Skip Navigation | |
Although the oldest members of the baby-boom generation will not become eligible for Social Security retirement benefits until 2008, when they reach age 62, many of them already have left the labor force. This paper--prepared as part of the Congressional Budget Office's (CBO's) ongoing examination of the future composition of the labor force and the performance of the economy--uses data from a national survey to provide information about the men and women who leave the labor force before age 62 and their sources of income. In keeping with CBO's mandate to provide objective, impartial analysis, the paper makes no recommendations. Ralph E. Smith of CBO's Health and Human Resources Division wrote the paper. Susan Labovich did the computer programming for the analysis, and Peter Richmond provided research assistance. The paper benefited from comments offered by June E. O'Neill, of Baruch College, City University of New York, and by Paul Cullinan, Molly Dahl, Noah Meyerson, Carla Tighe Murray, and Bruce Vavrichek, all of CBO. Loretta Lettner edited the paper, and John Skeen proofread it. Ronald Moore produced drafts of the manuscript. Maureen Costantino designed the cover and prepared the paper for publication. Lenny Skutnik produced the printed copies, and Annette Kalicki prepared the electronic versions for CBO's Web site. Douglas Holtz-Eakin
Introduction and SummaryMembers of the leading edge of the baby-boom generation--the large number of people born between 1946 and 1964--turn 58 this year. Most of them will become eligible for Social Security retirement benefits when they reach age 62. And, at age 65, they will qualify for Medicare. Considerable attention has been paid to whether boomers have saved enough to afford to retire and to whether they will decide to continue working once they become eligible for Social Security and Medicare.(1) Many boomers, however, are not waiting until age 62 or 65 to stop working. Over 4 million already have left the labor force either because they are disabled or because they have retired. If they follow in the footsteps of workers now in their early 60s, perhaps one-third of the men and nearly half of the women will be out of the labor force before their 62nd birthday (see Figure 1).
Labor Force Participation Rates of Men and Women, by Age, 2004 (Percentage of population)
Source: Congressional Budget Office based on data from the Census Bureau's Current Population Survey for the first nine months of 2004. This paper examines the characteristics of men and women who leave the labor force before reaching age 62 and analyzes their income sources given that they no longer work for pay. Most of the analysis concentrates on men and women ages 50 to 61 who were not in the labor force at any time in 2001--a group that includes not just the oldest boomers (those ages 50 to 55 in 2001), but also people born just before the baby boom began. (Information about the latter group offers insights into what could be in store for boomers as they age.) The empirical findings presented here are based largely on the Congressional Budget Office's (CBO's) analysis of data from the Survey of Income and Program Participation (SIPP), conducted by the Census Bureau. Because those data have several limitations, the qualitative findings of CBO's analysis are more meaningful than the precise estimates. One drawback is that the data are based on survey responses, which are not always accurate. Moreover, some of the questions--such as those that attempt to identify why respondents were not in the labor force or those that try to determine the presence of a disability--call for subjective judgments on the part of the respondents, rather than strictly objective facts. In addition, care should be taken in extrapolating the results presented here to the future activities or well-being of younger baby boomers. For example, probably fewer of them will receive defined benefit pensions when they leave the labor force, but more of them will have participated in 401(k) or other defined contribution plans. CBO's analysis indicates that, overall, the men and women in their 50s and early 60s who were not in the labor force in 2001 had much lower median family income, fewer assets, and higher poverty rates than did their contemporaries who were still in the labor force (see Table 1).(2) They also were much less likely to have finished high school.
