Printer Friendly

Institutional barriers to employment of older workers.

Diane E. Herz and Philip L. Rones are economists in the Division of Labor Force Statistics , Bureau of Labor Statistics. This article has been adapted from a broader report submitted to Congress in January 1989.

Social Security and private pension rules often encourage early retirement; those who wish to keep working typically face a part-time job market which provides few well-paid jobs

Retirement ages have fallen steadily since World War 11. Early retirement, a term often used to describe labor force withdrawal prior to age 65, has become the norm. By age 62, almost half of all men are out of the labor force (they are neither working nor looking for work), and by age 64, three-fifths are "retired" by that definition.1

The trend toward early retirement has generally been regarded as a positive one, as it primarily reflects improvements in retirement resources-Social Security, pensions, and wealth. Some retirements, however, may not be strictly voluntary; rather, they may be in response to actual or probable job loss, or to a lack of acceptable job opportunities for older workers. These retirements may be voluntary only to the extent that leaving the labor force is the best available option. Some workers might prefer another choice-either phased retirement or a "second career" upon job loss or pension acceptance. But this choice is often not feasible because of institutional rules and job market realities. Usually, workers must choose between continued full-time employment (during which some pension benefits often are lost), part-time work for relatively low wages, or complete retirement. Given these limitations, many workers choose complete retirement.

Anticipating a dramatic decline in the ratio of workers to retirees as the baby-boom generation becomes eligible for retirement early in the next century, Federal policy has been directed toward encouraging workers to extend their worklives. 2 Social Security regulations have been altered to encourage later withdrawal from the labor force, and mandatory retirement has been prohibited-regardless of age. Also, researchers are examining ways in which work schedules and environments could be designed to address the needs and desires of older persons.

At the same time that Federal efforts have been directed at extending worklives, however, an opposite and, for some workers, more dominant force has influenced retirement decisions; many employers have made early retirement possible or have liberalized options already offered in their pension plans. Such efforts may be the result of labor management negotiations, employers' desires to attract workers by offering the prospect of generous and early pensions, or employers' attempts to reduce employment by inducing older workers to retire earlier than they might have planned. Both by liberalizing provisions in their normal pension plans and by offering Early Retirement Incentive Plans (ERIP'S), employers have made retirement increasingly affordable for many older workers.

This article discusses various institutional obstacles faced by older persons who might want to work. It is divided into three sections: (1) the impact of Social Security regulations and pension policies on work activity; (2) the market for part-time jobs; and (3) age discrimination.

Social Security

Social Security benefits are the major source of income for the elderly. In 1986, 9 of 10 nonmarried persons or married couples in which the husband was age 65 or older received some portion of their income from Social Security, and 60 percent relied on Social Security for more than half of their total income.' Thus, changes in Social Security policies could affect the majority of older persons.

Researchers have examined the effects of Social Security rules on the work activity of older persons for many years. While findings are sometimes contradictory, most studies have concluded that Social Security regulations provide substantial disincentives to work at older ages. Policymakers concerned about reducing financial burdens on the Social Security system have recognized these disincentives to work; several reforms designed to encourage older persons to remain in the labor force were passed as part of the Social Security Amendments of 1983.4 Operations of the current program, evidence that work disincentives exist, and long-term reforms to the program and their potential effects are discussed below.

How the current program works. Social Security benefits are based on lifetime eamings in covered employment. While the process of determining benefits is complex, in general, annual earnings between 1951 and the year of eligibility for retirement first are averaged and adjusted for inflation to derive an Average Index of Monthly Earnings (AIME). A benefit formula is then applied to the AIME to determine an individual's full benefit amount-or Primary Insurance Amount (PIA). The percentage of the PIA that an individual actually receives depends on both age at retirement and current earnings, if the individual continues to work.5

Currently, individuals are eligible to receive full benefits (equal to 100 percent of PIA) at age 65--the "normal" retirement age as defined in the Social Security program. Reduced benefits equal to 80 percent of the PIA are available at age 62. For every month after age 62 that receipt is deferred, the 20-percent early retirement penalty is reduced by 0.56 percent (or 6.67 percent per year) so that the full PIA level is earned at age 65. If an individual chooses to continue working beyond age 65, he or she receives a delayed retirement credit of 3 percent per year. For example, a person working (and deferring Social Security receipt) to age 68 could expect to receive benefits equal to 109 percent of his or her determined PIA.

Social Security rules have been based on a longstanding policy that benefits are insurance against earnings lost due to retirement. Therefore, recipients who work may only earn up to a specified exempt amount before their Social Security benefits are reduced. Currently, Social Security recipients younger than age 65 can earn up to $6,480, beyond which point their benefit amount is reduced by $1 for every $2 earned, Workers ages 65 to 70 can earn $8,800 before benefit reductions begin. After age 70, these reductions no longer apply.

