[Home][Banking][Community Development][Newsroom][Education Resources][Publications][About Us][Research]
Home > Publications > The Regional Economist > April 2011 > Are Small Businesses the Biggest Producers of Jobs...

The Regional Economist | April 2011

Are Small Businesses the Biggest Producers of Jobs?

It is often claimed that small firms are responsible for a disproportionately large share of new jobs that are created in the U.S. economy. If true, this speaks well of the entrepreneurial spirit of the U.S. economy, whereby newcomers introduce new ideas or production processes that lead to new and improved products or services. The rise of global companies like Wal-Mart, Microsoft and Google from small beginnings is a testament to the importance of small businesses and the economic forces they sometimes unleash. However, the claim that small businesses generate a large percentage of new jobs must be evaluated carefully. First, there isn't a universal agreement on the definition of a small business. Furthermore, the failure rates of small business are quite high. According to the Bureau of Labor Statistics, only about half of the businesses that opened in 1994 were still operating five years later. Thus, when one accounts for job destruction, small businesses appear to account for a significantly smaller share of net new jobs created in the private sector than many people might believe.

What Do Past Studies Reveal?

The importance of small businesses to job creation has been part of the economic policy narrative for some time. In 1979, then-Massachusetts Institute of Technology Professor David Birch claimed that firms with 20 or fewer employees accounted for two-thirds of all new jobs created between 1969 and 1976; firms with 100 or fewer employees accounted for 82 percent of all new jobs created. Conversely, he found that large firms (500 or more employees) accounted for only 15 percent of net job growth. Birch's finding challenged the conventional wisdom about job creation at the time and, accordingly, had enormous influence on policymakers and researchers.1

Some economists soon began to challenge Birch's findings. Using the same data as Birch, Catherine Armington and Marjorie Odle found in 1982 that businesses with 100 or fewer employees accounted for only 39 percent of net new jobs. Several years later, Charles Brown, James Hamilton and James Meddoff pointed out that 40 percent of jobs created in small businesses in 1980 no longer existed in 1986. A more up-to-date assessment of the job-creation characteristic of small businesses can be found in work published by Stephen Davis, John Haltiwanger and Scott Schuh in 1996.

These authors noted that "a common confusion between net and gross job creation distorts the overall job creation picture and hides the enormous number of new jobs created by large employers."2 The authors found that although gross job creation is high for smaller firms (100 or fewer employees), so is job destruction. Slowly, researchers were coming to the conclusion that small businesses did create a lot of new jobs, but the high failure rate of these businesses suggested that their net job creation was much lower.

Earlier this year, a study designed to look at the entire economy was published.3 The researchers found that small firms create more net jobs than do large firms, which is consistent with the conventional wisdom but generally not the thrust of past research. However, they concede that Birch overestimated the importance of small business in job creation and found that there is a much smaller difference between the net number of new jobs created by large firms and small firms than Birch originally suggested.

Business Employment Dynamics

Researchers who want to assess the claim that small businesses account for a disproportionate percentage of new jobs must first confront several issues. First, what is the best data source for the hypothesis to be tested? Second, how should a small business be defined? (The Small Business Administration says a business is small if it employs fewer than 500 people. However, it may not be wise to lump together a Silicon Valley startup with a relatively large, established manufacturer.) Third, should the focus be on the gross number of jobs created or the net number of jobs created? The research suggests the latter. Why? Because even during the depths of the 2007-09 recession, businesses were still adding an average of nearly 800,000 new jobs a month. But they were shedding an even larger number of jobs per month—about 971,000.

In this article, we use the Business Employment Dynamics (BED) dataset from the Bureau of Labor Statistics.4 One drawback of the BED is that it has less than 20 years of history, which may limit the ability to draw firm conclusions. The analysis in this article uses the following breakdown of firm size: 1-19 employees; 20-99 employees; 100-499 employees; and 500 or more employees.

Job Gains by Firm Size

The table shows average gross and net job gains at all private business establishments from the third quarter of 1992 through the first quarter of 2010.5 Over this roughly 18-year period, gross job gains per quarter averaged a little less than 2.8 million, or about 929,000 per month. Since the 2007-2009 recession was extremely severe, the table includes a separate column that excludes the data from that period. The lower half of the table shows that businesses with fewer than 20 employees provided the largest percentage of gross job gains (about 30 percent). Businesses with between 20 and 99 employees accounted for the next largest share (about 27 percent), with the largest firms (500 or more) accounting for a somewhat smaller percentage (about 26 percent). The remaining category—businesses with between 100 and 499 employees—accounted for a smaller percentage of gross job gains. All of these percentages are little-changed if we exclude the recession period.

