- Federal Budget
- Trump Administration’s Overbroad Public Charge Definition Could Deny Those Without Substantial Means a Chance to Come to or Stay in the U.S.
- Federal Tax
- State Budget and Tax
- Social Security
- Poverty and Inequality
- Food Assistance
- Family Income Support
- Climate Change
- View all
Trump Administration’s Overbroad Public Charge Definition Could Deny Those Without Substantial Means a Chance to Come to or Stay in the U.S.
May 30, 2019
The Trump Administration’s proposed public charge rule unveiled last October could result in large numbers of individuals being denied lawful permanent residence status, the ability to extend their stay, to change their status, or to enter the United States, despite extensive research on the benefits of immigration to the country and immigrants’ demonstrated upward mobility.
Under longstanding immigration law, certain individuals can be denied entry to the United States or permission to remain here if they are determined likely to become a “public charge,” which for decades has been defined as being primarily dependent on government for monthly cash assistance or long-term institutional care. The proposed rule would significantly alter the public charge definition and, in turn, change the character of the country to one that only welcomes those who already have substantial wealth and income.
Under the proposed rule from the Department of Homeland Security (DHS), individuals who are determined likely to receive even modest assistance from a far broader set of benefits — including benefits that help many workers like SNAP (formerly known as food stamps) and Medicaid — at any point over their lifetimes would be considered a public charge. Immigration officials would look at many factors to determine the likelihood of benefit receipt, including whether the immigrant’s current family income is above 125 percent of the federal poverty level.
The proposed rule would significantly alter the public charge definition and, in turn, change the character of the country to one that only welcomes those who already have substantial wealth and income.The proposed policy is so radical and would change the public charge definition to one so broad that more than half of all U.S.-born citizens could be deemed a public charge — and by extension and implication, considered a drag on the United States — if this definition were applied to them. The proposed rule does not apply to U.S. citizens. It is instructive, however, to consider the share of U.S.-born citizens whom the proposed rule would characterize as a public charge when considering the reasonableness of the standard.
- If one considers benefit receipt of the U.S.-born citizens over the 1997-2017 period, some 43 to 52 percent received one of the benefits included in the proposed public charge definition.
- In just a single year, 3 in 10 U.S.-born citizens receive a benefit included in the proposed public charge definition.
- If data allowed us to look at U.S.-born citizens over the course of their full lifetimes, benefit receipt would exceed 50 percent of the population.
- A significant share of individuals working in the United States — 16 percent — receive one of the benefits included in the proposed definition in just a single year. These are workers upon whom our economy relies.
The current definition is, by contrast, far narrower. In a single year, just 5 percent of U.S.-born citizens and 1 percent of individuals working in the United States meet the current benefit-related criteria in the public charge determination.
The proposed public charge criteria are not only broad, but would discriminate against individuals from poorer countries, regardless of their talents, because the incomes of the vast majority of people from many countries fall below the new 125 percent-of-poverty threshold included as a consideration in the public charge determination under the proposed rule. This criterion would have racially disparate impacts, as people from countries with low incomes are disproportionately people of color. This threshold would be particularly problematic for immigrants from poor countries seeking entry to the United States, even if they have some family already here, because their own income is likely to be very low compared to U.S. poverty standards. In addition, given the more complex prediction that immigration officials would have to make, their discretion, which could be influenced by implicit (or explicit) racial/ethnic bias, would likely affect the outcome for more people. This bias could lead immigration officials to keep out large numbers of people from certain countries or racial/ethnic groups, and to deny adjustment or entry to people of color at higher rates than similarly situated white individuals.
The proposed rule is a shortsighted attempt to remake the U.S. immigration system — without congressional approval — into one that welcomes only those with significant wealth. Immigrants fill important jobs and contribute to economic growth, and research has shown that immigrants raise children who demonstrate substantial upward mobility, attaining more education than their parents and moving up the economic ladder. Had this rule been in effect in prior decades, the United States would have been deprived of the talents of many hardworking immigrants who moved to this country to build a better life for themselves and their children and, in turn, made important contributions to their communities and the United States as a whole.
