COVID-19 relief options and additional resources

Understanding Self-Employment Dynamics Among Individuals Nearing Retirement

Full Report

Research Summary

April 2014                                     No. 422

Understanding Self-Employment Dynamics Among Individuals
Nearing Retirement

By Bradley T. Heim Consulting Bloomington, IN 47401, 76 pages.
Under contract number SBAHQ-13-M-0055.


Baby boomers, who were born between 1946 and
1964, are entering their retirement years. Since rates
of self-employment rise with age, a disproportionate
share of the self-employed is composed of middle
aged or older workers. Some of these workers have
been self-employed for much or all of their working
lives while others have transitioned to self-employment
later in their careers, often as a way of moving
into retirement.

Future predictions of baby boomers as a key catalyst
for small business growth in the next decade and
beyond have tended to neglect an important trend.
The self-employment rate among those nearing
retirement (defined as individuals aged 55-64) has
dropped substantially in the past 20 years. This study
addresses two questions about this decline. First,
has the decline in self-employment rates among
near-retirees been driven by: 1) lower rates of individuals
in this age group choosing to become self-
employed (often called the entry rate) or remaining
self-employed (called the continuation rate), 2) an increase
in number of current self-employed choosing
to retire or find a job with a firm (called the exit
rate), or 3) fewer individuals being self-employed
when entering these years? Second, what economic
and policy variables help to explain the change in
these factors over time? This study explores almost
20 years of U.S. Census Bureau data from 1994-
2012 to describe in greater detail the continuation,
exit, and entry rates with respect to self-employment.


The self-employed constitute a substantial and
important part of the workforce in the United States.
Research has indicated that the annual total nonagricultural
self-employment rate (including unincorporated
and incorporated businesses) has generally
exceeded 10 percent over the past 20 years . In addition,
several papers have shown that the rate of self-employment
increases as individuals near retirement. However, since 1994, the
self-employment rate among those nearing retirement
has steadily declined. The self-employment rate
among near retirees was above 18 percent in 1994,
but dropped to around 16 percent in the early 2000s,
and dropped further to 14.3 percent in 2012
No existing study has examined the reasons for
this substantial decline in self-employment among

Overall Findings

The study’s key finding is that the decline in the
self-employment among 55- to 64-year-olds over the
1994-2012 period was driven by three factors:

(1) The self-employed are choosing to find a job
at a firm rather than being self-employed. The movement
(i.e., exit rate) to wage and salary employment
increased over the sample period, particularly among
the unincorporated self-employed, and particularly to
the private sector. The increase is apparent
across all regions of the country and across retail and
service industries.

(2) The rate of self-employment among 55-yearolds
(who comprise about 12 percent of the self-
employed aged 55-64) decreased over the sample
period; these declines were particularly notable
among the unincorporated self-employed and those
in the retail sector. Further, the self-employment
rate among this group is lower than in the full 55- to
64-year-old cohort.

(3) The share of 55-year-olds in the 55- to
64-year-old cohort increased.

Other important findings indicate that a higher
after-tax price of self-employed health insurance
is associated with decreases in entries into self-
employment among wage and salary workers (see
“Tax Treatment of Self-Employed Health Insurance
Premiums” box). The study also finds the following
demographic factors are important among those
nearing retirement:

• Women and African-Americans are less likely
to enter into self-employment and are more likely to
leave self-employment for a wage-and-salary job or

• Older individuals are less likely to leave self-
employment for wage-and-salary employment and
more likely to leave for retirement; they are less
likely to enter self-employment from retirement.

• Being married is positively associated with exits
to retirement.

In terms of income and wealth variables:

• Individuals with greater wealth (measured by
investment income) are less likely to leave self-
employment for wage-and-salary work although they
are more likely to leave for retirement.

• Individuals with more labor and transfer income
are more likely to leave self-employment for wage-
and-salary work but less likely to leave for retirement.

• Having an employer pension plan in the prior
year is negatively related to entry into self-employment
from a wage-and-salary job, but positively associated
with entry from not working.

In terms of state-level policy and economic variables,
neither an improving economy (as measured
by lower unemployment rates or higher average
wages) nor stricter disability insurance policy significantly
impacts either the entrance to self-employment
or the rate of exit from self-employment for
those aged 55-64 or those at age 55.

Regarding the propensity to be self-employed at
age 55, these findings apply:

• Women and African-Americans have a lower
propensity to be self-employed at age 55.