Labor Force Status of Men and Women Ages 50 to 61 and the Main Reason for Nonparticipation, 2001
CBO found that those men and women who had left the labor force even though they were not yet old enough to receive Social Security retired-worker benefits did so for a small number of reasons: The most frequent reason they offered for not working was that they were disabled--accounting for almost two-thirds of the men who were not in the labor force and two-fifths of the women. Most of the other men said that they were retired. Most of the other women said that they were retired, caring for others, or not interested in working. Survey responses indicate that the circumstances of men and women not in the labor force because of a disability were quite different from those of the men and women who had retired. Among CBO's findings:
Participation in, and Withdrawal from, the Labor ForceSince the first baby boomers were born, major changes have occurred in the labor force participation patterns of older men and women (see Figure 2). In the late 1940s and early 1950s, 9 out of 10 men ages 55 to 64 participated in the labor force, whereas fewer than 1 in 3 women did. Since the mid-1980s, however, only about two-thirds of men in that age group have participated in the labor force. Meanwhile, the labor force participation rate of women in that age group rose appreciably: in recent years, well over half have been in the labor force.
Labor Force Participation Rates of Men and Women Ages 55 to 64, 1948 to 2003 (Percent)
Source: Congressional Budget Office based on data from the Bureau of Labor Statistics. Why do some people stop working or looking for work before they become eligible for Social Security retirement benefits while others stay in the labor force long afterward? Individuals stop participating in the labor force if they decide that the benefits of working or seeking work no longer outweigh the costs of doing so. Those benefits include not just after-tax wages and other job-related remuneration, but also nonfinancial benefits (such as personal satisfaction and a social network). Likewise, the costs go well beyond the out-of-pocket expenses related to working (such as those for commuting and clothing). For most workers, the major cost is the value of the activities forgone while working--that is, the benefits they would have derived from whatever they could have done instead. An extensive body of literature on retirement decisions highlights the disincentives or barriers that lead many workers to decide to leave the labor force well before they become eligible for Social Security retirement benefits. The availability and structure of defined benefit pension plans, in particular, have been linked to early retirement. In those plans, when workers reach a certain age and have been with their employer a specified number of years, they qualify for a pension. Specific features of defined benefit plans place a large effective tax on people who, once eligible for a pension, remain with the same employer. Those features include less-than-actuarially fair accrual rates for additional pension benefits and legal restrictions that limit the ability of a worker to draw a pension while continuing to work for that employer. Although workers can respond to those disincentives by changing employers rather than retiring, the compensation from their next-best job may be well below what they currently earn. For employers, seniority-based systems may result in wages for older workers that exceed their actual or perceived productivity, discouraging employers from hiring or retaining such workers. Higher average costs of health insurance for older workers may further reduce employers' incentives to employ them. Likewise, older workers who lose their job may have considerable difficulty finding a new one that pays nearly as much as the one they lost, and they might respond by leaving the labor force.(3) Researchers have linked the long-term decline in the labor force participation rate of older men to the nation's greater affluence.(4) Pensions, Social Security, and private savings have enabled many workers to exit the labor force without being financially dependent on their children. The early-retirement incentives commonly found in defined benefit pension plans, noted previously, may further encourage workers to leave the labor force before they qualify for Social Security retirement benefits. In recent years, however, the decline in defined benefit pension plan coverage and the rise of 401(k) and other defined contribution plans have reduced the fraction of the workforce facing those incentives.(5) The future course of the labor force participation rate of older men is difficult to predict, in part because of different expected trends in its determinants. The nation's economy is likely to continue to grow, which could facilitate early retirement. However, the switch from defined benefit pension plans to defined contribution plans, along with increasing life spans, could discourage early retirement. CBO projects that the participation rate of men in their late 50s and early 60s will remain near its current level during the next decade.(6) By contrast, CBO anticipates that the participation rate of women in that age group will continue to rise as the cohorts with a greater attachment to the labor force enter their late 50s and early 60s.
