Information in this report is based on "Early Retirement Decisions in Non‑Metropolitan Communities" by Robert B. Enright Jr. and Lawrence A. Weiser in Proceedings of the 9th Conference on the Small City and Regional Communities, University of Wisconsin‑Stevens Point, March 30, 1990.

Introduction

A substantial reduction in the labor force participation of older and middle‑aged men is one of the most significant demographic trends affecting the U.S. economy. A generation ago, only half of elderly men, that is men 65 and older, could be classified as retired, currently however, over 80 percent can claim that status. Retirement in the United States is often viewed as a period of well deserved leisure following many years of employment. In the last two decades many Americans have availed themselves of this "golden pond" life‑style at ages well below the "normal" retirement age. Approximately one third of men between the ages of 55 and 64 are not in the labor force at all, and many more are no longer involved in the occupation which served as their principle career activity.

This report will describe the phenomenon of early retirement in the U.S., in Wisconsin, and in our local area. Some of the reasons being advanced to explain the increase in early retirement will be analyzed, and the implications of changes in retirement behavior will be explored. The important relationship between retirement decisions and the social security system will be discussed. 

Retirement Trends

Figure 1 shows the dramatic decrease in the labor force participation rate for men in the 55‑64 age range. As recently as 1960, close to 90 percent of men in this age group were working or actively seeking jobs. In 1989 only 67 percent were in the labor force leaving one‑third to be classified as retired. Although this trend appears to be bottoming out, there is some evidence that it could resume its downward path and even spread into younger age groups. Figure 1 clearly indicates that the retirement behavior of Wisconsin men roughly parallels the national pattern with a lag of approximately five years. The response to the early retirement "window" for Wisconsin public employees is a good example of this phenomenon. Retirement applications from state employees in 1990 are running 65 percent above previous years.

Trends in women's retirement patterns are more difficult to identify since their labor force participation rates have been rising for all age groups. Married women who work outside the home have traditionally retired at the same time as their husbands, and since they are two years younger than their husbands on average, it would be reasonable to expect a trend toward early retirement for women. However, some early retirement decisions by men may be influenced by the fact that their wives will continue to work and receive health insurance benefits for the family.
 

Note that the percent of men who are no longer in the labor force is somewhat lower in Central Wisconsin than in the State as a whole. Partly this can be explained by the fact that manufacturing and agriculture are highly important economic activities m our communities as compared to tourism and recreation which dominate the economies of Northern Wisconsin.
 

Reasons for Increased Early Retirement
 

Social scientists have not been able to devise one simple explanation for the trend toward earlier retirement. Instead, they suggest several factors which seem to play important roles in determining retirement age. These reasons can be divided into push and pull factors. Are American men pushed out of the labor force by layoffs, plant closures, forced retirements, a lack of employment opportunities, obsolete skills, age discrimination, and poor health; or are they pulled into early retirement by the attraction of additional years of leisure and by financial incentives such as social security, the availability of disability payments, private pensions, and the income from spouse's jobs? Obviously, there are: individuals who have decided to retire for each of these reasons, but which causes are most responsible for the social trends that have been observed? Researchers have been unable to successfully isolate the influences of each of these variables. 

To illustrate the complexity of this process, consider the effect of income on retirement decisions. It is a widely held view that Americans retire as soon as they are financially able to do so. The public perception of the early retiree is a physically fit professional who possesses the financial means to leave the work force and purchase well deserved leisure. His high income enables him this freedom, which was denied to all but the elite in earlier times. Consistent with this view, economists have shown that in the United States, real per capita disposable income has increased by more than 90 percent since 1955. This "income effect" hypothesis proposes that higher incomes have led to the increase in early retirement by making it more affordable than in the past. Contrary to this hypotheses, studies have shown that men in low paying jobs, not high paying jobs, are most likely to retire early and Wisconsin county data confirms this. Low paying jobs are more likely to be physically demanding, leaving workers vulnerable to disability and health benefits and employment income hazards. Moreover, the difference between disability are smaller for men in low paying jobs, and this provides poorly paid men an additional incentive to opt for early retirement. Alternatively, high income individuals are aware that they are giving up a great deal by choosing to retire. In pure dollars and cents, they are sacrificing more than the average worker, and therefore may delay retiring. The latter is called the "substitution effect." As the preceding analysis has demonstrated, the effect of income on retirement decisions is far from clear cut. 