The Trump Administration’s proposed public charge rule has not been finalized yet, and the government is required to review and consider the evidence and views presented in the more than 266,000 public comments it received before finalizing it. Moreover, the proposed rule indicates that immigration officials will not apply the new definition of public charge until the rule becomes effective (likely 60 days after the rule is finalized). Benefits that the public charge determination previously excluded (such as Medicaid and SNAP) will be considered only if applicants receive them after the final rule becomes effective. Nonetheless, many families that include immigrants already have forgone needed services due to extensive media coverage about the proposed rule and confusion caused by the Administration implementing policy changes similar to those in the rule when considering applications for entry into the United States. Additional Administration actions, such as a reported forthcoming proposed rule by the Department of Justice (DOJ) on deportability, could further increase fear and confusion in immigrant communities. (See box, below, for more details on the potential DOJ rule.)
Proposed DHS Rule Significantly Expands Definition of “Public Charge”
Under federal law back to the late 1800s, immigration officials can turn down people seeking to enter the United States and/or become lawful permanent residents (also known as green card holders) if officials determine that they are, or are likely to become, a “public charge.” Longstanding federal policy considers someone a public charge if they receive more than half of their income from cash assistance programs, such as Temporary Assistance for Needy Families (TANF) and Supplemental Security Income (SSI), or receive long-term care through Medicaid.
The proposed rule significantly expands the definition of public charge in two major ways. First, it broadens the list of public benefit programs considered in a public charge determination to also include health coverage through Medicaid, food assistance through SNAP (food stamps), housing assistance, and Medicare Part D low-income subsidies to help beneficiaries afford prescription drugs. Second, instead of looking at whether more than half of a person’s income comes (or would likely come in the future) from cash assistance tied to need, as they do now, immigration authorities would consider whether the individual received, or is likely to receive, modest amounts of any of these benefits — even if the benefits reflect only a small share of an immigrant’s total income.
The proposed rule creates new criteria and standards for immigration officials to use when evaluating whether an individual is likely to become a public charge. Particularly concerning is a new income criterion that would count as a negative factor in the public charge determination. Under this “income test,” having family income below 125 percent of the poverty line — currently about $31,375 for a family of four — would count against an individual in the public charge determination. Many low-wage workers have earnings below this level and could be deemed “likely to become a public charge” under the proposed rule, even if they receive no benefits. And many seeking admission to the United States from a poorer country would be unable to have current earnings (in their home country) above this level.
Department of Justice Will Likely Seek to Conform to DHS’ Public Charge Definition
A recent media report indicates that the Department of Justice (DOJ) plans to propose a rule related to grounds for deporting individuals determined to have become a public charge.a That rule would likely conform the public charge definition for deportability purposes to the definition used in the Department of Homeland Security’s rule on inadmissibility discussed in this paper. The details of the DOJ proposed rule are not known but the scope of the changes would be limited by certain statutory requirements. To be deportable as a public charge, a person would need to have received the relevant benefits within the first five years after entry based on circumstances that predated their entry. Most immigrants are not eligible for the major benefits during their first five years in the country, And, immigrants could still show that they received the benefits based on conditions that arose after entry, e.g., they lost their job, had an accident, became pregnant, or lost their housing. Still, some immigrants and their families would be affected and the publication of such a rule is certain to generate more confusion and fear in immigrant communities, and lead families to forgo assistance that they need and are eligible for.
a Yeganeh Torbati, “Exclusive: Trump administration proposal would make it easier to deport immigrants who use public benefits,” Reuters, May 3, 2019, https://www.reuters.com/article/us-usa-immigration-benefits-exclusive/exclusive-trump-administration-proposal-would-make-it-easier-to-deport-immigrants-who-use-public-benefits-idUSKCN1S91UR.
More Than Half of U.S.-Born Citizens Likely to Participate in Programs Included in Proposed Definition During Their Lifetimes
The breadth of the rule’s expansive definition of public charge is clear when one considers the share of U.S.-born citizens who would be considered a public charge if the proposed definition were applied to them. The rule, of course, applies only to individuals seeking entry or adjustment of status, but it is instructive to consider the share of U.S.-born citizens whom the proposed rule would characterize as a public charge when considering the reasonableness of the standard.
Looking at the U.S.-born citizen population in 2017 and considering benefit receipt over the 1997-2017 period, some 43 to 52 percent received one of the benefits in the public charge definition. If data allowed us to look at U.S.-born citizens over their full lifetimes, benefit receipt would exceed 50 percent of the population.
