The Decline in Medicaid Use by Noncitizens since Welfare Reform

Health Policy Online No. 5

The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.

Note: This report is available in its entirety in the Portable Document Format (PDF).

Introduction

This brief looks at whether the enrollment of noncitizens in Medicaid has decreased since the enactment of welfare reform. The common expectation has been that noncitizen enrollment would decrease, given that welfare reform significantly restricted Medicaid eligibility of noncitizens, barring most of them from receiving Medicaid during the first five years of living in the country unless states chose to cover them through state programs.¹ The analysis detailed in this brief finds that Medicaid enrollment among noncitizens did in fact decrease after welfare reform as intended by law, and that noncitizens are much less likely than native citizens to receive Medicaid. However, a report released by the Center for Immigration Studies (CIS) in March 2003 suggested the opposite—that enrollment of noncitizens in Medicaid increased after welfare reform and remains much greater than that of native households.²

The CIS finding regarding Medicaid was large enough to offset the decreases they found in use of TANF/General Assistance and food stamps by immigrant households, and led them to conclude that immigrant households still used welfare programs significantly more than native citizen households. A closer examination of their analysis shows that their methods overstate the percentage of the population receiving Medicaid and the share of immigrants on Medicaid, resulting in misleading conclusions about welfare use among immigrants. There are two main reasons for this. First, the CIS looked at Medicaid enrollment on a household level, which meant if one person in the household received Medicaid, the entire household was counted as receiving Medicaid. Since Medicaid only serves the individual eligible for the program and not the entire household, this method would overstate the share of the population on Medicaid. For example, in 2001, 14.4 percent of households had at least one individual on Medicaid, but only 11.4 percent of individuals reported Medicaid coverage.³ This method would also mask changes in individual Medicaid enrollment.

Secondly, the immigration status of a household was assigned according to status of the head of the household, without regard to the status of the individual recipients. This misrepresented the number of immigrants with Medicaid because households were identified as immigrant households receiving Medicaid even if all program recipients were native citizens. In fact, in the CIS report, 37 percent of those identified as immigrant households receiving Medicaid in 2001 were households in which all the Medicaid recipients were native citizens.⁴ A recent report found that the increase in immigrants' Medicaid use that the CIS reported was actually due to an increase in use of Medicaid or SCHIP by native citizen children in households headed by foreign-born individuals.⁵ Thus, assigning immigration status by household overstated the extent to which immigrants use Medicaid, especially because Medicaid is an individual benefit. Medicaid enrollment is more accurately studied by assigning citizenship status on an individual level and analyzing the percentage of people, not the percentage of households, who receive the benefit.

We analyzed the same data the CIS used, the March Current Population Survey (CPS) data from 1997 to 2002, and looked at Medicaid coverage by citizenship status on an individual level. In examining Medicaid coverage, we included anyone who reported coverage from Medicaid, SCHIP, or a state program. We divided the analysis into two time periods, 1996-2000 and 2000-2001, and provided two data points for 2000 because changes in survey questions and weighting techniques make comparisons across time difficult. The 1996-2000 data are weighted to the 1990 Census and represent respondents who did not verify their health insurance information, while the 2000-2001 data are weighted to the 2000 Census and represent respondents who were asked to verify their insurance information. These two time periods are key aspects of the analysis: starting from 1996 allows us to observe the impact of welfare reform provisions, while 2000 marks the beginning of the economic downturn.

Note: This report is available in its entirety in the Portable Document Format (PDF).

Usage and reprints: Most publications may be downloaded free of charge from the web site and may be used and copies made for research, academic, policy or other non-commercial purposes. Proper attribution is required. Posting UI research papers on other websites is permitted subject to prior approval from the Urban Institute—contact [email protected].

If you are unable to access or print the PDF document please contact us or call the Publications Office at (202) 261-5687.

Disclaimer: The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. Copyright of the written materials contained within the Urban Institute website is owned or controlled by the Urban Institute.