Central to any evaluation of the costs and benefits of immigration is immigrant use of government-provided services. One of the most contentious issues surrounding the immigration debate is immigrant use of means-tested programs. Partly out of concern for immigrant use of welfare programs, Congress curtailed welfare eligibility for some recent legal immigrants as part of its overhaul of the welfare system in 1996. Even with the changes made in 1996, most immigrants continue to be eligible for these programs because the changes primarily affected recent arrivals. Moreover, immigrants can retain welfare eligibility by naturalizing. Even if the immigrant himself is not eligible because of legal status (illegals are barred from using most programs) or length of residence in the country (one of the conditions of welfare reform), immigrant families can still receive benefits on behalf of their U.S.-born children, whose welfare eligibility is the same as any other native-born American. Also, in many cases state governments have chosen to provide benefits to otherwise ineligible immigrants. Therefore, changes made by Congress in the rules governing welfare use by newly arrived immigrants may not have had as large an effect on long-term welfare use by Mexican immigrants as policy makers may have thought.
Mexican Use of Means-Tested Programs Remains High Even After Welfare Reform. Figure 12 reports use of means-tested programs for households headed by natives, all immigrants, and Mexican immigrants. Despite welfare reform and a strong economy in the latter half of the 1990s, the figure shows that immigrants in general and Mexican households in particular use every major means-tested program at higher rates than natives. While use of Supplemental Security Income (SSI) by Mexican households is only slightly higher than that of natives, their use of TANF/general assistance, food stamps, Medicaid, and the subsidized school lunch program is dramatically higher than households headed by natives. All of these programs are very large in size. In 1999, more than $300 billion was spent on the means-tested programs listed in Figure 12. Even the school lunch program, the smallest of the programs listed in the figure, costs more than $5 billion annually.
Not only do immigrants use welfare programs at higher rates than natives, their use of the Earned Income Tax Credit (EITC) is also substantially higher than that of natives. With an annual cost of $32 billion, the EITC is the nation’s largest means-tested cash assistance program for workers with low incomes. Persons receiving the EITC pay no federal income tax and instead receive cash assistance from the government based on their earnings and family size.
The values for the EITC in Figure 12 and all subsequent figures almost certainly overstate program use by both immigrants and natives, because unlike the other programs listed in the figure, the Census Bureau assigns use of the credit to respondents based on their income and family characteristics, not based on their response to a specific question on the survey. All persons who file a return should receive the EITC; the IRS will process it automatically if you qualify. However, persons whose employment is not reported to the IRS (that is, they work under the table) or who do not file an income tax return will not receive the credit. Reasearch indicates that in most years, roughly 85 percent of those eligible for the EITC do receive it. Moreover, even if a person does not receive the credit, the fact that they are eligible means that they have no federal income tax liability. It is also important to note that residing in the country illegally does not necessarily preclude use of the credit.
Figure 13 reports the average payment received by households using means-tested programs. With the exception of SSI, Figure 13 shows that the percentage of Mexican households using cash assistance programs and food stamps is somewhat larger on average than that of natives who use these programs. This is because payments for public assistance, food stamps, and the EITC typically reflect the number of people in the households. Because Mexican households are larger on average (primarily because of higher fertility), the size of their average payment is also larger. This means that not only are Mexican households more likely to use means-tested programs, the size of the payment they receive is larger on average.
Mexican Use of Means-Tested Programs Over Time. As we have seen, income among Mexican immigrants rises the longer they live in the country. This improvement in income over time does result in a modest decline in their use of means-tested programs. Figure 14 reports the percentage of households using at least one major welfare program or the Earned Income Tax Credit for households headed by Mexican households based on how long the household head has lived in the country. The figure shows that Mexican households use means-tested programs at double the rate of natives, even when they have lived in the United States for many years.
While the poverty and income figures show significant progress over time, a somewhat different picture exists with regard to welfare use. While the rise in income that occurs over time should reduce welfare use, it may also be the case that Mexican immigrants become more familiar with the welfare system the longer they live in the United States. This increased awareness of welfare may offset, at least in part, a reduction in welfare use that should come with higher incomes. The fact that welfare does not drop dramatically over time is troubling because it indicates that even after having lived in the country for many decades, many Mexicans immigrants are still not self-sufficient.
Use of Means Tested Programs by Working Mexicans. While it is commonly believed that most people who use means-tested programs do not work, low-income workers or their children are eligible for most means-tested programs. In fact, some programs, such as the EITC, can only be used by those who work. Moreover, many individuals use welfare programs while they are looking for employment. These individuals show up in the CPS as having worked and as having been on welfare during the same calendar year. Thus, work and use of means-tested programs are not in any way mutually exclusive.
Most households, immigrant or native, have at least one adult in the household who works. Based on the 2000 CPS, 92 percent of households headed by a Mexican immigrant had at least one person 18 years of age and older who worked during 1999. For households headed by natives, 80 percent had at least one adult worker. Figure 15 reports use of means-tested programs for households headed by Mexican immigrants and natives in which at least one person works. The results show that among working households, use of means-tested programs by Mexican immigrants is dramatically higher than that of natives. Thus, high rates of welfare use by Mexican immigrants are not the result of a lack of work. Rather it stems from the fact that a large share of Mexican households have low-incomes and unstable employment histories and thus use a great deal of public services. This, of course, should come as no surprise given the very low skill level of a large share of Mexican immigrants.
The heavy use of welfare by working Mexican immigrants suggests that a guestworker program may create significant fiscal costs. Because the modern American economy offers very limited opportunities for workers with little education, it may simply not be possible to allow large numbers of unskilled people into the country without increasing the number of people using the nation’s welfare system. Of course, the guestworker program envisioned by some of its advocates would likely place severe restrictions on use of means-tested programs by the guestworkers. However, as we will see in the next section, although illegal aliens also are barred from using welfare programs, illegal aliens often make use of such programs, mainly by receiving benefits on behalf of their native-born children.
Use of Means-Tested Programs by Legal and Illegal Immigrants. Using the same method to distinguish illegal aliens as before, Figure 16 reports estimated welfare use for households headed by legal and illegal Mexican immigrants. Not surprisingly, Figure 16 shows that almost without exception, use of means-tested programs is higher for households headed by legal Mexican immigrants than for those headed by illegal Mexican immigrants. For many programs, in fact, use rates for legal Mexican immigrants are more than twice that of native-headed households. However, the results also indicate that illegal immigration imposes significant costs on public coffers. Whereas 14.8 percent of native households use at least one of the five major welfare programs, among households headed by an illegal alien from Mexico the figure is 24.9 percent. Illegal immigrants from Mexico primarily receive welfare benefits on behalf of their American-born children. Overall, the results in Figure 16 indicate that whether legal or illegal, immigrants from Mexico make heavy use of means-tested programs. The findings in Figure 16 also suggest that one possible unintended consequence of legalizing Mexican illegals already in the country would be to substantially increase their use of means-tested programs.
Use of means-tested programs by illegal immigrants from Mexico points to a fundamental problem that would almost certainly exist with any guestworker program. Even if guestworkers are made technically ineligible for means-tested programs, it seems almost certain that they would make use of them anyway by receiving benefits on behalf of their native-born children. After all, the findings in Figure 16 indicate that despite an outright ban on their use, illegals from Mexico actually use such programs at higher rates than natives in many cases. Any guestworker program would bring in large numbers of immigrants from Mexico with very low skill levels and resulting low incomes. Their low incomes coupled with their children’s eligibility would mean very significant costs to taxpayers even if the guestworkers themselves are successfully barred from using such programs.