Effects on work incentives. Studies have examined how these regulations may affect the work activity of older persons. Most studies have indicated that disincentives to work do exist, but a few have not. After examining the benefit stream available to an "illustrative" worker at various retirement ages in 1982, Gary Fields and Olivia Mitchell concluded that, although the level of benefits increased for each year of additional work, gains from additional benefits were more than offset by the smaller number of years of benefit receipt. 6 In fact, the present value of total future benefits for a person who worked until age 68 was only 90 percent of that for a worker who retired at age 60. This "penalty" for retiring at age 68 is largely the result of the 3 -percent delayed retirement credit being far below the actuarially neutral level (the level at which the value of benefits for an average worker would be the same regardless of when he or she retires). Thus, Fields and Mitchell concluded, the current system provides incentives to retire before age 65 rather than after. Studies by other researchers have produced similar results.7

The effect of the earnings test on work activity has also been widely examined. Using data from the Social Security Newly Entitled Beneficiary Survey (NBS), Howard M. lams found that male beneficiaries in 1982 had median earnings of $4,391 -just below the $4,400 earnings test level that year. Social Security benefit recipients ages 65 to 71, who could earn up to $6,000 before benefits were reduced, consistently had higher earnings than recipients younger than age 65.8 Median earnings for men in the older group were $5,460. In another study on the effect of the earnings test, Gary Burtless and Robert Moffitt found that workers kept post-retirement hours to the level at which total earnings equaled the exempt amount.9

Avoidance of earnings in excess of the exempt amount is understandable. Not only would half of any excess earnings be lost through Social Security reductions, but, in addition, all earnings would be subject to Federal, State, and local income taxes as well as Social Security withholdings. Another effect of the earnings test, although impossible to quantify, is that some persons might work "off the books" rather than pay Social Security and income taxes on their earnings.

For retirees with pension incomes, Social Security benefits, and a lifelong accumulation of savings, the earnings test is probably of little or no consequence. These workers often prefer early retirement. Some retirees may choose to work part time or part year to supplement their retirement income. For those who limit their activity so as not to exceed the exempt amount, the liberalization or elimination of the earnings test could increase their work effort. However, among the smaller group of part-timers who earn more than the exempt amount while receiving reduced benefits, work effort may actually be reduced, as they could work fewer hours without a loss in income.

What is still not clear, however, is the extent to which Social Security encourages retirement or discourages continued work. Some researchers believe that the method of calculating Social Security benefits may cause some workers to postpone retirement. Because the most recent (and presumably highest) years of earnings are averaged in benefit computation, some researchers believe that workers may choose to work longer, replacing low earnings years with higher ones, and therefore increasing their Social Security benefit.10

Changes in the program. The Social Security Amendments of 1983 contained several long-term provisions designed to remove work disincentives. These included the following:11

1. An increase in the normal retirement age. Beginning in the year 2000, the retirement age at which beneficiaries are eligible to receive full benefits will increase gradually from 65 to 67. The normal retirement age will remain at 67 for those reaching age 62 after 2022.

2. An increase in the early retirement penalty. Reduced benefits will continue to be available at age 62, but reduction factors will be revised to a maximum of 30 percent (for workers entitied at 62 when normal retirement age is 67) compared to the prior 20-percent reduction.

3. An increase in the delayed retirement credit. The delayed retirement credit will increase by half a percentage point every other year, from 3 percent for workers age 62 prior to 1987 to 8 percent per year for workers age 62 after 2004.

4. A decrease in the withholding rate under the earnings test. Beginning in 1990, the withholding rate will decrease from $1 of every $2 above the exempt amount for persons who attain full retirement ageto $1 of every $3. Beginning in 2000, the age at which this occurs will increase as the normal retirement age increases.

Potential effects of the changes. Despite the fact that Social Security is an important income source for almost all older persons, most analysts believe that changes in retirement ages, as a result of the Social Security amendments, will be small. This is particularly clear when the changes are dissected. For example, while earnings above the exempt amount will be subject to a one-third offset under the new law (rather than the present one-halo, those earnings will continue to be subject to Federal, State, and local taxes and Social Security withholdings. Thus, the system will still provide disincentives to exceed the exempt amount.