Who Stops Working Before Age 62, and What Do They Live On?The analysis presented here is based largely on information obtained from the 2001 Survey of Income and Program Participation. The sample relevant to this analysis consists of about 8,500 men and women ages 50 through 61 in 2001. The majority of that group were born during the baby boom. The others (ages 56 and older) were born earlier, but their inclusion provides additional information about the characteristics and resources of people who leave the labor force before becoming eligible for Social Security retirement benefits. For the purposes of this analysis, particular attention is paid to the income, assets, and characteristics of the 14 percent of men and 24 percent of women ages 50 through 61 who reported that they had not worked or looked for work at any time during 2001. (Additional information about the SIPP is provided in the appendix.) Those who were not in the labor force during that period were classified according to the main reason they provided for not working: 32 percent of the men and 26 percent of the women not in the labor force indicated that they were retired; 64 percent of the men and 40 percent of the women said that they were disabled; and, 4 percent of the men and 34 percent of the women said that they were taking care of others, not interested in working, temporarily ill or injured, could not find work, or were out of the workforce for other reasons.(7) On the basis of information that the respondents provided about their previous employment history and about their activities during the following year (2002), most of the men and women who were not in the labor force at any time in 2001 appear to have worked earlier but to have since totally withdrawn from the labor force. Only 3 percent of the male respondents who were not in the labor force in 2001 and 12 percent of the women said that they had never worked. Among the respondents for whom information was available for all of 2002, only 5 percent of the men and 6 percent of the women reported any subsequent earnings, and most of them earned less than $6,000 that year. Income and AssetsThe men and women in their 50s and early 60s who were not in the labor force had much lower median family income than did the men and women who were still in the labor force (see the last two columns of Table 2). They also had far fewer assets.
Income and Assets of Labor Force Participants and Nonparticipants Ages 50 to 61, 2001 (Dollars)
Among the men and women who were not in the labor force, those who had retired generally were in a much stronger financial position than those who were out of the labor force because of a disability. (Retired workers typically had lower income than did those who were still in the labor force, but more assets.) Retired men had a median family income of about $30,000 and a net worth, including home equity, that exceeded $200,000.(8) The median income of disabled men was only about $20,000, and their net worth (at $19,000) was less than one-tenth that of retired men. Such variations also were found for retired and disabled women. For many people, equity in their house constitutes their largest single asset. Excluding home equity from calculations of net worth made the differences between retired and disabled men and women starker: excluding home equity, the household net worth of the typical retired man or woman was about $90,000. The household net worth of the median disabled man or woman, however, was only about $1,000 or $2,000. For two reasons, the differences in average well-being may well be larger than even those estimates of income and net worth suggest. First, most of the retired men and women could anticipate becoming eligible for Social Security retired worker benefits when they reach age 62, which will add to their other income. But the majority of disabled men and women already were receiving a Social Security disability benefit and consequently will not become eligible for any additional Social Security benefit at age 62.(9) Second, workers who decided to retire because they felt that they could afford to do so and wanted to do something else with their time presumably were better off than if they had remained at work, assuming that their expectations were realized. Even though they no longer had the earnings from their former job, they had more time to do other things. Those who were not working because of a disability may have had less of a choice. The higher average income of the men and women who had retired, as opposed to having left the labor force because of a disability, is associated largely with the greater likelihood of their having a pension. Nearly three-quarters of the men and about one-third of the women who had retired were receiving a pension in 2001, accounting for a substantial portion of their average income (see Tables 3 and 4).(10) Most of the average income of retired men came from their pension, earnings of family members (usually a wife), and income from assets. For retired women, their husband's pension and Social Security also were major sources.(11)
Sources of Income for Men Ages 50 to 61 Not Participating in the Labor Force, 2001
Sources of Income for Women Ages 50 to 61 Not Participating in the Labor Force, 2001
Benefits from Social Security Disability Insurance and Supplemental Security Income (SSI), along with the earnings of family members, were the main sources of income for disabled men and women. Almost two-thirds of the disabled men received Social Security disability benefits and more than one-third received SSI. A slightly smaller portion of the disabled women received Social Security disability benefits, but more received SSI. Few disabled men or women received pensions, and, among those who did, their average pension was only about half that of retired pensioners. Those women not in the labor force because they were caring for others or for other reasons had a much higher median income than did women who said they were retired or disabled. The difference is largely attributable to the fact that a much higher percentage of them had husbands still in the workforce and that their husbands' average earnings were substantially higher than the earnings of retired or disabled women's husbands. Care must be taken in drawing conclusions about the future economic well-being of people not in the labor force in 2001 or the well-being of workers who subsequently leave the labor force. In particular, the value of various assets--including homes and stocks--could well be different in the future. Also, it is likely that a smaller percentage of future retired workers will have participated in defined benefit pension plans, as coverage in those plans gives way to coverage in defined contribution plans, such as 401(k) plans.(12) PovertyAnother gauge of a group's economic status is the percentage who are poor. In 2001, an individual under age 65 was considered poor by the Bureau of the Census if his or her family's cash income for the year was below about $9,200. The threshold for a married couple was about $11,900. In general, people in their 50s and early 60s who continue to work are at or near their peak earnings years. Thus, it is not surprising that very few of those who remained in the labor force were poor. Likewise, it is not surprising that the men and women who were not in the labor force had a much higher poverty rate than did those still in the labor force: 21 percent versus 3 percent (see Table 5).
