انتقال به خود اشتغالی در سنین بالاتر: نقش ثروت، بهداشت، بیمه درمانی و سایر عوامل

Relatively little research has been devoted to studying self-employment among older workers although they make up a disproportionate share of the self-employed workforce. This study uses 5 waves of panel data from the Health and Retirement Study to investigate the determinants of labor force transitions to self-employment at older ages. We estimate a multinomial logit model of transitions from wage and salary employment to self-employment, retirement or not working. Results are compared with those found in previous studies. New findings on the impact of health on transitions to self-employment are highlighted.

In 2002, 14.4 million U.S. workers, or 10.5 percent of the workforce, were self-employed in incorporated or unincorporated businesses.1 Given that rates of self-employment increase with age, a disproportionate share of the self-employed are middle-aged and older workers. Slightly more than half (54 percent) of the self-employed (in unincorporated businesses only) in 2002 were ages 45 and above although this age group represented just over one third (38 percent) of the workforce in total.2 Some of these older workers have been self-employed for much or all of their working careers while others have made the transition to self-employment later in their careers, often as part of the transition to retirement. Data from the March 2001 Current Population Survey (CPS), plotted in Fig. 1 show that self-employment rates start out very low among the youngest workers, increase steadily until traditional retirement ages (60 to 64), and then begin to rise even more sharply around retirement (65 and above).3 This pattern of rising self-employment rates at the end of the labor market career may reflect higher rates of retirement out of wage and salary work compared to self-employment, as well as transitions to self-employment at older ages. Indeed, tabulations from the 1998 cross-section of the Health and Retirement Study (HRS) reported by Karoly and Zissimopoulos (2004) reveal that nearly one third of workers age 51 and above who are self-employed made the transition to self-employment at or after age 50. Transitions to self-employment among older workers have implications for current levels of well-being (e.g., earnings) among those nearing retirement, and may also have implications for the accumulation of retirement benefits and the accumulation or decumulation of other financial assets. Income from self-employment is more variable compared with wage and salary work, with both upside and downside risk reflected in a more dispersed earnings distribution among the self-employed (Hamilton, 2000). The self-employed, including older self-employed workers, are also less likely to be covered by a pension plan or employer-provided health insurance compared with their wage and salary counterparts (Karoly and Zissimopoulos, 2004). Moreover, entry into self-employment may require investment capital, potentially placing current or future retirement assets at risk. At the same time, self-employed workers report being more satisfied with their work, perhaps due to greater autonomy and flexibility (Hundley, 2001a). Thus, self-employment may be an important component of phased retirement from the labor force. Despite the importance of self-employment in the labor market behavior of older workers and the implications for retirement income security, it is a relatively understudied area. The goal of this paper is to fill an important gap in our knowledge base about the labor force transitions of older workers generally and in particular, workers' transitions to self-employment in middle and late life. As we detail in the next section, there is a small but growing number of recent analyses of self-employment behavior among all workers, but far less attention has been paid to self-employment among workers in middle age and beyond. Longitudinal data from the 1992 to 2000 waves of the HRS, described in the third section, are used to examine patterns in labor force transitions after age 51. Using these data, in the fourth section, we compare transitions between employment classes and into retirement for self-employed workers and their wage and salary counterparts. We employ a decomposition (Fuchs, 1982) to reveal the source of increases in self-employment with age after which we describe our methods and results. We use detailed data on demographic characteristics, job characteristics, income and wealth from various sources, health status, access to health insurance coverage, retirement expectations and a spouse's characteristics to identify the determinants of who becomes self-employed later in life, who leaves the labor force, and who continues to work in the wage and salary sector. The final section concludes the paper.

