The Safety Net and Employer-Provided Health Insurance
Grant Description: How does the health care safety net affect the private insurance market? The researchers examined how the structure and characteristics of the safety net (i.e., hospitals and federally qualified health centers [FQHCs]) affect employees’ decisions to accept coverage for themselves and their dependents and employers’ decisions to offer coverage. The researchers posited that a stronger safety net may lead employees to accept jobs without health insurance or to refuse coverage if offered. At the same time, they suggested, a stronger safety net may prompt employers not to offer coverage, especially for smaller employers with many low-wage, low-skill workers. The researchers used Current Population Survey (CPS) data from 1988 to 1999 and Medical Expenditure Panel Survey (MEPS) data from 1996 to 1999. The objective of the study was to inform policymakers about the decision of employees to accept and employers to offer health insurance coverage and how the safety net influences those decisions.
Policy Summary: There is extensive literature on the extent to which public health insurance coverage through Medicaid induces less private health insurance coverage. However, little is known about the effect of other components of the health care safety net in crowding out private coverage. The researchers’ goal was to examine the impact of the health care safety net on the health insurance coverage of children. They conducted an individual-level analysis of health insurance coverage using data from the Current Population Survey for the years 1990-2000 combined with a long panel of state level data on hospital uncompensated care and free and reduced price care offered by Federally Qualified Health Centers (FQHCs). They focused on two distinct groups for the analyses: children aged 14 and under and single, childless adults aged 18 to 64. Because measures of the safety net are likely to be confounded with insurance coverage, the researchers used an instrumental variables approach in estimating the impact of the safety net on health insurance coverage. The instruments for the measures of the safety net included measures of tax appropriations and state and local support received by hospitals and FQHCs, state Disproportionate Share payments, state uncompensated care pool dollars, and the amount of state budget surplus or deficit. They also controlled for Medicaid eligibility of sample members with standard techniques. By additionally including state fixed effects the researchers avoided the bias that occurred in previous cross-sectional designs because of unobserved differences across states that resulted in, for example, high levels of both uninsurance and safety net dollars. Including state fixed effects let them restrict the analysis to the within-state variation in health insurance coverage and the health care safety net measures. When the researchers omitted state fixed effects, they observed that high levels of hospital uncompensated care were associated with statistically significantly higher levels of uninsurance and lower levels of both public insurance and private insurance. Their results implied that a one standard deviation increase in hospital uncompensated care per population (a 46 percent increase) would be associated with a 3 percentage point increase in the uninsurance rate among children. However this result, which generally conforms to prior findings, did not control for unobserved state heterogeneity. When the researchers included state fixed effects, their results provided mixed evidence on the extent of crowding-out. Hospital uncompensated care does not appear to crowd-out health insurance coverage and health center uncompensated care appears to crowd-out private coverage for adults and, in some specifications, children. Less crowd-out for hospital uncompensated care may be plausible given that most hospital uncompensated care pays for big ticket items rather than more routine care that individuals may think of when making coverage decisions. Most of the arguments about the exogeneity of their uncompensated care measures suggest that their estimates should overstate the extent of crowd-out. Similarly, the likely potential endogeneity concerns about their instruments would also suggest that they should overstate the extent of crowd-out. That the researchers did not find strong evidence of crowd-out suggests that the effects may be small if present. Further study of the determinants of uncompensated care provision is called for, and would shed light on the validity of potential instruments for uncompensated care.