EVENTS

Date:	Friday, November 12, 2004
Time:	9:30 AM -- 11:30 AM
Location:	Wohlstetter Conference Center, Twelfth Floor, AEI 1150 Seventeenth Street, N.W., Washington, D.C. 20036

November 2004

"Success Taxes," Entrepreneurial Entry, and Innovation

The impact of tax policy on entrepreneurial activity is at the center of the debate over the efficacy of recent tax reductions. A recent paper coauthored by AEI visiting scholar R. Glenn Hubbard and William Gentry of Williams College explores whether high taxes or more progressive taxes affect the decision to become an entrepreneur. At a November 12 AEI conference, the authors also analyzed whether entrepreneurs react to these options in a substantially different way than the general population.

William Gentry
Williams College

The role of "entrepreneurial entry" in innovation raises questions about the influence of tax policy. Gentry examined why marginal tax rates (MTRs) have a differential effect on entrepreneurs relative to other activities, how market entry is affected by changes in MTRs, and whether entrepreneurs in different industries and occupations are affected equally.

MTRs can significantly affect whether individuals engage in risky entrepreneurial behavior. When discussing taxes, "convexity" is the ratio of the weighted average of MTRs to successful entrepreneurial entry minus the ratio of the weighted average of MTRs to unsuccessful entrepreneurial entry. Gentry found that decreasing the tax schedule convexity by one standard deviation would increase the rate of entrepreneurial entry by 25 percent. He found little evidence for the importance of tax shelters, and no evidence that individuals from "innovative" industries and occupations were more likely to become entrepreneurs. The other main conclusions were that tax effects are traceable to "success" taxes and that the level and convexity of the tax schedule has a negative correlation to entrepreneurial activity.

William Randolph
U.S. Treasury Department

Progressive MTRs can discourage risky self-employment. Additionally, the convexity effect dominates the insurance effect in Gentry and Hubbard's results. The convexity of the tax schedule discourages risk taking. The insurance effect reflects after-tax incomes having a lower variance and the government sharing risk, which encourages risk taking.

Randolph cautioned that there were three possible sources of measurement error: First,
prospective entrepreneurs know more about their future earnings than do the economists studying them. Second, entrants into the market select themselves, and this causes the MTRs they face to be systematically understated. The understating increases with the convexity of the tax schedule. Finally, it is difficult to separate entrants' uncertainty and the researchers' uncertainty about entrants. This means that Gentry and Hubbard's estimates cannot separate the negative effect of convexity due to entrant behavior toward uncertainty from the negative effect of convexity due to "observational uncertainty," self-selection, and incorrect measurement of MTRs.

Kevin A. Hassett
AEI

Hassett began by graphically portraying Gentry's findings as a basic utility curve for entrepreneurs. He then pointed out that it might be better to think of new market entrants as pursuing a "Gulf Stream" curve. This curve is named after the $40 million airplane and implies greater consequences for success and failure than does the basic utility curve. However, it is not obvious that encouraging people to pursue the G-5 curve is beneficial to society. What is optimal to society is maximizing the percent of entrepreneurial success.

There were two problems with Gentry's tax measurements. First, Hassett questioned the use of MTR to outcome ratios to measure convexity. Second, it is necessary to consider more of the tax code than merely MTRs to understand the incentives for potential entrepreneurs. Hassett concluded by suggesting the next paper incorporate more areas of the tax code and that Gentry look for evidence of new entrepreneurs systematically making mistakes about their likelihoods of success.

AEI intern James Moore prepared this summary.