Seniors' self-employment

View the most recent version.

By Sharanjit Uppal

Full article in PDF

• Many working seniors are self-employed
• Self-employment growing at slower pace than paid jobs among seniors
• Long-term growth among the incorporated self-employed
• Self-employed seniors concentrated in a few industries
• Factors associated with seniors' self-employment
• Modelling self-employment among senior workers
• Higher-income seniors are more likely to be self-employed
• University degree increases probability of self-employment
• Seniors' self-employment is often a family affair
• Summary
• Data source and definitions

The potential impact of workforce aging is widely discussed as baby boomers enter their retirement years. The Minister of Human Resources and Social Development Canada established the Expert Panel on Older Workers in 2007 in response to two issues that could affect the standard of living in Canada. First, population aging could reduce the growth potential of the Canadian economy since income generally declines with retirement. The second question is how to provide for the older workers who are displaced as the economy adapts to a changing environment.

The risks associated with an increase in the old-age dependency ratio—defined as the ratio of retired individuals to the number of working people—are often debated. On the one hand, many believe that the increase in the number of retirees will put a strain on public resources, and possibly also lead to labour shortages in certain areas. Others argue that recent cohorts are likely to work longer since they tend to be healthier, better-educated, and more entrepreneurial than previous generations of retirees.

Indeed, the employment rate among seniors has increased in recent years (Uppal 2010). Between 1996 and 2006, the share of working seniors (65 and over) climbed from 11.8% to almost 14.8% among men, and from 4.0% to 5.8% among women. However, the fact that many of these employed seniors are self-employed has not been widely reported. According to the latest census data, 44.1% of senior men and 28.6% of senior women who had a job in 2006 were self-employed. Moreover, self-employment among older Canadians increased by more than 100,000 during the recent economic downturn (LaRochelle-Côté 2010).

Self-employment is typically seen as providing more flexibility and imposing fewer constraints on retirement timing, which could explain why many working seniors choose self-employment (Quinn 1980 and Hochguertel 2010). In addition, seniors typically have higher levels of human and financial capital to invest in a small business, two conditions thought to stimulate entrepreneurial activity (Beaucage and Najem 2006, and Zissimopoulos and Karoly 2007). Alternatively, some seniors may be pushed into self-employment through a lack of paid employment opportunities.

Despite the high incidence of self-employment among the senior population, little has been published on the topic recently.¹ The first objective of this study is to present new information on self-employment trends among seniors and examine their industrial and occupational profiles. The second is to examine factors associated with self-employment after age 64. Since a large sample is required to obtain a detailed description of seniors' self-employment, this study uses census data (see Data source and definitions).

Many working seniors are self-employed

Although participation in the job market drops significantly at age 65, many of those who remain on the job are self-employed. In 2006, 14.8% of senior men held a job (Table 1). As a proportion of senior men, 8.2% were paid employees and 6.6% were self-employed. Among senior women, the employment rate was 5.8%, consisting of 4.0% who were paid employees and 1.7% who were self-employed.² Thus, among seniors, 44.1% of working men and 28.6% of working women were self-employed.

The proportion of workers who are self-employed is even higher among older seniors.³ For example, the proportion of the self-employed among working men age 65 to 69 was 39.4% in 2006, but was greater than one-half of those who were still working after age 75.⁴ Among women, the self-employed made up one-quarter of workers age 65 to 69, rising to more than one-third of working women age 70 and over.

Self-employed seniors are more likely to work full year, full time than paid employees (Chart A). Among men, 43.8% of self-employed seniors worked full year, full time in 2005 compared to 39.8% of senior employees. Similarly, 32.0% of self-employed senior women worked full year, full time versus 29.6% of their paid counterparts. In contrast, paid employees age 25 to 54 were more likely than the self-employed to work full year, full time in 2005 (data not shown). Therefore not only is self-employment more prevalent among senior workers than younger workers (see Comparisons with younger age groups), but self-employed seniors are also more likely to report that they work full time for the full year.