Alan Gustman and Thomas Steinmeier studied the potential effect of 1983 reforms and concluded that "in comparison with the previous rules, the 1983 rules, when they take full effect, should have a fairly small impact on the number of people working full time and the number retired before age 65, but at age 65 and thereafter, the percentage of individuals working full time would be noticeably increased [largely due to the scheduled increase in the delayed retirement credit] and the percentages working part time and retired would both decline."12 Fields and Mitchell found that increasing the normal retirement age from 65 to 68 (legislation raises it to age 67) could be expected to increase average retirement age by only 1.6 months. They also found that the largest increase (still only 3.1 months) occurred when the percentage of total benefits received at age 62 was reduced from 80 to 55 percent (the reforms only reduced the percentage to 70 percent) and the delayed retirement credit was increased to 20 percent (from the current 9 percent) at age 68. 13

Other pensions

Retirement decisions are rarely based on Social Security benefit levels alone; they also depend, among other things, on preferences for leisure over work, on health status, and on other income sources. Although most workers can no longer be forced to retire because of their age, many other provisions in pension plans encourage workers to retire at specific ages, often well before the normal retirement age of 65 in the Social Security program.

Pension policies do not affect all workers. In fact, of those persons receiving Social Security benefits (of any type) in 1980-81, only about 57 percent of men and 31 percent of women were either receiving or expecting to receive a pension." For those who do receive pensions, however, plan policies greatly affect retirement decisions. And, in general, it does not take a large pension to induce workers to retire. 15

Retirement provisions. While individuals are not eligible for full Social Security benefits before age 65, normal (full-benefit) retirement ages in private and governmental pension plans tend to be much lower. In recent years, retirement programs have become increasingly liberal, allowing full benefits at earlier ages. Seventy-nine percent of pension plans surveyed by the Bureau of Labor Statistics in 1983 had no minimum retirement age or provided full benefits at age 62 or earlier, up from 55 percent in 1974. And 37 percent of those plans allowed for full-benefit retirement as early as age 55, usually with 30 years of service.16

Almost all private pensions surveyed by BLS in both 1974 and 1983 permitted early retirement, although at reduced benefits. Over the 1974-83 period, however, both age and years-of-service requirements for early retirement declined. In 1983, the length of service required for early retirement (with reduced benefits) at age 55 averaged 7 years and 2 months, down from 10 years and 3 months in 1974.17

Are pensions actuarially neutral? Individuals who opt for early retirement usually receive reduced benefits. However, reduction percentages are not always actuarially neutral; the greater number of years of pension receipt (due to early retirement) often more than offset any decline in benefits. In a study of 10 pension plans surveyed in the 1978 Benefits Amounts Survey, Fields and Mitchell found that, in half of the plans, the present value of net pension benefits was greatest for workers retiring at age 60.18 Four of the 10 plans paid the highest benefits at age 61 or 62, and the remaining plan at age 66. Similarly, Lawrence J. Kotlikoff and David A. Wise, in a study of more than 2,000 pension plans, found that plan provisions strongly discouraged work after a normal retirement age-some after an early retirement age.19 Continued work does provide additional earnings; however, the foregone pension benefits (as with deferred Social Security benefits) result in an implicit tax on earnings which may be as high as 100 percent. 20

Some pension provisions penalize continued work activity. Since passage of the Omnibus Budget Act of 1986, pension plans are no longer allowed to deny continued pension accrual for persons over age 65. Caps are still permitted, however, on years of service that may be counted toward a pension and on total benefit levels.21 Also, rates of accrual may be less than actuarially neutral for those who delay retirement. These provisions, in effect, reduce total compensation levels for persons who continue working after reaching either the maximum levels of credited service or pension benefits.

Combining pensions and Social Security. Not only do pension plans provide different options and retirement incentives than Social Security, but, in many cases, pension benefits are derived using a formula that accounts for Social Security benefits. Thus, changes in Social Security policy designed to alter work patterns may be undermined by the structure of pension plans.

A 1986 survey of employee benefits in medium- and large-sized firms found that 62 percent of all full-time participants in defined-benefit pension plans were in plans "integrated," or combined in some way, with Social Security.22 Sixty-nine percent of the employees in these integrated plans had offset provisions; pension benefits were derived as a function of Social Security payments (usually pension levels were reduced by 50 percent of an individual's Social Security benefit). For example, workers with expected pension benefits of $8,000 and expected Social Security benefits of $2,000 would actually receive pension benefits of $7,000 [$8,000 - (.50 x $2,000)] in addition to their $2,000 Social Security benefit. In this way, public policy efforts to increase incentives to work by reducing benefits would be countered by a 50-percent increase in private benefits. A reduction of $1,000 in Social Security benefits, for example, would be countered by a $500 corresponding increase in pension benefits. Plans with excess formulas also recognize the structure of Social Security benefits and attempt to increase benefits to workers with higher earnings (whose Social Security benefits replace a smaller share of earnings). This is accomplished by applying higher benefit accrual percentages to earnings above a specified limit-usually equal to the Social Security taxable maximum.23

Some retirees receive supplemental benefits to their pensions to compensate for retiring prior to eligibility for Social Security payments. A 1984 BLS survey of pension plans found that 11 percent of all plan participants could receive supplements upon early retirement. Ten percent were eligible for supplements on top of their full benefits if they retired "normally" before age 62 -when they would become eligible for reduced Social Security benefits.24 These supplemental payments are often equal to or greater than the Social Security benefits they would later receive. Each of these pension provisions that integrate Social Security and pension benefits may mitigate any changes in incentives that Social Security reforms are intended to produce.

