 years and over.. 44 50 21 20 34 40 14 11 34 38 15 32 15 18 6 6 50 63 26 20 41 53 21 3 45 58 22 22 29 37 12 7 112 months ending with March for Los Angeles, December for Seattle, and June for other areas. Data for establishments with 8 or more workers covered by State unemployment insurance laws or by the Railroad Retirement Act. Source : Bureau of Employment Security, Older Worker Adjustment to Labor Market Practices, (BES Bulletin R151, 1956), pp. 265-266. 8 in unionized firms and informal length-of-service considerations in nonunion firms form one of the principal attractions of the long-service worker’s present job. They also reduce the chance that the long-service worker will be laid off or dis charged and thereby forced to change his job. Once laid off, however, long service in a particular occupation or industry is unlikely to be cast aside lightly, and most displaced long-service em ployees will probably be reluctant to change occupations or industries, particularly if the change entails accepting considerably lower pay. Long service may also be associated with extreme specialization of work experience, so that the worker is unable to earn as much outside of his firm, industry, or occupation. Such specialization is likely to be a significant cause of the decreasing probability of job change with increasing length of service, especially among managerial and clerical employees whose knowledge of the routine and system in a given firm makes them highly pro ductive, but whose knowledge may not be easily transferable to other firms. The immobility of long-service employees may be the result of factors which predispose these workers in this direction. The low mobility of long-service employees may simply reflect the fact that they tended toward immobility to begin with. It is known, for instance, that some workers never settle down to a steady job, but remain highly mobile throughout their working lives. The probability that a worker will change jobs in a given year is not independent of his previous job changing experience. In 1951, for instance, 18 percent of a sample of workers covered by OASDI in both 1951 and 1952 changed employers in both years and these highly mobile workers were 51 percent of the workers who changed employers in 1952.6 Similar results are found in the Seven City Study. About three-fourths of the separa tions of workers under 25 years old were workers with less than 1 year of service. The ratio decreased slightly with age, but two-thirds of the separated workers 55 to 64 years had less than 1 year of service. Overall, about seven-tenths of the workers separated in 1955 by the firms in cluded in the sample had less than 1 year of 6 Bureau of Old-Age and Survivors Insurance, Division of Program Analy sis, “ Incidence of Employer Change,” Analytical Note No. 80, April 18, 1956. T able 2.4. S e p a r a t io n s of W o r k e r s b y L e n g t h S e r v ic e a n d A g e , S e v e n A r e a s , 1955 of [Percent] Length of .service (in years) Age T ota l L ess th a n 1 1 -4 5 -9 10-14 15 a n d over A l l a g e s................................................ 1 00 .0 6 9 .8 2 2 .1 4 .6 L6 1 .8 U n d e r 25 y e a r s _______________ 2 5 -3 4 y e a r s ............... ........ ................. 3 5 -4 4 y e a r s ......................................... 4 5 -5 4 y e a r s ________________ _____ 5 5 -6 4 y e a r s ________________ _____ 65 y e a r s a n d o v e r ............................ 100 .0 1 0 0 .0 100 .0 1 00 .0 100 .0 100 .0 7 5 .9 6 9 .4 6 9 .5 6 9 .7 6 5 .2 4 0 .5 2 2 .9 2 4 .3 2 2 .2 1 9 .2 1 8 .6 1 3 .4 1 .2 5 .3 5 .4 5 .9 7 .1 9 .2 .9 1 .8 2 .1 3 .7 1 1 .9 .1 1 .1 3 .0 5 .4 2 5 .0 Source : Derived from unpublished tabulations furnished by the Bureau of Employment Security. service (table 2.4). It may be concluded from these data that most of the workers of all age groups who leave jobs have short service in their last jobs.7 Information on separation rates classified both by length of service and by age (table 2.5) suggests that length of service is the most important factor associated with the lower mobility of older longer service workers. The very high separation rates of workers 65 years and older in all length-of-service categories is attributable primarily to retirement. The separation rates for workers with less than 1 year of service include a large proportion of workers in industries where employment is casual or seasonal, and these rates are naturally very high. For these workers, there is a slight tendency for separation rates to increase with increasing age and to reach a peak in the 45- to 54-year age group. In the 1- to 4-year length-of-service group, the differences in separation rates associated with age are small (except for those under 25 years old, or 65 years and older), and the same observation applies to the longer length-of-service groups. Certainly, the figures demonstrate a far more definite pat tern of declining separation rates varying with length of service (holding age constant) than with age (holding length of service constant).8 Occupation. Most jobseekers confine their search to local labor markets. This geographical limita tion of job search often requires that the job changer change either his industry or his oc cupation or both if he is to find a suitable job. Occupational attachment naturally varies with the amount of specialized training required in the occupation. Occupational mobility of ex perienced workers into the licensed professions or unionized crafts is likely to be small, and a large proportion of the new recruits for such oc cupations will come from schools and on-the-job training programs rather than from other oc cupations. Job changers in such occupations are likely to restrict themselves to changes within the occupational group. Workers highly at tached to their occupations may exhibit high geographical mobility, however, as is evidenced by the high frequency of such changes in a sample of scientists studied by the Bureau of Labor Statistics.9 Mobility differs among occupations (table 2.6). Only 9 job shifts were completed per 100 workers in professional and technical occupations, but there were 16 job shifts per 100 sales workers, and 27 job shifts per 100 laborers. The rates of job leaving by workers among managers, officials, and proprietors and farmers were quite low. In part, this reflects the high average ages of people in these occupation groups. Almost half the job shifts involved a change to a different occupation. This varied considerably according to occupation. Shifts by managers, officials and proprietors, laborers and sales workers were more than likely to involve a different occu pation while professional workers and craftsmen were less likely to shift to a different occupation. T a b l e 2.5. S e p a r a t io n R a t e s , b y A ge S e r v ic e L ength of [Separations per 100 employees] Length of service (in years) Age 7 Because the type of separation is not available, it cannot be concluded that all or most of this job changing is voluntary. The decrease in the fre quency of voluntary reasons for job changing with increasing age in the survey cited in table 2.2 suggests that a smaller proportion of the separations of older workers with short service is voluntary. s In a similar analysis, Lloyd George Reynolds observed that voluntary separation rates of workers decreased slightly with increasing age when length of service was held constant. The Structure of Labor Markets (New York, Harper, 1951), p. 21. 9 Bureau of Labor Statistics, Occupational Mobility of Scientists (BLS Bulletin 1121,1953), p. 4. and Total Less thanl All ages................ ................... 54 150 33 11 8 7 Under 25 years_____________ 25-34 years........ ...................... 35-44 years.................. ............ 45-54 years............................ . 55-64 years................ .............. 65 years and over..................... 95 60 47 39 31 57 134 146 150 173 160 225 48 31 28 25 27 39 24 13 9 10 10 27 8 7 5 7 34 7 4 4 4 36 1-4 5-9 10-14 15 and over S ource : Derived from unpublished tabulations furnished by the Bureau of Employment Security. 9 T a b l e 2.6. P e r c e n t of J ob S hifts in 1961 I n v o l v in g C h a n g e of M a jo r O ccu p a tio n G r o u p , b y O c c u p a tio n G r o u p of J ob L eft Job shifts in 1961 Major occupation group of job left All occupations............... ........................... Professional, technical, and kindred workers_____ ____ _____ ____ _________ Farmers and farm managers______ ______ Managers, officials, and proprietors, ex cept farm ................................................ Clerical and kindred workers.................... Sales workers______________ ______ ____ Craftsmen, foremen, and kindred workers. Operatives and kindred workers................ Private household workers_____________ Service workers, except private householdFarm laborers and foremen_____________ Laborers, except farm and mine------------- Percent of job Num Percent of shifts to ber persons different (thou with work occupation sands) experience 10,869 13.5 811 50 8.9 1.8 436 1,457 859 1,678 2,238 181 1,302 645 1,212 5.5 12.2 15.6 18.1 16.6 5.3 16.2 15.0 27.3 47.2 30.8 (9 65.6 39.1 59.1 32.7 47.5 93.9 49.2 52.4 58.4 i Figures not shown where base is less than 100,000. S ource : Preliminary data from a forthcoming BLS study of job mobility of workers in 1961. Industry. In contrast to occupational moves, movement of workers between industries is seldom the result of career development. Job changes resulting from layoff often lead to industry changes because of the worker’s attachment to a locality. Shifts from one industry to a different industry are more common than occupational shifts. About three-fifths of the job shifts in all industries in 1961 were to jobs in a new industry (table 2.7), while only one-half of the job shifts involved changes in major occupation group. Quit rates for a more detailed classification within manufacturing industries vary inversely with gross hourly earnings. Industries charac terized by relatively high earnings generally have low quit rates. Industries with relatively small proportions of production workers and relatively small proportions of female employees also have somewhat lower quit rates on the average.10 Trends in Labor Mobility Recent changes in mobility have been con ditioned by changes in economic conditions. Occupational mobility during the depression decade of the 1930’s was markedly lower than during the prosperous decade of the 1940’s. Manufacturing quit rates were low during the 1930’s but high in the prosperous 1920’s and 1940’s. Judging by these rates, there has been a 10 For an analysis of these relationships in 1956, see Robert M. Shaw, “ The Nature of Industries with High and Low Quit Rates,” Employment and Earnings, September 1957. 10 long-term down-trend in mobility, but it is subject to reversal during periods of marked prosperity. Most of the discussion of trends in labor mobility has of necessity centered on trends in manu facturing quit rates, which have heretofore been the only statistical series of comparable observa tions. Evaluation of trends in manufacturing quit rates must include consideration of the level of unemployment. Even when this factor is taken into account, however, it appears that quit rates during the relatively prosperous years of 1951-53 and 1955-57 did not approximate the very high quit rates of the World War II period. The downtrend in the quit rate has been explained in terms of the following influences : 1. Growth of unions 2. Development of seniority provisions 3. Development of fringe benefits (especially pensions) 4. Government and supplementary unemployment benefits 5. Growth of large corporations 6. Aging of the labor force 7. Stability of manufacturing employment T a b l e 2.7. P e r c e n t of J ob S hifts in 1961 I n v o l v in g C h a n g e of I n d u st r y G r o u p , b y I n d u st r y of J ob L eft Job shifts in 1961 Industry group of job left Num ber Percent of Percent of persons job shifts with work to different experience industry Total............................ ............................... 10,869 13.5 61.7 793 708 10.6 57.2 52.5 Agriculture------- ------------------------------Wage and salary workers------ ----------Self-employed and unpaid family workers............. _............._................ 25.5 85 1.8 Nonagricultural industries_____________ 10,076 9,729 Wage and salary workers___________ 159 Forestry, fisheries, and mining__ Construction___________________ 1,555 Manufacturing_________________ 2,222 1,249 Durable goods_____________ 973 Nondurable goods__________ Transportation, and public 451 utilities........ ............................... Trade_____ ____________________ 2,512 382 Wholesale__________________ 2,130 Retail____ _________________ Service industries_______________ 2,544 Finance, insurance, and real 391 estate___________ ____ ____ 343 Business and repair services.. Private households2....... ........ 248 Personal services, except pri 433 vate households................... 356 Educational services............... 773 Other services...................... 286 Public administration................... 13.8 15.1 20.4 38.0 Self-employed and unpaid family workers........................ — ................ 347 12.2 12.4 11.8 10.0 (9 62.1 61.1 62.9 41.0 61.7 60.0 64.0 19.3 15.5 20.1 12.6 69.4 60.1 84.8 55.7 70.2 12.7 23.3 6.3 70.3 83.1 2 90.7 20.2 66.0 4.2 90.5 8.7 14.4 7.7 62.4 63.8 78.3 1 Figures not shown where base is less than 100,000. 2 Domestic service work in private households for several different em ployers is not counted among the job shifts. Source: Preliminary data from a forthcoming BLS study of job mobility of workers in 1961. T a b l e 2.8. T u r n o v e r R a t es in M a n u f a c t u r in g , A n n u a l A v e r a g e s , 1948-61 Year Turnover rates per 100 employees in manufacturing 1 Separa tions 5.4 5.0 4.1 5.3 4.9 5.1 4.1 3.9 4.2 4.2 4.1 4.1 4.3 4.0 1948____ ____ ________ 1949.............................. 1950____ ____________ 1951...................- ........ 1952....... ...................... 1953....... - ................... 1954____ ____________ 1955_________________ 1956_________________ 1957__________ ______ 1958_________________ 1959____________ ____ 1960_________________ 1961____________ ____ Quits Accessions 3.4 1.9 2.3 2.9 2.8 2.8 1.4 1.9 1.9 1.6 1.1 1.5 1.3 1.2 5.4 4.3 5.3 5.3 5.4 4.8 3.6 4.5 4.2 3.6 3.6 4.2 3.8 4.1 New hires (>) (») (*) 4.1 4.1 3.6 1.9 3.0 2.8 2.2 1.7 2.6 2.2 2.2 Unemploy ment rate * 3.8 5.9 5.3 3.3 3.1 2.9 5.6 4.4 4.2 4.3 6.8 5.5 5.6 6.7 1 Bureau of Labor Statistics. 2 Percent of civilian labor force unemployed. * Not available before 1951. Each of these factors undoubtedly has played a role in the decline, but no one of the factors explains the downtrend over the entire period for which turnover data is available.11 During the postwar period, with which this analysis is pri marily concerned, it is likely that only the develop ment of fringe benefits, aging of the labor force, and stability of manufacturing employment could have had much influence. Unions, seniority practices, unemployment benefits, and large cor porations were prevalent during both the war and immediate prewar period. During the post war period it can scarcely be said that there has been enough of an increase in prevalence of these factors to have had much influence on the quit rate. The relative stability of employment in manu facturing over the period 1946-50, and over the period 1953-62 has probably played the major role in the downtrend of the quit rate over this period. When accession rates are low, relatively few highly mobile younger workers and shortservice workers enter manufacturing, so that quits of younger workers fall off also, and consequently total quits decrease. These relationships are shown in turnover data for the period 1948-61 (table 2.8). In years of ii Some of the principal articles discussing the downtrend in mobility are Ewan Clague, “ Long-Term Trends in Quit Rates," Employment and Earnings, December 1956; Arthur M. Ross, “ Do We Have a New Indus trial Feudalism," American Economic Review, December 1958; Joseph Shister, “ Labor Mobility: Some Institutional Aspects," Industrial Relations Re search Association, Proceedings, 1950; Paul F. Brissenden, “ Labor Mobility and Employee Benefits," Labor Law Journal, November 1955. high unemployment, such as 1949, 1958, and 1961, both the quit rates and the new hire rates are relatively low; while in prosperous years, such as 1951 and 1955, both quit rates and new hire rates are relatively high. Even so, the quit rate in 1949 (a recession year) is as high as the quit rate in any year after 1953. Apparently the quit rate in manufacturing has shown a down trend in recent years, as has the new hire rate. Comparison of data on labor mobility in 1961 with data relating to 1955 provides some evidence of the nature of changes in mobility patterns in periods of differing economic conditions. The year 1955 was a year of relatively high prosperity (unemployment rate— 4.4 percent) compared with 1961 (unemployment rate— 6.7 percent). As expected, sharp differences are clearly evident in the mobility patterns of the 2 years. In general, voluntary job changing was sharply reduced in the latter year while job shifts for economic reasons were substantially higher. The net effect of these two forces was a drop in over all mobility rates except for women workers 18-34. In the older age groups, the 1961 data show sharper changes for men than for women. The number of voluntary job shifts in 1961 for “ im provement in status” of men 45-64 was only about half the rate of 1955. The decline for women in these age groups was relatively slight (table 2.9). T a b l e 2.9. N u m b e r of V o l u n t a r y a n d I n v o l u n t a r y J ob S hifts p e r 100 P e r so n s W ho W o r k e d , 1955 a n d 1961 Reason for leaving job Age and sex Improvement in status 1961 1955 Economic 1961 1955 Male, 14 years and over............... ............. 