Poor health is the most controversial explanation offered for the increase in early retirement. When early retirees were surveyed concerning the reasons they stopped working before age 65, roughly two‑thirds cited poor health as the major cause, and Wisconsin county data supports this relationship. Paradoxically, during the last several decades when early retirement has been increasing, health as measured by declines in mortality rates has improved markedly. Possibly modern medicine is successful at postponing death, but leaves many survivors too frail to work. One recent study based on data from the National Center for Health Statistics concludes that most of the increase in life expectancy between 1970 and 1980 was in years with a disability which limits primary activity. Of course "poor health" may not be an objective reason for retirement as much as it is a rationalization for forced retirement. Moreover, the increased availability and acceptability of disability benefit programs may have contributed to this trend. An interesting footnote to the debate on the relationship of health status and retirement is the reduced physical requirements of jobs in our economy. As the proportion of the labor force that is involved in agriculture and manufacturing has declined, jobs have become less physically demanding, and this process would be expected to extend the age of labor force participation.

Table 2 cross‑classifies all 72 Wisconsin counties by economic and demographic characteristics. The non‑metropolitan counties have much higher early retirement rates than the metropolitan counties. This is explained by the very high early retirement rates in Wisconsin's twenty recreational counties which are primarily due to net migration into these counties by retirees who have lived in other parts of Wisconsin and adjoining states. Also, the groups of non‑metropolitan counties which are characterized by manufacturing and agricultural activity have early retirement rates which are significantly higher than those of metropolitan counties. 

Means of early retirement rates in Wisconsin counties by classification of counties. Categories are not mutually exclusive. Number of counties in each category is in parentheses.
 

Walworth County is classified as both recreational and university. The following five counties are classified as both recreational and manufacturing: Door, Forest, Marinette, Menominee, and Oconto. 

Social Security and Retirement Decisions

Why should we be concerned about the trends in retirement that have been analyzed above? One obvious reason is to anticipate and understand the retirement plans of our employees, co‑workers, and, from a personal perspective, our families and ourselves. Clearly, the "normal" retirement age has dropped below 65, and may continue to decline. From the standpoint of social policy, we must analyze whether our national retirement system will be able to cope financially with the changes that are taking place. Finally, will the U.S. be able to maintain the productive work force necessary to provide an improving standard of living for our population in the decades ahead?
 

Social Security (OASDHI) is the foundation of our nation's retirement system, and it has proved to be one of the government's most successful programs. Currently, over 80 percent of the population is fully insured by social security, and 90 percent of middle‑aged workers are covered. 11 percent of the U.S. population are receiving retirement benefits, and those benefits average well above the poverty threshold for the typical retired worker and spouse. Consequently, the poverty rate for older Americans is lower than for those under 65. Over 40 percent of the income of elderly Americans consists of social security benefits.
 

As the number and proportion of elderly Americans continues to increase, it will become more difficult to finance the large payments required under social security benefit schedules. In 1983 the regulations of the system were amended to maintain the financial integrity of the system. The age at which full retirement benefits may be paid will be increased gradually from 65 to 67. Taxes have already been raised several times to create large surpluses in the trust fund with the plan to draw those reserves down as benefits begin to exceed revenues.
 

A great deal of controversy surrounds the size and timing of these trust fund balances, but the rhetoric, pro and con, has masked the more fundamental economic issue. That question is whether a smaller, older labor force can produce enough goods and services to provide a rising standard of living for the American people of all ages. A recent study indicates that a continuation of the per capita real GNP growth rate at its historical average of 1.85 would require massive increases in capital investment and technical progress. This scenario does not appear possible. However, an acceptable per capita GNP growth of 1.5 percent is achievable with reasonable increases in capital investment. This is not a rosy prospect for many Americans who are struggling to provide a decent standard of living for their families, but it is far from a doomsday prediction. The arithmetic of our economic future indicates that with hard work, careful planning, and a little luck, retirees and their grandchildren will be able to maintain and enhance their standard of living.