Recently, researchers have begun to compare the incentive effects of Social Security and private pension provisions on individual retirement decisions. A study by James H. Stock and David A. Wise modeled the retirement behavior of employees in one large firm and simulated the effects of changes in Social Security and of alternative pension plans. The researchers found that increases in the firm's early retirement age dramatically reduced the number of workers retiring by age 60. In contrast, the effects of changes in Social Security rules were minimal.25 Also, the researchers concluded that "Changes in Social Security provisions that would otherwise encourage workers to continue working can easily be offset by countervailing changes in the provisions of the firm's pension plan."

Early Retirement Incentive Plans

Early withdrawal from the labor force has expanded with the increasing use of Early Retirement Incentive Plans (ERIP'S). These plans allow workers to retire earlier than the normal terms of their pension plans would allow. Typically, ERIP'S either liberalize the requirements for pension eligibility or provide employees with richer pension benefits. Some also offer early retirees either a continuation of or improvement in medical coverage after their separation from service. ERIP'S are typically offered for only a short period, after which the normal plan rules apply.

ERIP'S, in many ways, are simply an extension of the trend toward early retirement made possible by pension plan provisions already discussed in this article. The key issue related to ERIP'S is whether they are truly voluntary: Do workers perceive turning down these offers as being a viable option? Are workers satisfied with the early retirement decision?

Two facts are critical to the discussion of ERIP'S. First, no one knows how prevalent they are. The few surveys of employers conducted to date often are not representative samples of all employers and often have low response rates; hence, the results based on data from reporting firms may not reflect the experience of all firms.

Second, and probably the most important for policy considerations, it is difficult to determine the extent of involuntary separations resulting from these plans. study by Phyllis H. Mutschler and others indicates that companies, workers, and unions have embraced them.26 At the same time, a study conducted for the Public Policy Institute of the American Association of Retired Persons (AARP) concludes that such plans are primarily "older worker termination programs," and that neither the individuals involved nor the Nation's interests are well served by them.27

The plans seem to be voluntary, given that the majority of eligible workers do not accept them. A study by Hewitt Associates indicates that, on average, about I in 3 workers accept ERIP'S when offered.28 Indeed, some companies have had far more workers accept these offers than they had expected, causing a damaging loss ,of experienced personnel. On the other hand, numerous lawsuits related to these plans suggest that some older workers view them as forced retirement schemes. 29

The Hewitt Associates' analysis of the prevalence of ERIP'S is the most extensive to date. Of the 529 companies responding to their 1985 survey, a third reported that they had used early retirement windows (whereby the employee is given a specific period of time in which to decide whether to retire with the improved benefit package) or other types of voluntary separation plans. About 40 percent of the companies using ERIP'S had offered them more than once. Plans were offered far more frequently by the larger firms than by smaller ones; over half of the companies employing 25,000 persons or more had used them. And, as mentioned, about a third of all eligible employees accepted these offers, although about 1 in 4 plans had acceptance rates of at least 75 percent.

Employees' views of such plans are difficult to interpret. The AARP report makes no mention of ERIP'S as a welcome offer to many older workers who may view retirement quite positively. Yet, substantial numbers of workers welcome the opportunity to retire earlier than "normal." In a survey of workers age 40 and older, conducted for AARP by the Gallup Organization, 41 percent of all workers surveyed responded that they would be likely to accept incentive offers for early retirement.30 Affirmative responses were most common among workers with high levels of income and education.

Mutschler and others studied persons who had retired from an unidentified Fortune 500 company with and without special incentives. They found that employees clearly responded to the economic incentives of the ERIP'S under study-the better the retirement package, the more likely workers were to accept it.31 Also, there was no evidence that those accepting the office had suffered financially as a result (which would have suggested coercion); however, the authors did express some concern over the long-term effects of inflation on the value of retirees' pensions. Other than this study, little is known about the conditions under which workers accept ERIP'S and the outcomes of their decisions.