14 to 17 years...................................... 18 to 24 years............................ ........... 25 to 44 years....................... ................ 45 to 64 years............................ ....... 65 years and over_______ ___________ 6.1 3.0 14.0 7.9 2.1 .9 7.8 5.7 18.6 9.1 4.1 .7 6.8 2.8 13.0 7.6 4.6 2.8 5.3 2.7 11.1 5.6 4.0 1.5 Female, 14 years and over.................. ...... 14 to 17 years............................ ........... 18 to 24 years.................... ................... 25 to 44 years______________________ 45 to 64 years________ _____________ 65 years and over _______ ____ 4.1 1.4 9.4 4.2 2.3 .1 4.8 5.7 10.5 4.3 2.6 .8 2.8 1.4 5.5 2.8 2.1 .7 2.3 2.3 3.0 2.3 2.1 .3 Source : Bureau of the Census, Current Population Reports, Labor Force Series P-50, No. 70, February 1957, and preliminary data from a forthcoming BLS study of job mobility of workers in 1961. 11 673173— 63 3 T a b l e 2.10. R e a s o n for L e a v in g J o b , by O c c u p a t io n : D ist r ib u t io n O c cu p atio n G r o u p of J obs L eft in 1961 and 1955, by M a jo r Reason for leaving job 1961 1955 Major occupation group of job left Economic Termina Improve ment in tion of tem status porary job Other Economic Termina Improve ment in tion of tem status porary job Other Total______________________________________________________ 32.1 32.6 12.9 22.5 23.5 37.6 17.9 20.9 Professional, technical, and kindred workers__________________ Farmers and farm managers_________________________________ Managers, officials, and proprietors, except farm_______________ 21.7 34.8 16.1 27.4 15.9 18.5 35.8 15.4 12.7 29.2 31.4 33.3 38.5 41.6 38.3 51.8 37.9 40.0 43.8 40.0 18.8 30.2 33.4 33.9 32.0 12.7 10.4 31.1 23.4 15.2 18.7 27.6 28.8 16.1 15.4 mp.ripa! and kindred workers _ . . . Rales workers_______________________________________________ Craftsmen, foremen, and kindred workers_____________________ Operatives and kindred workers_____________________________ Private household workers___________________________________ Service workers, except private household_____________________ Farm laborers and foremen . __ _ . ... Laborers, except farm and mine______________________________ (9 36.0 16.9 18.4 56.2 38.5 27.0 19.2 13.2 46.9 1 Percent not shown where base is less than 100,000. Changes in the level of economic activity between 1955 and 1961 increased the proportion of job changes due to economic reasons. About 29 percent of craftsmen, foremen, and kindred workers who changed jobs, did so for economic reasons during 1955 compared with 56 percent in 1961. An increase also occurred in the opera tive group— from 31 percent in 1955 to 38 percent in 1961 (table 2.10). The proportion of job changes for economic reasons among nonfarm laborers also increased from 33 percent to 47 percent during this period. The lower level of economic activity in 1961 compared to 1955 prevents any conclusion regard ing a trend in recent years from being more than suggestive. The decline in job changing for reasons of improvement of status and the increase in job changing for economic reasons between 1955 and 1961 is almost certainly attributable to the relatively high level of unemployment in 1961. While the sharp drop in voluntary job changing for men 45 to 64 points up the difficulties which older workers experience in shifting to better jobs, it is impossible to indicate to what extent the in creasing prevalence of pensions may have been responsible. (9 37.3 36.2 47.3 28.7 34.7 31.0 33.6 18.4 23.6 (9 6.9 14.9 14.0 3.9 7.4 11.5 10.8 53.5 9.2 (9 19.8 31.9 20.3 11.2 19.4 30.5 36.4 15.0 20.3 10.1 17.2 10.0 21.0 12.1 15.1 12.0 17.6 9.9 18.6 13.9 52.8 17.3 Source : Bureau of the Census, Current Population Reports, Labor Force Series P-50, No. 70, February 1957, and preliminary data from a forthcoming BLS study of job mobility of workers in 1961. effect attributable to pensions alone, and not to seniority provisions, age composition of the work force, size of firm, wages, or industry is almost impossible because firms with pensions frequently are large firms and often have high wages, strict seniority provisions, and are older firms with relatively large numbers of older workers. The classification of firms into pension and nonpension groups and the comparison of total separation and hire rates therefore throws little light on the in dependent effect of pensions on turnover. Despite the paucity of supporting data, many writers assert that pensions do reduce mobility independently of other influences, although they recognize that pensions are closely related to other labor market institutions which reduce labor mobility. Clark Kerr writes: Private pension plans, except where they provide full and immediate vesting of both the employee’s and firm’s contribution, retard such movement. They tend to tie the worker to the company while employed; and hold him in a company-attached labor pool when unemployed.12 Robert Tilove writes: Effects of Pensions on Mobility . . . the conclusion seems inescapable that most private pension plans, in the form in which they commonly exist today, exercise a restraining influence on labor mobility. They often involve forfeiting accrued pension benefits upon any shift in employment; and they inhibit Pensions are only one of the influences which tie workers more closely to their jobs with increasing age and length of service. Identification of the ^ “ Social and Economic Consequences of the Pension Drive,” Handbook on Pensions (National Industrial Conference Board, Inc., Studies in Person nel Policy No. 103, 1950), p. 85. 12 hiring in the upper ages, either because pension costs are thought to be greater or because the older worker may be reluctant to take a job on which he may not have a sufficient prospect of accruing pension rights.13 2. The expectation of additional pension benefits to be earned in the future may induce workers to continue their present jobs rather than to change to another firm which does not offer pension coverage. The possible effects of pensions on mobility may be summarized briefly: importance to different workers will vary with the char acteristics and attitudes of the workers. ^ T . Rigid maximum hiring ages are often found in firms with pension plans; by reducing the opportunity for older workers to change jobs, such practices may inhibit mobility (this problem is dealt with in chapter III). and S e p a r a t io n R a t e s , by Older workers may be especially reluctant to move from a job with pension coverage to an uncovered job, even if benefits earned by completed service are fully vested. The im mobilizing effects of benefits to be earned in the future do not appear to differ in substance from the immobilizing effects of higher wages or better working conditions, and 11Robert Tilove, Pension Funds and Economic Freedom (Fund for the Republic, New York, 1959), p. 23. T a b l e 2.11. H ir e Pensions are but one component of total compensation, and their therefore should be no more subject to criticism. I n d u st r y , A g e , and P e n s io n C o v e r a g e , 1955,1 S ix A r e a s 2 Annual hires and separations per 100 employees Industry division and pension class A ll ages A ll I ndustries Pension: Hires........... ......................................... ....................................................... Separations_____________________________________________________ N o pension: Hires......... .......... ........................................................................................ Separations.......................................................... ...................... ............. C onstruction Pension: Hires______ __________ __________________________________________ Separations_____________________________________________________ N o pension: Hires................ ............................ ............ ................................ ................. Separations.____ _______ _____________ __________________ ______ M anufacturing Pension: Hires................ ................................ ........................................................... Separations__________________ ____________________________ ____ _ N o pension: Hires________________________________ _______________ __________ Separations........................................................ .......... .............. ............. T ransportation , C ommunication , an d W holesale and R etail T rade Pension: Hires......... ..................... .......... .......... .............. ........................................ Separations_____________________________________________________ N o pension: Hires________________________ _______________________ __________ Separations_____________________________________________________ and 35 to 44 years 45 to 54 years 55 to 64 years 65 years and older 37 34 91 74 42 39 30 29 18 18 12 13 11 50 67 62 115 92 81 74 62 58 53 51 38 37 31 50 97 98 194 173 107 134 108 104 84 65 72 71 48 99 146 136 135 113 186 173 143 136 139 132 144 148 157 170 30 27 83 64 39 34 24 27 15 14 9 10 8 49 53 46 115 81 64 59 47 41 34 30 25 20 14 36 25 25 57 41 29 29 28 29 10 12 5 8 7 70 34 22 32 23 45 41 50 40 20 16 2 4 3 1 60 55 125 106 63 62 49 46 28 28 21 15 8 26 68 65 123 104 86 81 54 56 42 44 34 36 19 38 29 29 52 48 30 31 18 25 12 7 4 6 7 79 96 83 135 120 118 101 91 76 84 70 49 38 34 49 65 62 179 157 70 63 60 61 34 36 41 38 44 68 67 69 111 106 65 59 60 63 78 84 49 52 50 67 R eal E state Pension: Hires............................................................ ........... - ............................... Separations.......... ....................................................................... ........... N o pension: Hires......................... .......... ............ .......... ................................................ Separations................... .......... ................................................................... Services Pension: Hires....... .................................................................................................... Separations................................................................ .............................. . N o pension: Hires......... .................................................................................................. Separations................................................................................................. 1 In establishments with 50 workers or more covered by State unemploy ment insurance laws or by the Railroad Retirement Act. Years ending June 1955, except Los Angeles (March 1955) and Seattle (December 1955). 25 to 34 years U tilities Pension: Hires.................................. ................................................. ....................... Separations______________________________________________ ______ N o pension: Hires___________________________________________________ _______ Separations_____________________________ ■_______ ____ __________ F inance , I nsurance , Under 25 years 2 Detroit, Los Angeles, Minneapolis-St. Paul, Philadelphia, Seattle, and Worcester. Source : Derived from unpublished tabulations furnished by the Bureau of Employment Security. 13 3. Unvested benefits may be a large part of the employee’s wealth, and thereby may make job changing extremely costly to him. The greater the length of service, the larger will the benefits be; hence, unvested pensions tend to reduce the mobility of workers with long service (who are usually older workers) much more than that of shortservice workers. Even multiemployer plans which allow workers to keep their benefits while shifting among firms within the plan may tie employees closely to the industry and union which control the pension plan. These plans seldom include vesting (other than early retirement which is often accompanied by restrictions on other employment) and usually require long periods of covered employment for receipt of benefits. T a b le 2.12. A n n u a l S e p a r a t io n a n d H ir e R a t e s , b y S iz e of E sta b l is h m e n t , A g e , a n d P e n s io n C o v e r a g e , 1955,1 Six A r e a s 2 Workers under 45 years: Pension: Hires....... ...... Separations__ No pension: Hires.............. Separations. __ 55 51 60 53 43 37 44 38 44 40 47 42 93 86 82 72 75 62 51 52 4 80 71 Turnover and Pension Status. There is little data on the effect of pension coverage on labor mobility. Herbert S. Parnes of Ohio State University inter viewed matched samples of workers in plants with and without pension plans and concluded that pensions reduce labor mobility slightly, but that seniority provisions are far more important in reducing mobility.14 The most comprehensive data dealing in the turnover rates for firms with and without pensions comes from the Seven City Study conducted in 1956 by the Bureau of Employment Security. Annual turnover data for 1955 were collected for establishments of 50 or more employees classified by pension plan coverage. Results are shown below: Workers 45-64 years: Pension: Hires_______ Separations... No pension: Hires_______ Separations... 25 25 24 23 16 13 14 15 12 12 16 16 64 66 48 44 30 22 26 32 A nnu al turnover rates by pension plan coverage and age o f worker (per 100 employed), six areas Pension plan Age All ages______ Under 45 y e a r s .___ 45 years and o v e r ._ 4 5 -6 4 years____ 1 No pension plan Hires Separa tions Hires 37 47 16 16 34 42 18 16 67 80 45 47 Separa tions 62 71 46 46 1 Bureau of Employment Security, Older Worker Adjustment to Labor Market Practices, op. cit., p. 66. Firms with pension plans had lower turnover than other firms and total employment in these firms increased by 3 percent, compared to 5 percent in the nonpension firms. Separation rates for workers 45 years and older were higher than hiring rates in both groups of firms, but the excess was accounted for by the more frequent separa tions of workers 65 years and older in both groups of firms. A similar pattern of lower turnover rates in firms with pensions persists when firms are classi fied by major industry division (table 2.11). 14 Hires and separations per 100 employees Pension class and age Number of employees in establishments 5099 100499 500999 1,0004,999 5,000 and over Total 47 46 1 See footnote 1, table 2.11. 2 See footnote 2, table 2.11. Source : Derived from unpublished tabulations furnished by the Bureau of Employment Security. Firms with pensions had lower turnover in most age brackets in all industry divisions. Hiring and separation rates are nearly equal in size for each age, industry, and pension class except for workers 65 years and older. Thus there is no noticeable tendency for firms with pensions to separate more older workers than they hire. The ratio of the separation rate in nonpension firms to the rate in pension firms appears to increase with increasing age in most industries, being particularly high for workers 45 years and older. For workers 55 to 64 years old, the ratio of separation rates ranges from about 1.4 in services to 6.3 in finance, insurance, and real estate (ignoring transportation, communication, and utilities, in which the ratio is 0.5). The sharper decrease in separation rates in firms with pensions would be expected if pensions made an important contribution to lower turnover. When firms are classified by number of employ ees it is found that firms with pensions have lower turnover rates than firms without pensions in each size-of-firm and age class (table 2.12). Turnover rates generally are inversely related to size of firm. The same patterns prevail in quit rates for pension and nonpension firms classified by in dustry division (table 2.13). The quit rates are actually higher in pension firms for workers 25 to m Unpublished study to be published in 1962. 44 years old in construction and for workers under 25 years old in service industries and in trans portation. On the whole, however, pension firms have lower quit rates than nonpension firms. Quit rates by age for pension firms classified by size of firm also are lower than the quit rates of similarly classified nonpension firms (table 2.14). The quit rates are inversely related to size of firm. These data do not prove that pensions reduce mobility, but they clearly demonstrate that firms with pensions have lower turnover and lower quit rates, independently of industry or of size of firm, although the lower mobility of workers in pension firms cannot be laid to pensions alone. There are three major reasons that pensions may not be the principal immobilizing influence in firms with pensions. First, characteristics other than industry or size of firm which affect mobility may differ between pension and nonpension firms. Pensions are more common in high wage firms; hence, pension firms probably have higher wages than nonpension firms. Pensions are also much more common in unionized firms, and such firms are T a b l e 2.13. A n n u a l Q u it K a t e s , b y I n d u s t r y , A g e , an d P e n s io n C o v e r a g e , 1955,1 Six A r e a s Quits per 100 employees Industry division and pension class Under 25 to 34 25 years years 35 to 44 years 45 to 54 years 55 to 64 years 65 years All and ages2 older All industries: 1 Pension________ No pension......... 