The benefits included in the proposed definition serve a far broader group of low- and moderate-income families than those served by cash assistance and institutional care programs (those considered under the current definition). Looking at just one year of program participation shows that 28 percent — nearly 3 in 10 — of U.S.-born citizens receive one of the main benefits included in the proposed definition. By contrast, about 5 percent of U.S.-born citizens meet the current benefit-related criteria in the public charge determination.
Under the rule, immigration authorities are tasked with predicting whether someone will ever, over the course of their lifetimes, receive one of the benefits included in the public charge definition. To understand the breadth of this definition, we’d ideally want to look at U.S.-born citizens over their lifetimes and measure the share who receive one of the named benefits. Unfortunately, data limitations preclude that. But we can look at the share of U.S.-born citizens who receive these benefits both in a single year using Census data and over a 19-year period using the Panel Study of Income Dynamics (PSID), a longitudinal data set.
Approximately 43 to 52 percent of U.S.-born individuals present in the PSID survey in 2017 participated in either SNAP, Medicaid, TANF, SSI, or housing assistance over the 1997-2017 period. If we were able to capture more years and a higher share of people’s childhoods with data that are corrected for underreporting, we estimate that more than half of the U.S.-born population participate in SNAP, Medicaid, TANF, SSI, or housing assistance over their lifetimes.
Additional PSID analyses make this clear. Benefit receipt is higher during childhood than during adulthood, so capturing childhood years increases the share receiving benefits at some point. We find that 59 percent of children born during 1999-2017 (in non-immigrant PSID households) received one of the five benefits over the period. This makes clear that a majority of U.S.-born citizens will receive one of these benefits at some point over the course of their lives.
Many Workers Participate in Programs Included in Proposed Definition
Another way to examine the breadth of the rule’s definition of public charge is to apply it to all individuals working in the United States, regardless of citizenship status. If all U.S. workers were subjected to a public charge determination, a significant share would be considered a public charge under the proposed rule. Looking at just one year of program participation shows that 16 percent of U.S. workers receive one of the main benefits included in the proposed definition. By contrast, just 1 percent of U.S. workers meet the current benefit-related criteria in the public charge determination.
The reality of the current U.S. labor market is that many workers combine earnings with government assistance to make ends meet. Table 1 shows that a significant share of workers in all major industry groups would be defined as a public charge if the definition were applied to them, despite the important role that these workers play in these industries and in the economy.
|Many U.S. Workers Use Benefits Considered Under Proposed Public Charge Definition|
|Percent that use benefits under current definition*||Percent that use benefits under proposed definition**|
|Leisure and hospitality||1%||28%|
|Other services (repair and maintenance, private household workers, etc.)||1%||20%|
|Wholesale and retail trade||1%||20%|
|Agriculture, forestry, fishing and hunting||4%||19%|
|Transportation and utilities||1%||15%|
|Educational and health services||1%||15%|
|Professional and business||1%||14%|
|Information (publishing, broadcasting, telecommunications, etc.)||0%||10%|
Income Test Likely to Keep Many out of United States
The proposed rule creates a variety of new criteria and standards for immigration officials to use when evaluating whether an individual is likely to become a public charge. Particularly concerning is a new income criterion that would be considered as a negative factor in the public charge determination. Under this “income test,” having family income below 125 percent of the poverty line — about $31,375 for a family of four, which is more than twice what full-time work at the federal minimum wage pays in the United States — would count against an individual in the public charge determination.
Many low-wage U.S. workers have earnings below this level and could be deemed “likely to become a public charge” under the proposed rule, even if they receive no benefits. This test could prevent individuals with low or modest incomes from being granted status adjustment or lawful entry/re-entry to the United States.
That standard could also be out of reach for many people seeking to enter from a country where incomes in general are much lower than in the United States. The 125 percent test would disproportionately affect immigrants from poor countries (especially those who are not in families already living and working in the United States) and have a racially disparate impact on who is allowed into the United States. The World Bank provides an online data tool that allows users to estimate the percent of the population from various countries that’s below different poverty thresholds. To approximate 125 percent of the U.S. poverty line, one can use a $20 per-person, per-day poverty line in the World Bank online tool. According to the tool, 13 percent of the U.S. population is below the $20 per-person, per-day poverty line. (Similarly, 15 percent of the U.S. population is below 125 percent of the U.S. poverty line.)