The hazy distinction between voluntary and involuntary retirement makes analysis of this labor force issue difficult. The voluntary nature of ERIP'S may not even be a static concept. An individual who had positive views about accepting an "early out" at the time of the offer may have a very negative view after the fact, or vice versa. While it certainly is possible to better quantify the use of ERIP'S than has been done so far, to evaluate the effect of these programs on workers' financial and nonfinancial well-being would be far more complicated.

In summary, the incentives in Social Security and private pension policies often operate in opposite directions. It is unclear exactly what long-term effect Social Security reforms will have on the work activity of older persons; however, it is clear that private pensions have not followed Social Security's lead in encouraging later retirement. While pension policies that allow retirement well before age 65 are undoubtedly attractive to many older workers, those who might prefer to continue to work part time often do not or cannot. Some reasons for this are discussed in the next section.

The market for part-time jobs

Many observers have noted that older workers often are faced with a choice between continuing full time in a long-held job or withdrawing from the labor force; the majority reject part-time employment, which usually pays relatively low wages and provides very few benefits. Whether part-time work by older persons will become more common in the future depends on many diverse factors, including pension and Social Security regulations that determine levels of nonwage income and place restrictions on employment; the characteristics of part-time jobs; and the preference for leisure over work at older ages,

Relatively few older persons choose to work after first receiving retirement benefits, and those who do usually have very low levels of nonwage retirement income. 32 However, many older men still do not seek employment despite very small pension benefit levels. 3 3According to data from the Social Security New Beneficiary Survey, fewer than 1 in 4 persons was employed at all 18 months to 2 years after first receiving retired-worker benefits. 34 And, among those who were not in the labor force, 95 percent responded in the Current Population Survey that they "do not want a job now, either full or part time."35

Does this low level of work activity beyond retirement mean that older persons simply do not want to work, or is it a reflection of poor employment options? While the preference for leisure over work is very strong for many older persons, it is also possible that many say they do not want to work because they see only very limited choices. As discussed previously, substantial institutional barriers-especially the Social Security earnings test -provide strong disincentives to full-time work at later ages. And pension provisions often make continued work for one's employer unjustifiable. Older workers, then, tend to be funneled into the part-time job market, where options are frequently limited to low-paid employment. The solution, many argue, would be to expand opportunities for parttime work to include jobs that are well paid and provide nonwage benefits.36

There is little doubt that part-time work done by older workers is relatively low paid. Jondrow and others, cited earlier, have found that hourly wages tend to decline about 30 to 40 percent when weekly hours are reduced from 35 to 20. 3 7However, low pay is not necessarily evidence of age discrimination. In fact, the same researchers found that " . . . the scarcity of well-paid part-time jobs is not a matter of discrimination against older workers; such jobs are scarce throughout the economy." 38

A primary reason for the scarcity of well-paying parttime jobs is the high cost of such schedules to employers. Training costs, for example, are virtually identical for fulland part-time workers, as are many administrative costs. A short workweek raises the hourly costs to employers for such expenses. In contrast, jobs that generally require little training do not significantly raise the costs to employers offering part-time schedules, especially if the benefit packages are more limited than those given fulltime workers. These jobs, by their nature, are usually low skilled and provide low pay.

Whether such a restrictive market for part-time jobs for older persons is the only possible scenario is still open to question. Hilda Kahne, in her book Reconceiving PartTime Work, distinguishes between "New Concept" parttime work and "Old Concept" work." The latter was described above- work at very low pay rates, often in low-wage industries, and with few benefits. The former, Kahne envisions, would be work done in the full range of industries and occupations and would generally provide prorated full-time wages and benefits. Kahne presents a convincing argument for the potential interest in such employment from the older workers' point of view; however, she does less to explain how such jobs make sense for employers, particularly those not facing labor shortages. For now, at least, it appears that such "New Concept" job market offers lag behind workers' desire for them.

As the younger population declines and the growth rate in the female labor force slows,40 some service-sector employers are beginning to target jobs to older workers. Such employment will attract a narrow range of elderly persons, however, as it is typically part-time work with few fringe benefits. While widespread worker shortages may occur in the future,41 their effect on employment opportunities for older workers is difficult to predict, particularly in the context of today's institutional structures that strongly favor early retirement. Age discrimination

When any group's labor market experiences are found to be inferior to another's, the issue of discrimination is a subject for discussion. However, discrimination is one of the most difficult labor market issues to identify and quantify. This is because it is difficult to discern whether between-group differences in earnings or unemployment, for example, are the result of discrimination or of real differences in productivity or labor market goals. These measurement problems have limited the amount of research conducted on age discrimination in employment. It would appear, however, that discrimination does take place in the job market-that hiring, training, and promotion decisions involving older persons are not entirely age, sex, and race neutral.

