This project has three distinct purposes. 1. Estimate the effect the tax preference for employer-provided health insurance has on the allocation of resources. This will be accomplished in part by using a two-period model to derive a demand function for health insurance, then using a maximum likelihood procedure to estimate the model parameters. Household demand for health insurance with and without the tax preference can then be estimated. 2. Analyze how consistent this implicit subsidy is with obtaining an efficient allocation of resources. Three plausible cases will be considered. In each case the tax preference appears to be inefficient. For example, in one case it is assumed that there are paternalistic altruists in society who care about how much health insurance some other people have. In this situation, an in-kind transfer is necessary to obtain an efficient allocation of resources, but, in order to do so, it must be consistent with the preferences of the paternalistic altruists. There is a problem with this implicit subsidy because the higher a household's taxable income, the larger the tax break that household receives on each dollar of employer-provided health insurance. Therefore, the tax preference tends to lower the price of health insurance more for high-income households than for low-income households. This is an attribute the paternalistic altruists are likely to object to, so the tax preference does not appear to be consistent with their preferences. If this assertion is true, the tax preference is inefficient. It seems reasonable that the paternalistic altruists would prefer for the most assistance to be directed towards the neediest households. The extent to which the tax preference is inconsistent with the preferences of the paternalistic altruists will be analyzed in two ways. First, the households in a sample will be placed in a subsample with households of the same size whose incomes are in the same bracket. The mean effect the tax preference has on the amount of health insurance consumed by households will be estimated for each subsample. Second, the mean of the measures of equivalent variation due to this implicit subsidy will be estimated for each subsample. 3. Consider each of the cases separately and develop income tax and health insurance programs that are more consistent with obtaining an efficient allocation of resources under the assumption that each of the cases is reality.
