اصلاحات رفاهی، عرضه نیروی کار، و بیمه سلامت در جمعیت مهاجر

Although the 1996 welfare reform legislation limited the eligibility of immigrant households to receive assistance, many states chose to protect their immigrant populations by offering state-funded aid to these groups. I exploit these changes in eligibility rules to examine the link between the welfare cutbacks and health insurance coverage in the immigrant population. The data reveal that the cutbacks in the Medicaid program did not reduce health insurance coverage rates among targeted immigrants. The immigrants responded by increasing their labor supply, thereby raising the probability of being covered by employer-sponsored health insurance.

The number of immigrants entering the United States grew rapidly in recent decades. During the 1950s, only 250,000 legal immigrants entered the country annually. By the 1990s, nearly 1 million persons entered the country legally each year and another 300,000 entered—and stayed in—the country illegally.1 An increasing number of the new immigrants fall in the lower range of the skill and income distributions. In 1960, the typical immigrant earned 4% more than the average native worker. By 1998, the typical immigrant earned 23% less (Borjas, 1999). The trends in the size and skill composition of the immigrant population sparked a contentious debate over the economic and demographic impact of immigration.2 For instance, there has been a great deal of concern over the possibility that immigrants do not “pay their way” in the welfare state (Smith and Edmonston, 1997). And, in fact, the evidence suggests that immigrant households are now much more likely to receive public assistance than in the past.3 Concurrent with the resurgence of large-scale immigration, there has been an increase in the number of persons who lack health insurance coverage.4 Recent research suggests there may be an important link between these two trends. Despite the relatively high participation rate of immigrants in the Medicaid program, Camarota and Edwards (2000) report that immigrants are also disproportionately more likely to be in the population of uninsured persons: although persons in immigrant households make up only 13% of the population, they make up 26% of the uninsured. Camarota and Edwards conclude that “immigrants who arrived between 1994 and 1998 accounted for 59% of the growth in the size of the uninsured population” during that period (p. 5). The 1994–1998 period coincided with the enactment of the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA). The 1996 welfare reform legislation specified a new set of rules for determining the eligibility of foreign-born persons to receive practically all types of federal aid. In rough terms, PRWORA denies most means-tested assistance to non-citizens who arrived after the legislation was signed in 1996, and limited the eligibility of many non-citizens already living in the United States. The available evidence indicates that the rate of welfare participation in immigrant households declined sharply—relative to the decline in native households—in the aftermath of PRWORA (Borjas, 2001 and Fix and Passel, 1999). This paper uses data drawn from 1995 to 2001 Current Population Surveys (CPS) to examine the impact of PRWORA on health insurance coverage among immigrants. Because PRWORA reduced immigrant participation in welfare programs (including Medicaid), it seems reasonable to suspect that the welfare cutbacks should have increased the size of the foreign-born uninsured population. Remarkably, this expected increase did not occur. In fact, the fraction of immigrants who were not covered by health insurance remained roughly stable (or fell) during the period. The immigrant provisions in PRWORA could potentially affect only a subset of the immigrant population, depending on the immigrant’s state of residence, on the type of visa used to enter the United States, and on the immigrant’s naturalization status. This variation in eligibility rules can be exploited to examine how immigrants responded to the cutbacks in public assistance. It turns out that the immigrants most adversely affected by PRWORA significantly increased their labor supply, thereby raising the probability that they were covered by employer-sponsored health insurance. In fact, the evidence indicates that the increase in the number of immigrants covered by employer-sponsored health insurance was large enough to completely offset the impact of the Medicaid cutbacks. The study, therefore, provides evidence of a strong crowdout effect of publicly provided health insurance among immigrants.5 It is important to note, however, that my results differ in an important way from the evidence typically reported in the crowdout literature. The welfare reform legislation affected immigrant participation in a vast array of public assistance programs, not just Medicaid. For example, PRWORA also restricted immigrant receipt of cash benefits and food stamps. The crowdout effects documented in this paper, therefore, measure the total immigrant response to a generalized cutback in public assistance, rather than the immigrant response to eligibility changes in the Medicaid program.

The 1996 welfare reform legislation contained a number of provisions that greatly limited the eligibility of many immigrants (particularly non-citizens and non-refugees) to receive many types of public assistance. In response to the federal legislation, many states chose to protect their immigrant populations from the presumed adverse impact of PRWORA by offering state-funded assistance to these groups. I use data drawn from 1995 to 2001 Annual Demographic Supplements of the Current Population Surveys to examine the relation between the immigrant-related provisions in PRWORA—as modified by the subsequent state responses—and health insurance coverage in the immigrant population. In the absence of any behavioral response, one would have expected that health insurance coverage rates would have been sharply curtailed in the population most adversely affected by the restrictions, the non-citizens living in states that did not offer state-funded assistance to their immigrant populations. In other words, as the Medicaid cutbacks took effect, the proportion of those immigrants covered by some type of health insurance should have declined. The empirical analysis indeed indicates that the targeted immigrant population experienced a decline in Medicaid coverage as the PRWORA provisions took effect. However, the analysis also reveals that the expected decline in health insurance coverage rates did not materialize. If anything, health insurance coverage rates actually rose slightly in this group. The resolution to this conflicting evidence lies in the fact that the affected immigrants responded to the welfare cutbacks. The immigrants most likely to be adversely affected by the new restrictions significantly increased their labor supply, thereby raising their probability of being covered by employer-sponsored insurance. In fact, this increase in the probability of coverage through employer-sponsored insurance was large enough to completely offset the Medicaid cutbacks. The empirical analysis, therefore, provides strong evidence of a sizable crowdout effect of publicly provided health insurance among immigrants. In an important sense, the state programs were unnecessary. In the absence of these programs, the targeted immigrants themselves would have taken actions to reduce the probability that they would be left without health insurance coverage.