41 47 22 34 14 22 6 16 3 12 9 11 17 24 Construction: Pension.............. No pension......... 38 21 36 24 23 21 11 21 5 18 11 9 27 20 Manufacturing: Pension..... ......... No pension......... 38 43 21 34 12 19 5 11 2 9 8 10 14 23 Transportation, com munication, and public utilities: Pension............... No pension......... 31 7 16 24 15 12 2 5 8 2 6 13 9 Wholesale and retail trade: Pension............... No pension......... 51 60 30 47 20 30 10 21 4 17 6 12 25 35 Finance, insurance, and real estate: Pension_________ No pension......... 37 90 24 59 16 48 4 22 3 19 22 16 21 49 Service: Pension............... . No pension_____ 70 46 19 29 15 22 6 28 6 13 18 16 18 26 1 See footnote 1, table 2.11. 3 See footnote 2, table 2.11. S ource : Derived from unpublished tabulations furnished by the Bureau of Employment Security. T a b l e 2.14. A n n u a l Q uit R a t e s , b y S iz e of E sta b l ish m e n t , A g e , an d P e n s io n C o v e r a g e , 1955,1 Six A r e a s 2 Quits per 100 employees Number of employees in establishments Pension class and age Workers under 45 years Pension-------------- No pension Workers 45-64 years: Pension............... No pension______ 1,0004,999 5,000 and over Total 50-99 100-499 500-999 23 33 29 39 27 33 21 39 20 22 23 5 15 6 21 7 14 3 12 4 7 4 1 See footnote 1, table 2.11. 2 See footnote 2, table 2.11. Source : Derived from unpublished tabulations furnished by the Bureau of Employment Security. likely to have strict seniority rules and effective grievance procedures which minimize the necessity of workers changing jobs in order to obtain satis factory work situations. Firms in seasonal in dustries are less likely to have pension plans than firms with fairly steady year-round employment; hence, the nonpension firms probably include most of the seasonal firms which characteristically have high labor turnover. In other words, pension firms are likely to offer better compensation, working conditions, and job security than non pension firms, and might be expected to have lower turnover regardless of the effect of pensions in holding workers. Second, pension firms have lower accession rates than nonpension firms. It was shown above that about 70 percent of workers separated in 1955 had less than 1 year of service. If the patterns of separation of pension firms and of nonpension firms are similar to those of all firms, then the lower accession rates of pension firms would mean that fewer short-service workers who are prone to quit or are frequently discharged are employed. The lower accession rates of pension firms may be a result of lower separation rates induced by pensions, but they are also related to the lower rate of growth of employment in firms with pensions. Third, pension firms have lower separation and quit rates in most age groups in all industry and size-of-firm classes although the differences are largest for older age groups. Since the effect of pensions on the mobility of young workers is probably quite small, this finding suggests that it is factors other than pensions which account for 15 much of the lower mobility of workers in firms with pensions. Turnover among young workers is quite high in pension firms, and quit rates in pension firms in most industries drop below 10 percent only after age 45. The willingness of workers under 45 to quit jobs is probably sufficiently high to accom modate necessary employment shifts even in pen sion firms. In short, the immobilizing effects of pensions which might be significant for manpower policy are likely to be important principally for older workers who are also likely to be long-service workers; for it is older workers who stand to lose large unvested benefits, are close to retirement, and face special difficulties in finding new jobs. Pension Trends Affecting Mobility Three recent developments in pension plans may have had the effect of reducing the immobilizing effects of pensions: (1) vesting has become more com m on;15 (2) early retirement and disability retirement are being provided in an increasing number of plans; and (3) collectively bargained multiemployer plans have grown in importance. Whatever effect pension plans have in reducing mobility would be mitigated by any of these three developments. Vesting. In establishing a pension plan, an employer is likely to concern himself initially with pensioning long-service employees only. Sim ilarly, a collectively bargained plan is likely to serve the organizational goals both of companies and unions if benefits are restricted to workers with long service. In the early years of a pension plan, uncertainties related to the cost and security of the plan will usually dictate a conservative policy in paying benefits. In 1952, when col lectively bargained plans were relatively new, only one-fourth of the 300 plans studied by the Bureau of Labor Statistics included vesting pro visions, while in late 1958 almost six-tenth of the 300 plans included vesting.16 Vesting was more common in single-employer plans than in multi employer plans. The trend toward vesting ob served in the Bureau of Labor Statistics study is also shown by Daniel M. Holland's analysis of the likelihood of vesting in 124 large company plans included in four studies of pensions conducted by the Bankers Trust Co.17 The likelihood of 16 vesting for workers with 10 or more years of service has increased steadily over the 1947-59 period. In the 1956-59 period, for instance, the likelihood of vesting of workers 55 to 59 years old with 20 or more years of service was 55.0 con trasted to 21.3 in the 1947-49 period (table 2.15). The likelihood of vesting is quite small for workers with less than 10 years of service, regardless of age. It is clear from table 2.15 that 10 years is the shortest length of service which is common, and the low likelihood of vesting for workers 30 to 39 years old with 10 years of service suggests that attainment of age 40 is also a common requirement. Both Holland's sample and the Bureau of Labor Statistics sample show changes in vesting in long-established plans and are, therefore, unlikely to be representative samples of all current plans. An analysis of several surveys of new or amended pension plans showed little evidence of a trend toward more liberal vesting provisions.18 The trend toward more liberal vesting in established plans and the absence of such a trend in new plans reflects the reluctance of plan authorities to provide liberal vesting initially. It seems likely that once a given vesting provision is established in a plan, any further change will be in the direction of liberali zation, except in those plans which face serious financial difficulties. These considerations sug gest that vesting provisions will become more liberal and more common as plans mature. How significant is the trend toward vesting for labor mobility? It was pointed out above that the immobilizing effects of pensions are is If employment earns the employee a right to a future pension or earns a right to a cash refund of both the employee's contribution (if any) and the employer's contributions, the pension is said to be vested. Vesting may be graded or full. If the pension rights are such that the employee receives a pension at retirement which accurately reflects the pension benefit formula, vesting is said to be full; if the benefits are less than the actuarial value of the service or earnings, vesting is said to be graded. If pension rights are vested as soon as they are earned, vesting is said to be immediate. If rights become vested only after a minimum number of years of service or after the attainment of a certain age, vesting is said to be deferred. 16 Bureau of Labor Statistics, Pension Plans Under Collective Bargaining (BLS Bulletin 1259,1959), p. 4. n By likelihood is meant the proportion of workers in these plans with given age and service who would have a vested pension. The likelihood is computed in this instance by weighting the vesting provisions of the plan by the number of workers of all ages and lengths of service included in the plan. Thus if 50 percent of the workers covered by the 124 plans were in plans which provided vesting at age 40 with 10 years of service, the likelihood for this age and service class would be 50. is Thirty-ninth Annual Report of the National Bureau of Economic Research, Inc. (New York, 1959), table 22, p. 66. 2.15. L ik e l ih o o d of W o r k e r s of G iv e n A ge a n d L e n g t h of S e r v ic e H a v in g V e sted P e n s io n R ig h t s , 124 L a r g e C o m p a n y P l a n s T able Age Length of service (in years) Under 30......... 30-39............ . . 1947-49 1950-52 1953-55 Less than 5________ 5-9___________ ____ 0.2 .8 0.1 2.2 0.02 1.8 0.02 1.9 Less than 5 5-9....... ..................... 10-14......................... .2 1.3 4.0 .l 2.6 6.7 .02 2.3 5.7 . 02 2.3 6.3 40-44................ Less than fi 5-9....... ................... . 10-14......................... 15-19......................... .2 1.3 5.4 7.5 2.6 7.4 9.8 45-49....... ........ Less than 5 _______ 5-9......................... . 10-14........................ 15-19______________ 20 or more_________ .2 1.3 6.9 9.3 13.1 2.6 8.5 11.1 18.4 50-54................ T/P.ss than 5 5-9.............. ............ 10-14........................ 15-19....... .................. 20 or more................. .2 1.4 7.2 9.8 18.3 T^.ss than 5 5-9............ ................ 10-14......... .............. . 15-19....... .................. 20 or more_________ 60 and over___ Less than 5________ 5-9________ _______ 10-14_______ ______ 15-19_______ ______ 20 or more................ 55-59...........— 1956-59 0 2.3 25.5 26.9 0 2.3 30.6 38.4 .02 2.3 31.0 32.6 34.9 .02 2.3 35.6 43.9 47.4 2.7 8.8 12.0 20.2 .02 2.5 31.7 34.6 37.8 .02 2.6 37,2 46.7 50.8 2.2 9.5 12.5 21.3 1.1 3.7 12.0 15.5 25.0 2.6 34.2 38.2 42.7 .3 2.9 40.0 49.9 55.0 10.8 12.0 20.1 25.0 36.1 1.5 4.1 23.1 29.5 41.5 2.0 4.5 37.7 43.4 51.2 1.9 4.5 42.0 59.6 82.3 1.0 .1 .1 .1 .1 Source : Daniel M. Holland, “ The Pension Structure,” in A Respect for Facts, Fortieth Annual Report of the National Bureau of Economic Research, Inc. (New York, 1860), p. 45. likely to be most important for older workers with long service. Workers with vested pensions do not stand to lose benefits, so their reluctance to change jobs is perhaps less than that of sim ilarly situated workers with unvested pensions. The provision of vesting with 10 or more years of required service can probably do much to de crease the immobilizing effects of pensions on the age and service groups for whom pensions are likely to have the most significance. Never theless, the propensity of long-service workers to change jobs is quite low even in nonpension firms. It is unlikely, therefore, that even liberal vesting provisions requiring attainment of age 35 or 40 and as little as 5 years of service would do much to increase the total amount of job changing of workers with vested pensions. If the independent effect of pensions in reducing mobility is small, then the effect of vesting in increasing mobility is also likely to be small. However, the increase in mobility which results from the spread of vesting may be quite significant « Bureau of Labor Statistics, Pension Plans Under Collective Bargainingt op. cit., pp. 11 and 12. for senior white-collar workers to whom unvested pensions may be important reasons for not chang ing jobs, since job-security provisions are uncom mon in this group. As pointed out in the section on occupational mobility, managerial and profes sional workers appear to exhibit low job mobility. Early Retirement. Many plans which make no other provision for vesting provide for retirement before the normal retirement age (usually age 65) at the option of the employee, sometimes with the employer's approval. In one or two instances, early retirement is at the option of the employer. The service required for early retirement is often lengthy, as much as 30 years. When early retire ment provisions are included in pension plans, the minimum age is often 55 years or 60 years. Occa sionally, early retirement is accompanied by a larger pension between retirement age and age 65 when the worker becomes eligible for OASDI (or age 62 for women, and for men since 1961), and many plans permit income equalization options. Early retirement may make an important con tribution to decreasing the immobilizing effects of pensions, since early retirement makes it possible for the worker who is only a few years away from normal retirement age to leave his job without forfeiting his pension. Early retirement provisions have undoubtedly become more common in the past few years. Of 300 collectively bargained plans in force in 1958, 218 included early retirement, and only 71 did not provide for either vesting or early retirement.19 Early retirement nearly always results in the worker receiving a smaller pension than he would receive if he worked to the normal retirement age. For some industries, however, the results are favor able to workers as long as retirement is at his op tion. Early retirement may permit a worker to quit a heavy job and find a new light job at which he may be able to work for several years past the normal retirement age. Disability retirement provisions are unlikely to have much effect on mobility, since disability standards are usually quite restrictive. The worker who draws a disability pension, however, does have his pension rights protected. Multiemployer Plans. Most multiemployer pen sion plans are negotiated by a union with a number of employers or with an employer's 17 association and cover workers in one industry either nationally or in some area. Such plans provide a limited degree of pension portability. As long as workers shift between employers who are members of such plans, the workers’ pension rights are protected. Multiemployer plans in clude vesting less frequently than do other collectively bargained plans. It might appear that the relatively high fre quency of industry shifts in job changes discussed in the section on industry mobility patterns might limit the pension protection provided by multi employer plans to a small proportion of job changers who remain in their original industry of attachment. The frequency of industry change in the national data is probably irrelevant for most multiemployer plans, however, since they are concentrated in industries to which workers show strong attachment. In 1960, for instance, about 23 percent of the workers covered under multiemployer plans were employed in apparel and other finished textiles industries; about 20 percent were in contract construction; and 15 percent were in motor transportation.20 These plans frequently cover a large proportion of the jobs in a given industry. Illustrating this point, the trustees of the Electrical Workers Pension Fund (IBEW) reported: The very nature of electrical construction work often makes changing from job to job and traveling from city to city imperative. Thus our National Plan as it is con structed, benefits all workers, since it assures them of a pen sion no matter where they go nor how often they are called upon to change jobs. The fact that a worker is not restricted to one area because of pension considerations is definitely an advantage to him in job opportunities and it is a boon to the contractors who must have trained electricians on the job wherever they are located. 