Duke 2008 | ASHE | Equity and Efficiency in Health and Healthcare

The overwhelming majority (61%) of Americans receive health insurance through their employers. Since coverage is often linked to full-time wage/salary employment, a larger portion of self-employed individuals are uninsured. Beginning in the 1980s, attempts were made to level the playing field between wage/salary and self-employment. The 1986 Tax Reform Act (TRA86) allowed the self-employed to deduct 25% of their premium from their taxable income. Prior to this, the selfemployed paid for health insurance with after-tax dollars. Previously, the 1942 Stabilization Act and the 1954 Internal Revenue Code granted the same rights to employers and employees for their individual contributions to their employees' plans. A series of amendments were made to the TRA86 between 1996 and 2003 which gradually increased the credit to 100%. The link between insurance and wage/salary employment in the U.S. has received considerable attention in the literature. Primarily, the literature has sought to determine whether there is any 'job lock'. Madrian's (1994) seminal work was motivated by the passage of COBRA and addressed whether job turnover was affected by the possible loss of employer-sponsored insurance. Madrian (1994) and Gruber and Madrian (1997), for example, find evidence of job lock while others (e.g., Holtz-Eakin et al., 1996) do not. The only study which focuses specifically on self-employed individuals and their job turnover is Gurley-Calvez (2006). She studies the original TRA86-mandated 25% credit and finds positive (negative) effects of this credit on entrepreneurial survival (exit).

Our paper concerns the TRA86 amendments and investigates whether they affected entrepreneurial survival rates from 1995-2005. Specifically, we focus on the transitions from selfemployment into different types of wage/salary employment (e.g., with/without employer-provided health insurance) or exits from the labor force altogether. We use data from the March Supplements of the Current Population Survey (CPS) to analyze the effects of the most recent changes in the tax code. We exploit the longitudinal feature of the CPS and construct a panel using the Outgoing Rotation Group (ORG) series which is composed of two-year cross-sections. We also have information on self-employment tenure from the January and February Supplements. The empirical framework is based on a multinomial logit and hazard models. The analysis is performed for primeaged self-employed individuals?male, female, single, and married.

The first paper to analyze the original TRA86 with respect to the elasticity of demand for health insurance coverage is Gruber and Poterba (1994). They show that the subsidy increased the demand for insurance by self-employed entrepreneurs, with marginal statistical significance. More recently, Gumus and Regan (2007) examine the TRA86 amendments. Their estimates of the price elasticity of demand suggest that the series of tax credits did not provide sufficient incentives for selfemployed entrepreneurs to obtain coverage. In light of these findings, we analyze whether the TRA86 amendments improved the chances of entrepreneurial survival despite their ineffectiveness in encouraging uninsured self-employed entrepreneurs to purchase insurance. Our results indicate that having coverage, either in one's own name or through an alternative source (e.g., spouse), plays an important role in explaining transition patterns for entrepreneurs.