Work, Retirement, and the Encore Career: Elders and the Future of the American Workforce

Abstract

Retirement trends have altered dramatically in the past few decades, toward working longer. How and when will current workers, especially baby boomers, retire? The United States is undergoing a dramatic transformation in retirement trends and patterns-when and how older Americans leave their career jobs, and eventually exit the labor force entirely. This article will address the long-running trend, going back more than 100 years, document a recent break from this experience, and ask why the trend has ended. It will also describe the way in which people leave their career jobs-all at once, or gradually, in stages (Quinn, 2000, 2002). Will future retirement patterns resume the long-run trend or extend the different patterns observed recently? Have we entered a new retirement era? (For those in a hurry, the answer is yes!) American men had been retiring earlier and earlier since the 1880s, reflecting the increasing wealth of the nation. In the mid-1980s, however, this trend stopped and then reversed, and American men (and women, even more so) began leaving the labor force later than had their predecessors. Many retire in stages, using bridge jobs after leaving career employment. These new trends have now lasted a quarter-century, suggesting they are not just a temporary exception to the prior long-run story. In addition, most of the likely causes of this dramatic shift are permanent alterations in the retirement environment. In future decades, older Americans will live longer and healthier lives. They will face different financial incentives and more economic uncertainty than in the past, and are likely to respond in the easiest way-by working a few more years. Some will work at their career jobs and others on post-career bridge jobs, often in different occupations or industries, and sometimes in different locales. Job exit will look more like initial job entry (with different opportunities sampled) than it used to. This is generally a good thing-for older workers themselves, for the firms that employ them, and for the nation as a whole. The Historical Record Dora Costa (1998) has estimated employment rates for American men ages 55 to 64, and 65 and older, from 1850. For both groups, there has been a steady decline, with a few short-lived exceptions, from 1880 until the late twentieth century- more than 100 years of earlier and earlier retirement (Costa, 1998, Figure 2.1).1 Considerable research has analyzed why this labor supply decline occurred. The primary economic answers are wealth and, later, the financial incentives by public and employer pension plans. As citizens became wealthier over time, they spent these additional resources in many ways, including staying in school longer, entering the labor force later, working fewer hours per year, and purchasing more leisure late in life; i.e., retiring earlier. Defined-benefit (DB) employer pension plans and Social Security grew in importance, but often declined in value for workers who remained in the labor force long. As a result, many older Americans could afford to retire earlier and were often penalized financially if they did not. Table 1 documents the decline in male labor force participation rates (LFPR) from 1950 through 1985. For men ages 55 and 60, who are too young for Social Security retirement benefits, the declines are modest, only 8 percent and 16 percent. For those age 62, rates dropped by more than one-third.2 For men of ages 65, 68, 70, and 72, the declines were much larger, on the order of 60 percent to 70 percent. If one defines the average age of retirement as the age at which the participation rate declines to 50 percent, it fell from age 70 in 1950 to age 65 in 1970 and to age 62 by 1985, an eight-year decline in only thirty-five years. For older women, with data since 1965, there is much less change over time (see Table 2). Participation rate changes for women too young for Social Security (ages 55 and 60) were plus-9 and 0 percent between 1965 and 1985. …

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 69 citations exist upstream.