The Impact of State Taxes on Self-Insurance: Background

2. Background
2.1. Premium tax
Premium taxes are the primary state tax levied on the insurance industry. Every state taxes insurance premiums. Thirteen states also levy at least one other tax (income, franchise, privilege, or capital) on the industry. Insurer state taxes are not trivial. The industry has paid more in state taxes than in U.S. Federal taxes in at least some years. For example, in 1990 insurers paid $7 billion in state taxes and $5.7 billion in Federal taxes (U.S. Department of Commerce, Bureau of the Census, State Government Tax Collections, 1992, Internal Revenue Service Statistics of Income Corporate Sourcebook, 1990, Sangha and Neubig, 1994). In addition, because the premium tax is levied on gross receipts and profit margins are thin in the insurance industry, the effective state tax rate can be substantial. For example, although this study finds that the mean and median state taxes as a percentage of premiums were only 1.6 percent in 1993, Sangha and Neubig, (1994) report that state taxes as a percentage of income was 31 percent, 24 percent, and 61 percent in 1990, 1991, and 1992, respectively.
2.2. Sales tax
As a gross receipts tax, the premium tax resembles a sales tax. Accordingly, this paper is related to those in the economics literature addressing the extent to which consumers bear the incidence of the sales tax (see review in Poterba, 1996). Poterba (1996) models the price effects of the sales tax and shows that in perfect competition, consumers should bear its full burden.  Consistent with this prediction, the sales tax studies find little evidence that retailers bear the incidence of the sales tax. In fact, Besley and Rosen (1999) report that for some goods, prices inexplicably overshift, i.e., an increase in tax revenue of one dollar per unit increases the price by more than one dollar.
The usefulness of Poterba’s (1996) model for understanding the impact of insurer premium taxes on self-insurance may be limited because of structural differences in the two gross receipts taxes. Unlike the sales tax rate, which is identical for all retailers within a locale, a policy’s premium tax rate varies depending on the insurer’s state of incorporation. Specifically, if state i’s tax rate is • and the tax rate in insurer j’s state of incorporation is •, then insurer j’s premium tax rate in state i is the greater of • and •. These “retaliatory” taxes favor in-state insurers and discriminate against out-of-state insurers. Because most policies are written by out-of-state insurers, policies written within a single state face a range of premium tax rates.
Sales tax analogues to within-state premium tax variation suggest some insurers may bear the incidence of the premium tax. For example, retailers are more likely to bear some tax incidence of the sales taxes along political borders. Theoretical and empirical studies (e.g., Mikesell, 1970, Sidhu, 1971; Gordon, 1983; Fox, 1986, Mintz and Tulkens, 1986; Braid, 1987; Rappaport, 1994, Trandel, 1994, Gordon and Neilsen, 1997, among others) are consistent with smaller retail price responses to sales tax changes along political borders. Likewise, Goolsbee (1999) reports that residents in high sales tax locations purchase more on the Internet, where sales are often not subject to sales tax. Although we are unaware of any empirical evidence, similar anecdotes have been advanced about mail-order sales, which also often escape sales taxes. In summary, the evidence suggests retailers bear some tax incidence when consumers are proximate to less heavily taxed jurisdictions or can access delivery systems (e.g., Internet or mail-order) that bypass the sales tax. These findings would imply that when consumers face multiple premium tax rates in a state, insurers bear some tax incidence.