The following is the public comment I submitted, which may be of interest to readers.1 (Also see my colleague Steven Camarota's comment, here).
As a researcher who has spent a decade working on immigration issues with a variety of think tanks in Washington, D.C., I applaud the administration's efforts to enforce the law against admitting or adjusting the status of any alien who is likely to become a public charge. The proposed rule defining the basis for a public-charge determination is consistent with our nation's goal to prioritize self-reliant immigrants, and to reserve welfare benefits for refugees and others who have arrived specifically for humanitarian reasons. In my opinion, the proposed rule is a significant improvement over the status quo, but it does not go far enough. Many ordinary immigrants will still consume welfare under the new rule.
I have organized my comment into five points. The first two points show that the proposed rule is a necessary reform, and the final three points argue that the proposed rule does not go far enough.
The Proposed Rule Addresses a Real Problem
1) Immigrant welfare use is high. The federal government does not keep administrative data on welfare consumption by immigration status. However, we can generate estimates from the Survey of Income and Program Participation (SIPP), which is designed to capture welfare use and has the least undercount of all the major Census surveys. I commend DHS for using the SIPP in its own analyses rather than using a less comprehensive dataset such as the Current Population Survey.
In the most recent SIPP data analyzed by the Center for Immigration Studies, 55 percent of immigrant-headed households received welfare in the form of cash, food, medical care, or housing assistance, compared to 35 percent of native households. Welfare use rises to 63 percent for households headed by noncitizen immigrants, and to a remarkable 80 percent for noncitizen households with children.2 Clearly, immigrants not only consume welfare, but do so at high rates.3
Critics have suggested the welfare use rates are inflated because they are based not on individuals but on households, which often include U.S.-born children. Indeed, DHS's own SIPP estimates are based on individuals only. But if we limit the analysis to individuals without considering their dependents, then an impoverished immigrant could sign up her U.S.-born child for a means-tested benefit and have it count as native use of the welfare system. That is clearly incorrect, as any program for children benefits the parents who are otherwise responsible for them. As a parent myself, I must pay for my children's food, shelter, and medical care. If the government steps in and does that for me, that frees up space in my budget to spend on other things. The same is true for immigrants who benefit when taxpayers cover their children's basic needs.
2) The public-charge law is rarely enforced. Although it has long been against the law to admit or adjust the status of any alien who is likely to become a public charge, enforcement of this law has been ineffective. Remarkably, even as a majority of immigrant households have ended up on welfare, the number of applicants disqualified by the public-charge law appears to be miniscule.4 Part of the problem is that the current system considers only cash handouts and long-term institutionalization as grounds for a public-charge determination. As a result, an immigrant could receive a wage subsidized by the EITC, shop at the grocery store with food stamps, pay for doctors' visits through Medicaid, live in a subsidized rental unit, and heat it with energy assistance — all without being a public charge in the eyes of the government.
The current system is based on the idea that a public charge must be "primarily" dependent on the government. In other words, if an immigrant is receiving public assistance but could theoretically do without it, then that person should not be considered a public charge. But whatever one thinks of the welfare state — I personally would like to see it shrink — the public has decided through its elected representatives that most Americans in poverty should have the support of government programs to alleviate their condition. Therefore, with few exceptions, any poor person is "likely to become a public charge." Administrative rules should reflect that fact.
Furthermore, in 1996 Congress declared "a compelling government interest to enact new rules for eligibility and sponsorship agreements in order to assure that aliens be self-reliant in accordance with national immigration policy" (emphasis added).5 Self-reliant is an unambiguous descriptor. Nowhere does the law qualify that aliens need to be "primarily" self-reliant — a formulation that might even be seen as a contradiction in terms. DHS is correct to jettison the "primarily" qualifier from its definition of a public charge.
The Proposed Rule Is a Significant Improvement, but Does Not Go Far Enough
3) Any receipt of means-tested anti-poverty benefits by immigrants or their dependents should count toward the public-charge determination. The proposed rule admirably expands the list of welfare programs that could qualify an immigrant as a public charge if he or she participates in them. In addition to cash welfare such as TANF and SSI, the list would also include food stamps, housing assistance, Medicaid — perhaps including CHIP, as it should — and means-tested subsidies associated with Medicare prescription drugs. The proposed list is much more comprehensive than the current one, but it still leaves out several important welfare programs — namely, WIC, free school lunch, the refundable portion of the EITC, and the ACTC. All of these are means-tested, anti-poverty benefits just as the other programs are. Excluding the refundable tax credits is especially problematic given their cost. The refundable portion of the EITC and the ACTC cost over $80 billion combined in 2016.6 As cash transfers to the poor, these tax credits meet any definition of a means-tested, anti-poverty benefit.
Unfortunately, the proposed rule allows immigrants to use the programs on the list to a limited extent without counting against them in a public-charge determination. Allowing any welfare use at all is unnecessary, complicated, and contradictory to the administration's proposed definition of a public charge as "an alien who receives one or more public benefits". As I explained at the end of point No. 2, there is no meaningful distinction between dependency and "partial dependency" when it comes to defining a public charge, and there is no basis for such a distinction in the relevant statutes.
Finally, the proposed rule does not appear to count welfare use by immigrants' dependents. This is a critical omission. As explained in point No. 1, if an immigrant relies on the government to take care of the individuals he or she is supposed to be taking care of, then the immigrant is effectively receiving that government aid himself or herself. By failing to consider dependents, the proposed rule will allow a substantial number of immigrants to continue to benefit from welfare. To illustrate, about 21 percent of immigrant adults were on Medicaid in 2014, but fully 58 percent of children in immigrant-headed households were on Medicaid at the same time.7 Taking an immigrant's household size into account, as DHS proposes to do, will mitigate some of the problem of welfare use by dependents, but DHS should more directly address it.