Self-employment growing at slower pace than paid jobs among seniors⁵

Uppal (2010) found that the proportion of seniors who work past age 64 has increased since 1996, following 15 years of decline. For example, the employment rate for senior men increased from 11.8% to 14.8% between 1996 and 2006. Over the same period, the share of paid employees as a proportion of senior men increased from 5.4% to 8.2% (Chart B). Similar patterns were found among women. Did self-employment also contribute to the recent increase in employment among seniors?

The answer is yes, but the number of paid employees increased faster than the number of self-employed seniors and the proportion of senior workers in self-employment fell as a consequence (Chart C). Between 1996 and 2006, the share of the self-employed among working seniors declined from 53.5% to 44.4% among men, and from 33.7% to 29.4% among women.^6,7

By historical standards, self-employment was still relatively high among seniors in 2006. Among senior workers in 1981, 37.8% of men and 13.0% of women were self-employed. The relative importance of self-employment for seniors increased rapidly over the next 15 years, especially among men: by 8.5 percentage points from 1981 to 1986 and by 7.9 points from 1991 to 1996. Both periods were characterized by relatively weak labour markets. Other studies suggest that many workers are 'pushed' into self-employment during periods of economic stagnation (Kuhn and Schuetze 2001).

Long-term growth among the incorporated self-employed

The self-employed can be classified into two categories: the 'incorporated,' who own a separate business entity, and the 'unincorporated,' who do not. Both the incorporated and unincorporated may have paid employees resulting in four categories (Table 2). The majority of self-employed seniors were unincorporated: about two-thirds of self-employed men and three-quarters of self-employed women. And the vast majority of those who were unincorporated did not have paid employees: 80.4% among women and 74.2% among men.⁸ In contrast, more than one-half of incorporated seniors had paid employees in 2006.

While the unincorporated without paid help—also referred to as own-account workers—still comprise the majority of self-employed seniors, their share has declined. Between 1981 and 2006, the share of self-employed men who were unincorporated without paid help fell steadily from 67.7% to 50.5% (Chart D). There was also a modest decline among women—from 61.8% to 58.9%.

In contrast, the share of incorporated self-employed men more than doubled, from 15.1% in 1981 to 31.9% in 2006. Incorporated women also increased their share of self-employed seniors: from 16.9% in 1981 to 26.8% in 2006. Thus self-employment among seniors is increasingly comprised of incorporated business owners, as opposed to own-account workers.

Self-employed seniors concentrated in a few industries

In comparison with paid employees, self-employed seniors were more concentrated in a few industries. More than three-quarters of self-employed seniors could be found in three industries: primary goods, consumer services and business services (Table 3). One-third of self-employed men were in primary goods and one-third of self-employed women were in consumer services. By way of comparison, senior men who were paid employees were typically employed in consumer services, business services and manufacturing (63.8%). Senior women working as paid employees were likely to work in consumer services, health and business services (72.9%).

Self-employed seniors were also more likely to be concentrated in a few occupations (Table 4). Almost one-half were employed in the top 10 occupations (out of 520). The most frequently reported occupation was farmer or farm manager, accounting for 24.2% of self-employed men and 17.2% of self-employed women. Concentration in the top 10 occupations was much lower for paid employees: 28.4% for men and 37.6% for women.

Factors associated with seniors' self-employment⁹

Research suggests that self-employment is related to a number of factors, including financial capital, education, and personal characteristics (Fuchs 1982, Bruce et al. 2000, and Zissimopoulos and Karoly 2007).

The financial capital hypothesis suggests that individuals in wealthier families are more likely to be self-employed because the associated risks and investments are more easily addressed when individuals are financially sound (Georgellis et al. 2005). Even though the census does not contain any information on financial wealth, it is possible to test that hypothesis by using "adjusted family income" as a proxy.¹⁰ It is calculated by subtracting the employment income of the respondent from the total family income,¹¹ and next adjusting to account for the size of the family (see Data source and definitions). Individuals can then be classified across quintiles in order to verify whether those with higher financial capital also have higher self-employment rates.

