The Washington Post
By: Glenn Kessler
January 27, 2014
The senator made this statement as he criticized what he described as President Obama's approach to job creation: "The president thinks that you collect money from the rest of the country, bring it to Washington, and then we re-pass it out." As an example, he pointed to the federal government's ill-fated $534 million loan guarantee to solar-panel manufacturer Solyndra, which in 2011 ceased business operations. (His comment about "one of the richest men in America" appears to refer to George Kaiser, a billionaire who owned a big chunk of the company.)
For the purposes of this fact check, we are interested in Paul's claim that nine out of 10 businesses fail -- which he claimed ensured that nine times out of 10, the wrong people would benefit from government largess intended to boost new businesses.
As far as we can tell, there is no statistical basis for the assertion that nine out of 10 businesses fail. It appears to be one of those nonsense facts that people repeat without thinking too clearly about it. Here are some basic questions to ask when assessing such a factoid:
1. What's the time frame? Two years, five years, 10 years? That can make a big difference.
2. Does "fail" mean that it goes out of business because it was not financially viable? Or does that also include data about successful enterprises that merge with another company?
3. Wouldn't failure rates be different for some industries than others? Does it make sense to lump all businesses together?
There have been a number of studies that have looked at this issue. This chart, from Web site designer smallbusinessplanned.com, summarizes the results of three different studies. Basically, after four years, 50 percent of the businesses are open. As time goes on, the success rate decreases, but it never gets to a failure rate of "nine out of 10."
One of the most cited studies on this issue (No. 3 in the chart above) was published in 1989 by Bruce D. Phillips of the National Federation of Independent Business and the late Bruce A. Kirchhoff, director of the technological entrepreneurship program at New Jersey Institute of Technology. They concluded that new establishments with 500 or fewer employees show an average survival rate of 39.8 percent after six years, but that there were wide variations among types of firms.
The Small Business Administration, in its informative frequently asked questions on small businesses, provides this answer on the survival rate of new businesses:
"About half of all new establishments survive five years or more and about one-third survives 10 years or more. As one would expect the probability of survival increases with a firm's age. Survival rates have changed little over time."
The source for the SBA statement is the Bureau of Labor Statistics, which offers this chart on survival rates.
Even this does not show the whole picture. As Brian Headd, an economist at the Small Business Administration, demonstrated in a 2002 study for Small Business Economics, about one-third of closed business were actually successful when they "failed."
"The significant proportion of businesses that closed while successful calls into question the use of 'business closure' as a meaningful measure of business outcome," the study says. "It appears that many owners may have executed a planned exit strategy, closed a business without excess debt, sold a viable business, or retired from the work force."
This pie chart also from smallbusinessplanned.com, demonstrates this phenomenon.
There is recent research by Harvard University's Shikhar Ghosh that three out of every four venture-backed firms fail, which was newsworthy because the failure rate was higher than normally cited by the venture capital industry. But here, again, the use of the phrase "failure" is a bit misleading. Only about 30 percent result in a total loss of the venture-capital investment; the rest simply did not exit with a full return of the initial investment, such as through an acquisition or initial public offering.
We sought a response from Paul's aides but were told an immediate answer was not possible. We will update this column if we get more information.
The Pinocchio Test
Paul is basing his critique on Obama on a fallacy. Rather than nine out of 10 businesses failing, about 50 percent survive four or five years. Even that percentage may be overstated depending on the definition of "failure."
Investing in a new business does have a certain level of risk, but it is not nearly as high as Paul claimed. A senator should avoid repeating such economic myths on television, thereby perpetuating false impressions.