This paper tests whether the effect of tax-based subsidies for self-employed health insurance on the level of self-employment differs with the type of non-group insurance regulatory regime at the state level. Using a panel of tax returns from 1999 to 2004, we estimate fixed effects instrumental variable regressions for the probability of being self-employed, allowing the effect of the after-tax price of self-employed health insurance to differ by regulatory regime. Our results suggest that states with community rating and guaranteed issue regulations had significantly smaller increases in the fraction of taxpayers reporting some amount of self-employment income as a result of a decrease in the after-tax price of self-employed health insurance. However, there is suggestive evidence that heavily regulated states experienced a larger increase in exclusive self-employment, particularly among older taxpayers. (JEL J24, H24, I18)