If we apply that $20 a day threshold to the rest of the world, many individuals would fall below that threshold, including:
- 80.8 percent of the world population;
- 99.2 percent of the population of South Asia;
- 98.5 percent of the population of Sub-Saharan Africa; and
- 79.0 percent of the population of Latin America and the Caribbean.
Of course, the figures are much different in wealthy countries. In countries the World Bank defines as “high income,” 14.4 percent of people in those countries would fall below the 125 percent threshold.
The map below color codes countries based on the percent of their populations with income below the $20 per-person, per-day poverty line. (These calculations use the March 2019 update of 2013 data because they are available for more countries. Currently, the World Bank tool includes 2015 data for a more limited number of countries.)
These data show that the application of the 125 percent threshold to potential immigrants living abroad could have a dramatic effect on who would be allowed to come in to the United States lawfully. To be sure, many immigrants seeking to rejoin family in the United States will be joining families that also have income that can count toward this 125 percent of poverty test. The test will remain hard, however, for those joining family of modest means, because the arriving individual will have income on the wage scale of their home country.
A country’s low wage rates are not determinative of a potential immigrant’s core traits and skills or their ability to develop skills and succeed in the United States. Indeed, throughout our history, poor individuals have come to the United States and have achieved significant upward mobility for themselves and their children, helping to grow the nation and its middle class, its industries, and its innovation sector.
Broadened Public Charge Definition Could Lead to Racial Bias in Immigration Decisions
Broadening the definition of public charge opens the door to increased discrimination in the adjudication of adjustment and lawful entry applications based on race, ethnicity, and country of origin.
Under current policy, immigration officials are trying to answer a very narrow question: Is someone likely to become primarily dependent on a narrow range of benefits that only a small share of Americans receive? Individuals who are determined likely to become a public charge under the current policy can generally overcome the finding with a legally enforceable affidavit of support from a sponsor.
In contrast, under the proposed rule, immigration officials would be asked to predict whether an individual is likely to receive at some point in the future any of a much broader range of benefits that a significantly larger share of Americans receive. (It also appears likely that fewer individuals would be able to overcome a public charge determination through an affidavit of support, raising the stakes of the determination for individuals seeking entry or permission to remain in the United States.)
Given the more complex prediction that immigration officials would have to make, their discretion, which could be influenced by implicit (or explicit) racial/ethnic bias, would likely affect the outcome for more people. Some immigration officials, faced with the difficult task of predicting future benefit use, may assume that immigrants from certain countries or racial/ethnic groups will be more likely to receive benefits than similarly situated white immigrants and use that assumption to deny adjustment or entry to people of color at higher rates than their circumstances justify.
Higher rates of poverty and benefit receipt in the United States among people of color are due to (among other factors) a history of slavery and discrimination leading to a large and persistent racial wealth gap; unequal education, job, and housing opportunities; and for some recent immigrants, lower educational opportunities in their home countries. These opportunities are more plentiful in the United States, resulting in higher educational attainment among the children of lower-skilled immigrants.
The rule could have a discriminatory impact and result in applicants for status adjustment or lawful entry being denied at higher rates based on their race or ethnicity, all else being equal.
Immigrants’ Children Tend to Be Highly Upwardly Mobile
By using such a broad public charge definition, the proposed rule appears to presume that both immigrants themselves and by extension and implication, U.S.-born citizens who receive government assistance contribute little to the economy which, as the data above indicate, is untrue. In the case of immigrants, the proposed rule is saying that the nation would be better off without their offspring, as well. Yet when immigrants’ children are considered, the economic case for this rule is even harder to support.
Studies have long found that the children of immigrants tend to attain more education, have higher earnings, and work in higher-paying occupations than their parents. Economist David Card observed in 2005 that “Even children of the least-educated immigrant origin groups have closed most of the education gap with the children of natives.” The National Academy of Sciences’ 2015 immigration study similarly concluded that second-generation members of most contemporary immigrant groups (that is, children of foreign-born parents) meet or exceed the schooling level of the general population of later generations of native-born Americans. Even for immigrants without a high school education, the overwhelming majority of their children graduate from high school. According to the National Academy of Sciences 2017 report, 36 percent of new immigrants lacked a high school education in 1994-1996; two decades later, only 8 percent of second-generation children (i.e., children of foreign-born parents) lacked a high-school education.