Self-employment rates can also vary by educational attainment. Higher education may give individuals the skills to start and remain in business. Certain fields of study, like law and medicine, also lead graduates into occupations with relatively high rates of self-employment. Past studies have produced mixed results on the link between education and self-employment.

Other personal characteristics are also known to influence the probability of being self-employed. For example, those with another self-employed family member (usually the spouse) tend to be self-employed themselves. The link is less clear for other characteristics. Might those with activity limitations be more likely to be self-employed in order to work around their constraints? Are new immigrants more entrepreneurial than the Canadian-born or more established immigrants?

Modelling self-employment among senior workers

The probability that a working senior would be self-employed as opposed to a paid worker was estimated using probit models. In addition to the aforementioned factors, a number of other demographic variables were included as controls.

Since farmers and farm managers comprise the largest sub-group of self-employed seniors and their characteristics differ from those of other self-employed workers, alternative models were estimated excluding this sub-group. A third set of models, using industry controls, was also estimated to control for this heterogeneity without subdividing the sample.

These models are estimated using cross-sectional data. The results are thus descriptive in nature—they do not address the probability of becoming self-employed. Such inferences would require longitudinal data. Currently available longitudinal data sets lack either the sample size or the range of variables to conduct such an analysis focusing on seniors.

The results are presented as marginal effects which measure the change in the odds of being self-employed for a certain characteristic in comparison to a reference group (Table 5). These marginal effects can generally be interpreted as the difference in probability between the groups being compared. For example, the value of -0.04 in the upper-left-most cell in Table 5 indicates that those in the first quintile of "other income" are 4% less likely to be self-employed than those in the third quintile.

Higher-income seniors are more likely to be self-employed

According to the model, adjusted family income was positively related to self-employment among seniors. For working men, the probability of being self-employed as opposed to being a paid employee was higher by 0.04 in the fourth income quintile and by 0.11 in fifth income quintile than for those in the middle quintile. For those in the bottom quintile, it was lower by 0.04, while the difference between those in the second and the third quintiles was not statistically significant. Among working women, those in the fourth and fifth quintiles were also more likely to be self-employed than those in the middle quintile. However, the differences were not statistically significant between the bottom two quintiles and the middle quintile.¹²

University degree increases probability of self-employment

With respect to education, men and women with university degrees were more likely to be self-employed than those with less than a high school education. The predicted probabilities of self-employment were higher by 0.02 for university-educated men and by 0.07 for university-educated women. On the other hand, men who were either high school or postsecondary graduates, but without a university degree, were less likely to be self-employed than those who had not completed high school. Among women, the differences between middle levels of education and high school graduates were either small or not significant.

Evidence for the education hypothesis was stronger, especially for men, when farmers were removed from the sample. In this sample, both postsecondary and university graduates had higher predicted probabilities than high school graduates. These results suggest that factors other than education play a large role for farmers and farm managers.

Seniors' self-employment is often a family affair

For many seniors, self-employment is a family affair. Men and women with another self-employed family member were more likely to be self-employed themselves than those who had another family member working as a paid employee. The probability was higher by 0.17 for men and by 0.18 for women.

Conditional on the fact that they were working, senior men age 75 and over were more likely to be self-employed than younger seniors. If they were working, women age 65 to 69 were less likely to be self-employed than those 80 and over. Similar patterns held when farmers were excluded from the sample: the relative probability of self-employment peaked in the 75-to-79 age group for both men and women.

Controls for other demographic variables yielded some interesting results. Men and women with activity limitations were more likely to be self-employed than those without limitations. On the other hand, recent immigrants (who immigrated in the preceding 10 years) and Aboriginal peoples were less likely to be self-employed.