18 In industries where workers show high attach ment, multiemployer plans can provide substantial protection of pension benefits and such plans need not reduce interfirm or interarea mobility, al though they perhaps restrict interindustry and occupational mobility. Even these immobilizing effects are likely to be small if the plans provide that workers can continue membership in the plan by making contributions after ceasing to work in covered employment. Such provisions are not common, although the IBEW plan mentioned above allows workers to continue membership in the fund after leaving covered employment. There is some tendency for several plans in the same industry which are established by a single international union to provide reciprocal transfer of credits or to merge into a single plan. The coordinated plans or single plans which result nat urally provide additional protection of pension benefits for covered workers who change jobs. The immobilizing effects of multiemployer plans are probably much smaller than the effects of unvested single-employer plans. Nevertheless, portability of pension credits among employers in the same industry is an inferior substitute for vesting in many industries. In others, it pro vides protection of pension credits which is often superior to vesting. As pointed out above, rela tively few single-employer plans provide vesting before age 40 and 10 years of service, but multi employer plans permit young workers with short service to keep their pension credits when they move to other employers who are members of the plan.3 0 30 Bureau of Labor Statistics, Multiemvloyer Pension Plans Under Collective Bargaining (BLS Bulletin 1326,1962), table 3, p. 98. Chapter HI. Employment Opportunities of Older Workers Summary Older workers have long faced difficulties in finding employment, although these problems are less severe during periods of high employment. The growth of seniority rules has strengthened the position of older workers, and relatively few older workers with long service experience unemploy ment during a given year. Employers usually rate their older workers highly in terms of steadi ness and output, and there is no evidence that the output of older workers in general is significantly lower than that of younger workers. Younger workers usually have higher unem ployment rates than do older workers, and larger proportions of younger workers than older workers experience unemployment during a given year. Those older workers who experience unemploy ment tend to be unemployed for longer periods than younger workers. It is lengthy unemploy ment of older workers that gives the older worker problem its special importance for public policy. The unemployment rates of older workers are less sensitive to changes in the general level of unemployment than the rates of younger workers, and the unemployment rates of older workers have not worsened relative to the general level of unemployment in recent years; hence the unemployment problem of older workers is probably related more to personal characteristics of unemployed older workers and to discriminatory practices which bar many jobs to older workers than to cyclical influences. In 1956, unemployed older workers were more frequently skilled and had somewhat higher past earnings than younger jobseekers, but they had less formal education and were more frequently physically handicapped. In 1961, about threefourths of the unemployed older workers were last employed in industrial occupations. Aged men who are seeking full-time work after being retired from their regular jobs find extreme difficulty in getting jobs. Many firms impose maximum age limits in hiring and thereby close important parts of the job market to older jobseekers. Studies in 1950 and 1956 showed that the proportion of job orders filed with public employment offices including maximum ages ranged from one-fourth to three- fourths. Restrictions were especially common in professional, managerial, and clerical jobs. Em ployers frequently mentioned higher pension costs and closeness to involuntary retirement age as reasons for not hiring older workers, but the desire to promote from within (often imposed by seniority rules) and a conscious policy of keeping the work force young are perhaps more important reasons. The importance of pension costs in the hiring decision depends on the provisions of the pension plan. Multiemployer plans nearly always have fixed contributions, so that there is no additional cost to the employer of hiring an older worker. The additional costs of older workers to firms with fixed benefit plans depend on the mortality and turnover experience of the firm. Fixed benefit per year of service plans probably have higher additional costs for older workers than percentage of compensation plans, but the additional cost for a 45-year-old worker compared to a 35-year-old worker would seldom exceed 1 percent of total compensation for the average employee. Despite the reluctance of some firms with pen sion plans to hire older workers, it appears that as a group, firms with pensions hire as many 45- to 64-year-olds as they separate; thus the presence of maximum hiring ages associated with pension plans does not appear to have had unfavorable effects on the employment opportunities of these older workers. Introduction Many unemployed older workers have difficulty in finding jobs even during periods of high em ployment.21 Corporate pensions are related to the older worker problem because many firms with pensions have maximum hiring ages which are rigidly enforced. It is said that pension costs are higher for older entrants and that firms are reluc tant to hire workers who will shortly be retired without a full pension. Older workers faced serious employment prob lems long before pensions were common. During the 1920?s, few older workers were protected by seniority provisions against layoff; hence many 21 By “ older workers” is meant workers 45 years and older largely as a matter of statistical convenience. Age problems in getting and retaining jobs begin at different ages in different occupations. 19 T a b l e 3.1. U n e m p l o y m e n t R a t e s 1 9 4 8 -6 1 1 of M en, by these older workers from the labor market, and (2) taking steps to assure the continued usefulness of these age groups as productive workers.22 A ge, [Annual averages] Percent of civilian labor force unemployed by age in years Year 1961....... ......... 1960................. 1959................. 1958................. 1957................. 1956................. 1955................. 1954................. 1953................. 1952................. 1951................. 1950................. 1949_________ 1948................. 14 and over 6.5 5.4 5.3 6.8 3.8 3.5 3.9 4.9 2.4 2.4 2.6 4.9 5.5 3.3 14 to 19 20 to 24 15.8 14.0 13.2 15.2 11.3 9.6 9.9 11.2 6.8 7.6 7.0 11.0 11.9 8.3 10.7 8.9 8.7 12.7 7.8 6.3 7.0 9.8 4.3 4.0 3.5 7.7 9.9 6.3 25 to 34 5.7 4.8 4.7 6.5 3.3 2.9 3.0 4.4 1.9 1.8 2.0 4.2 4.7 2.5 35 to 44 4.6 3.8 3.7 5.1 2.8 2.3 2.8 3.7 1.7 1.7 1.7 3.3 3.8 2.1 45 to 54 4.9 4.1 4.1 5.3 3.3 2.7 3.0 3.9 1.9 1.8 2.1 3.9 3.9 2.3 55 to 64 5.7 4.6 4.5 5.5 3.5 3.2 4.1 4.9 2.4 2.1 2.5 4.7 4.9 2.8 65 and over 5.5 4.2 4.8 5.2 3.4 3.3 3.7 4.2 2.0 2.1 3.3 4.6 4.9 3.0 1 Old definition of unemployment for 1956 and earlier years. Soubce: Bureau of the Census and Bureau of Labor Statistics. suffered unemployment. The growth of unions and collective bargaining during the 1930’s and 1940’s led to the more widespread adoption of seniority practices, providing greater security to employed older workers. The job security pro vided by union-negotiated seniority provisions also spread to nonunion firms. The unemployment problems of older workers in the 1930’s disappeared during the high employ ment years of World War II and the immediate postwar period. Ewan Clague, writing in 1947, predicted that— The problem of the older worker will rise when the first postwar recession in business occurs. A deep and prolonged depression need not be envisaged; the assump tion may be made that not more than 5 million persons would be out of work at any one time and that the business setback would not last more than a year or two. Unem ployment of that dimension would undoubtedly emphasize the difficulties of the older worker in industry. It would bring to the fore again the issue of adequate old-age retire ment benefits and the question of the suitable retirement age. If this were the whole problem, the lessons would be simply that older workers can and do get jobs in full prosperity periods, but they are the chief sufferers in business depressions; and that special legislative consider ations, therefore, should be given to them under social security . . . Meantime, the first business recession will cause deep concern as to the impact of unemployment on the older workers. In succeeding periods of recovery and prosperity this type of unemployment will not disappear, but will remain as an increasingly stubborn and difficult question. The growing numbers of the aged will speedily convert this into a national problem of the greatest importance. The decision will then have to be made between (1) lowering the retirement age and attempting to remove 20 These predictions have been borne out. The 1948-49 recession gave impetus to the 1950 revisions of the Social Security Act and to the unions’ drive for negotiated pensions. During the 1950’s, interest in older workers and their problems grew. Research by government and private groups sought solutions to the older worker problem. The three recessions of the postKorean period brought the problem into promi nence. Older jobseekers tended to be unemployed for lengthy periods, frequently exhausted their unemployment benefits, and posed special prob lems in declining industries and depressed areas. The older worker problem has two aspects: (1) keeping older workers on the job; and (2) getting older workers jobs if they become un employed. Most of the recent concern over the older worker problem has concentrated on the problems of getting a job, but obviously this problem would be less severe if older workers were more successful at holding jobs. Older workers usually have the job security which arises from seniority provisions in col lective bargaining, favorable employer attitudes, and satisfactory performance records. Older workers are usually the principal beneficiaries of seniority rules, and collective bargaining provides ways of tailoring jobs to the abilities of older workers, as well as protecting their transfer rights in instances of departmental shutdowns or plant transfers.23 A survey of employer attitudes showed that a majority of employers thought that older workers were superior to younger workers in terms of “ steadiness” and output, but that older workers were more difficult to train. A majority also thought that older workers were absent less often than younger workers, but that they tended to be absent for longer periods.24 Bureau of Labor Statistics studies of productivity show that for pieceworkers in four footwear factories and in 22“ Employment Problems of the Older Worker,” Monthly Labor Review, December 1947, pp. 661-663. 2* Unions usually oppose variation in pay rates for older workers, but stress job changes to meet the problems of failing ability. See Melvin K. Bers, Union Policy and the Older Worker (Berkeley, Institute of Industrial Relations, University of California, 1957), pp. 34-46. See also, Bureau of Labor Statistics, Older Workers Under Collective Bargaining, Part I (BLS Bulletin 1199-1,1956), pp. 13-25. *4“ The Older Worker,” Factory Management and Maintenance (McGrawHill, March 1958), pp. 85-96. four clothing factories, older workers had slightly lower average output. Older office workers had about the same output as younger workers. A Canadian study showed that older retail sales workers had slightly higher output than younger workers. In all of these studies, however, variability was high in each age class, so that age was not an important variable in explaining differences in worker productivity.25 Unemployment of Older Workers Unemployment rates for men commonly are highest for workers in the younger age groups, decrease sharply and are lowest for men 35 to 44 years old, then increase slowly with increasing age. In 1961, for instance, the unemployment rate was 10.7 percent for men 20 to 24 years old, 4.6 percent for men 35 to 44 years old, and 5.7 for men 55 to 64 years old (table 3.1). The unemploy ment rate of men 65 years and older is usually little higher than that of men 45 to 54 to 64 years old, but the rate is computed on a labor force made up in large part of farmers and self-employed managers and proprietors whose unemployment rates are very low. Thus the unemployment rates of aged men understate unemployment as a pro portion of those who are exposed to unemployment.5 * 25 Bureau of Labor Statistics, Job Performance and Age: A Study in Measurement (BLS Bulletin 1203, 1956) and Comparative Job Performance by Age: Office Workers (BLS Bulletin 1273, I960); and Age and Performance in Retail Trade (Economics and Research Branch, Department of Labour, Ottawa, Canada, 1959). 2« In the regression of the unemployment rate of men 55 to 64 years old on the unemployment rate of men 35 to 44 years old, the intercept shows the hypothetical unemployment rate of men 55 to 64 years old which would be associated with the absence of unemployment of men 35 to 44, and the regres sion coefficient shows the average change in the unemployment rate of men 55 to 64 years old which is associated with a 1 percentage point increase in the unemployment rate of men 35 to 44 years old. The regression coefficients for the older age groups are close to 1, and this suggests that recessions cause about the same absolute changes in unemploy ment rates for the older age groups. Since the intercept coefficients for the older age groups are greater than zero, however, the unemployment rates of the older age groups increase proportionately less with recession than does the rate of men 35 to 44 years old. It may be observed from charts 3.4-3.6 that the observations of more recent years do not fall disproportionately above the regression lines. There is no trend in the unemployment rates of older men once account is taken of the variation in unemployment rates of men 35 to 44 years old. Regression coefficients are shown below: Standard error of regression Regression Intercept coefficient coefficient coefficient Regression age 14 to 19................ - ......... . 20 to 24____ ____ _________ 25 to 34________ __________ 45 to 54................... ............. 55 to 64_____ _____________ 65 and over____ _____ ____ 2.98 .37 .37 .30 .71 1.25 2.55 2.38 1.33 1.00 1.05 .86 0.22 .15 .04 .04 .08 .10 The increase in unemployment rates for age groups older than 45 years is related to the greater duration of unemployment among unemployed older men. In 1961, for example, the average duration of unemployment increased regularly with age: Average duration of unem ployment, in weeks, 1961 Age 14 18 25 45 65 All ages__ __ _______ ____ to 17________________________ ____ to 24________________________ ____ to 44________________________ ____ to 64____ _ _ _ ____ ____ and over____ _ _ _ Men 16. 9. 13. 16. 20. 30. 9 1 1 8 9 4 Women 13. 6. 11. 13. 16. 5 9 1 8 1 _. The greater average duration of unemployment of older workers is the principal characteristic of such workers that makes the older worker problem of major public concern. Relatively few older workers are unemployed in the course of the year. In 1960, for instance, the proportion of persons in the labor force at some time during the year who were unemployed at any time during the year decreased with age after age 20: Percentage of persons in labor force in 1960 with unemployment Age 14 18 20 25 35 45 65 ____ All a g e s _ _ __ _ to 17________________________ ____ to 19________________________ ____ to 24________________________ ____ to 34________________________ ____ to 44________________________ ____ to 64________________________ ____ a n d over___ _____ __ _ .____ Men 18. 19. 33. 34. 20. 15. 14. 10. 4 3 0 8 1 3 4 8 Women 15. 3 13. 7 29. 9 19. 5 16. 2 14. 5 12. 