The effect of marginal tax rate on the probability of employment-based insurance by risk group.

RESEARCH OBJECTIVE

To evaluate the effect of the tax subsidy on participation in employment-based health insurance for high- and low-risk individuals. The total exclusion of employer-paid health insurance premiums from taxable income has frequently been seen as contributing to excess insurance and hence welfare loss. However, less attention has been paid to quantifying the extent to which the tax subsidy mitigates the deleterious effects of adverse selection on the health insurance market. Adverse selection reduces pooling in an insurance market, so that high-risk individuals are either unable to obtain coverage or are forced to pay premiums that are unaffordable to all but the wealthiest. If there is an external benefit to society of an individual's purchase of medical care, then the presence of adverse selection may reduce the purchase of health care below the socially optimal level. Therefore, a mechanism for enhancing access to insurance and ultimately to medical care for high-risk individuals may be socially desirable.

STUDY DESIGN

Data from the March 1996-March 1998 Current Population Survey (CPS). For each observation in the sample, state and federal income tax liability is calculated using code based on the ACIR Significant Features of Fiscal Federalism. The probability of having employment-based coverage in either one's own name or as a dependent is evaluated as a function of demographic variables such as age, education, marital status and family size, family income, type of employment, employer size, occupation, location, marginal tax rate, risk group (determined by self-assessed health status), and an interaction between risk group and tax rate. CPS data do not identify individuals who have declined offered coverage. Under alternative models of employer group decision making, the tax subsidy will have an important influence on the employer's decision to offer coverage. If offered, high-risk individuals accept coverage, while some low-risk individuals may decline coverage.

PRINCIPAL FINDINGS

For all individuals, the probability of having coverage is an increasing function of the marginal tax rate. Those classified as high-risk because their own or a family member's self-assessed health status is fair or poor are less likely to have coverage than those considered low-risk. The effect of the tax subsidy on insurance coverage is greater for high-risk individuals than for individuals classified as low-risk.

CONCLUSIONS

These preliminary results indicate that high-risk individuals benefit from the tax subsidy by increased access to employment-based coverage. Therefore, welfare loss from excess levels of health insurance may be mitigated by welfare gain through expanded access to health insurance and hence to health care for high-risk individuals.

IMPLICATIONS FOR POLICY, DELIVERY, OR PRACTICE: Elimination or reduction of the tax exclusion of health insurance premiums may have the unintended consequences of disproportionately reducing the probability of obtaining coverage in the employment-based market for high-risk individuals.