Our study of self-employment among workers age 51 and above was motivated by the relative paucity of studies examining the patterns and determinants of self-employment among older U.S. workers. Although self-employment rates rise steadily with age, our knowledge of self-employment behavior is largely confined to younger workers or analyses of the self-employed workforce as a whole. There has been little effort to identify the factors associated with transitions to self-employment at older ages, and to determine how the determinants might vary for men and women. We compare our findings to those that do exist in the literature, particularly that of Fuchs (1982) who examined transitions to self-employment using an earlier birth cohort from the RHS. Longitudinal data from the HRS reveal that self-employment rates peak at 24 percent for women and 38 percent for men at ages 66 and 65, respectively. This pattern reflects the higher rates of retirement out of wage and salary work compared with self-employment, as well as transitions from wage work to self-employment at older ages. There is also evidence in the HRS that self-employed workers are more likely to transition to part-time work, perhaps as a bridge to retirement, compared with their wage counterparts. This pattern suggests that there may be greater flexibility among the self-employed to reduce their hours of work rather than completely retiring from the workforce. Our multivariate models of the determinants of transitions to self-employment from full-time wage and salary work for men and women ages 51 to 67 identifies both push and pull factors that may drive the decision among older workers to become self-employed. For example, poor health-as measured by the presence of a work-limiting health condition-appears to be a push factor. This result differs from Fuchs' (1982) earlier study, which found no impact of health on transitions to self-employment for men. Moreover, health has not been identified as a prominent push factor for younger workers to move to self-employment. This result may indicate that older workers with a work-limiting health condition are better able to accommodate their condition and continue working if they are self-employed compared with employment in the wage sector. Higher wealth holdings may be a pull factor. Men and women in the highest wealth quartile were significantly more likely to move to self-employment than their counterparts in the lower wealth quartiles. This result is consistent with prior research that shows that access to capital is a significant determinant of becoming self-employed although our study is one of the first to show this result for transitions to self-employment at older ages. In addition, we find that transitions to self-employment are more likely for men that have ever received an inheritance but less likely if their spouse had received an inheritance in the past. Among the measures of employer-provided benefits, we find that having pension coverage in the current job reduces the likelihood of becoming self-employed-a result that is consistent with prior literature. We find that health insurance coverage may be capturing job quality or possibly a form of job lock that deters transitions that would lead to a loss of health coverage. At the same time, men with a spouse who has health insurance on their current job or women who spouse has retiree health coverage are less likely to move to self-employment, a result that goes against the job-lock hypothesis. In terms of job characteristics, we find that workers with more flexibility on their wage and salary job are more likely to become self-employed, which does not accord with the view that workers who desire more flexibility later in their career move into self-employment. It is possible that great flexibility in the wage job allows workers to combine wage and self-employment work simultaneously, thereby easing the transition to self-employment. Alternatively, workers who would find self-employment appealing may also be drawn to wage jobs with more flexibility in hours. In addition, transitions to self-employment are higher for certain occupations-a result that is consistent with earlier research (on older men) suggesting that wage workers in jobs with more autonomy or pay tied to performance were more likely to become self-employed. Controlling for other factors, demographic characteristics were generally not strongly associated with transitions to self-employment for men or women. Our results on a spouse's characteristics are suggestive of the fact that self-employment decisions may be best reviewed from the perspective of the household rather than the individual. Transitions to self-employment for older men and women appear to be driven by many of the same factors yet there are some important differences, particularly in the role of spousal variables. Our models also confirmed many of the main results of prior studies of the determinants of transitions to retirement. For example, movements to retirement rise with age, financial wealth, poor health, access to retiree health benefits and, particularly for men, are somewhat lower for the more educated. Retirement expectations are also highly predictive of future retirement. Whether or not future cohorts of older workers move into self-employment at the same rates as the cohorts we observe in the HRS remains to be seen. Such transitions are likely to be affected by the state of the economy, the value of individual wealth portfolios, and other factors. To the extent that self-employment requires an upfront investment, older workers may draw down their assets to invest in a new business, an investment that may or may not be paid back during the retirement years. Their ability to draw down assets to invest in a new business may depend on the level of accumulated financial assets, and on their ability to cash-out pension benefits. Public policies and market forces that we have not considered may also affect self-employment transitions for future cohorts of older workers. For example, policies with respect to the Social Security system such as early and normal retirement ages, the treatment of earnings among those drawing benefits, and the structure of the Social Security self-employment tax may have implications for whether older workers choose to move to self-employment from the wage sector, and when workers choose to retire from self-employment. Other aspects of state and local policy pertaining to unemployment insurance, worker's compensation, disability insurance, and other social insurance programs may also differentially influence the self-employment decisions of older workers compared to younger workers. For example, some states have begun experimenting with offering lump sum unemployment benefits that could be used to start a new business instead of searching for another wage and salary job. Further advances in technology, especially communications and information-based technologies, may further facilitate increased rates of self-employment through home-based enterprises.