چرا سهم بیمه سلامت کارکنان افزایش یافت؟

We explore the causes of the dramatic rise in employee contributions to health insurance over the past two decades. In 1982, 44% of those who were covered by their employer-provided health insurance had their costs fully financed by their employer, but by 1998 this had fallen to 28%. We discuss the theory of why employers might shift premiums to their employees, and empirically model the role of four factors suggested by the theory. We find that there was a large impact of falling tax rates, rising eligibility for insurance through the Medicaid system, rising medical costs, and increased managed care penetration. Overall, this set of factors can explain more than one-half of the rise in employee premiums over the 1982–1996 period.

The dominant feature of the health insurance market in the US is the provision of private health insurance through the workplace. But the past two decades have been a period of substantial reduction in both the scope and generosity of employer-provided health insurance. In 1982, roughly 80% of workers were covered by employer-provided health insurance. By 1998, this had fallen to 73%. Similarly, in 1982, 44% of those who were covered by their employer-provided health insurance had insurance that was fully financed by their employer. But by 1998, this had fallen to 28%.1 There has been a voluminous literature in recent years on the causes and consequences of the decline in employer-provided health insurance coverage. But there has been virtually no work on the parallel time trend of declining employer payments for health insurance. This is a particularly glaring omission in light of recent evidence which shows that most of the time trend in private insurance coverage appears to be reductions not in employer offering of insurance, but in employee takeup of insurance conditional on offering (Cooper and Schone, 1997 and Farber and Levy, 1999). Thus, the key dimension along which employers appear to be adjusting their health insurance spending is through the generosity of what they contribute. Moreover, this raises the possibility that it is reductions in employer generosity that are responsible for declining insurance coverage. In this paper, we attempt to model the set of factors that may be driving employers to shift their health insurance costs to their employees. We begin by discussing the theory of why employers might shift premiums to their employees. There are two classes of explanations. The first is that employers are shifting premiums in order to induce employees to choose the most cost effective option from the range of insurance choices offered by the employer. The second is that premium sharing results from imperfect worker sorting across firms; with heterogeneity in tastes among co-workers, premium contributions become a useful tool for separating worker types. By requiring contributions, the firm can provide insurance only to those who demand it, and can pass the savings back to employees in the form of higher wages. We then turn to estimating the role of a number of factors which fit into these categories of explanations: managed care penetration (which is correlated with the range of choices offered by the employer); the expansion of eligibility for the public Medicaid insurance program for women and children (since the imperfect sorting model predicts that such outside options should lead to increased employee contributions); health insurance costs (since the imperfect sorting model predicts that rising medical costs will increase the pressure on firms to shift the cost of insurance to their workers); and marginal tax rates (since employer contributions are tax subsidized, but employee contributions are often not tax subsidized, higher tax rates increase the incentive for employers to finance insurance costs). We investigate the role of these four factors using the only nationally representative annual data on premium sharing that covers this period of rapidly rising employee premium contributions: the Current Population Survey (CPS). These data provide only a noisy measure of premium sharing, based on a question of covered employees as to whether their employers pay all, some, or none of premiums. Compared to more comprehensive sources available for particular years, however, these data capture both variation across job/places and over time in the propensity to share costs between employers and employees. Moreover, this disadvantage is counterbalanced by the significant advantage that we can match to these data job and locational variation in our measures of interest. Based on these matches, we can investigate the role of these factors in driving the rise in employee premium sharing. Our results suggest that each of these factors is strongly related to employer contribution decisions. We find that the time trend in these influences corresponds quite strikingly to that of employee contributions, and that overall these factors can explain more than one-half of the rise in employee contributions over the entire 1982–1996 period. Our paper proceeds as follows. We begin, in Section 2, by providing background on employer and employee contributions for health insurance. We also discuss heuristically the theoretical issues involved in thinking about the tradeoff between employer and employee-financed insurance payments. Section 3 lays out our data and empirical framework for testing hypotheses about this shift. Section 4 presents our results, and assesses the extent to which the factors we investigate can explain this time series trend. Section 5 concludes.

The large and growing literature on the determinants of health insurance coverage of the US population has been focused primarily on the decision of employers to offer health insurance. But there is a growing recognition in health economics that employee takeup decisions may be the more important margin for explaining the large declines in coverage that we have witnessed over the past two decades. This contention is bolstered by the fact that there was such an enormous shift in premium costs from employers to employees over this time period. Yet, to date, there has been no explanation for this dramatic and potentially important trend. In this paper, we have investigated four possible determinants of this trend, drawing on the theoretical arguments for why, in the face of tax subsidized employer premiums, employers would shift premium costs to employees. We find that, for all four factors, we obtain the expected relationship with employee financing, although the relationship between Medicaid expansions and employee financing is somewhat sensitive to specification. The time trends in these predictors match the time trend in overall employer contributions quite well, and they explain more than one-half of the overall time series trend. These findings, particularly the strong effect for tax incentives, suggest that premium financing is a price sensitive decision for firms. Thus, policies that subsidize the employer-provision of health insurance may not only increase insurance offering, but also reduce the burden of premium payments for employees. This finding provides an additional factor that must be included in cost–benefit analysis of employer versus individual subsidies as a means of expanding insurance coverage.