An alternative model included industry as a control. Similar to the effect of dropping farmers from the sample, adding industry controls strengthened the relationship between a university degree and self-employment.

Looking at particular industries, seniors working in business services, construction and utilities, and primary goods industries were more likely to be self-employed than those in consumer services. The opposite was true for men and women employed in manufacturing, transportation, and public administration and education. The relationship was strongest between working in primary goods and self-employment, but weakened considerably when farmers were excluded.

Summary

Recent studies have documented the increasing employment among seniors in Canada. However, much less is known about the extent of self-employment among working seniors. Using detailed information from the census, this article presented new information on self-employed seniors. It also examined the factors associated with self-employment among working seniors.

The self-employed comprised a substantial portion of the employed labour force among seniors. Among those who had a job in 2006, more than 1 in 3 seniors were self-employed. Although the number of self-employed seniors continues to increase, between 1996 and 2006 the number of employed seniors increased even faster. As a result, the proportion of self-employed seniors declined. Since self-employment increased rapidly among seniors in the 1980s and the 1990s, its share of working seniors in 2006 was still relatively high by historical standards.

In 2006, the majority of self-employed seniors were unincorporated without paid help. Over the past few decades, however, a new class of self-employed seniors—those with incorporated businesses—became increasingly prevalent.

Self-employed seniors were concentrated in a few industries and had a much less diverse occupational profile than younger self-employed workers. Farmers and farm managers accounted for one-quarter of senior men and one-sixth of senior women in self-employment.

This study also looked at factors associated with self-employment among seniors. Self-employment was positively associated with other family income, indicating that individuals with more financial capital were more likely to be self-employed. The presence of a self-employed family member (most often the spouse) and having a university degree were other factors associated with a higher probability of being self-employed. Although these results persisted in models that excluded farmers and included controls for industry, they varied somewhat in magnitude.

Data source and definitions

This study uses data on men and women, 65 years of age and over, from the censuses of 1981, 1986, 1991, 1996, 2001 and 2006. Census data are required to conduct detailed analyses for relatively small population groups, like self-employed seniors. The census is conducted every five years. One-fifth of households receive the long form which, in addition to basic demographic information, asks more detailed questions including some on labour market activities. The 20% sample is weighted to represent all Canadians.

Variable definitions

Employed: a person is considered to be employed if he or she had a job in the week preceding the census, including those who were temporarily absent for the entire week because of vacation, illness, a labour dispute at work, maternity/parental leave, bad weather, fire, family responsibilities, or some other reason.

Employment rate: the number of employed persons expressed as a percentage of the relevant population.

Employee: paid worker – working for wages, salary, tips or commission.

Self-employed: includes individuals who had a job in the reference week and belonged to one of the following categories: self-employed without paid help, incorporated; self-employed with paid help, incorporated; self-employed without paid help, not incorporated; or self-employed with paid help, not incorporated.

Unpaid family worker: worked without pay for a relative in a family business or on a farm.

Work activity: based on data prior to the census year since data on weeks worked are for the previous year. An individual was classified to be working full year, full time if he or she worked 49 to 52 weeks full time (30 hours or more per week).

Other family income: this variable is calculated by first subtracting employment income (if any) from total economic family income and then adjusting for family size by dividing it by an adjustment factor that takes the lower relative needs of additional family members, compared to a single person living alone, into account. Income quintiles are then calculated using the adjusted other family income. Information on income variables is for the year prior to the census year.

Education: education levels are constructed using the highest certificate, diploma or degree variable. The lowest level, Level 1, is below high school graduation certificate or equivalency diploma. Level 2 is high school graduation certificate or equivalency diploma. Level 3 includes other trades certificate/diploma or registered apprenticeship certificate. Level 4 consists of college, collège d'enseignement général et professionnel (CEGEP) or other non-university certificate or diploma from a program of 3 months to less than 1 year, college, CEGEP or other non-university certificate or diploma from a program of 1 year to 2 years, college, CEGEP or other non-university certificate or diploma from a program of more than 2 years, or certificate or diploma below bachelor. The highest level, Level 5, includes bachelor's degree, certificate or diploma above bachelor, degree in medicine, dentistry, veterinary medicine or optometry, master's degree, or earned doctorate degree.