The United States remains a country with a dynamic economy and opportunity for upward mobility, educational attainment, creativity, and entrepreneurship. Given the inevitable inaccuracies in immigration officials’ predictive capabilities, removing individuals or keeping them out of the country based on an extremely broad definition of “public charge” would cost the United States many needed workers, including those who care for children and seniors and build homes as well as those who start businesses, go to college, and have children who become teachers, inventors, and business leaders. Forfeiting this talent would weaken the entire nation and our local communities.
To calculate the percent of U.S.-born citizens that participate in a single year in programs included in the Administration’s proposed rule, we used the Current Population Survey. Our calculations include SNAP, TANF, SSI, Medicaid, housing assistance, and state General Assistance programs. We corrected for underreporting of SNAP, TANF, and SSI receipt in the Census survey using the Department of Health and Human Services/Urban Institute Transfer Income Model (TRIM). The figures are for 2016, the latest year for which these corrections are available. Our estimates understate the share of U.S.-born citizens who participate in a single year in programs included in the definition because we do not correct for the underreporting of Medicaid or account for low-income subsidies in the Medicare Part D program, which are also included in the rule.
The PSID, a longitudinal dataset, shows that 24 percent of the U.S.-born population interviewed in 2017 recently participated in at least one of the five programs. This estimate is lower than CBPP’s single-year figures presented above (28 percent) because the PSID data are not corrected for survey respondents’ tendency to underreport receipt of government benefits.
The PSID-based figures undoubtedly would have been higher if we could have corrected for the underreporting of benefit receipt in the PSID. The CPS/TRIM-based estimate of the share of individuals who participated in one of the benefit programs in 2016 is about 1.2 times as large as the PSID-based estimate. We use this adjustment factor to estimate that as many as roughly 52 percent of U.S.-born citizens participated in SNAP, Medicaid, TANF, SSI, or housing assistance in at least one year over the 1997-2017 period.
But underreporting is only one reason that the 43 percent estimate described above is lower than the share of U.S.-born citizens who receive benefits in at least one year over this period and well below the figure for the share of U.S.-born citizens who receive one of these benefits at some point over their lifetimes.
In looking at benefit receipt over the 1997-2017 period, the PSID only provides data on benefit receipt for most programs every other year. The PSID dataset thus lacks any measure of participation in alternate years for some programs such as Medicaid.
More importantly, these data do not measure benefit receipt over individuals’ entire lives. Using PSID data for 1997-2017 is an important improvement over using a single year of data to analyze the share of U.S.-born citizens who receive one of the benefits included in the proposed rule’s public charge definition, but it still captures only a portion of most respondents’ lifetimes and significantly underestimates the share of U.S.-born citizens who receive a benefit at some point during their lives. If we were able to capture more years and a higher share of people’s childhoods with data that are corrected for underreporting, as described above, we estimate that more than half of the U.S.-born population participate in SNAP, Medicaid, TANF, SSI, or housing assistance over their lifetimes.
To be sure, not all citizens who participate in the programs listed in the proposed rule would technically meet the proposed definition of a public charge. The rule directs immigration officials to disregard program participation if the benefit amounts or durations fall below thresholds established in the rule. Due to data limitations, we cannot appropriately model all of those provisions. However, we think that those provisions would be extremely difficult for officials to apply when making a prospective determination, so any projected future receipt would likely bar an applicant from status adjustment or entry.
Finally, when the Census Bureau asks about health coverage in the Current Population Survey, it asks about Medicaid and the Children’s Health Insurance Program (CHIP) together, so the data on Medicaid also include CHIP recipients. CHIP is not included among the benefits in the proposed rule, however; in many states Medicaid and CHIP programs are so closely aligned that parents wouldn’t be able to tell whether their children were Medicaid or CHIP recipients, and it is unclear whether immigration officials projecting future benefit receipt would be able to distinguish either.
 The full text of the administration’s proposed rule can be found here: https://www.govinfo.gov/content/pkg/FR-2018-10-10/pdf/2018-21106.pdf.
 Lawful permanent residents are not reevaluated for public charge as part of the application process to become a U.S. citizen.
 See, for example, David Card, “Is New Immigration Really So Bad?” National Bureau of Economic Research Working Paper 11547, Revised August 2005, https://www.nber.org/papers/w11547. For a more recent and comprehensive review of the literature, see: National Academy of Sciences, “The Economic and Fiscal Consequences of Immigration,” 2017, https://www.nap.edu/read/23550/chapter/2.