Regarding the earnings of older workers, Richard A. Wanner and Lynn McDonald, using National Longitudinal Survey (NLS) data for mature men, found that as the men in the sample aged between 1966 and 1976 (and gained tenure and experience), they had a substantial decline (in real terms) in earnings." This occurred during a period of sizable increases in earnings among all workers. The poor earnings performance among older workers was determined to be unrelated to any decrease that might have been associated with job changing.

The authors identified three theoretical explanations for this. First, human capital theorists in economics would attribute the lower earnings of older workers primarily to lower productivity, perhaps related to skill obsolescence and employers' reluctance to invest in the upgrading of those skills. This would seem reasonable, given the relatively short payoff time for such an investment. Second, equity theorists in sociology would argue that wages reflect not only productivity but also the workers' need for income and that declining wages at older ages describe a legitimate lifetime earnings profile. Workers' preferences for increased leisure (largely associated with declining financial need) may partly explain the earnings profiles of older workers found by Wanner and McDonald. Although their methodology accounts for reductions in the number of weeks worked, by their own admission, they may have missed some of the hours effect, such as not incorporating older workers' lower propensity to accept overtime work.

The researchers prefer a third explanation: employers assume that older workers will accept lower levels of salary increases, or fewer of them, because older persons' ability to find comparable alternative employment is quite low. Certainly, human capital theorists would agree that workers accumulate extensive firm-specific human capital for which a new employer would be unwilling to compensate them, Stephen R. McConnell, in his assessment of age discrimination, also highlights this decline in leverage for older workers brought about by their high cost of job switching.43

Using the National Longitudinal survey to examine the wage decline for workers who were forced to look for a new job, David Shapiro and Stephen H. Sandell found little evidence of discrimination. In fact, they determined that about 90 percent of any loss in earnings in workers' new jobs reflected a loss of their firm-specific human capital.44 While this is a real and important earnings loss for many workers, it cannot be said that such declines in earnings are the result of unfair practices by employers, who would not be expected to pay for skills, knowledge, or experience that are not transferable to a new job.

It is interesting to examine workers' own perceptions of discrimination. In the previously mentioned survey conducted by the Gallup Organization in 1985, a sample of workers age 40 and older was asked whether they had experienced age discrimination.45 Only about 6 percent answered in the affirmative, mostly with the response that they had been denied a promotion or a chance for advancement because of their age. The perception of age discrimination increased with age; 4 percent of workers in their forties believed they had experienced age discrimination, compared with 10 percent of those age 63 and over. It is not clear whether the greater affirmative response for the older group represents increased discrimination with age (although it seems reasonable that this would be the case) or the greater number of years over which they could have experienced discrimination. A shortcoming of this survey, and most others, is that respondents are employed persons only; thus, those who may be unemployed or out of the labor force who have been victims of age discrimination are not included. These may be the groups of older persons who have been most affected by discriminatory employment practices.

Further insight into age bias comes from employers' perceptions of older workers. In a 1985 study, Benson Rosen and Thomas H. Jerdee found that many managers exhibited age discrimination in their personnel decisions.46 They asked 6,000 readers of Harvard Business Review (most of whom were in management positions) to make management decisions in seven hypothetical cases. In half of the respondents' questionnaires, the worker in question was a younger person; in the other half, an older one. Except for the age of the workers, the scenarios were identical. In the almost 1,600 returned survey forms, respondents consistently made different hiring, promotion, discipline, and training decisions based on the stated age of the worker in question. Yet, in a final set of questions, respondents indicated a very high level of support for nondiscriminatory business practices. Interestingly, respondents age 50 and over were consistently more supportive of the older workers than were younger respondents, from which the authors concluded that an older worker's best prospect for fair treatment appeared to be working for an older supervisor.

The above research and other similar work suggest that age discrimination exists regarding older workers' employment and advancement opportunities. Nevertheless, relatively few older workers state that they have been the victims of age discrimination. Few older workers search for a job, whereby they would be most exposed to discrimination. Also, experience on the job may provide many older workers with the skills and abilities that prevent them from being marginal employees. In addition, the promotion expectations (or desires) of some workers may decline with age, often because of the desire to stay in a "comfort zone" toward the end of a career.

WHILE RECENT Federal policy has been directed toward extending worklives, substantial disincentives to work at older ages still exist. Changes in Social Security policies designed to increase work activity will not substantially alter the structure of the Social Security system and, as a result, may have only minor impacts on the labor force behavior of older persons. In addition, any changes that do occur may be overshadowed by decreases in work activity resulting from greatly liberalized pension plan provisions.

