Over the past few decades, policy makers have considered employer mandates as a strategy for stemming the tide of declining health insurance coverage. In several states such laws passed in initial votes but subsequently were voided or overturned, and an employer mandate was a key component of the Clinton Administration's failed health-care reform proposal in the early 1990s. This debate has been invigorated of late: employer mandates appear to varying degrees in the health-care reform strategies proposed by several candidates in the 2008 presidential race. To evaluate these proposals it is important to understand not only their impact on insurance coverage but also their likely effects on labor market outcomes such as wages, hours, and employment.
The only employer health insurance mandate that has ever been enacted is Hawaii's Prepaid Health Care Act, which was established in 1974 but faced legal challenges and uncertainties until 1983. Hawaii's law requires all private employers to provide insurance to employees working at least 20 hours per week. Based on a standard supply-demand framework, we investigate the effects of Hawaii's law by comparing insurance coverage and labor market outcomes in Hawaii and the rest of the United States for the period from 1977 to 2006. Our main source of data is the March Current Population Survey (CPS). We combine the CPS data with information on health insurance premiums to account for the rising cost of the mandate that occurred as the relative cost of health insurance increased over time. We adapt propensity-score methods from the literature on quasi-experimental design to test for heterogeneous treatment effects.
Our results show that in the years following the implementation of Hawaii's mandate, employer-sponsored health insurance remained relatively constant in Hawaii while it fell significantly in the rest of the US. Since the early 1990s, the coverage rate has been roughly 11 percentage points higher in Hawaii. The gap is substantially larger for less educated workers and others groups that tend to have low rates of coverage in the voluntary market, with little or no gap evident for workers with a high propensity to obtain employer-sponsored insurance.
Consistent with the expected impact of the mandate on employer costs, we find evidence of corresponding differences in wage growth. Among worker types with a low propensity to receive health benefits (but not for worker types with a high propensity to receive health benefits) wage growth over the three decade analysis period is lower in Hawaii by a small but statistically significant amount. We find no evidence that the law has led to an increased reliance on part-time workers who are exempt from the mandate, nor do we find that the mandate has reduced employment probabilities.
This is joint work with John DiNardo (U of M) and Robert Valletta (San Francisco Fed).
|