In defending his opposition to tax rate increases for the affluent, Speaker of the House John Boehner continuously says, "Small business is the engine of job creation in America." Even in a news conference, no one questions him. Yet, the statement is overly simplistic, under researched and misleading.
Perhaps no politician of either party wants to risk explaining the economic impacts of small business are limited, and "small business" is not what most voters think it is.
When looking at jobs in the American economy, investigators classify business as small, medium or large by establishment size. They calculate the number of new jobs produced in each tier and conclude small business does indeed produce the bulk of them. It seems to make sense. At any one time, there could be thousands of restaurants hiring. On the other hand, there are only 120 automobile assembly plants in the United States.
The conclusion, however, glosses over a couple of realities, both of which independently refute, or at least complicate, the speaker's adage. One is the very nature of business in the job-creation arena. The other is the difference between establishments (the unit of analysis) and firms.
First, the nature of business as it pertains to jobs:
Job destruction, job creation is a continuous process. Destruction occurs when an incumbent leaves a position. In calculating job destruction, the underlying reason for departure (death, promotion, termination, furlough, resignation, etc.) is not considered.
Creation is the opposite: A person enters or begins a job. Again, the underlying rationale (new hire, recall, transfer, etc.) does not influence the calculation.
The number of "new jobs," whether across the economy, in an industry or at an establishment, is a simple calculation: Job creation minus job destruction equals new jobs. (In the worst of times, this could be a negative number.)
So, for instance, if the restaurant down the street hired two cooks, and a dishwasher quit, analysts would say this establishment experienced two job creations and one destruction. Therefore, it generated one new job.
Every first Friday, the Bureau of Labor Statistics cumulates individual establishment numbers and issues its monthly "Employment Situation Report" – America's official "jobs numbers." Instead of parsing establishments by size, however, number crunchers could analyze jobs data by establishment age; that is, how long has an establishment existed? This analysis produces findings that muddy the establishment-by-size analysis.
Establishments in their startup year generate about two-thirds of all new jobs. In other words, startup establishments produce two new jobs for every one new job generated by all the establishments — regardless of size — that have existed for at least a year.
With this finding, the speaker might conclude, "startups are the engine of job creation in America." Instead of advocating tax breaks for business owners and managers, he might pursue strategies to encourage innovation and entrepreneurship.
Second, the difference between establishments and firms:
To produce valid findings, investigators must collect analogous data from comparable sources. "Establishments" are comparable sources. They are the places where people conduct business. They produce analogous data, their payrolls.
In the case of the "jobs numbers," the methodology is sound. Calculations are valid. Findings are accurate. Nonetheless, results can be misleading. The culprit is the unit of analysis, establishment size.
Consider two pharmaceutical retailers a block apart. Both the family drugstore on Elm Street and the CVS Pharmacy on Main Street conduct business onsite and have payrolls. Therefore, both are establishments for data collection purposes. However, even though CVS employs about 200,000 people with nearly 7,460 retail stores nationwide, its establishment on Main Street is a small business by definition — less than 50 employees. More to the point, the CVS Main Street store is small business in the data as is each of its other 7,500 establishments across the country.
The family drugstore is not just an establishment. It is also a firm by definition. The single establishment on Elm Street encompasses the totality of the family's business activity.
One the other hand, the CVS retail store on Main Street is a very small part of a much larger firm. The CVS firm is clearly very large business. The family drugstore firm is just as clearly small business.
Using firm size instead of establishment size as the unit of analysis flips findings on their head. When new jobs from the 12-month period from February 2010 through March 2011 were calculated — data beyond March 2011 were not available at the time of this analysis — the dichotomy was remarkable:
With "establishment" as the unit of analysis, small business produced 44 percent of the new jobs. Large business generated just 5 percent.
But with "firm" as the analysis unit, 17 percent of the new jobs came from small business. Large business spawned 56 percent.
Over 20 years (1990-2011), small firms produced 16 percent of the new jobs. Large firms generated 65 percent.
With these findings, the speaker might conclude, "Big business is the engine of job creation in America." Instead of advocating tax breaks for business owners and managers or pursing strategies to encourage innovation and entrepreneurship, he might suggest increasing subsidies to large corporations.
Poul Anderson, famed science fiction author and winner of the Libertarian Futurist Society's first Special Prometheus Award for Lifetime Achievement, summed up the complexity of problems in a sentence widely known as "Anderson's Law": "I have yet to see any problem, however complicated, which, when you looked at it in the right way, did not become still more complicated."