For those persons who wish to continue working at older ages, a scarcity of well-paying, part-time jobs may limit work activity. And, for some workers, age discrimination may provide barriers to employment. As the number of young workers continues to decline, however, it is possible that employers will begin to provide more attractive work options for older persons.


1 The concept of retirement is more complex for women. That is because women in their fifties and sixties today often had very little work experience throughout their lives. As a result, they often do not choose a retirement age based on their own work history or pension resources. Labor force participation rates for women ages 65 and over have followed the same trend as those for men -they have declined from a high of about 11 percent in the early 1960's to about 7 percent today. Participation rates for women ages 55 to 64 have changed little over the last two decades.

2 John A. Svahn and Mary Ross, "Social Security Amendments of 1983: Legislative History and Summary of Provisions," Social Security Bulletin, July 1983, pp. 3 - 48. See Older Americans in the Workforce: Challenges and Situations (Washington, Bureau of National Affairs, Inc., 1987). Also see appendix: Text of the Age Discrimination in Employment Act of 1967, as amended. Tbe 1986 amendments removed the upper limit (which had been age 70) in the law's proscription against age discrimination.

3 Susan Grad, Income of the Population 55 and Over 1986 (Social Security Administration, 1988), Publication No. 13 - 11871, pp. 1, 87.

4 Svahn and Ross, "Social Security Amendments," pp. 3 -48.

5 For a detailed explanation of how benefits are derived, see the Social Security Handbook, 1988 (Social Security Administration, 1988), Publication No. 05-10135, eh. 7; see also Social Security Bulletin, Annual Statistical Supplement, 1987 (Social Security Administration, 1987).

6 Gary S. Fields and Olivia S. Mitchell, "Restructuring Social Security: How Will Retirement Ages Respond?" in Stephen H. Sandell, ed., The Problem Isn't Age (New York, Praeger Publishers, 1987), pp. 192-205.

7 See, for example, Arden Hall and Terry R. Johnson, "The Determinants of Planned Retirement Age," Industrial and Labor Relations Review, January 1980, pp. 241 - 54; and Richard V. Burkhauser, "The Early Acceptance of Social Security: An Asset Maximization Approach," Industrial and Labor Relations Review, July 1980, pp. 484-92.

8 In 1982, workers were subject to an earnings test until age 72. Currently, benefits are not reduced for earnings after age 70. See Howard M. Iams, "Jobs of Persons Working After Receiving Retired-Worker Benefits," Social Security Bulletin, November 1987, pp. 4- 19.

9 Gary Burtless and Robert A. Moffitt, "The Joint Choice of Retirement Age and Postretirement Hours of Work," Journal of Labor Economics, April 1985, pp. 209 - 36.

10 Alan S. Blinder, Roger H. Gordon, and Donald E. Wise, "Reconsidering the Work Disincentive Effects of Social Security," National Tax Journal, December 1980, pp. 431-42.

11 Svahn and Ross, "Social Security Amendments," pp. 3 -48.

12 Alan L. Gustman and Thomas L. Steinmeier, "The 1983 Social Security Reforms and Labor Supply Adjustments of Older Individuals in the Long Run," Journal of Labor Economics, April 1985, pp. 237-53.

13 Fields and Mitchell, "Restructuring," pp. 192 - 205.

14 John R. Woods, "Retirement Age Women and Pensions: Findings From the New Beneficiary Survey," Social Security Bulletin, December 1988, pp. 5-16.

15 See Scott H. Beck, "Determinants of Labor Force Activity Among Retired Men," Research on Aging, June 1985, pp. 251 - 80.

16 Donald Bell and William Marclay, "Trends in retirement eligibility and pension benefits, 1974-83," Monthly Labor Review, April 1987, pp. 18-25.

17 Ibid., p. 19.

18 Gary S. Fields and Olivia S. Mitchell, Retirement, Pensions, and Social Security (Cambridge, MA, The MIT Press, 1984), pp. 33 - 51.

19 Laurence J. Kotlikoff and David A. Wise, The Incentive Effects of Private Pension Plans (Cambridge, MA, National Bureau of Economic Research, Inc., 1984), Working Paper Series 1510.

20 Laurence J. Kotlikoff and David A. Wise, Employee Retirement and a Firm's Pension Plan (Cambridge, MA, National Bureau of Economic Research, Inc., 1987), Working Paper Series, 2323.

21 Patricia M. Moore, "Age-Related Changes in Retirement Policies and Plans," Compensation and Benefits Management, Spring 1988, pp. 217-20.