Gross and Net Job Gains by Firm Size

Average job gains (in thousands) per quarter, 1992:Q3 to 2010:Q1

 Gross Job Gains
Net Job Gains
Size
Total Sample Period
Excluding 2007-09 Recession

Total Sample Period
Excluding 2007-09 Recession
1 to 19
821
828

16
28
20 to 99
747
758

25
40
100 to 499
496
505

25
37
500 or more
722
739

40
68
TOTAL
2,787
2,831

105
173
 
Percent of Total
1 to 19
29.5%
29.3%
 
15.0%
16.1%
20 to 99
26.8%
26.8%
 
23.6%
23.1%
100 to 499
17.8%
17.8%
 
23.4%
21.3%
500 or more
25.9%
26.1%
 
37.9%
39.4%
TOTAL
100.0%
100.0%
 
100.0%
100.0%

SOURCE: Authors' calculations based on Bureau of Labor Statistics' Business Employment Dynamics dataset. Some percentages do not total 100 due to rounding.

The analysis in the table seems consistent with the conventional wisdom that small businesses are the largest source of job creation in the economy. However, as suggested by previous studies, the conclusion tends to change when the focus switches to net job creation.

The two right-hand columns in the table examine net job gains. Net job gains are defined as job gains minus job losses. Three findings are apparent from the table. First, net job gains were significantly smaller than gross job gains. The net gains per quarter averaged only 105,000, or 35,000 per month. Second, the table shows that the recession dramatically reduced the rate of net job creation. Once net job losses during the recession are removed from the calculation, the number of net jobs rose to 173,000 per quarter (about 58,000 per month). Finally, and perhaps most importantly, the BED data show that since 1992, net job creation tended to be largest among the largest firms: These firms accounted for about 38 percent of the total. The smallest firms showed the smallest percentage of net jobs created. This result does not change if the past recession is excluded from the sample.

In short, small businesses showed higher rates of gross job creation, but they also exhibited high rates of job destruction. Looked at from this standpoint, net job creation matters most.

Linpeng Zheng provided research assistance.

Write a Letter to the Editor | Read Letters to the Editor

Endnotes

  1. Birch followed up his original study with several subsequent studies (not cited herein). [back to text]
  2. One drawback of this study is that it focused on the manufacturing sector, which is a relatively small share of the economy and, thus, probably not a good representation of total job creation. [back to text]
  3. See Neumark, Wall and Zhang. [back to text]
  4. The BED is a quarterly series that is based on the Quarterly Census of Employment and Wages, which uses state unemployment insurance records. See Spletzer et al. for more information about the BED. [back to text]
  5. Changes in employment can arise from opening or expanding businesses, or closing or contracting businesses. Gross job gains include the sum of all jobs added at both opening and at expanding establishments. Gross job losses, then, include the sum of all jobs lost at both closing establishments or contracting establishments. [back to text]

References

Armington, Catherine; and Odle, Marjorie. "Small Business: How Many Jobs?" The Brookings Review, Winter 1982, Vol. 1, No. 2, pp. 14-17.

Birch, David L. The Job Generation Process. Cambridge, Mass.: MIT Program on Neighborhood and Regional Change, 1979.

Brown, Charles; Hamilton, James; and Medoff, James. Employers Large and Small. Cambridge, Mass.: Harvard University Press, 1990.

Davis, Stephen J.; Haltiwanger, John C.; and Schuh, Scott. Job Creation and Destruction. Cambridge, Mass.: MIT Press, 1996.

Haltiwanger, John .C.; Jarmin, Ron C.; and Miranda, Javier. "Who Creates Jobs? Small vs. Large vs. Young." NBER Working Paper 16300, August 2010. See www.nber.org/papers/w16300

Neumark, David; Wall, Brandon; and Zhang, Junfu. "Do Small Businesses Create More Jobs? New Evidence for the United States from the National Establishment Time Series." The Review of Economics and Statistics, February 2011, Vol. 93, No. 1, pp. 16-29.

Spletzer, James R.; Faberman, R. Jason; Sadeghi, Akbar; Talan, David M.; and Clayton, Richard L. "Business Employment Dynamics: New Data on Gross Job Gains and Losses." Monthly Labor Review, April 2004, pp. 29-42.

 

<< back to top