Shikha Dalmia argues that the notion that less-skilled immigrants rely heavily on the social-welfare system is more myth than reality. She does not, in my view, make a convincing case.
1. She claims that the reason very few visa applications are rejected on the grounds that the applicant might become a public charge is that existing visa rules largely screen out individuals who are likely to rely heavily on the social-welfare system. As Byron York discussed in a recent column, however, consular officials are not allowed to consider whether an applicant is likely to be eligible for SNAP, Medicaid, and a wide range of other means-tested benefits.
Consider the following from a 2011 report by Curtis Skinner of the National Center for Children in Poverty:
In 2009, some 24 percent of children in immigrant families lived below the official poverty line and 51 percent below double the poverty line; the respective figures for children in native families were 18 and 38 percent. Although labor force participation and employment rates are very high among immigrant fathers, many work in low-wage jobs. Among immigrants, about 29 percent of children lived in a low-income working family in 2009 compared to 17 percent of native-family children. Research shows that the immigrant-family child poverty rate is negatively associated with parental education, English proficiency, length of U.S. residence, and citizenship status. In 2009, almost one-quarter of all children in immigrant families lived in “linguistically isolated” families in which no household member over age 14 speaks the English language very well. Moreover, 53 percent of immigrant-family children lived with at least one parent who had not graduated from high school, compared to 44 percent of native-family children. [Emphasis added]
As Skinner goes on to observe, it is true that immigrant families tend to access means-tested benefits at lower levels than their citizen counterparts. But he observes that this reflects a number of factors, including linguistic isolation and the fact that members of mixed-status families are reluctant to interact with the authorities. Efforts to regularize the status of unauthorized immigrants will likely change this landscape, increasing the number of eligible individuals who access benefits.
It is possible that having large numbers of poor immigrants in the U.S. who do not access means-tested benefits is not something we ought to fret over, as the immigrants in question are better off than they were in their native country. This is a perfectly reasonable, but it is important to be explicit about it.
The Kaiser Family Foundation described the health coverage landscape for non-citizens in a recent report:
Similar to native citizens, the majority of naturalized citizens have employer or other private coverage. However, non-citizens are nearly three times as likely to be uninsured relative to U.S.-born citizens, reflecting more limited access to both private and public coverage (Figure 1). Although non-citizens are as likely as citizens to work, they are often in jobs and industries that do not offer coverage. Further, non-citizens are subject to immigrant-specific eligibility restrictions in Medicaid and CHIP. Since 1996, lawfully residing non-citizen immigrants have been barred from enrolling in Medicaid and CHIP during their first five years in the United States. States were recently provided the option to eliminate this “five-year bar” for pregnant women and children, but do not have the option to waive the waiting period for other adults. Undocumented immigrants are prohibited from enrolling in Medicaid and CHIP. [Emphasis added]
Had KFF separated less-skilled immigrants from skilled immigrants, one assumes that the picture would change in important ways. KFF notes that a key barrier to expanded Medicaid coverage is lack of transportation and language barriers. That is, one reason why the number of non-citizens accessing Medicaid is not higher is that a large share of the non-citizen population in the U.S. does not have even limited English language proficiency or the means to travel. And so the fact that these individuals are not public charges flows from the fact that their resources and skills are so limited that they can’t navigate a complex bureaucracy. It should go without saying that this is unfortunate, yet as these individuals gains resources and skills, more are likely to become eligible for public assistance.
2. Dalmia observes that Latinos are settling in states that tend to be less generous in terms of social service provision:
The dearth of proof for the view that people flock to the U.S. for welfare is long-standing. In fact, according to the Agriculture Department, which administers food stamps, Latinos in recent years have increasingly flocked to states such as Tennessee, Arkansas, Alabama, Texas and the Carolinas, which have stingy benefits and plentiful jobs, instead of to traditional gateways, such as New York and California, which have relatively generous programs.
The 10 states that experienced the largest percentage increase in their foreign-born population from 2000 to 2009 spent far less on public assistance per capita compared with the 10 states with the slowest-growing foreign-born populations.
One needn’t maintain that immigrants are flocking to the U.S. for welfare to believe that poor immigrants with limited skills and limited English language proficiency are likely to eventually become eligible for means-tested benefits, and the fact that profound poverty and linguistic isolation are barriers to accessing means-tested benefits does not mean that there is no reason for concern. Moreover, high-poverty states receive more generous federal transfers, e.g., more generous Medicaid match rates, etc.
3. Dalmia makes the valid point that we need to consider educational and labor market outcomes for the children of immigrants:
Of course, even if immigrants don’t come to the U.S. to live off the welfare state doesn’t mean they don’t end up doing so. The best evidence for this claim came in the 2007 Heritage Foundation study, which found that even though immigrants have been barred since 1996 from receiving federal means-tested benefits, their households still obtain about $20,000 more in benefits and services (such as schools and emergency medical care) than they pay in taxes. The study estimated that these costs imposed in 2004 a net burden of about $90 billion annually and a whopping $1 trillion over a decade.