Activity limitations: are based on questions that refer to conditions or health problems that have lasted or are expected to last six months or more:

1. "Does this person have any difficulty hearing, seeing, communicating, walking, climbing stairs, bending, learning or doing any similar activities?"
2. "Does a physical condition or mental condition or health problem reduce the amount or kind of activity this person can do: (a) at home? (b) at work or at school? (c) in other activities, for example, in transportation or leisure?"

Reponses to either question indicating "yes, often" and "yes, sometimes" are used to create the corresponding activity-limitation variables.

Occupation: Based on National Occupational Classification (520 occupations).

Industry: Based on the 2002 North American Industry Classification System.

Recent immigrants: Individuals who immigrated to Canada between 1997 and 2006.

Established immigrants: Individuals who immigrated to Canada before 1997.

Aboriginal peoples: Self-reported aboriginal status.

Comparisons with younger age groups

Self-employment as a percentage of the total population is higher for younger age groups than seniors because more individuals are working in the former age range.

The rate has been more stable among senior men than younger age groups. While the rate among senior men remained around 6% between 1981 and 2006, for those age 25 to 54 it decreased from 13.1% in 1981 to 11.9% in 1991, increased to 13.6% over the next five years, and fell to 12.5% in 2006 (data not shown). For 55- to 64-year-olds, after registering a small increase between 1981 and 1986 and a decline to 12.9% during the next five years, it was on the rise until 2001 and remained at 15.5% in 2006.

Among women, all three age groups experienced a steady increase over time. The rate increased from 0.7% to 1.7% for seniors, from 2.1% to 6.6% for women age 55 to 64, and from 3.2% to 6.9% for 25- to 54-year-olds.

When self-employment among seniors was looked at as a proportion of those working, it was found to be much more prominent than among the younger age groups (Chart E). Employed senior men and women were almost twice as likely to be self-employed as workers age 55 to 64, and almost three times as likely as 25- to 54-year-olds. Also, regardless of age, men were much more likely to be self-employed than women.

---

Notes

1. Existing studies tend to focus on the population age 15 to 64 (see, for example, Lin et al. 1999 and Moore and Mueller 2002). Gardner (1994) and Turcotte and Schellenberg (2007) provide some general numbers on self-employment among seniors but do not delve into the details of that specific population.
2. Farm self-employment constitutes an important portion of total self-employment. Among men, 35,660 out of the 123,670 self-employed were farmers. The corresponding numbers for women were 8,890 out of 40,400.
3. This is not surprising since paid employees typically retire earlier than the self-employed. In addition, some people who retire from paid employment enter self-employment.
4. Out of 158,455 (84,950) employed men (women) age 65 to 69, 62,370 (21,385) were self-employed. Among those age 75 and over, of the 53,130 and 23,845 men and women who were working, 28,480 and 8,220 were self-employed, respectively.
5. In the remainder of the paper, 'employed' refers to the paid employed and self-employed. Unpaid family workers are excluded.
6. The numbers (44.4% and 29.4%) differ from those mentioned earlier (44.1% and 28.6%) since unpaid family workers are excluded here.
7. Examining non-farm self-employment is also important since it removes the effect of the relative decline of the agriculture sector within the ranks of the self-employed over that period. As a proportion of working senior men, total self-employment declined by 9.1 percentage points between 1996 and 2006, but the proportion of the non-farm self-employed declined by only 1.8 percentage points.
8. Of the 84,205 (29,610) men (women) who were unincorporated, 62,430 (23,805) did not have paid employees.
9. This section pertains only to 2006.
10. This proxy was also used by Uppal (2010).
11. It is necessary to remove the employment income of individuals from our definition of adjusted family income since there is a direct relationship between self-employment (and paid employment) and employment income. However, other sources of income from all family members were included (e.g., pension income, transfers, dividends, and capital gains) as well as the employment income of other family members.
12. Similar results were found when self-employed farmers were removed from the sample.

References

Beaucage, André and Elmustapha Najem. 2006. "Une analyse longitudinale de la pérennité des expériences des travailleuses et des travailleurs autonomes canadiens." Industrial Relations.Vol. 61, no. 2. Spring. (accessed December 20, 2010).

Bruce, Donald, Douglas Holtz-Eakin and Joseph Quinn. 2000. Self-employment and Labor Market Transitions at Older Ages. Center for Retirement Research (CRR) Working Paper 2000-13. Center for Retirement Research at Boston College. Chestnut Hill, Massachusetts. 29 p. (accessed December 20, 2010).

Fuchs, Victor R. 1982. "Self-employment and labor force participation of older males." The Journal of Human Resources. Vol. 17, no. 3. Summer. p. 339-357. (accessed December 20, 2010).

Gardner, Art. 1994. The Self-employed. Statistics Canada Catalogue no. 96-316E. Ottawa. 52 p.

Georgellis, Yannis, John G. Sessions and Nikolaos Tsitsianis. 2005. "Self-employment longitudinal dynamics: A review of the literature." Economic Issues. Vol. 10, no. 2. 39 p. (accessed December 20, 2010).

Hochguertel, Stefan. 2010. Self-employment Around Retirement Age. Tinbergen Institute Discussion Paper Tinbergen Institute (TI) 2010-067/3. Amsterdam, The Netherlands. Tinbergen Institute. 47 p. (accessed December 20, 2010).

Kuhn, Peter J. and Herb J. Schuetze. 2001. "Self-employment dynamics and self-employment trends: A study of Canadian men and women, 1982–1998." Canadian Journal of Economics. Vol. 34, no. 3. August. p. 760-784. (accessed December 20, 2010).

LaRochelle-Côté, Sébastien. 2010. "Self-employment in the downturn." Perspectives on Labour and Income.Vol. 11, no. 3. March. Statistics Canada Catalogue no. 75-001-X. p. 5-13. (accessed December 20, 2010).

Lin, Zhengxi, Garnett Picot and Janice Yates. 1999. The Entry and Exit Dynamics of Self-employment in Canada. Statistics Canada Catalogue no. 11F0019MPE – No. 134. Analytical Studies Branch Research Paper Series. Ottawa. 32 p. (accessed December 20, 2010).

Moore, Carol S. and Richard E. Mueller. 2002. "The transition from paid to self-employment in Canada: The importance of push factors." Applied Economics. Vol. 34, no. 6. March. p. 791-801.

Quinn, Joseph F. "Labor-force Patterns of Older Self-employed Workers." 1980. Social Security Bulletin. Vol. 43, no. 4. April. p. 17-28. (accessed December 20, 2010).

Turcotte, Martin and Grant Schellenberg. 2007. A Portrait of Seniors in Canada, 2006. Statistics Canada Catalogue no. 89-519-XIE. Ottawa. 301 p. (accessed December 20, 2010).

Uppal, Sharanjit. 2010. "Labour market activity among seniors." Perspectives on Labour and Income.Vol. 11, no. 7. July. Statistics Canada Catalogue no. 75-001-X. p. 5-18. (accessed December 20, 2010).

Zissimopoulos, Julie M. and Lynn A. Karoly. 2007. "Transition to self-employment at older ages: The role of wealth, health, health insurance and other factors." Labour Economics. Vol. 14, no. 2. April. p. 269-295. (accessed December 20, 2010).

Author

Sharanjit Uppal is with the Labour Statistics Division. He can be reached at 613-951-3887 or perspectives@statcan.gc.ca