4) Immigrants without at least some college should bear a heavy burden in proving they will not become public charges. Having a job is not enough. As required by statute, the proposed rule would consider an immigrant's education as part of a broader public-charge determination. However, I am concerned that not enough weight will be given to this highly significant predictor of welfare consumption. While 37 percent of households headed by noncitizens with at least some college use welfare, the rate rises to 81 percent for households headed by noncitizens with only a high school diploma or less.8 The latter rate is so high that any potential immigrant without some college training should bear a heavy burden in proving he or she will not become a public charge.
Having just any job should not be enough to overcome that burden. Contrary to images of welfare users as layabouts, most means-tested anti-poverty benefits are available to people who work. To illustrate, 55 percent of immigrant-headed households containing at least one worker are on some form of welfare — a rate that is virtually identical to the rate for all immigrant-headed households.9 Wisely, the proposed regulation would not weigh income heavily in the immigrant's favor unless it is at least 250 percent of the poverty line. In fact, not having an income of at least 250 percent of the poverty line should count heavily against the immigrant.
5) Case-by-case discretionary power should be limited. The proposed rule would allow DHS officials broad discretion in making a public-charge determination. While some discretion is always necessary given the wide range of circumstances that immigrants may face, the administration should consider limiting the opportunities for it to be applied. Broad discretion would allow future administrations to weaken enforcement of the public-charge rule without even giving notice that the rules are changing.
Needless to say, the immigration bureaucracy has a history of interpreting restrictions as leniently as possible. An example discussed above is the "primarily dependent" concept — an approach to the public-charge law that was mostly ineffective in accomplishing the intentions of Congress, and perhaps deliberately so. USCIS Director Cissna alluded to this institutional problem in explaining why he changed the organization's mission statement:
Referring to applicants and petitioners for immigration benefits, and the beneficiaries of such applications and petitions, as "customers" promotes an institutional culture that emphasizes the ultimate satisfaction of applicants and petitioners, rather than the correct adjudication of such applications and petitions according to the law. ... All applicants and petitioners should, of course, always be treated with the greatest respect and courtesy, but we can't forget that we serve the American people.10
The tendency for the bureaucracy to adopt a serve-immigrants culture rather than a serve-Americans culture is exactly why administrative rules should be clear and predictable, with discretion limited to the most difficult cases only.
Conclusion
In summary, the proposed rule is a major step in the right direction. By expanding the list of welfare programs that may count toward a public-charge determination, the administration will be more faithfully executing the law against admitting or adjusting the status of an immigrant who is likely to become a public charge. However, more can be done to ensure that we have immigrants who are self-reliant. The new list of welfare programs should be expanded further. Any use of those programs (rather than use beyond a certain threshold) by immigrants or their dependents should count toward a public-charge determination. Low levels of education should count more heavily against immigrants, and reliance on DHS discretion should be reserved for truly marginal cases.
End Notes
1 I received a Ph.D. in public policy from Harvard University in 2009. I have since worked on immigration policy with the American Enterprise Institute, the Heritage Foundation, and the Center for Immigration Studies. This comment is my own and does not necessarily reflect the views of any organization.
2 Steven A. Camarota and Karen Zeigler, "63% of Non-Citizen Households Access Welfare Programs", Center for Immigration Studies, December 2, 2018, Table 1 and Table 4. This analysis uses Wave 2 of the 2014 SIPP panel. DHS appears to use Wave 1.
3 One might be tempted to dismiss these statistics on the grounds that they come from the Center for Immigration Studies (CIS), an organization that advocates less immigration. But CIS simply tabulated publicly available data from the Census Bureau — something anyone with a rudimentary background in data analysis can do. Indeed, when a researcher with the progressive organization Demos re-analyzed the data from an earlier CIS study of the SIPP, he came up with numbers that "essentially match the CIS figures." (Matt Bruenig, "Welfare Use by 'Immigrant' and 'Native' Households", Demos, September 10, 2015. Bruenig wanted to include Social Security and Medicare, which are not means-tested programs, as welfare.) In addition, the Migration Policy Institute, an organization generally supportive of immigration, found that "nearly half" of all noncitizens live in families that use one of the four major welfare programs. (Migration Policy Institute, "Nearly Half of All Noncitizens in U.S. Could Be Affected by Proposed Trump Administration Public Charge Rule, Up from Current 3 Percent", June 12, 2018.)
4 Although the government does not appear to have good data on the question, CIS expert Jessica Vaughan has estimated that merely "hundreds" of applicants are rejected on public charge grounds, despite well over a million new entrants or applicants each year. (Jessica Vaughan, "Proposed Public Charge Rule: Few Will Be Barred, Billions Will Be Saved", Center for Immigration Studies, September 27, 2018.)
5 8 U.S. Code § 1601 (5).
6 Robert Rector and Vijay Menon, "Understanding the Hidden $1.1 Trillion Welfare System and How to Reform It", Heritage Foundation, April, 2018, Appendix Table C.
7 Author's calculations using the SIPP.
8 Camarota and Zeigler, Table 3. The figure for "at least some college" is imputed using the weighted N.
9 Camarota and Zeigler, Table 6.
10 Megan Keller, "Immigration Agency Chief Defends Striking 'Nation of Immigrants' from Mission Statement", The Hill, August 15, 2018.