5 8 .3 From the foregoing data it may be concluded that relatively few older workers are unemployed during the year, but once employed their unem ployment tends to be of longer duration. When older workers are displaced, it is usually because they are laid off or discharged. Older workers tend to be unemployed longer on the average be cause they are unemployed predominantly because of economic reasons, rather than voluntarily and because fewer opportunities for employment are open to them. The unemployment rates for men of different ages tend to respond in similar fashion to economic changes. When the unemployment rates of men in various age groups are plotted against the un employment rate of men 35 to 44 for the period 1948-62, the observations show very close linear relationships 26 (charts 3.1-3.6). Hence, while un21 Relation o f Unemployment Rates for Men o f D ifferent Ages to Unemployment Rates o f Men 35-44 Unemployment rate Unemployment rate of men 14-19 of men 20-24 Unemployment rate o f men 25-34 Unemployment rate Unemployment rate Unemployment rate o f men 55-64 of men 65 and over Unemployment rate of men 35-44 22 of men 45-54 employment of older men has grown more severe as unemployment has increased in recent years, it does not seem to have grown worse relative to the general level of unemployment. Unemployment rates of older workers and the rate of workers 35 to 44 years old change by about the same absolute amounts with fluctuations in the level of economic activity. Since unemployment rates of older work ers are proportionately higher in years of low un employment, the unemployment rates are rela tively less sensitive to recession and recovery than are the rates of younger workers. Understanding the behavior of unemployment of older workers depends primarily on understanding the reasons that the unemployment of older workers is higher than that of the prime age groups even during periods of prosperity. Two principal reasons are usually mentioned— special characteristics of older workers which disqualify them for the jobs which are available, and discrimination against qualified older workers. Characteristics of Unemployed Older Workers Like any other group of people, older jobseekers are diverse. Some older jobseekers are on tem porary layoff, and can expect to be recalled when employment increases. These face no special problems, although they may have difficulty in finding temporary work. Seniority systems and pensions are likely to hold these workers closely to the firm even though they are on layoff, espe cially if other regular jobs are closed to older workers. The displaced older worker who cannot expect to return to his regular job may face more dif ficult problems. He often restricts his job search to the local labor market area. If employment in the area or in his industry of attachment is declining, the displaced older worker may not be able to find any job. Workers displaced from high-wage jobs may find it impossible to get comparable jobs even if they are willing to move. Consequently, many displaced workers in their fifties and sixties withdraw from the labor force. The personal characteristics and work experi ence of older workers often contribute to their difficulties in finding suitable work. A study of work applicants at public employment offices con ducted as part of the Seven City Study in 1956 27 Bureau of Employment Security, Older Worker Adjustment to Labor Market Practices (BES Bulletin R151, 1956). found that 30 percent of the jobseekers were 45 to 64 years old, while 10 percent were 65 years and older. In contrast, a matched sample of employed workers included 32 percent 45 to 64 years old and only 3 percent 65 years and older. About two-fifths of the older jobseekers were last employed in manufacturing. The older jobseekers had on the average higher skills and higher past earnings than those under 45 years, but had less formal education and were more frequently physically handicapped.27 In 1961, 70 percent of the unemployed older men were last employed in industrial occupations (craftsmen and foremen, operatives, nonfarm laborers), compared to 74 percent of the un employed men 25 to 44 years old. There were only slight differences in the unemployment rates of labor force groups of different ages in various occupations in 1961: Unemployed as percent of experienced civilian labor force in occupations group (men) Occupation group All occupations_________________ Professional and technical_____________ Farmers and farm managers___________ Managers, officials, and proprietors___ Clerical and kindred workers__________ Sales workers___________________________ Craftsmen and foremen________________ Operatives_______________________________ Service workers_________________________ Farm laborers and foremen____________ Laborers, except farm and mine_______ 25 to 44 years 45 years and older 5. 1 5. 2 1. 7 .5 1. 6 3. 4 2. 9 5. 2 8. 0 5. 9 6. 8 14. 6 2. 0 .4 1. 6 4. 0 3. 6 7. 3 7. 6 6. 4 7. 7 13. 7 Source : Employment and Earnings, June 1962, table SA-37, p. 90. Some aged workers retire while they are still capable of working, and many of these remain attached to the labor force. Retired workers often seek part-time or seasonal jobs in order to keep their earnings under the maximum allowed for OASDI benefits, but some seek full-time jobs at their customary rates of pay. Firms which do not impose rigid age limits in hiring are usually reluctant to hire workers who have re tired and are receiving sizable pensions, and with two-fifths of the private work force covered by pensions, relatively few jobs are open to the aged retired workers, even if they are in good health and are fully qualified. Indeed, it may be diffi cult to justify special efforts to find jobs for pen sioners when other qualified unemployed persons are seeking work. 23 Discrimination Against Older Workers Many firms refuse to hire older workers and impose rigid maximum age limits in hiring and thereby close important parts of the labor market to older workers. In 1950, a study con ducted by State employment services found that 72 percent of the job orders placed with the public employment service in Columbus, Ohio, and 52 percent of those in Houston, Tex., specified maximum hiring ages. In Lancaster, Pa., 60 percent of the job openings, and in New York City, 25 percent of the job openings included upper age limits. The proportions of job orders and job openings with age restrictions were commonly higher in professional and managerial and clerical jobs and job orders, and somewhat lower in other occupation groups.28 Similar results were found in the Seven City Study conducted by State employment security agencies and the Bureau of Employment Security in 1956. More than one-half of the job openings in the seven cities specified upper age limits of less than 55 years, and in 20 percent of the openings, the age specified was less than 35 years. Again, managerial and professional and clerical jobs were more frequently restricted than other oc cupations, and in this study, unskilled jobs were also more frequently restricted than other kinds of jobs. The frequency of upper age limits ap peared to increase with size of establishments. Slightly more than one-half of the job openings in firms with fewer than 20 employees specified maximum ages, and the proportion increased until almost four-fifths of the job openings in establishments with 1,000 or more employees specified age limits. Very often, however, the age limits specified by establishments in the largest size group were higher than those of estab lishments in the 100- to 999-employee class. Of the largest establishments, 19 percent of the age specifications were for ages above 54 years. According to the Bureau of Employment Se curity, the greater frequency of age limits in large firms was “ presumably due to pension and insurance plans.” The reasons most frequently given for not hiring older workers in the Seven City Study included (1) Inability to maintain normal pro duction standards (22 percent of all reasons); (2) inability to meet physical requirements (21 124 percent); (3) inflexibility (13 percent); (4) pension and insurance costs (10 percent); (5) too close to compulsory retirement age (in the usage of this report, involuntary retirement age) (7 percent); and (6) simply prefer younger workers (5 per cent). No other reason provided more than 4 percent of the responses.29 Reasons mentioned by employers for not hiring older workers are not necessarily the most important reasons for such practices. Probably most firms prefer to fill vacancies with qualified workers who are already employed. Indeed, in many firms seniority rules provide that workers may bid on openings in order of seniority. In such firms, job openings are predominately open ings at entry grades, usually requiring little experience and carrying relatively low pay. Since many older jobseekers are accustomed to the advantages and pay which came with seniority in the firm they left, jobs at entry grades may not be acceptable, even if the older jobseeker is acceptable to employers. Finally, many firms have a definite policy of “ keeping the firm young” or at least preventing the average age of the work force from rising any more than necessary. Firms which are not expanding rapidly normally will encounter the problem of an aging work force, and pensions are often adopted to induce retirement of aged workers. Under such circumstances, the hiring of older workers goes against the firm’s welldefined policy. Despite the frequency of age limitations, there is evidence that employers are likely to relax age limits if qualified applicants are referred, although employers are especially reluctant to relax age limits associated with pension plans.30 Age prefer ences or restrictions are also found much less frequently in States which have enacted legislation to bar hiring discrimination based on age. As of June 1962, laws prohibiting discrimination against older workers were in force in 15 States, including New York, California, Pennsylvania, Ohio, Massachusetts, and New Jersey. These laws forbid discrimination against workers of specified ages in most industries, although some 28 Bureau of Employment Security, Workers are Young Longer, a report of the Findings and Implications of the Public Employment Service Studies of Older Workers in Five Cities, third printing, 1953. 29 Bureau of Employment Security, Counseling and Placement Services for Older Workers (BES Bulletin E152,1956),pp. 28-60. so Ibid. of the State laws exclude small employers and farm, domestic service, and noncommercial em ployment. Most of the laws apply to workers between the ages of 40 or 45 to 60 or 65 years, although New Jersey’s law applies to workers 21 years and older, and New York’s law has been interpreted to apply to discrimination based on “ overage,” regardless of the worker’s chronological age. Among the employment practices outlawed are refusal to hire, discrimination in pay or working conditions, and discharge of workers on account of age alone. Employment agencies and labor unions are also restrained from discrimina ting in referral or service to older workers.31 The effectiveness of the State laws against age discrimination is difficult to judge. Enforcement varies between the States, of course, but even in the States with strong laws which are vigorously enforced (such as New York and Pennsylvania), few cases have been sustained against employers. The laws are probably effective in eliminating or reducing the employer’s reluctance to consider older workers. In 1956, for instance, age prefer ences were found in only 24 percent of the job orders in Worcester, Mass, (where age discrimina tion was prohibited), but age restrictions or preferences in other cities were found in 34 to 79 percent of the job openings.32 What relation is there between pension plans and the imposition of maximum hiring ages? It was previously noted that about 10 percent of reasons offered by employers for not hiring older workers were related to higher pension and insur ance costs, and 7 percent were due to closeness to involuntary retirement ages. The relationship between pension plans and maximum hiring ages is illustrated by a survey of San Francisco area firms which found— Virtually all the firms with rigid upper age limits in hiring had pension plans, all accompanied by compulsory retirement provisions. M ost of the companies with less rigid age limitations in hiring had pension plans, but some of the plans were not accompanied by compulsory retirement provisions, and in fact, a majority of these firms did not have compulsory retirement. Some of the 31 Bureau of Labor Standards, Division of State Services, “ Brief Summary of State Laws Against Discrimination in Employment: Older Workers,’ ' Fact Sheet No. 6-B, June 1962. 32 Bureau of Employment Security, Counseling and Placement Services for Older Workers, op. cit. 33 Margaret S. Gordon, “ The Older Worker and Hiring Practices,” Monthly Labor Review, November 1959, p. 1201. 34 For a more detailed discussion of these points, see Bureau of Employ ment Security, Pension Costs in Relation to the Hiring of Older Worker (BES Bulletin E150, 1956), pp. 19-21. firms with no upper age limits in hiring had pension plans, but almost none had compulsory retirement. In this study, it also appeared that both policies were more often found together in older firms and in firms which were not unionized.333 4 A number of factors tend to operate together. A firm which is willing to pay the higher pension costs which are associated with involuntary re tirement probably considers that its aging work force constitutes a serious problem. In such an instance, it would scarcely be rational to hire additional older workers whose involuntary re tirement in a few years would only cause higher pension costs. It is true, of course, that a firm with an aging work force is unlikely to hire older workers whether it has a pension plan or not; hence, the independent impact of pension plans on hiring practices is difficult, if not impossible, to determine. Pension and Insurance Costs and Hiring Practices Pension and insurance costs have been men tioned as one reason for employers’ reluctance to hire older workers. Certain issues need only brief consideration. Insurance costs (including workmen’s compensation, group medical and hospital insurance, and life insurance), taken as a whole, are unlikely to differ very much between older entrants and younger entrants. If depend ents are covered by hospitalization and medical care, the smaller number of maternity benefits and the smaller number of younger children of the older entrants may offset their slightly higher morbidity. Group life insurance rates reflect the mortality experience of the covered group, so that a marked increase in the average age of the covered group would lead to higher mortality and perforce to higher insurance costs. Unless the number of older workers hired is a substantial fraction of coverage, however, the costs are un likely to increase significantly with the selective hiring of older workers. Workmen’s compensation costs also depend on experience. There is no reason to believe that injuries are either more frequent or more severe for older workers than for younger workers.31 Money-purchase pension plans are also free from higher costs for older entrants. In a study of multiemployer plans in 1960, for instance, the Bureau of Labor Statistics found that all but 8 of the 611 plans which required employer con25 tributions specified contributions as either rates per period of time worked or as percentages of payroll.35 In multiemployer plans, at least, the age of the worker has no direct effect on the employer's pension costs. It is obvious, however, that the hiring of older workers does increase the cost to the multiem ployer plan of providing retirement benefits on either a flat benefit or a flat benefit per year of service formula. If benefits are held constant, then contributions would have to be increased. For the individual employer who is a member of a multiemployer plan and who is faced with the decision to hire an older worker, however, this potential contribution increase is insignificant. Multiemployer plans are commonly found in industries where unions have considerable influ ence on hiring; hence, the employer may not be able to discriminate between workers of different ages if he wishes. Single-employer money-purchase plans (in which the contribution rate is fixed and the benefit variable) clearly have employer costs which are no different for new entrants of different ages. For single-employer plans, however, moneypurchase plans have probably been becoming less common, because the benefits from such plans weight early years and contributions more heavily, hence benefits sometimes lack a reason able relationship to the earnings of the years immediately preceding retirement. The costs of pensions for older entrants are relevant in plans of the definite benefit type in which contributions are variable. In negotiated single-employer plans (pattern plans), there ap pears to be a trend toward flat benefits for each year of covered service. In the Bankers Trust Co. study of plans adopted or amended in 195659, 87 percent of such pattern plans had some form of flat benefit per year of covered service, often $2 or $2.50 benefit per month for each year of covered service.36 To finance the expected cost of such a benefit, the employer must con tribute more for a year close to retirement than for a year more distant from retirement. An example of the annual charges of a plan providing a benefit of $2.50 per month for each year of covered service is given below under the following assumptions: (1) Two and one-half percent interest rate (a higher rate would make the annual charges 26 lower and increase the percentage difference between younger and older workers). (2) 1951 Group Annuity Mortality Table (which assumes decreases in mortality below that experienced by group annuitants in the base period of the study on which the table is based). (3) No turnover (turnover would increase the relative difference in costs and lower total costs but it is partially offset if the plan is vested). Annual charges for a flat benefit of $2.50 per month at age 65 for each year of covered service Age of entry 25___________________ 30___________________ 35___________________ 40___________________ 45___________________ 50___________________ 55___________________ 60___________________ Monthly pension $100. 87. 75. 62. 50. 37. 25. 12. 00 50 00 50 00 50 00 50 Annual charge $158. 87 171. 92 187. 07 197. 69 218. 92 243. 87 269. 32 299. 58 Obviously there are differences in cost of pen sions for new entrants of different ages. How ever, the additional pension cost of hiring a worker 35 years old compared to a worker 45 years old would be only $32 (in this example), or less than 1 percent of annual compensation for an average factory worker. Considering the wide variation in worker productivity that exists within a single job and the uncertainties attached to hiring new workers, it does not appear that such a difference should be considered very significant. If in the foregoing example a higher interest rate and normal turnover had been assumed, this would increase the proportion which the older worker’s costs bear to younger worker’s costs. However, turnover savings in the plan might be large enough to reduce the differential costs in absolute terms and so make the difference less as a per centage of total compensation. In contrast to the flat benefit per year of service benefit formula common in negotiated plans, nonnegotiated plans usually base benefits on compensation, especially compensation during the entire period of credited service. It is im possible to estimate the cost of pensions under such plans without detailed assumptions con cerning the worker’s future compensation. Never 35 Bureau of Labor Statistics, Multiemployer Pension Plans Under Cob lective Bargaining (BLS Bulletin 1326, 1962), p. 11. 