 For more details, see: National Immigration Law Center, “How to Talk About Public Charge with Immigrants and Their Families,” updated January 2019, https://www.nilc.org/issues/economic-support/how-to-talk-about-public-charge-pif/.
 National Immigration Law Center, “Changes to ‘Public Charge’ Instructions in the U.S. State Department’s Manual,” updated August 7, 2018, https://www.nilc.org/wp-content/uploads/2018/02/PIF-FAM-Summary-2018.pdf. For evidence on immigrants forgoing assistance, see Hamutal Bernstein et al., “One in Seven Adults in Immigrant Families Reported Avoiding Public Benefit Programs in 2018,” Urban Institute, May 22, 2019, https://www.urban.org/research/publication/one-seven-adults-immigrant-families-reported-avoiding-public-benefit-programs-2018.
 The rule directs immigration officials to disregard projected program participation if the official believes the benefit amounts or durations would fall below thresholds established in the rule. However, those provisions would be difficult for officials to apply when they are trying to predict whether or not an individual is likely to become a public charge in the future. To apply those provisions, an immigration officer would need to calculate the amount of benefits that an individual immigrant might receive in the future which would require in-depth knowledge about program benefit rules and predictions about the income and characteristics of an immigrant’s future household members. That is so difficult that as a practical matter, immigration officials will likely default to only determining whether there is a likelihood of receiving any amount of benefits. Therefore, any projected future receipt would likely result in a person being deemed “likely to become a public charge.”
 See methodological appendix for further details.
 The current definition is modeled as: Personally receiving more in TANF, SSI, and General Assistance than in earnings, or a member of a family that receives more in TANF, SSI, and General Assistance than earnings. Due to data limitations we did not include participation in institutional care programs.
 The PSID, conducted by the University of Michigan’s Institute of Social Research, began in 1968 and follows about 5,000 families (and the families that branched off from the original survey respondents) annually.
 This is based on a CBPP update of an analysis done by Diana Elliott from the Urban Institute using a PSID dataset created by Sara Kimberlin from the California Budget & Policy Center and Noura Insolera from the University of Michigan’s Institute of Social Research, which runs the PSID. The survey data were collected between 1999 and 2017, but the program participation questions generally ask about participation in the current and previous two calendar years. The PSID does not include data on Medicare Part D Low-Income Subsidies. We also did not include General Assistance in our PSID analysis due to concerns about the quality of the data for that variable. The inclusion of those programs would increase our estimates of the share of U.S.-born citizens who receive benefits included in the proposed rule.
 See methodological appendix for a detailed explanation of how we reached this estimate.
 PovcalNet: the online tool for poverty measurement developed by the Development Research Group of the World Bank: http://iresearch.worldbank.org/PovcalNet/povDuplicateWB.aspx.
 Throughout this PSID analysis, “U.S. born” refers to individuals in the PSID’s main sample, and excludes a later, supplemental sample of immigrants added to the PSID in 1997-1999 and in 2017. The main sample actually includes a small number of immigrants, including some who were present in the United States since 1968 when the PSID began or those who joined existing PSID households in later years. PSID respondents in 2017 were asked about current receipt of Medicaid and housing assistance, receipt of TANF and SSI in the past year, and receipt of SNAP last month, last year, and two years ago.
 Using the Current Population Survey and baseline data from the Health and Human Services/Urban Institute Transfer Income Model version 3 (TRIM3) to correct for the underreporting of TANF, SSI, and SNAP, we find that 28 percent of the U.S.-born population participated in one of the five programs in 2016. The CPS/TRIM figure would be even higher if we were able to correct for the underreporting of Medicaid.
 To calculate the adjustment factor of 1.2 we divide the CPS/TRIM share of U.S.-born citizens participating in 2016 (28.4 percent before rounding) by the comparable point-in-time figure from the PSID (23.6 percent). In calculating the latter figure, we include people who reported receiving SNAP during the last month, last year, or two years ago. If we only count those in the PSID who report receiving SNAP last year (as our TRIM figure does), the PSID point-in-time participation rate for the five programs would be 22.1 percent, the undercount adjustment factor would be 1.3, and our adjusted estimate of the share of U.S.-born citizens ever participating over the 1997-2017 period would be even higher at 56 percent.
 We estimate this upper bound by applying the annual underreporting factor (1.2) to the estimate of benefit receipt over the full period (43.4 percent).