Author: Joshua Wright

With EMSI’s new data categories, we can now more closely parse data on the major classes of workers in the labor market. This is a significant shift in how we present employment data, and one of the valuable applications is being able to track and analyze self-employed workers — those whose primary job, their chief source of income, is working on their own.

In this piece, we’ll use EMSI’s self-employment data to dig into a segment of the workforce that has previously been hard (or impossible) to get at for local researchers and planners. But first, it’s important to distinguish between EMSI’s two proprietors datasets and why we’re only focusing here on the self-employed, the third of our four class of workers.

We also track what we call extended proprietors. This is distinct, miscellaneous set of workers/income earners, and the fourth category included for subscribers to our web-based labor market research tool. Extended proprietors do side gigs or earn income through a sole proprietorship or partnership, most often (but not always) in addition to their primary job. If surveyed, they would not list this extra work as their main source of income. On the other hand, our self-employed dataset uses the American Community Survey and other publicly available sources to track proprietors who work for their own unincorporated business, practice, or farm. (People with incorporated businesses are considered wage and salary workers for their own companies, and are thus not considered proprietors).

In short, EMSI’s self-employment data gives an estimate of the true self-employed in the workforce (i.e., those who are primarily self-employed and consider themselves as such). And it does so at the county, ZIP code, MSA, and state level.

A Note on Monitoring Entrepreneurial Activity

EMSI’s two new proprietor datasets offer a window into entrepreneurial activity for any level of geography, but we caution against labeling all workers in the self-employed or extended proprietor classes as entrepreneurs. More accurately, inside the extended proprietors dataset are those who pursue extra work opportunities while maintaining their day job, while the self-employed dataset includes those who have taken the additional step and are primarily on their own. Once start-up owners incorporate their business, they fall under the traditional wage and salary worker datasets.

Another thing to keep in mind: As you compare EMSI’s self-employment numbers at the national level with other sources, you might find higher estimates of the self-employed workforce than EMSI’s. The Current Population Survey, for example, tracks workers’ primary and secondary self-employment, while the ACS — which again is what EMSI uses — keeps tabs on only primary self-employment. A teacher who mows lawns in the summer could fall in the secondary self-employed category in the CPS, but for the purposes of EMSI data, that teacher’s lawn mowing would be found in extended proprietors because if you asked him what he does for a living, he would say teacher.

Overview

Note: The data in this post is from EMSI’s 2012.2 Class of Worker dataset.

Proportion of All Workers

Self-employed workers account for 7.1% of all U.S. workers, up from 6.4% in 2001. This percentage is lower that what the Current Population Survey reports, but again, we are looking at those who are primarily self-employed. Note that this proportion does not include peripheral proprietor activity through our extended proprietor dataset.

Among major sectors, the share of self-employed jobs is greatest in administrative and support services (particularly landscaping and janitorial services); agriculture, forestry, fishing and hunting; construction; and transportation and warehousing. In each of these sectors, 20% or more of the workforce is classified as self-employed.

As shown in the chart below, agriculture, forestry, fishing and hunting has made the biggest jump in the proportion of self-employed workers since 2001 (from 19.3% to 26%), followed by transportation and warehousing (16.2% to 20.2%).

While just over 10% of all jobs in the real estate industry are categorized as self-employed, nearly 70% of jobs (more than 5.5 million) in this sector are in the extended proprietors dataset. In this case, extended proprietors include part-time agents or people drawing income in a real estate partnership. The 11% in the self-employed category encompasses those who would list their real estate work as their main source of income.

Top and Bottom States for Self-Employed

Driven by a larger-than-average proportion of self-employed jobs in the arts and entertainment sector (as well as in forestry and logging), Vermont has the highest share of self-employment (11.6% of all workers) among all states and the District of Columbia. Maine (10.3%) and Montana (10.1%) are grouped closely together in the second and third spots, followed by California (9.7%), South Dakota (9%), and Idaho (9%).

By far the smallest share of self-employment is found in Washington, D.C. (2.1%), which is no surprise given that governments workers — regardless of the industry or agency — are considered wage and salary workers. After D.C., Delaware (4.4%), Virginia (5.4%), and New Jersey (5.4%) have the next-smallest shares.

The following table also shows 2001-2012 self-employment job growth and decline, and no state has expanded its self-employed workforce more than Arizona (36%). The growth of entrepreneurs has also been impressive in Texas (31%), Nevada (31%) and Florida (25%), while Nebraska has lost the largest percentage of self-employed workers (-11%, a loss of almost 9,000 jobs).

Since the heat of the recession in 2008, Vermont (8%) and Arizona (7%) have led the way in growth among the self-employed.

A Look at MSAs

For the largest metropolitan statistical areas, Riverside, California has the highest percentage of self-employed workers (12.4%), followed by Los Angeles (10.5%). None of the other 30 largest MSAs has a double-digit presence of self-employed workers. Just missing that mark is Miami (9.7%), while San Francisco (9.3%) is also close.

New York City is right at the national average — 7.1% of its workforce is self-employed. Chicago is at 5.7%.

For all MSAs regardless of size, Guymon, OK (35.4%) and Rio Grande CIty-Roma, Texas (21.4%) have the largest percentage of self-employed workers. The lowest are Williston, North Dakota (2.6%) and Hinesville-Fort Stewart, Georgia (2.9%).

Takeaways

1. Recession’s Toll

There are almost 400,000 fewer self-employed jobs in the U.S. than in 2006, and the proportion of self-employed workers to the entire workforce is below pre-recession levels.

2. Highest-Paying Industries are Declining in Self-Employment

Of the 20 highest-paying industries with at least 100 self-employed jobs at the start of the recession, 17 have fewer self-employed workers in 2012 than in 2008. This includes offices of physicians and offices of dentists, both of which have declined 3%. Almost as pronounced is the drop in self-employment in legal services. In 2008, more than 214,000 self-employed jobs were classified under offices of lawyers; in 2012, it’s estimated to be 209,494, a 2% decline.