This would be cause for concern — if those numbers were the whole story. The study was criticized for counting government spending on the (American-born) children of immigrants but then ignoring the taxes these offspring paid when they grew up. By that standard, most middle-income families in the U.S. with three or more children in public schools would be a net burden. [Emphasis added]
First, immigrants have not been barred from receiving all means-tested benefits since 1996, as the National Immigration Law Center explains. Exceptions were carved out of the 1996 law and eligibility has been restored in the years since for a number of other programs. States can allow “qualified” lawful permanent residents to access TANF, Medicaid, CHIP, school breakfast and lunch programs, WIC, a variety of in-kind services, and SNAP.
Moreover, note that Sessions et al. are making a narrow point about less-skilled immigrants who are likely to become eligible for means-tested benefits, not all immigrants. As George Borjas and Lawrence Katz observe in their paper on “The Evolution of the Mexican-Born Workforce in the United States,” the lagging performance of less-skilled immigrants relative to other immigrants and the native-born population is transmitted to future generations. So when Dalmia compares the native-born children of all immigrants to the children of middle-income native-born families, she’s not making a comparison that is relevant to the concerns raised by conservatives critical of permitting less-skilled immigration. Rather, the relevant comparison is between the children of skilled immigrants as opposed to the children of less-skilled immigrants.
4. Dalmia notes that even if there is a net fiscal cost associated with immigration, it is outweighed by the larger social benefits.
State-level studies that have taken both into account consistently find that the economic contributions of these immigrants dwarf their fiscal costs. A 2006 analysis by the Texas comptroller estimated that low-skilled unauthorized workers cost the state treasury $504 million more than they paid in taxes in 2005. Without them, however, the state’s economy would have shrunk by 2.1 percent, or $17.7 billion, as the competitive edge of Texas businesses diminished. Likewise, a 2006 study by the Kenan Institute at the University of North Carolina found that although Hispanic immigrants imposed a net $61 million cost on the state budget, they contributed $9 billion to the gross state product.
Once again, we’re using a broad, diffuse category of Hispanic immigrants rather than less-skilled immigrants, regardless of ethnicity. The pressing policy question is whether it would be wise for U.S. policymakers to reduce the influx of less-skilled immigrants while commensurately increasing the influx of skilled immigrants who are deemed unlikely to become eligible for means-tested benefits. There is good reason to believe that such a substitution would yield a larger contribution to gross state product.
5. Dalmia goes on to make a very similar point:
The Heritage Foundation study also implied that a homegrown working class would be cheaper for the country because households headed by low-skilled immigrants consumed $10,000 more in government services than those headed by Americans.
The trouble is that the study compared the welfare use of low-skilled immigrant households with average American households, rather than with low-skilled American households.
In comparing welfare use by immigrants with that of Americans in the same socioeconomic stratum, a different picture emerges, as a study by Leighton Ku and Brian Bruen of George Washington University for the Cato Institute found recently. Low-skilled foreigners, including adults and their U.S.- born children, were generally less likely than Americans to receive public benefits, such as from Medicaid, the Supplemental Nutrition Assistance Program and Supplemental Security Income. This is partly because many adults are in the U.S. illegally or on temporary visas or haven’t held a green card long enough to qualify for most means-tested benefits besides emergency health care. But the value of benefits they receive is usually lower, too.
“The combination of lower average utilization and smaller average benefits indicates that the overall cost of public benefits is substantially less for low-income non-citizen immigrants than for comparable native-born adults and children,” the Cato study concluded. [Emphasis added]
To reiterate, lower average utilization reflects the severity of linguistic isolation, among other factors. As this condition is alleviated, some immigrants will earn incomes high enough to become ineligible for means-tested benefits while others will become eligible for means-tested benefits. The crucial question is whether we ought to shift our immigration policy to reduce less-skilled immigration while encouraging more skilled immigration.
I’ll make a few very simple claims, several of which are normative claims that not everyone will embrace:
(a) in the U.S., less-skilled individuals with limited English proficiency are likely to have lower lifetime incomes than skilled individuals who speak English fluently;
(b) individuals with low lifetime incomes are more likely to be eligible for and to rely on means-tested public benefits, particularly as they acculturate and learn to navigate U.S. institutions;
(c) the children of less-skilled individuals are likely to have somewhat lower skill levels and lifetime incomes than the children of skilled individuals;
(d) it is not incoherent to suggest that U.S. policymakers have more of an obligation to address the needs of less-skilled individuals, whether native-born or foreign-born, currently residing in the U.S. as opposed to those of less-skilled individuals residing elsewhere and who would like to access the U.S. labor market;
(e) and so an immigration policy that expands the population of less-skilled individuals is not the ideal course of action, particularly if we can instead craft an immigration policy that substitues an increase in the number of skilled immigrants for less-skilled immigrants. The Sessions effort to limit the influx of immigrants likely to become eligible for means-tested benefits is a reasonable start.
The labor market position of less-skilled individuals in the U.S. has been deteriorating for at least 30 years. If this development is somehow reversed in the coming years, I would be very amenable to embracing a different approach.