Poverty Rates of Labor Force Participants and Nonparticipants Ages 50 to 61, 2001 (Percent)
The retired workers fared much better than did the men and women who were not in the labor force because of a disability. Fifteen percent of the retired men and 14 percent of the retired women had income below the poverty threshold, compared with 24 percent of the disabled men and 34 percent of the disabled women. One limitation of the way poverty is measured is that it does not take into account assets owned by individuals and their families except to the extent that the assets produce current income (for example, interest and dividends). Two people may have the same cash income, but if one owns a house and has an IRA and the other does not, their actual economic situations are quite different. The assets are available to meet future spending needs, whether or not they produce current income. The extent to which the retired workers with low income but substantial assets might be better off than their annual income suggests can be gauged by translating those assets into the annual income they would produce if converted into an annuity. Such a calculation, using each retired worker's net worth, excluding home equity, reduces the estimated number of poor retired workers: the percentage of retired men with family income below their poverty threshold falls from 15 percent to 11 percent; the percentage of retired women counted as poor falls from 14 percent to 11 percent.(13) Annuitizing each retired worker's home equity reduces the estimated poverty rates of retired men and retired women to about 5 percent and 9 percent. Including the annuity value of the assets of those men and women not in the labor force because of a disability makes very little difference because they had so few assets--especially those with low cash income. Even allowing for the equity in their homes, about one-quarter of the disabled men and one-third of the disabled women still would have income below the poverty threshold. Health InsuranceBesides causing a decline in cash income, withdrawal from the labor force also may put at risk a worker's access to health insurance. Most adults under age 65 obtain health insurance coverage through their own or their spouse's employer. The cost of employer-sponsored insurance generally is much lower than the cost of insurance that a worker can obtain in the individual health insurance market. Moreover, employers typically pay the majority of the premium, which is not counted as taxable income to the worker. Most men and women ages 50 through 61 were covered by a health insurance policy at the end of 2001, whether or not they were in the labor force (see Table 6). Most of those in the labor force were covered by an employer-sponsored plan--either through their own current or former employer or that of their spouse.