• Self-employment at age 55 increases with educational
level and income.

• Having an employer pension in the prior year is
not significantly related to self-employment at age 55.

Policy Recommendations

The significant declines in self-employment rates
suggest a need for policies to reduce the exits of
near-retirees from self-employment to private wage
and salary employment and to increase entrepreneurship
among younger cohorts.

The author makes several suggestions to this end.

• Increasing the availability and affordability
of private health insurance may decrease the rates
of exits from self-employment, increase the rates
of entry to self-employment, and increase self-
employment rates among new entrants into the 55- to
64-year-old cohort. Such changes in health insurance
would counter the “job lock” phenomenon particularly
detrimental during a recession, e.g., individuals
not wanting to leave jobs that provide health insurance.

• Reductions in the after-tax price of health insurance
while self-employed may slow or reverse the
declining trend in self-employment at age 55.

An important area for future research, then, will
be to evaluate the extent to which recent changes
in health insurance cost and coverage due to the
implementation of the Affordable Care Act of 2010
affected rates of self-employment in this cohort.

The author points out that the changes in self-
employment rates may also be due to other economic
and social factors, such as increased productivity and
efficiency among larger businesses, making small
businesses less able to compete. If this proved to be
the case, it is unclear what policy response would be
desirable for the economy as a whole.

Tax Treatment of Self-Employed
Health Insurance Premiums

Unlike employees who get health insurance
through their employer, the cost of health insurance
premiums for the self-employed is not
excludable from both income and payroll taxes.
Under current law, self-employed health insurance
premiums are deductible from income
taxes, but are not excludable from the calculation
of self-employed payroll taxes (formerly
called Self-Employed Contribution Act [SECA]
taxes). This increases the after-tax cost of health
insurance compared to that of an employee. The
SECA tax is 15.3 percent of taxable income
however the self-employed are able to deduct
half of this tax paid.

Scope and Methodology

Data for this study come from the 1994-2012 waves
of the Annual Social and Economic (ASEC) supplements
to the Current Population Survey (CPS),
commonly known as the March CPS. The CPS is
a nationally representative survey of households
administered by the Census Bureau on a monthly
basis. It is possible to match respondents across two
adjacent March surveys, so that one can observe
whether an individual continued in, entered, or exited
from self-employment between their first March
interview and their second.

There are several advantages to using the CPS for
this study. First, the CPS is the dataset that is used in
the most commonly referenced tabulations of self-
employment rates by other researchers. Second, the
March CPS is collected annually, so that annual rates
of entry and exit can be examined (in contrast to a
data source like the Health and Retirement Study
(HRS) in which respondents are interviewed only
every other year). Third, the sample size is sufficiently
large that it is possible to tabulate rates overall
and for a number of subsamples (including incorporation
status, industry, and region of the country).
However, individuals are observed for at most one
two-year pair, and so it is not possible to track individuals
over longer periods of time. In addition,
although many variables of interest are included in
the study, the CPS does not include information on
respondents’ wealth.

To address the questions about the decline in self-
employment among those nearing retirement, the
researcher uses a Markov Chain model to relate the
self-employment rate in a particular year to rates of
continuation within self-employment, entry into self-
employment, exit from self-employment, and the
self-employment rate among individuals who turn 55
that year. To examine what demographic, economic,
and policy variables are associated with entries to or
exits from self-employment during the pre-retirement
years, and which variables are associated with the
level of self-employment among 55-year-olds, a set
of multinomial logit estimations is run. Simulations
are then performed to examine the extent to which
changes in economic and policy variables may have
led to changes in rates of continuation, entry, exit,
and self-employment at age 55. Counterfactual simulations
are presented to examine what trends would
have looked like had these variables not changed over
the roughly twenty-year period.

This report was peer-reviewed consistent
with Advocacy’s data quality guidelines. More
information on this process can be obtained by
contacting the director of economic research by
email at or by phone at (202)

Additional Information

This report is available on the Office of Advocacy’s
website at To receive
email notices of new Advocacy research, news
releases, regulatory communications, publications,
and the latest issue of The Small Business Advocate
newsletter, visit and subscribe
to the Small Business Regulation & Research

This document is a summary of the report identified above,
developed under contract for the Small Business Administration, Office
of Advocacy. As stated in the report, the final conclusions of the full report
do not necessarily reflect the views of the Office of Advocacy. This summary
may contain additional information, analysis, and policy recommendations
from the Office of Advocacy.