Immigrants Use Less Welfare at Every Income Level

Poor immigrants use less welfare than poor native-born Americans

An astute reader of our recent Cato policy brief on immigrant welfare use suggested stratifying our data by income since many welfare programs are intended to help the poor. Doing so would be a return to how we used to report immigrant welfare use in the olden days. We do that below and find that immigrants consistently report lower welfare use for nearly every program, regardless of poverty or income.

Here are our methods before the results. You can skip this paragraph if you just want to see the results. Like our brief, we use the Survey of Income and Program Participation (SIPP) to evaluate welfare use by program for immigrants and native-born Americans by poverty and at different income quartiles. Our brief looks at individual welfare use, which is almost always the best variable for any analysis because individuals always contain the same number of people: one. The other choice would be a household-level analysis, which would include many native-born Americans and different numbers of people. Here, like our brief, we use the individual as the unit of analysis. SIPP is an unfortunately small survey, so we are limited to reporting only welfare programs where the sample sizes are large enough to be statistically significant.

We first use the poverty variable from the SIPP. This variable indicates whether an individual belongs to a household in poverty by providing the household income divided by the poverty threshold for that household. Immigrants, both in and out of poverty, report less welfare consumption than native-born Americans across almost all programs, with a few exceptions. Immigrants who are not in poverty report slightly higher consumption of Medicaid and WIC, and immigrants in general report higher rates of EITC recipiency. However, immigrants generally use less welfare than native-born Americans with the same level of poverty (Figures 1 and 2).

This pattern is also consistent across the income distribution. SIPP allowed us to break out welfare use by income quartile. We find that immigrants report lower welfare participation than native-born for nearly all programs in every income quartile. This holds not only among people in poverty, but also among those not in poverty. The only exceptions are that immigrants in the fourth (highest) income quartile report more Medicaid and SNAP use than natives (Table 1). There’s no good reason for that, as nobody in the highest quartile of income should be receiving those benefits, even if they are old-age entitlement programs, to say nothing of food stamps.

Another exception to this pattern is the rate of earned income tax credit (EITC) recipiency. Immigrants across all poverty and income subgroups consistently receive EITC at a higher rate than native-born Americans. Unlike means-tested and old-age benefit programs, EITC is framed as a tax credit and does not entail the same kinds of eligibility constraints as the traditional programs. That likely explains its higher use among immigrants (Table 2). Regardless, there is no compelling public policy reason for this, even if you otherwise like redistribution.

Interestingly, the only groups that reported statistically significant Temporary Assistance for Needy Families (TANF) use were native-born in both the lowest income quartile and out of poverty. No grouping of immigrants reported significant TANF use. These results, taken together, show the boundaries of reported TANF use: Immigrants tend not to get TANF, and anyone above the bottom 25 percent of incomes tends not to get it at all.

Note that the Figures above are not directly comparable to those in our brief. This analysis here uses only the weighted means for the demographic subgroups in question, but it does not adjust for undercounting using demographic or administrative data. Thus, the Figures here should not be used to calculate aggregate usage amounts across groups and should only be compared to each other (i.e., immigrants in poverty compared with native-born Americans in poverty).

The data consistently show that immigrants report less welfare use than native-born Americans across income levels and poverty statuses. Whatever one’s views on immigration policy, the claim that immigrants systematically consume more welfare than native-born Americans isn’t supported in the data. Rather than vilifying immigrants or closing the border, the more fiscally responsible policy position is to build a wall around the welfare state and thus limit welfare to US citizens only. The optimal policy is to remove the welfare state entirely.

We’re lucky to have smart and dedicated readers who closely examine our research. Thank you to the reader for emailing us this question. To him and our other readers, please keep the questions coming.