36 1960 Study of Industrial Retirement Plans (New York, Bankers Trust Co., 1960), p. 19. theless, the pension costs accruing from a given month’s compensation will be higher for an older worker than for a younger worker, since for the latter, a smaller sum at compound interest will suffice to buy the annuity earned by the credited compensation. Again, interest rates and turn over are important considerations. Thus, it appears that for an employer in a multiemployer plan the additional pension cost of hiring an older worker is irrelevant, while for most employers with single-employer plans, pen sion charges will be higher for older entrants than for younger entrants. The specific difference in costs of hiring will vary with the actual mortality and turnover experienced by the plans. How significant are the additional pension costs of hiring older workers? For most workers in low or middle paid brackets, the additional annual costs are likely to be a very small percentage of annual compensation. Pension costs are but one of the variables which should be considered in the hiring decision. If older workers have lower turnover costs to the firm, these costs should also be considered. It was pointed out in chapter II that older workers generally have lower separation rates than young workers, and this is often thought to be an important offset to their higher pension costs. The lower turnover rates of older workers may or may not be significant, however. Newly hired older workers may have higher turnover rates than newly hired young workers, although this may be a characteristic of the temporary jobs for which they are often hired. The degree to which the older worker has lower turnover costs and other costs (such as insurance or absenteeism) will depend on the experience of the individual firm. In view of the complexity of the problems and the difficulty of generalizing, it does not seem justified to say, as many do, that additional pension costs of older workers are not or should not be significant to employers. It must be remembered, however, that hiring is subject to great uncertainty, and the decision to exclude arbitrarily workers past a certain age because their pensions would cost a cent or so more an hour may rob the firm of the opportunity to hire valuable workers. Effect of Pension Plans on Hires and Separations of Older Workers Previous sections have indicated that pension plans may adversely affect employment oppor tunities for older workers by leading to the estab lishment of maximum hiring ages and by raising the cost of placing older workers on the payroll. Almost certainly there are individual situations in which pension plans have had such adverse effects. The problem is to ascertain, if possible, whether these results have been so widespread that total employment opportunities for older workers have been affected. Only fragmentary evidence is available on this point. While firms with pensions hire proportionately fewer older workers than do firms without pen sions, it is also true that firms with pensions separate proportionately fewer older workers than do firms without pensions. Data from the 1956 study by the Bureau of Employment Security and State employment security agencies show that older worker separation rates are about equal to older worker hiring rates both for firms with pen sions and for those without pensions (chapter II). As pointed out in chapter II, no significant con clusions for mobility can be drawn from these facts since the composition of the two groups of firms by industry and by size of firm differs con siderably. For present purposes, however, it is important to note that firms with pensions hired 16 workers aged 45 to 64 per hundred workers of the same ages per year, which was the same as the separation rate for the same age group. Thus there was no tendency observable from these data for firms with pensions to force more older people into the labor market than they hired from it. This study considerably weakens any presumption that private pensions are detrimental to older jobseekers as a group. The following conclusion drawn by the Bureau of Employment Security must be interpreted with care: The older worker is more likely to be hired for a job which is not covered by a private pension plan than one in which he will have coverage. Workers 45 years and older account for 25 percent of the hires in employment 37 Bureau of Employment Security, Older Worker Adjustment to Labor Market Practices, op. cit., p. 4. without pension plans, contrasted to 14 percent in jobs with this advantage.37 27 The implications of this conclusion are modified considerably, if it is accompanied by the following statement: The older worker is more likely to be looking for a job because he was separated from a firm without pension coverage than because he was separated from a firm with pension coverage. Workers 45 years and older account for 27 percent of the separations in employment without pension plans, contrasted with 18 percent in jobs with this advantage.38 The conclusion is modified even more if workers 65 years and older are excluded from consideration. There is no difference between hire rates and separation rates of workers 45 to 64 for firms with pension plans. About 3 percent of the separations from firms with pension plans are workers 65 years and older, but only 0.6 percent of the new hires are workers in this age group. In firms without pensions, about 3 percent of both hires and separations consisted of aged workers. Hence, the detrimental effects of pen sions on the job opportunities of older workers are concentrated in the 65 years and older group. It is clear that firms with pension plans infre quently hire workers who are 65 years and older. Additional information bearing on this question is available in the age composition of employment of industries by pension plan coverage in 1956: 38 Ibid., p. 68. 28 Percent of employees 45 to 64 years old Industry All industries____________________ Construction____________________________ Manufacturing, total___________________ Durable____________________________ Nondurable________________________ Transportation, communication, and public utilities________________________ Trade____________________________________ Finance, insurance, and real estate____ Service___________________________________ Pension No pension 31 40 32 31 34 32 32 31 30 33 30 31 26 40 42 31 31 32 Source : Bureau of Employment Security, Older Worker Adjustment to Labor Market Practices, op. cit., table X X II, p. 255. Overall, there is little difference between the proportions of older workers employed, and in few industries is there much difference in propor tions. In all industries, however, the proportions of workers 65 years and older in firms with pen sions were considerably lower than the correspond ing proportions in firms without pensions. The limited evidence available indicates that while pension plans may have contributed to blocking employment opportunities for individual older workers, it does not seem that firms with pension plans are failing to employ their propor tionate share of workers 45 to 64 years old. The impact of pension plans on the unemployment of older workers is probably to increase the severity of the problem somewhat, but it cannot be con cluded that practices associated with pension plans are a major cause of the older worker problem. Chapter IV. Retirement and Withdrawal From the Labor Force Summary The proportion of men 65 years and older in the labor force has been declining steadily since 1890, but the decrease has been particularly sharp since 1950. Less than a third of the aged men were in the labor force in 1961. Employed aged men are concentrated in certain types of occupations and industries. Occupations in which the employment of all men has expanded greatly over the past decade have also shown increases in employment for aged men, but occu pations in which total employment was stable or contracting have been marked by sharp reductions in the number of employed aged men. Employed aged men are heavily concentrated in farming (despite the decline in number of aged farmers) and managerial occupations, where self-employ ment is common, and are proportionately much less important in the industrial occupations. The trend toward retirement has also been marked by an increasing proportion of all aged men employed at part-time jobs which provide additional income to supplement retirement pensions. The most important factor in stimulating increased retirement for older workers and reduc ing their participation in the labor force has been the growth of available retirement income through the public and private systems. The central position of retirement income in inducing retire ment is shown by the fact that pensioners with large pensions are more likely to be satisfied with retirement than those with small pensions. The need for income is the usual reason given by older workers who want to continue work beyond age 65. Private pensions, as a major supplement to income from public retirement programs, have served to induce voluntary retirement and with drawal from the labor force. At the same time, it has been asserted that through the practice of involuntary retirement, private pensions have also had the effect of forcing into retirement many who still wish to work and who remain capable of contributing effectively to the national output. Involuntary retirement provisions are included in single employer pension plans covering some what more than half of the workers in large negotiated plans and in about four-fifths of the large nonnegotiated plans. Involuntary retire ment rules usually apply to workers at age 65 in nonnegotiated plans, but the age is often 68 or 70 years in negotiated plans. Multiemployer plans seldom have involuntary retirement. Employers who favor involuntary retirement usually stress the usefulness of such rules in removing aging workers whose efficiency is decreasing and in opening opportunities for promotion of younger workers. Employers who oppose involuntary retirement stress either the costs of such provisions or the inappropriateness of age as a criterion for separation. The age composition of their own work forces and the available labor supply are probably influential in disposing employers to favor or to oppose involun tary retirement in their firms. Union leaders’ attitudes toward involuntary retirement are also conditioned by their specific needs. Although usually opposed to involuntary retirement in principle, leaders of* unions in indus tries suffering heavy unemployment often view retirement of aged workers as an equitable way to ease the unemployment problems of the union membership. Opinions of observers not directly involved in the negotiation or conduct of pension plans usually favor a flexible retirement age or urge that the age of involuntary retirement be raised. The effect of involuntary retirement is difficult to determine. According to a 1952 study, involun tary retirement rules were the original reason for leaving the labor force of about 13 percent of the retired men in 1952, although few of these men were interested in working. Involuntary retire ment is more common as a reason for retirement among men receiving larger pensions. On the whole, there appears to be little dissatisfaction with retirement among those involuntarily retired, or among those older workers for whom involun tary retirement lies in the pear future. From the point of view of manpower policy, the desirability of retirement and the resulting reduc tion in the labor force depends chiefly on the level of demand for labor. In wartime when manpower is severely limited, the appropriate manpower policy is to induce older workers to remain active workers. With unemployment at more normal 29 levels, retirement serves the interest of manpower policy by permitting many aged workers who are partially disabled or for whom work is especially burdensome to withdraw from the labor force, thereby opening greater opportunities for other employed and unemployed workers. In occupa tions and industries in which unemployment is especially severe, retirement reduces the level of unemployment and facilitates the employment adjustments essential in a rapidly changing economy. Changes in retirement programs, both public and private, will have manpower effects by making retirement more or less desirable and thus in fluencing the incentive for aged workers to remain in the labor force. One example of such a change is the recent adoption of early retirement provisions in both public and private programs. It is yet too soon to evaluate the effect of the recent change permitting early retirement at age 62 under the OASDI program. The decentralized nature of private pension systems provides an important element of flexi bility in meeting manpower problems. While special provisions applying to the needs of partic ular industries have been adopted in the public retirement programs of some countries, such measures would be difficult to justify in the OASDI system, in which uniformity of retirement age and benefit schedules for all covered workers is a well-established principle. Private pensions are well suited to meet such problems, since the content of the plans may be varied to meet the special needs of an industry or firm. Illustrative of the flexibility inherent in pensions negotiated through decentralized collective bargaining, are provisions for early retirement at employer’s request included in many of the United Auto mobile Workers agreements which provide twice the normal pension from the time the employee retires until he becomes eligible for OASDI benefits. Aged Workers in the Labor Force Nearly all men between the ages of 25 and 64 years work or look for work during the year, but women, young people, and aged people (persons 65 years and older) are less frequently in the labor force. The proportion of aged men in the labor force decreased from about two-thirds in 1890 to 30 less than one-third in 1961. The rate of decrease has been particularly rapid since 1950. Since 1890, the proportion of women in the labor force has increased steadily, while the proportion of young people has fallen off somewhat as high school and college attendance became more common. Public and private pensions played no part in the decline in labor force participation of aged men before 1940. Most of this earlier decline is attributable to rising levels of income, to the shift from rural to urban patterns of living, and to mass unemployment during the 1930’s which caused many aged men to cease looking for work.39 Wartime labor demands kept many aged men at work, and their labor force rate increased during the war. With peace, however, labor force participation of aged men resumed its decline (table 4.1). In 1950 when OASDI benefits were being received by 1.8 million retired workers (mostly men), and private pensioners numbered about 450,000, 46 percent of men 65 years and older were in the labor force. By 1960, the number of aged workers receiving OASDI benefits increased to 8.1 million, and approximately 1.8 million people (mostly aged men) received private pensions. The labor force rate of aged men had fallen to 32 percent. The improvements in OASDI benefits, coverage, and eligibility since 1950 now make retirement income available to practically all workers in private industry reaching age 65. Many men eligible for private pensions or the pensions provided by the Railroad Retirement Act and by Federal, State, and local governments often receive sufficient income to finance a comfortable retirement. The ready availability of retirement income has probably been the most important influence leading to the recent reduction in the labor force participation rate of aged men. The middle or lower income worker receiving OASDI benefits and a supplemental private pension will frequently have a retirement income greater than one-half of his preretirement earnings. Because of the favorable tax treatment of OASDI benefits 39 For a discussion of labor force trends since 1890, see Gertrude Bancroft, The American Labor Force (New York, Wiley, 1958); Clarence D. Long, The Labor Force Under Changing Income and Employment (Princeton, Princeton University Press, 1958); and John C. Durand, The Labor Force in the United States, 1890-1960 (New York, Social Science Research Council, 1948). T a b l e 4.1. P e r c e n t of P o p u l a tio n in t h e L abor F orce, by Se x and A g e , 1940-61 [Annual averages] Male Years 1961______________ 1960______________ 1959______________ 1958______________ 1957______________ 1956....... .................. 