22 Employee Benefits in Medium and Large Firms, 1986, Bulletin 2281 (Bureau of Labor Statistics, 1987), pp. 53 - 76.

23 Donald Bell and Diane Hill, "How Social Security Payments Affect Private Pensions," Monthly Labor Review, May 1984, pp. 15-20.

24 Bell and Marclay, "Trends in retirement eligibility," p. 24. 25 James H. Stock and David A. Wise, The Pension Inducement to Retire: An Option Value Analysis (Cambridge, MA, National Bureau of Economic Research, Inc., 1988), Working Paper Series, 2660.

26 Phyllis H. Mutschler, James H. Schultz, and Thomas D. Leavitt, What Price Retirement?: A Study of Early Retirement Incentive Programs (Washington, American Association of Retired Persons, 1984).

27 Elizabeth L. Meier, Early Retirement Incentive Programs: Trends and Implications (Washington, American Association of Retired Persons, Andrus Foundation, 1984).

28 Hewitt Associates, Plan Design and Experience in Early Retirement Windows and Other Voluntary Separation Plans (Lincoinshire, IL, Hewitt Associates, 1986).

29 Raymond C. Fay, The ADEA and Early Retirement Incentives, (Washington, Law Offices of Bell, Boyd, and Lloyd, undated).

30 American Association of Retired Persons, Work and Retirement: Employees Over 40 and Their Views (Washington, American Association of Retired Persons, 1986).

31 Mutschler and others, "What Price Retirement?"

32 See Stephen R. McConnell and others, Alternative Work Options for Older Workers: A Feasibility Study (Los Angeles, CA, The Ethel Andrus Gerontology Center, 1980).

33 Beck, "Determinants of Labor Force Activity," p. 276.

34 Howard M. lams, "Jobs of Persons Working After Receiving Retired-Worker Benefits," Social Security Bulletin, November 1987, pp. 4-19.

35 The Current Population Survey (CPS) is a monthly survey of about 55,000 households conducted by the Bureau of the Census for the Bureau of Labor Statistics.

36 See McConnell and others, Alternative Work Options;also see Hilda Kahne, Reconceiving Part-Time Work: New Perspectives for Older Workers and Women (Totowa, NJ, Rowman & Allanheld, 1985).

37 Jim Jondrow, Frank Brechling, and Alan Marcus, "Older Workers in the Market for Part-Time Employment," in Steven H. Sandell, ed., The Problem Isn't Age (New York, Praeger Publishers, 1987), pp. 84-99.

38 Ibid, p. 96.

39 Kahne, Reconceiving Part-Time Work.

40 See Howard N Fullerton, Jr., "Labor force projections: 1986 to 2000," Monthly Labor Review, September 1987, pp. 19- 29.

41 See, for example, Lawrence Olson, Christopher Caton, and Martin Duffy, The Elderly and the Future Economy (Lexington, MA, Lexington Books, 1981).

42 The National Longitudinal Survey of Labor Market Experience (NLS), a survey of several cohorts conducted by the Ohio State University, was specifically designed to measure many of the labor market problems experienced by older workers. The mature men's cohort was first interviewed in 1966, when respondents were ages 45 to 59. See Richard A. Wanner and Lynn McDonald, "Ageism in the Labor Market: Estimating Earnings Discrimination Against Older Workers," Journal of Gerontology, Vol. 38, No. 6, pp. 738-44.

43 Stephen R. McConnell, "Age Discrimination in Employment," in Herbert S. Parnes, ed., Policy Issues in Work and Retirement (Kalamazoo, mi, W. E. Upjohn Institute for Employment Research, 1983).

44 David Shapiro and Steven H. Sandell, "Age Discrimination in Wages and Displaced Older Men," Southern Economic Journal, July 1985, pp. 90-102.

45 American Association of Retired Persons, Work and Retirement.

46 Benson Rosen and Thomas H. Jerdee, Older Employees: New Roles for Valued Resources (Homewood, IL, Dow Jones-Irwin, 1985).
COPYRIGHT 1989 U.S. Bureau of Labor Statistics
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1989 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Rones, Philip L.; Herz, Diane E.
Publication:Monthly Labor Review
Date:Apr 1, 1989
Previous Article:Sources of increasing inequality in wages and salaries, 1960-80.
Next Article:Unemployment insurance in the United States and Europe, 1973 -83.

Related Articles
Self-employment among older U.S. workers: the 1990s showed a downward trend in self-employment rates, however, the fact that self-employment rates...
Employment and Training: Most One-Stop Career Centers Are Taking Multiple Actions to Link Employers and Older Workers.
Comparative indicators.
Comparative indicators.
Barriers get higher for older workers; 'Recession increases discrimination' - survey.
Over 50s facing jobs struggle.
Social area differences in VET participation.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters