Self-Employment

Self-employment has greater importance among older workers. The self-employed in 1994, for example, numbered 1.5 million at ages 55 to 64 and 1 million at ages 65 and older (Bregger, Table 4). Although a tenth of workers in the labor force was self-employed in the 1979 to 1996 period, the self-employment rate at ages 55 to 64 was about two and one-half times that at ages 25 to 34 (Manser and Picot, p.4 and Table 2). The self-employment rate in the labor force was about 18 percent at ages 55 to 64 years and about 25 to 29 percent at age 65 years and older.

Increasing percentages of self-employed older workers result from either the self-employed retiring later than employed workers or from workers shifting from wage and salary work into self-employment after retirement from a long-term, career job. Bruce, Holtz-Eakin, and Quinn examined the job shifting between 1992 and 1996 among persons born from 1931 to 1941 using the Health and Retirement Survey data. A net increase in self-employment occurred because more wage and salary workers shifted to self-employment than self-employed workers shifted to wage and salary jobs. Further, the self-employed in 1992 were more likely than wage and salary workers to be employed in 1996. Finally, 30 percent of the re-entrants into the labor force between 1992 and 1996 entered self-employment. Analyzing who was more or less likely to make these changes, the investigators found that women and African Americans were less likely to make the transition from wage and salary work to self-employment. Interestingly enough, the portability of health benefits from wage and salary jobs appears unrelated to the decisions to continue working or to move into self-employment jobs.

Are older workers attracted to self-employment or ‘‘pushed out’’ of wage and salary jobs at older ages? The answer is conjectural but several influences affect older workers. Some are attracted to self-employment by the flexibility and control over hours of work and wages. While wage-and-salary workers commonly work in full-time, full-year jobs, the self-employed are more likely working in part-time and/or part-year jobs (Devine). Part-time work would be particularly attractive to older workers because of limits set by Social Security legislation over the taxable earnings allowable while collecting benefits. Social Security limits the taxable earnings of some working beneficiaries, reflecting its goal of replacing lost income from retirement. Social Security legislation allows working beneficiaries at ages 62 to 64 to keep full benefits with earnings up to a threshold ($10,680 in 2001) and then reduces benefits $1 for each $2 of earnings over the threshold. Legislation in March 2000 eliminated a similar restriction on working beneficiaries at ages 65 to 69. Consequently, workers in full-time, full-year jobs would be unable to receive full Social Security benefits, while those who flexibly control their work hours could receive full benefits working part-time.

Consistent with the importance of reducing hours, about 84 percent of men and 74 percent of women reported reduced hours between their preretirement jobs and their postretirement jobs shortly after starting Social Security benefits in mid-1980 to 1981 (Iams 1987). In the decade after starting Social Security benefits, about one-third of retired men and one-fourth of retired women worked, with a much higher percentage self-employed on their ‘‘retirement job’’ than on their longest job before retirement (Iams 1995; Social Security Administration).

Although self-employment attracts many older workers, some may be ‘‘pushed’’ toward self-employment. Influences such as health and job loss could push wage-and-salary workers toward self-employment, and a major push probably would be pension plans, which seldom permit beneficiaries to continue working on the pension-covered job (Bureau of Labor Statistics 1999). Defined contribution plans may require that a person leave employment in order to receive an annuity or a lump sum payment from the pension plan. Defined benefit plans commonly require job termination for benefit receipt, and penalize part-time work prior to retirement by indexing the pension to the highest few years of earnings. Years of prior full-time earnings would determine benefits, and inflation would erode the value of the highest earnings for pension calculation of those who reduced to part-time work prior to retirement.

Who are the self-employed? Unfortunately, information on jobs of the self-employed must come from information for the total labor force rather than from information on older workers alone (Manser and Picot; Bruegger; Devine). This reflects small sample sizes in data and an infrequent focus on older self-employed workers. In 1975 and 1990, self-employed jobs were more likely held by men, non-Hispanics, the married, and the more educated (Devine, Table 4). Although men were more likely than women to be self-employed, the rate of self-employment among women increased after 1975 while remaining relatively constant among men (Devine). There are substantial differences in the self-employment rate among sixty ethnic/racial groups using 1990 census data (Fairlie and Meyer). Ethnic groups with higher earnings than average seem more likely to enter self-employment.

What kind of jobs do the self-employed generally have? The self-employed were more likely among workers in agriculture, construction, retail and wholesale trades, financial, and other service industries (Manser and Picot, Table 2). The self-employed in 1996 also were more likely among managers, sales workers, precision production or craft workers, and farming or related occupations (Manser and Picot, Table 2). Looking at recent Social Security beneficiaries, the jobs of the self-employed varied by gender (Iams 1987). Beneficiary women most commonly were working in sales; as general managers, hairdressers, or cosmetologists; and as bookkeepers or accountants. About a quarter of the beneficiary men were working as managers and professional employees, primarily managers, lawyers, accountants and auditors, and clergy, while another quarter were in sales and about a quarter also worked as farmers and grounds keepers or gardeners. These self-employed beneficiaries averaged lower median hourly wages than the beneficiaries working in wage and salary jobs.

Self-employment may be an effective method for increased earnings over time. Holtz-Eakin, Weathers and Rosen (2000) found self-employment increased earnings from 1969 to 1990 when looking at five-year intervals with the Panel Study of Income Dynamics. They found that a higher percentage moved up from lower income levels in a five-year interval than wage and salary workers when they were self-employed or when they became self-employed. This occurred for both men and women and for African Americans and non-African Americans. However, the self-employed in upper income levels were more likely to experience downward mobility relative to wage and salary workers.

Income varies between two clusters of self-employed workers: those legally incorporated as a small business with an owner and those who are sole proprietors or in a partnership (the unincorporated self-employed). The unincorporated self-employed receive lower income than wage and salary workers (Devine, Table 9 and Table 10). For example, unincorporated, self-employed men receive 85 to 90 percent and women receive only 54 to 59 percent of the median hourly income relative to wage and salary workers. In contrast, the self-employed with incorporated businesses received 25 to 49 percent higher hourly income than wage and salary workers depending on gender and year. The self-employed owners of incorporated businesses are disproportionately men and full-time workers (Devine, Table 9 and Table 10). A gender differences remains even after considering full-time work and incorporated self-employment, because self-employed women had far lower median earnings than men in both incorporated and unincorporated jobs of part-time and of full-time workers. The advantages of incorporation may have increased over time as evidenced by an increasing percentage of incorporation among the self-employed—from 24 percent in 1975 to 31 percent in 1990 for men, and from 8 percent to 18 percent, respectively, for women.

One possible reason for the rapid increase in incorporation of the self-employed could be the legal advantages of business ownership, such as provision of health and pension benefits. While employers provide health and retirement benefits through costs partly supported by tax advantages, the self-employed have to pay these costs themselves if they are unincorporated. The self-employed were less likely to have health insurance coverage in 1990 than the wage and salary workers, taking into account gender and full-time/part-time hours of work (Devine, Table 11). Among the self-employed, the incorporated self-employed were more likely to have health insurance than the unincorporated self-employed. About 90 percent of incorporated self-employed had coverage although only a quarter of women and half of men had jobs that provided coverage rather than coverage from other family members. In contrast, the majority of unincorporated self-employed had health coverage but seldom from their own job—23 percent of men and 9 percent of women had insurance through a job. The majority of men and women wage and salary workers have pension coverage provided by an employer. The self-employed can have retirement coverage by saving tax advantaged funds in Individual Retirement Accounts or Keogh plan accounts. In 1993, about 16 percent of the self-employed ages 30 to 54 had any pension or Keogh account coverage compared with 60 percent of wage and salary workers (Iams 1995, Table 1).

Given the marked economic advantages displayed by the incorporated self-employed, it is no surprise that those with assets in businesses are concentrated among the upper ten percent of the wealth distribution in America (Kennikell and Sunden 1997, Table 4 and Table 5). Among those aged sixty-five and older, the upper wealth decile owns 82 percent of all business assets and the upper 0.5 percent own 44.5 percent of business assets.