1955______________ 1954______________ 1953______________ 1952______________ 1951______________ 1950____________ 1949__________ 1948__________ _ 1947______________ 1946______________ 1945______________ 1944______________ 1940______________ Female 14 years and older 14 to 19 years 20 to 24 years 25 to 34 years 35 to 44 years 45 to 54 years 55 to 64 years 65 years and older 14 years and older 14 to 19 years 20 to 24 years 25 to 34 years 35 to 44 years 45 to 54 years 55 to 64 years 80.3 80.1 80.5 82.1 82.7 83.7 83.6 83.9 84.4 84.6 84.8 84.4 84.5 84.6 84.4 83.7 88.0 89.7 83.9 44.6 46.3 46.7 47.4 49.7 51.4 49.5 49.3 50.9 51.9 53.7 53.2 53.6 54.3 54.2 53.7 64.9 70.0 44.2 89.8 88.9 88.6 89.5 89.8 90.8 90.8 91.5 92.2 92.0 91.0 89.0 87.7 85.6 84.8 82.3 95.5 98.5 96.1 97.6 96.4 96.3 97.3 97.3 97.4 97.7 97.5 97.6 97.7 97.1 96 2 95.9 96.0 95.8 94.2 97.0 99.0 98.1 97.7 96.5 96.6 98.0 97.9 98.0 98.1 98.1 98.2 97.9 97.6 97.6 98.0 98.0 98.0 97.3 98.2 99.0 98.5 95.6 94.5 94.6 96.3 96.4 96.6 96.5 96.5 96.6 96.2 96.0 95.8 95.6 95.8 95.5 96.1 96.6 97.1 95.5 87.3 85.4 85.9 87.8 87.5 88.5 87.9 88.7 87.9 87.5 87.2 87.0 87.5 89.5 89.6 89.6 91.4 92.1 87.2 31.7 32.3 33.4 35.6 37.5 40.0 39.6 40.5 41.6 42.6 44.9 45.8 46.9 46.8 47.8 48.5 52.1 52.2 45.0 36.9 36.4 35.8 36.0 35.9 35.9 34.8 33.7 33.6 33.9 33.8 33.1 32.4 31.9 31.0 31.1 36.2 36.8 28.2 29.9 30.2 29.1 29.1 30.6 31.9 29.9 29.8 30.5 31.5 32.1 31.5 32.5 32.5 31.6 32.3 40.0 42.0 23.3 47.1 46.0 44.9 46.4 46.0 46.4 46.0 45.3 44.5 44.8 46.6 46.1 45.0 45.3 44.9 46.3 54.1 55.0 49.5 36.4 35.9 35.2 35.6 35.6 35.4 34.9 34.5 34.1 35.5 35.4 34.0 33.5 33.2 32.0 32.9 38.9 39.0 35.2 43.8 43.2 43.1 43.4 43.3 43.1 41.6 41.3 41.3 40.5 39.8 39.1 38.1 36.9 36.3 36.2 39.8 40.5 28.8 50.1 49.4 48.6 47.9 46.5 45.5 43.8 41.2 40.4 40.1 39.7 38.0 35.9 35.0 32.7 31.5 35.2 35.8 24.3 37.9 36.8 36.2 35.2 34.5 34.9 32.5 30.1 29.1 28.7 27.6 27.0 25.3 24.3 24.3 23.6 26.5 25.4 18.7 65 years and older 10.7 10.5 9.9 10.3 10.5 10.9 10.6 9.3 10.0 9.1 8.9 9.7 9.6 9.1 8.1 8.4 9.6 9.8 7.4 Source : Bureau of Labor Statistics and Bureau of tbe Census. and double exemptions for persons 65 and over, the after-tax income of the retired married worker is sometimes as large as his preretirement income. The proportion of aged men who remain in the labor force also depends on the opportunity to continue work. Unemployed aged workers who might prefer to work may retire because they are unable to find suitable jobs. The postwar period has been marked by relatively high employment, despite four recessions, and relatively few aged workers have suffered unemployment previous to retirement. Nevertheless, employment oppor tunities for aged men have been diminishing and undoubtedly this has effected the propensity of aged men to continue seeking work. The effect of employment opportunity on re tirement is also shown by the relation of the changes in the number of aged men employed in various occupations to the changes in the number of all men employed. In professional and tech nical, managerial, and sales occupations, in which the number of employed men has increased sharply since 1952, the number of aged men has also increased; while in the industrial occupations, more than one-half of the decline in employment of all men is accounted for by the drop in the number of employed aged men (table 4.2). Peter O. Steiner and Robert Dorfman, The Economic Status of the Aged (Berkeley, University of California Press, 1957), table 4.3, p. 41. 41 Bureau of Labor Statistics, Educational Attainment of Workers, 1959, Special Labor Force Report No. 1, Reprint 2333 (1960) from Monthly Labor Review, February 1960, table E, p. A-9. Employed aged men are concentrated dispro portionately in occupations where self-employment is common. In 1961, about two-fifths of the employed aged men were farmers or managers, while only one-fifth were in the industrial occupa tions (craftsmen and foremen, operatives, and nonfarm laborers). This represents a considerable change from 1952, when larger proportions of the employed aged were in industrial occupations, (table 4.3). Labor force data from a study of aged men conducted in 1952 show considerable variation between different occupation groups. About two-thirds of the aged men, whose lifetime occupation was in professional and technical work, were in the labor force, but only one-third of the aged men whose lifetime industrial occupations were in the labor force.40 Labor force participation of aged men is also related to educational attainment. About threefifths of the aged men with 5 years or more of college were in the labor force in March 1959, while only one-fourth of those with 4 or fewer years of school were in the labor force.41 An important labor force trend associated with pensions has been the tendency of retired workers to seek part-time employment to supplement their pensions. The proportion of all aged men who worked primarily at part-time jobs during the year increased from 12 percent in 1950 to 17 percent in 1960, and the proportion of aged men who worked primarily at full-time jobs decreased from 38 percent to 27 percent (table 4.4). Some 31 T a b l e 4.2. E m p l o y m e n t by O c cu p atio n G r o u p s of 1952 A ged M e n a n d 1961 and A ll M e n , and C hang es in E m ploym ent, [Number in thousands] Occupation group Men, 14 years and older Increase or decrease (—) Men, 65 years and older Increase or decrease (—) 1952 1961 Number Percent 1952 1961 Number Percent T otal1...................................... ......................................... 42,349 44,318 1,979 4.7 2,351 2,071 -280 -1 1.9 Professional and technical..................... ............... .......... Farmers....... .............. ........................ ............ ................... Managers.......................... .................................. ................ 3,137 3,688 5,089 4,955 2,581 6,003 1,818 -1,107 914 58.0 -3 0.0 18.0 118 495 235 185 423 399 67 -7 2 164 56.8 -14.5 69.8 Clerical............................ ................................................... Sales.................................. .................................................. Craftsmen..... ..................................................................... 2,798 2,247 8,478 3,120 2,737 8,407 322 490 -7 1 11.5 21.8 -.8 118 95 400 118 138 259 0 43 -141 0 45.3 -3 5.2 Operatives............................................................................ Service..................................... ........................................... Farm laborers...................................................................... Other laborers..................................................................... 8,817 2,585 1, 780 3,730 8,441 2,992 1,685 3,397 -376 407 -9 5 -333 -4 .3 15.7 -5 .3 -8 .9 283 259 95 259 165 233 74 111 -118 -2 6 -2 1 -148 -4 1.7 -1 0.0 -22.1 -57.1 1 Detail will not add to total because of rounding. Source : Derived from Peter O. Steiner and Robert Dorfman, The Eco nomic Status of the Aged (Berkeley, University of California Press, 1957), of this change is probably attributable to workers who shifted from a full-time schedule to a parttime schedule without changing jobs. Aged men who work part time usually prefer part-time work. In 1961, 35 percent of the aged men at work in the average week in non agricultural industries worked part time, and four-fifths of these usually worked part time because they did not prefer or could not accept full-time work.42 The increased importance of voluntary part-time work for aged men underlines the growing impor tance of pensions as the primary source of income for aged men, and the increasingly supplementary role of earnings. Relation of Pensions to Retirement Corporate pensions may induce retirement by two effects: (1) pensions provide income which, together with OASDI benefits, enable some work ers to retire voluntarily; and (2) involuntary re tirement practices often found in pension plans require some aged workers to retire who would otherwise continue work. Income. The provision of retirement income by pensions has a major effect on voluntary retire ment. A 1951 study of pensioners found that about two-fifths of those with incomes of less than $1,000 regarded retirement as satisfactory, while almost three-fourths of the pensioners with in comes of $5,000 or more found retirement satis factory.43 Between 1954 and 1958, almost fourfifths of the men retiring from the Eastman Kodak 32 table 4.4, p. 42; and Bureau of Labor Statistics, Labor Force and Employment in 1961, Special Labor Porce Report No. 23, Reprint 2395 (1962) from Monthly Labor Review, June 1962, tables C-6 and C -8, pp. A-23 and A-25. Co. (which has a generous pension plan) retired at age 65 or earlier, although later retirement was permitted.44 The need for income appeared to be the principal reason for men 64 years old wanting to continue work after age 65 in a survey of firms which had varying retirement policies.45 Obvi ously, corporate pensions play an important role in inducing retirement by providing necessary income. In most pension plans, 65 is the normal retire ment age; i.e., the age at which retiring workers receive full pensions. The selection of this partic ular age follows the precedent established by the public retirement system— 65 is also the normal retirement age for OASDI. Retirement at age 62 for men was adopted in the 1961 social security amendment, following earlier adoption of the same age for women. This amendment was in tended to permit unemployed men or partially disabled men to retire earlier by providing retire ment income. Similar provisions for earlier re tirement are included in many private pension plans. Early retirement clearly increases the flexi bility of private pension plans in meeting the needs of aging workers. Many private pension plans provide for retirement before age 62 and allow in 42 Bureau of Labor Statistics, Labor Force and Employment in 1961, Special Labor Force Report No. 23, Reprint 2395 (1962) from Monthly Labor Review, June 1962, tables C-6 and C -8, pp. A-23 and A-25. 43 John J. Corson and John W. McConnell, Economic Needs of Older People (New York, Twentieth Century Fund, 1956), p. 47. 44 Marion B. Folsom, “ Goals in Governmental and Private Plans for Social Security,” Address given on the 25th Anniversary of the Social Security Act, Washington, D.C., Aug. 15, 1960, p. 16. 45 National Committee on the Aging, “ Work Attitudes at Age 65“ (New York, 1959), p. 5. come equalization options which provide larger pensions during the period between the worker’s retirement from the company and the time at which the worker becomes eligible for OASDI. Although early retirement is usually at the worker’s request, retiring from a company may simply be the beginning of a search for another job. Few private pension plans provide early retirement benefits which are adequate to provide for comfortable retirement, especially if retire ment takes place several years before the worker is eligible for OASDI benefits. Although a large, but unknown, number of workers receive early retirement benefits from private pensions, there is little evidence that the spread of private pen sions with early retirement has led many men 55 to 64 years old to withdraw from the labor force. The labor force rate of men 60 to 64 years old, for whom early retirement provisions are most significent, has edged down in the past few years, but no steady trend is observable. Nevertheless, the early retirement in OASDI has not yet shown its full effect, and early retire ment accompanied by withdrawal from the labor force may become more common in the next few years. Involuntary Retirement Many pension plans specify that workers must retire at a given age (often 65, 68, or 70 years) without exception. Such rules are here termed automatic retirement, following the practice of the Bureau of Labor Statistics. Whenever exceptions may be made at the discretion of management, T a b le 4.3. E m pl o ym e n t of A ged M e n , b y M a jo r O ccu p a tio n G r o u p s , 1952 a n d 1961 Percent of employed men Occupation group 14 years and older 1952 1961 65 years and older 1952 1961 Total employed............................ 100 100 100 100 Professional and technical______ Farmers and farm managers........ Managers and proprietors............ 7 9 12 11 6 14 5 21 10 9 20 19 Clerical and related...................... Sales workers................................ Craftsmen and foremen............... 6 6 19 7 6 19 5 4 17 6 7 13 Operatives,................................... Service.......................................... Farm laborers............................... Other laborers............................... 21 6 4 9 19 7 4 8 12 11 4 11 8 11 4 5 Source : See table 4.2. T a b le 4.4 P e r c e n t of A ged M e n W ith P a r t - T im e F ull - T im e J o b s , 1950-60 Year Percent of aged men with work during the year and Percent of all aged men with— Part-time jobs Total Full-time jobs 50-52 1-49 Total weeks weeks 50-52 1-49 weeks weeks 1960...................... 43.1 16.6 6.7 9.9 26.6 16.8 9.8 1959..................... 1958____________ 1957..................... 1956............... ___ _ 1955_______ _____ 42.4 43.4 47.3 46.4 48.1 14.6 15.0 15.1 14.5 12.7 5.8 6.6 6.3 5.9 5.5 8.8 8.4 8.8 8.6 7.2 27.8 28.4 32.2 31.9 35.4 18.0 18.5 21.5 22.5 24.5 9.8 9.9 10.7 9.4 10.9 1954_____ ______ 1953____________ 1952____________ 1951____________ 1950........ ............ 45.9 48.2 50.3 51.1 49.3 11.6 14.8 11.3 11.3 11.6 4.3 3.8 4.5 4.1 7.3 11.0 6.8 7.2 34.3 33.4 39.0 39.8 37.7 23.3 24.6 26.0 28.0 25.8 11.0 8.8 13.0 11.8 11.9 0) (0 1 Not available. Source : Derived from Bureau of the Census, Current Population Reports, Labor Force, Series P-50, Nos. 43, Mar. 13, 1953; 48, Nov. 30, 1953; 54, Aug. 4, 1954; 68, .Tune 1956; 77, November 1957; 86, September 1958; and 91, June 30,1959, and Bureau of Labor Statistics, Work Experience of the Population in 1959, Special Labor Force Report No. 11, Reprint 2360 (1961) from Monthly Labor Review, December 1960; and Work Experience of the Population in I960, Special Labor Force ReportNo. 19, Reprint 2381 (1962) from the Monthly Labor Review, December 1961. the rules will be termed compulsory retirement• The frequency with which exceptions to com pulsory retirement rules are made varies among firms, of course, and sometimes compulsory retirement may be enforced as rigidly as if the firm had automatic retirement. In many collec tively bargained plans, retirements under com pulsory retirement rules are made subject to the grievance and arbitration machinery established under the contract. In firms with workers on layoff, even voluntary retirement rules may be virtually automatic in practice because of worker opinion favoring retirement of those who are eligible. If compulsory retirement is administered with latitude and with attention to the desires and abilities of the aged workers, or if there is no involuntary retirement age, the firm will be said to havoflexible retirement. The Bureau of Labor Statistics studied 300 collectively bargained pension plans covering 4.9 million workers in late 1958, and found that in voluntary retirement provisions were included in 179 plans covering about 2.7 million workers. Among the plans studied, involuntary retirement was found in all of the plans covering the chemical, products of petroleum, and electric and gas utility industries, and in most of the transportation plans. Involuntary retirement was found in none of the plans in construction and apparel and in few of the primary metals industry plans studied. 33 Of the 179 plans with involuntary retirement, compulsory retirement was found in 109, auto matic retirement (without earlier compulsory retirement) in 52 plans, and compulsory and automatic retirement at different ages in 18 plans. Involuntary retirement was found in about one-half of the 249 noncontributory plans but in more than nine-tenths of the 51 contributory plans. Of the 231 single-employer plans, about one-half (covering more than two-thirds of the 3 million workers) included involuntary retire ment; of the 69 multiemployer plans, about oneseventh (covering one-fourth of the 1.9 million workers) included involuntary retirement. Compared to a similar study in 1952, there was little evidence of any significant change in the prevalence of involuntary retirement provisions.46 The Bankers Trust Co. studies of pension plans found that 83 percent of the large con ventional (usually nonnegotiated) plans adopted or amended in 1956-59 included involuntary retirement provisions, while 94 percent of such plans in the 1953-55 period included involuntary retirement. Single-employer negotiated plans showed an opposite trend, with 72 percent of the new or amended plans in the later period having involuntary retirement in contrast to 67 percent of the 1953-55 plans.47 Involuntary retirement is found chiefly in single-employer plans, and is more often included in nonnegotiated than in negotiated singleT a b l e 4.5. N o r m a l , C o m p u l so r y , a n d A utom atic R e t ir e m e n t A ges in 300 P e n s io n P l a n s U n d e r C ollec t iv e B a r g a in in g , L ate 1958 [Workers in thousands] Age i Normal retirement Compulsory retirement Automatic retirement Plans Workers Plans Workers Plans Workers All plans with retire ment provisions_____ Age 55_______________ Age 60__ _ _ ___ Age 65__ _ ______ ____ Age 66______ ___ _ __ Age 67_________ _ Age 68_____ ______ Age 69 Age 70. _ _ Age 74 Age 75_._ __________ Other 2______ ______ 300 4,909.8 127 1,927.9 70 919.8 1 15 282 3.0 579.6 4, 289.2 82 1.006.9 24 2 4 17 170.0 19.7 21.8 169.9 1 7.8 22 1 453.4 85.0 1 35 1 8 1 888.1 1.0 28.0 3.9 30.2 1 An earlier normal, compulsory, or automatic retirement age for women was provided in some plans. 2 Normal retirement benefit provided when age plus years of service equal 80. Source : Bureau of Labor Statistics, Pension Plans Under Collective Bargaining (BLS Bulletin 1259, 1959), p. 23. 34 employer plans. Despite the opposition of many union leaders to involuntary retirement, a sub stantial majority of single-employer negotiated plans include involuntary retirement, and there does not appear to be any significant trend toward elimination of such provisions. Involuntary retirement is more frequently in large than in small firms. A 1955 study by the National Industrial Conference Board found that 58 percent of the plans of firms with fewer than 1,000 employees had involuntary retirement, while 90 percent of the plans in firms with more than 1,000 employees had such provisions. A study of California pension plans found the following relation between size of firm and the presence of involuntary retirement provisions:48 Number of employees 1 0 0 -1 9 9 __________________________________ 2 0 0 -4 9 9 __________________________________ 5 0 0 -9 9 9 __________________________________ 1,000 or m o re ___________________________ Percent of firms with involuntary retirement 54 65 80 69 Involuntary Retirement Age. The age at which involuntary retirement is required in pension plans varies among firms. In the BLS sample of 300 collectively bargained pension plans in late 1958, it was found that in no instance was involuntary retirement found at an age below 65. About twothirds of the 127 plans with one-half of the 1.9 million workers subject to compulsory retirement had age 65 as the compulsory retirement age. Age 68 was the next most common, but one fairly small plan had a compulsory retirement age of 75. Almost two-fifths of the 70 plans with auto matic retirement (or almost one-half of the 52 plans with automatic retirement and without com pulsory retirement at an earlier age) specified age 65 as the automatic retirement age. Of the 70 plans with automatic retirement, 22 plans with almost one-half of the 9 million workers subject to automatic retirement specified age 70 as the automatic retirement age (table 4.5). It is apparent that relatively few of the work ers covered in the plans studied are forced from their jobs at age 65 because of automatic retire« Bureau of Labor Statistics, Pension Plans Under Collective Bargaining (BLS Bulletin 1259,1959), p. 20. u a Study of Industrial Retirement Plans, 1960 (New York, Bankers Trust Company, 1960), pp. 11-12. Governor’s Commission on the Employment and Retirement Problems of Older Workers, Recommendations and Report of the Employment and Retirement of Older Workers, 1960, p. 11. T a b l e 4.6. N orm al a n d C o m p u l so r y R e t ir e m e n t A ges in C o n v e n t io n a l P l a n s , 1956-59 a n d 1953-55 Retirement age (men) 1 Normal Compulsory 60____________________ 65____________________ 65_ _________________ 65____________________ 65____________________ ______ 67 68____________________ 70 _____ None_________________ None______ _ ______ 65____________________ 68____________________ 70____________________ 67. _ _________________ 68____________________ 70__ ________________ Percent 1956-59, new plans 4 13 72 2 2 7 1953-55, new plans 6 74 5 8 2 3 2 i Retirement ages for women are occasionally lower. Source : 1960 Study of Industrial Retirement Plans (New York, Bankers Trust Co., 1960), p. 12. ment, and if compulsory retirement provisions are administered with much flexibility, relatively few are forced out by these provisions. Involun tary retirement at ages 68 or 70 is clearly of much less importance in compelling retirement than it would be if it were entirely concentrated at age 65. Nonnegotiated plans typically have compulsory or automatic retirement, and in most instances specify age 65 as the involuntary retirement age (table 4.6). Only 4 percent of the new plans adopted in 1956-59 and studied by the Bankers Trust Co. have a compulsory retirement age higher than the normal retirement age. Conventional plans are much more uniform in their approach to involuntary retirement than are negotiated plans. However, the plans studied by the Bankers Trust are all large plans, and smaller conventional plans probably include involuntary retirement less often than do larger plans. E m ployer Attitudes. Opinion concerning the de sirability of involuntary or flexible retirement differs among employers. Some firms vigorously resist efforts of unions to remove involuntary provisions from negotiated pension plans, occa sionally even accepting a strike rather than giving in, while other firms do not want involuntary retirement.49 Margaret S. Gordon found in a survey of retirement practices in the San Fran cisco Bay region that employers frequently gave the following reasons for adopting involuntary retirement: <9 For example, see Melvin K. Bers, Union Policy and the Older Worker, op. cit., pp. 71-74. 50 Paraphrased from Margaret S. Gordon, “ The Older Worker and Retirement Practices,” Monthly Labor Review, June 1960, p. 581. fii Paraphrased from Governor’s Commission, Report, op. cit., p. 119. 1. A uniform retirement policy avoids discrimination among employees. 2. An involuntary retirement policy improves employ ment and promotional opportunities for younger workers. 3. Older workers are less productive or less flexible. 4. An involuntary retirement policy reduces uncertainty and thereby encourages older workers to plan for retire ment.50 Employers favoring flexible retirement stress the following reasons: 1. Age alone is not an appropriate basis for separation. 2. The firm loses the services of desirable workers if involuntary retirement is practiced. 3. Flexible retirement reduces pension costs since some workers delay receipt of their pensions and continue work. 4. Employee morale is improved by flexible retirement.51* The diversity of employer practice in the matter of involuntary retirement is probably related to the kind of work which is performed and to the conditions of labor supply in the industry. During World War II, for instance, some firms abandoned involuntary retirement rules of long standing, but resumed them when the labor shortage subsided. Attitudes toward retirement practices are strongly influenced by the desire of employers to provide an orderly method of removing aged workers from their payrolls. The operation of seniority rules and the tendency of older employees to stay on the job naturally leads to an increasing proportion of older workers in any firm’s work force unless the firm’s employment is expanding rapidly. Thus the firm may find itself operating with reduced efficiency both through decreasing efficiency of workers as they age and by losing the more able younger workers whose promotion opportunities have been blocked. A principal advantage of a pension plan is that it provides a way of separating overage employees from employ ment without facing public disapproval or lengthy grievances. Rapidly expanding firms are unlikely to face these problems, especially if the firm is in an area of rapidly expanding population where the proportion of older jobseekers is low. Union Attitudes. Union leaders often view the retirement provisions of a pension agreement as one means of meeting the needs of diverse groups of the rank and file. Industries with heavy employment impose special problems for union 35 leadership, since the older workers usually have high seniority and may be eligible for early retirement, while younger members with low seniority may be on layoff. As a principle, union leaders have nearly always opposed involuntary retirement, viewing it as an arbitrary limitation on the worker's right to his job. The Steelworkers, for instance, opposed the return to involuntary retirement by the United States Steel Co. after World War II, and in the 1949 pension negotiations won the right to joint determination in instances when the company requests a worker to retire.52 The United Automobile Workers have opposed in voluntary retirement practices, although the automobile pattern includes compulsory retire ment at age 68.53 Despite union opposition to involuntary retire ment as a principle, however, unemployment in particular industries often leads to more receptive attitudes toward involuntary retirement or to efforts to make voluntary retirement more at tractive. Pensions have made up a large part of the packages negotiated in many industries during the past decade, and union members who are not near the retirement age may be expected to apply pressure for retirement practices which would reduce the competition for jobs. In some unions this has already occurred. Automatic retirement (for workers eligible for pensions) was included at the union's request in the Pacific Maritime Association-International Longshore men's and Warehousemen's Union pension plan.54 The problems of unemployment in the auto mobile industry have led unions and management to negotiate a provision for “ involuntary early retirement," which provides that workers between ages 60 and 65 retired at the employer's request or under mutually satisfactory conditions shall receive a double retirement benefit of $5.60 per month for each year of service until the worker becomes eligible for OASDI, at which time the benefit reverts to the normal benefit formula of $2.80 per month for each year of service. Other Opinions. Opposition to involuntary retire ment is often found in the recommendations of study groups concerned with the problems of aging. The policy statement on employment security and retirement of the 1961 White House Conference on Aging urged employers to review 36 their retirement policies “ for the purpose of finding ways and means of achieving greater flexibility in the range of time of retirement. . . " While urging flexible retirement, however, the Conference opposed legislation “ . . . limiting the independent determination of the time of retirement." 55 The Committee on Economic Needs of Older People, appointed by the Twentieth Century Fund, urged that employers and unions should consider— Substituting for requirements that workers retire at a fixed age methods to distinguish those who should be retired because of the decline in their working capacities— physical, mental, or both.56 The California commission charged with making recommendations concerning the employment and retirement of older workers criticized involuntary retirement as follows: . . . the fact that age 65 has long been regarded as the usual retirement is neither a necessary nor a sufficient reason for perpetuating the practice of compelling retire ment at age 65. It is true that a uniform and immutable rule that every employee in a given enterprise must retire at age 65 has the virtue of certainty; but to say that a rule is socially desirable because it avoids discrimination is to oversimplify the facts. W e have ample evidence to prove that all persons within a given enterprise who reach the chronological age of 65 are not the same in respect to working abilities, health, or preparation for retirement. Continuing, and pointing out that especially difficult personnel problems may arise where retirement is completely voluntary, the Com mission urged that either flexible retirement policies be adopted or the involuntary retirement age be increased.57 The informed consensus appears to urge recon sideration by concerned parties of the practice of involuntary retirement, but few observers urge the intervention of government to prohibit involuntary retirement at age 65 or higher. Effects of Involuntary Retirement. The effect of involuntary retirement practices associated with w Bers, Union Policy and the Older Worker, op. cit., p. 74. *3 Ibid. m Ibid., p. 80. 55 Special Staff on Aging, U.S. Department of Health, Education, and Welfare, The Nation and Its Older People (Washington, April 1961), pp. 144-145. fie Corson and McConnell, op. cit., p. 476. Governor’s Commission, op. cit., pp. 11-1 i pension plans in forcing retirement is difficult to assess. In April 1952, a survey of a national sample of the aged found that 13 percent of the aged men who were not in the labor force had been compelled to retire because of involuntary retire ment systems. Only 1 percent of aged men were in the labor force after being compelled to retire (12 percent of all men retired under compulsory systems). Of those retired under retirement systems who considered themselves well enough to work and were not working, only one-third were interested in work.68 In the 1951 survey of pensioners previously referred to, 45 percent of all pensioners were compelled to retire by their employer on account of age. Involuntary retire ment appears to be the predominant reason for retirement among those pensioners with incomes of $2,000 a year or more (about 65 percent of all reasons), while it was less important for pensioners with retirement incomes of less than $1,000 (46 percent of reasons).69 Since the higher income pensioners also tended to be more satisfied with retirement than the lower income pensioners, it seems likely that there is not much resentment of pensioners against firms which practice compul sory retirement. This is supported by two other surveys. A survey of industrial workers in the San Francisco Bay region found that 42 percent of the employed men 50 years and older accepted the idea of a fixed retirement age without qualifi cation, and only 12 percent thought that there should not be a fixed retirement age.*5 60 A survey 9 of workers 64 years old found that three-fifths of the workers both in firms without automatic retirement and in firms with automatic retirement at age 68 or above wanted to continue work, while only two-fifths of those in firms with automatic retirement at age 65 wanted to continue work.61 From this evidence, it appears that although involuntary retirement probably does induce some workers to retire who might otherwise continue work for several years, many workers who are compelled to retire would probably retire shortly anyway. At any point in time, many s8 Peter O. Steiner and Robert Dorfman, The Economic Status of the Aged, op. cit., pp. 48-49. 59 Corson and McConnell, op. cit., p. 75. 6° G. Hamilton Crook and Martin Heinstein, The Older Worker in Industry (Berkeley, Institute of Industrial Relations, University of California, 1958), p. 50. 6! National Committee on the Aging, op. cit., p. 4. 62 Governor’s Commission Report, op. cit., p. 115. Permission was granted infrequently in 25 percent, and very rarely in 53 percent. aged men who originally retired because they were compelled to are no longer willing to work, either because of illness or because they are now satisfied with retirement. The likelihood of the involuntarily retired worker withdrawing from the labor force is enhanced by the fact that he is separated from a job in which he usually has had long tenure, and at the same time is provided a pension which often enables him to maintain his standard of living fairly well. The spread of involuntary retirement would have fewer consequences for labor force trends if compulsory retirement provisions were adminis tered with considerable latitude, since automatic retirement rules are much less frequent than compulsory retirement rules. There is little information available on this question, but the California pension study suggests in only onefifth of the plans with compulsory retirement is permission to continue work granted frequently.62 If this represents practices across the Nation, then in many instances compulsory retirement and automatic retirement have equivalent effects in compelling the vast majority of covered workers to retire at the involuntary retirement age. Much of the future growth in pension coverage will be concentrated in smaller firms, where coverage is not now very high. Many smaller firms will be included in multiemployer plans, in which involuntary retirement provisions cover only one-fourth of the workers. Small firms with their own plans are less likely to include involun tary retirement than are large firms, perhaps because of the higher pension costs required by involuntary retirement rules. Moreover, those small firms which do include involuntary retire ment will probably administer compulsory rules with somewhat more flexibility than can be expected from a large firm. The foregoing comments suggest that the spread of pension coverage to a larger proportion of the private work force will be accompanied by the spread of involuntary retirement, but that the proportion of older workers subject to involuntary retirement may increase at a some what lower rate. The spread of involuntary retirement and the much wider availability of retirement income made possible by private pensions will probably serve to continue the downtrend in labor force participation of aged men in the immediate future. 37 U.S. GOVERNMENT PRINTING 0 F F I C E : I 9 6 3
@FedFRASER