11/01/2002
Summary of working paper 9015
Replacing the current personal and corporate income taxes with a 20 percent flat tax should virtually triple the amount of entrepreneurial activity.
In Taxes and Entrepreneurial Activity: Theory and Evidence for the U.S. (NBER Working Paper No. 9015), authors Julie Berry Cullen and Roger Gordon explore how taxes affect the decision to start a new business. Individuals can choose how much they work for themselves or work for someone else. Even if this choice does not affect their pretax earnings, it can well affect their tax liabilities. If business income is treated more favorably than earnings, then the tax system will encourage business activity.
Differences in the tax treatment of profits versus losses also can have strong effects on the incentives to choose a more risky occupation. For example, under a progressive tax schedule, profits will push the entrepreneur into higher tax brackets while losses will have the opposite effect. This implies that profits will be subject to a higher tax rate than the rate against which any losses can be deducted, making risk-taking less attractive. Similarly, profits from a partnership or proprietorship are subject to the payroll tax, but losses from such a business are not deductible under the payroll tax, again implying that entrepreneurs will keep fewer of their profits than of their losses. In contrast to these two mechanisms, the option to incorporate when combined personal income and payroll taxes become high enough makes risk-taking more attractive by reducing the tax rate on profits while leaving the treatment of losses unchanged. Finally, the asymmetric treatment of noncorporate profits from a partnership or proprietorship are subject to the payroll tax, but losses from such a business are not deductible under the payroll tax, again discouraging risk-taking,while small business normally finds it difficult to sell equity to outside investors, forcing the entrepreneur to bear substantial risks, it does implicitly share risks with the government, for example, by saving taxes when the business has losses.
The authors estimate the effects of taxes on entrepreneurial activity using an IRS dataset containing over two million individual tax returns spanning twenty-two years from 1964 to 1993. After eliminating people over 65, those who were listed as dependents on other people's returns, and people who filed jointly, the authors sorted the returns into six groups based on an individual's potential wages. Their estimates were made from random samples drawn from each of the six groups.
Each of the tax effects just described shows up clearly in the data. The results imply, for example, that a drop in personal tax rates in each bracket by 5 percentage points would lead to over a 20 percent fall in entrepreneurial activity. Allowing non-corporate firms a tax rebate for any losses beyond those sufficient to reduce their taxable income to zero, as under a negative income tax, is forecast to generate a 50 percent increase in entrepreneurial activity. Reform of the payroll tax, enabling losses to be deductible just as profits are taxable, should increase entrepreneurial activity. More dramatically, replacing the current personal and corporate income taxes with a 20 percent flat tax should virtually triple the amount of entrepreneurial activity, even with the remaining distortions under the payroll tax and the lack of tax rebates when taxable income becomes negative. Cullen and Gordon also find that a low unemployment rate and low real interest rates stimulate entrepreneurial activity.
-- Linda Gorman