Meanwhile, the largest self-employed occupations tend to be lower-skilled, lower-paying jobs — construction laborers, child care workers, etc.

3. Older Americans are Working for Themselves

In 2009, Dane Stangler of the Kauffman Foundation wrote, “Contrary to popularly held assumptions, it turns out that over the past decade or so, the highest rate of entrepreneurial activity belongs to the 55-64 age group.” While our self-employed dataset does not solely contain entrepreneurs (see note above), EMSI data backs up Kauffman’s research. As we showed earlier, over 30% of all self-employed workers are over 55, and another 28% are 45-54.

4. Majority of Self-Employed are Men

Of the estimated 10.6 million self-employed jobs in the US, more than 6.6 million (63%) are held by men. This is in contrast to a much more even male/female breakdown (51% male/49% female) when we consider all wage and salary and self-employed jobs.

Joshua Wright is an editor at EMSI, an Idaho-based economics firm that provides data and analysis to workforce boards, economic development agencies, higher education institutions, and the private sector. He manages the EMSI blog and is a freelance journalist. Contact him here.

Forget about all the perceived problems with the American higher education system, and ponder these two numbers: 12.8 million and 3.4 million.
The first is the estimate of Americans actively looking for work and unemployed. The second is the number of job openings in the U.S. as of the end of December, according to the Labor Department.

There are jobs to be had, and plenty of people to fill them – if only they had the right skills. But this is not yet another article on the nation’s well-documented skills mismatch (at least not directly). Rather, it’s on the educational component of this debate, which was recently brought to the surface in a new Georgetown study on the unemployment rates analyzed by students’ major field of study in college, and by columnist Virginia Postrel’s response to it for Bloomberg.

Georgetown’s Center on Education and the Workforce, analyzing 2009-10 data on graduates from the Census Bureau’s American Community Survey, showed that the worst job prospects are for students coming out of architecture programs (13.9% unemployment), followed by arts (11.1%), humanities and liberal arts (9.4%), social science (8.9%), and law and public policy (8.1).

Postrel acknowledges some of Georgetown’s findings but goes right at those who think the US needs students to be more career-minded. Looking at national education stats, she argues most Americans are already choosing a college major based on the prospects of landing a job upon graduation. And if the supply of "practical" workers increases, Postrel warns, the quality of the workforce will deteriorate and wages will be lower.

“Contrary to what critics imagine,” she writes, “most Americans in fact go to college for what they believe to be ‘skill-based education.’ A quarter of them study business, by far the most popular field, and 16 percent major in one of the so-called Stem (science, technology, engineering and math) fields. Throw in economics, and you have nearly half of all graduates studying the only subjects such contemptuous pundits recognize as respectable.” Further, Postrel says those who argue for initiatives to push students in STEM fields – or away from liberal arts – disregard “the diversity and dynamism of the economy, in good times as well as bad.”

So what’s better? Would the U.S. be better off if more students took a broad-based liberal arts approach to their education, or should more be concerned with learning a specific trade or set of skills?

Postrel cited the National Center for Education Statistics’ report entitled "2008-09 Baccalaureate and Beyond Longitudinal Study" (PDF here), which looked at graduates from the 2007-08 academic year. Two points should be made on this study: 1) the economy and employment prospects related to certain programs have changed drastically since 2007-08, and 2) the report includes only bachelor’s degree recipients.

With this in mind, EMSI tapped into the NCES database to get a more recent and thorough picture of the most popular programs areas for recent US grads. We looked at the total number of degrees given out in 2010 (the most recent year available from NCES) at the associate’s, bachelor’s, master’s, and doctorate level among major program categories.

Our analysis includes degrees at the four main postsecondary levels mentioned above, not certificates or postsecondary awards. We did this to better reflect graduates who have made a more-than-one-year investment in education and have therefore given their decision serious thought.
A few noteworthy items (see the full data here):

These stats point to a few telling short-term shifts. Yes, a significant number of all degree recipients have moved toward skill-based education. But more striking is the decline in students choosing STEM majors as a share of all graduates. In 2003, 21% of degrees awarded in the U.S. were from STEM-related programs. In 2010, that percentage dipped to 19%.

Postrel would prefer students not be pigeonholed into going the skill-based route, even when it’s STEM-related. Yet it’s hard to ignore the illuminating data from Georgetown’s report: Recent grads in humanities and liberal arts have a 9.4% unemployment rate – lower than architecture, which has taken a beating with the construction downturn, but considerably higher than engineering (7.5%), business (7.4%), and especially health (5.4%).

So, to go back to the question posed earlier, our economy would seem to be better off – or, to put it another way, unemployment would most improve — if more students earned a skills-based education over a liberal arts degree. At least in the short term, that appears to be the best solution to get people back to work. But outside training for a few fields, there’s no automatic path to a job. Even some new registered nursing grads – with nursing once seen as a lock to find a job – are having a hard time finding employment. And EMSI and others have written about the souring job prospects for lawyers.

The Great Recession affected every industry in one way or another. Nonetheless, it’s worth repeating: There are 3.4 million jobs openings out there – 39% more than in June 2009. But in a sure sign that something is out of whack, hiring is only up 12% over that same time. 

Flickr Photo by Jason Morrison: Ann Morrison gets her Masters of Science in Nursing, Pepper Pike, Ohio.

Joshua Wright is an editor at EMSI, an Idaho-based economics firm that provides data and analysis to workforce boards, economic development agencies, higher education institutions, and the private sector. He manages the EMSI blog and is a freelance journalist. Contact him here.