Health Insurance Coverage Among Labor Force Participants and Nonparticipants Ages 50 to 61, 2001 (Percent)
The main source of health insurance for people not in the labor force varied widely, depending on whether they were not working because of a disability, because they had retired, or because of another reason. About half of the disabled were covered by Medicare or Medicaid, whereas roughly three-quarters of the retired men and women were covered by employer-sponsored health insurance (either their own or that of a spouse). Women who were not in the labor force because they were caring for others or not interested in working or for other reasons were more likely than other women to be uninsured (19 percent, compared with 13 percent of disabled women, 9 percent of retired women, and 9 percent of women still in the labor force). Characteristics of Men and Women Not in the Labor ForceAs noted previously, 14 percent of the men and 24 percent of the women ages 50 to 61 were not in the labor force in 2001. They differed from their contemporaries who remained in the labor force in several ways (see Table 7). Moreover, among the men and women not in the labor force, there were further distinctions between those who had left the labor force because of a disability and those who had retired.(14)
Characteristics of Labor Force Participants and Nonparticipants Ages 50 to 61, 2001 (Percent)
Educational Attainment. Only 13 percent of the men ages 50 to 61 who were not in the labor force had graduated from college, compared with 34 percent of the men still in the labor force in 2001. Likewise, about one-quarter of the men no longer in the labor force had not completed high school, compared with only 1 in 10 of the men still in the labor force. Those differences are almost entirely attributable to the much lower educational attainment of men who were not in the labor force because of a disability. For instance, while only 5 percent of the disabled men had graduated from college, 34 percent had not finished high school. By contrast, the educational attainment of men who said that they had retired was akin to that of men still in the labor force. The situation for women is similar. Few of the women not in the labor force because of a disability had graduated from college, while 40 percent of the disabled women had not completed high school. Women who had retired also were more likely than women still in the labor force not to have completed high school, but the difference was much smaller. (Women not in the labor force for other reasons--because they were caring for others, for instance--were more likely to have completed high school than were the disabled women, but less likely than were the retired women.) Marital Status. A smaller percentage of men not in the labor force were married (55 percent), compared with men in the labor force (74 percent). That difference, for the most part, is associated with the men not in the labor force because of a disability: only half of the men not in the labor force as a result of a disability were married, compared with over two-thirds of retired men. Although women not in the labor force were as likely to be married as those in the labor force, there were major differences between those women not in the labor force because of a disability and other nonworking women. Less than half of the disabled women were married, compared with three-quarters of the women who had retired. Origin. Men not born in the United States were slightly more likely to be in the labor force than were native-born men; by contrast, foreign-born women were slightly less likely to be in the labor force. For men, the biggest difference was in the share that were retired: only 4 percent of retired men were foreign born, compared with 9 percent of men not in the labor force because of a disability and 11 percent of men in the labor force. Foreign-born women were much more likely than native-born women to be out of the labor force because they were taking care of others or not interested in working: they constituted 19 percent of that group, compared with only 10 percent of the women in the labor force. Age Last Worked. Nearly all (97 percent) of the men who were not in the labor force said that they had worked earlier. Those who left the labor force because of a disability were much more likely than the retired workers to have withdrawn before age 50. A lower percentage of women not in the labor force (88 percent) said that they once had worked, and many more (53 percent) said that they had stopped working before age 50. The women not in the labor force because of a disability or for other reasons were less likely than retired women to have worked at one time, and, if they did, were more likely to have stopped working before age 50. Disability Status. Respondents--whether or not they were working--were asked whether they had a physical, mental or other health condition that limited the kind or amount of work they could do. Predictably, virtually all who cited disability as the main reason they were not in the labor force answered the question affirmatively. In addition, about 15 percent of the men and women who were still in the labor force said that they had a work- limiting disability, as did almost 30 percent of the retired workers. That is, while having a work-limiting disability did not necessarily result in a person's leaving the labor force, it did increase the likelihood that he or she would do so (see Box 1).
The Survey of Income and Program Participation (SIPP) is a longitudinal survey of the population of the United States that has been conducted by the Bureau of the Census since the mid-1980s. Each panel comprises a nationally representative sample of households selected by the bureau and interviewed every four months for up to four years. The sample of the population used in this paper came from the panel begun in 2001, the most recent panel available. The panel originally consisted of about 35,000 households, but attrition reduced the size of the panel interviewed in subsequent waves of the survey. The sample relevant to the main part of the analysis presented in this paper consists of about 8,500 people--approximately 4,100 men and 4,400 women--who were ages 50 through 61 at the end of 2001 and for whom sufficient information existed for each month of that year (the first three or four waves of the survey, depending on when the respondents were first interviewed).
Characteristics and Labor Force StatusMost of the information about the personal characteristics of the respondents reported in the analysis of individuals ages 50 through 61 comes from responses to questions asked in the third or fourth interview. The Congressional Budget Office (CBO) calculated the age of respondents in December 2001 using their reported date of birth; in cases in which the year of birth was reported but not the month, July was used. Labor force status was determined on the basis of answers to questions about activities during each month of 2001. Individuals were counted as participating in the labor force during 2001 if they had worked or looked for work at any time during that year. Otherwise, they were counted as not in the labor force. Respondents not in the labor force were categorized according to their answer to the question "What is the main reason you did not work at a job or business between . . . and today?" Those who responded that they were retired or that they were unable to work because of a chronic health condition or disability were classified, respectively, as "retired" or "disabled." All others were classified as "other." They included those who said that they were temporarily unable to work because of an injury or illness, those out of the labor force because of pregnancy or childbirth, those taking care of children or others, those going to school, those unable to find work or who had been laid off, those not interested in working in the paid labor force, or those not working for other reasons. However, about 80 respondents who said that they were receiving Social Security Disability Insurance benefits did not give disability as their main reason for not working. CBO reclassified them as disabled.
Income, Poverty, and AssetsThe sources and amount of a respondent's annual income were calculated by summing the respondent's answers to the monthly-income questions asked in each interview. The annual incomes reported in this paper were calculated by summing the incomes reported during the twelve months of 2001. Individuals were counted as poor if their family income fell below the poverty threshold used by the Census Bureau for their family size. The Census Bureau collected asset information for each household in a set of supplementary questions asked during the third interview, which occurred in late 2001. Net worth is based on the sum of the market value of assets owned by every member of the respondent's household minus the liabilities owed by household members. Assets include homes, other real estate, cars, businesses, and financial assets. Individual retirement accounts are included, but the value of future Social Security and pension benefits is not. Unlike the information on income, the data on assets and liabilities include household members who are not related to the respondent. Respondents might report that they were receiving benefits from one public program when those benefits actually came from another source, or they might incorrectly report the amount of income they had received. For example, some of the respondents who said that they received Social Security retired worker benefits were not old enough to be eligible for those benefits. (That particular discrepancy could result from a mistake either about the actual source of their income or about their age.)
Annuitizing Net WorthSomeone who is not in the labor force and has considerable assets but very little income is in a better position to meet his or her spending needs than someone with the same income who has few assets. Some respondents who lived in households with substantial net worth reported little or no income from interest, dividends, or other asset-related sources. In many cases, the lack of reported income simply reflects the fact that some assets--notably the equity in owner-occupied homes--do not produce cash income. In some cases, the lack of reported income may be because the actual owner of the asset is someone living in the respondent's household who is not a relative. In other cases, the respondents may not report income from an asset because they do not consider that income as available for current consumption or because they do not remember that particular income source. Interest and dividends from assets held in a 401(k) or individual retirement account, for example, might not be reported because they are not considered current income for tax purposes. To get an indication of how much difference those assets might make, CBO calculated the annual income that each respondent's reported level of assets could generate if those assets were used to purchase an annuity. For single people, the annuity would provide an annual income for the remainder of the annuitant's life, adjusted each year for inflation, up to 3 percent. For married people, the annuity would provide an annual income for the remainder of the annuitant's life, or that of his or her spouse, also adjusted for inflation. The specific annuity rates used for those calculations were based on the age and marital status of the respondent, using rates quoted by the Thrift Savings Plan on its Web site in mid-September 2004. The relevant rates ranged from 3.7 percent for a married annuitant age 50 to 5.9 percent for a single annuitant age 61. For example, the annuity for a single person age 61 who reported a net worth of $100,000 would be $5,900 per year. If the amount from the annuity exceeded the interest, dividends, and other property income reported by that person, it was substituted for the reported amount of property income and used to produce an adjusted income. Two sets of estimates were made: one based on the annuitization of the respondent's entire net worth, including home equity, and the other based on net worth excluding home equity. Adjusted poverty rates then were calculated for each labor force status group on the basis of those adjusted incomes. As reported in the text, the adjustments were largest for those men and women who were not in the labor force because they had retired.
|