Donate
Independent, objective, nonpartisan research

R 1298TMR

Authors

R 1298TMR

Tagged with:

Publication PDFs

Database

This is the content currently stored in the post and postmeta tables.

View live version

object(Timber\Post)#3711 (44) { ["ImageClass"]=> string(12) "Timber\Image" ["PostClass"]=> string(11) "Timber\Post" ["TermClass"]=> string(11) "Timber\Term" ["object_type"]=> string(4) "post" ["custom"]=> array(5) { ["_wp_attached_file"]=> string(13) "R_1298TMR.pdf" ["wpmf_size"]=> string(6) "216583" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(134260) "iii Foreword Designing an efficient and effective social safety net is one of the most intractable aspects of pubic policy. For example, simple rules for eligibility are easy to enforce but highly inefficient because they tend to provide benefits to some individuals and families not really in need. More complex rules that isolate only those most deserving are difficult to administer and often create expensive bureaucratic nightmares. The 1996 welfare reform legislation included many sweeping changes, including time limits and restricted eligibility for benefits. In the wake of this wave of reform, however, it became clear that some of the most draconian decisions were not politically feasible. Subsequent legislation has already undone some of the 1996 reforms. In this report, Thomas MaCurdy and Margaret O’Brien-Strain examine one of the most controversial elements of the legislation—the denial and subsequent reinstatement of benefits for noncitizen immigrants under the federal Supplemental Security Income (SSI) and Food Stamp programs. The authors argue that both the original decision and its reversal did not meet the most important goals of welfare reform, which were to meet the needs of the poorest families while promoting self-sufficiency and family responsibility. Taking away all SSI and Food Stamps benefits would have plunged two-thirds of immigrant recipients into poverty. However, in simply reversing the reforms and reinstating the benefits, legislators missed a second opportunity to design a program that efficiently targets the benefits to those truly in need. Under the reinstated program, one in four SSI dollars goes to families with incomes substantially above the poverty line. And iv 20 percent of the total benefits paid out to immigrants under SSI goes to families with incomes higher than 60 percent of all California families. The authors presented the work reported here to Congressional staff, federal agency staff, and numerous members of nongovernmental agencies in June 1997. Although the decision to reinstate benefits is now behind us, many other reforms are just beginning to take effect. Thus, there will be numerous opportunities to rethink, perhaps more carefully, welfare program design over the coming years. This work is presented with that thought in mind, and also as a reminder that targeting of the social safety net is every bit as important as the social insurance that the net provides. This is especially true in an environment where government programs are being judged not only for their effectiveness but also for the efficiency of their design. David W. Lyon President and CEO Public Policy Institute of California v Contents Foreword................... iii Figures............. vi Tables.............. ix 1. INTRODUCTION........... 1 2. WHY DID WELFARE REFORM TARGET IMMIGRANTS?.. 3 3. HOW WERE THE REFORMS REVERSED?...... 7 4. HOW IMPORTANT WOULD THE INCOME LOSS HAVE BEEN FOR IMMIGRANTS?............ 11 Assessing the Consequences of Denying Benefits..... 11 Immigrants Losing SSI........... 12 Immigrants Losing Food Stamps........... 15 Summary of Key Findings............. 18 5. WHICH IMMIGRANTS GAIN FROM THE RESTORATION OF BENEFITS?................. 21 Restoring All SSI Benefits............. 22 Restoring SSI Only to the Disabled or Only to the Elderly..... 24 Restoring All Food Stamps.................. 26 Restoring Food Stamps to Children and the Elderly...... 27 Summary of Key Findings................... 28 6. SHOULD WE BE RETHINKING WELFARE REFORM?....... 31 Efficiency of Immigrant SSI Benefits............. 32 Efficiency of Immigrant Food Stamps Benefits....... 32 Were Immigrants a Special Case?............... 33 vi Appendix A. Data and Methodology............... 37 B. Primary Tables and Description........... 43 Bibliography.............. 73 About the Authors............. 75 vii Figures 4.1. Benefit Losses for Families Receiving SSI........... 14 4.2. Change in Income Relative to Poverty Level Assuming Benefit Denial: SIPP Data............... 15 4.3. Income Sources for Immigrant SSI Families: SIPP Data..... 16 4.4. Benefit Losses for Families Receiving Food Stamps..... 17 4.5. Change in Income Relative to Poverty Level Assuming Loss of Food Stamps: SIPP Data.............. 19 5.1. Share of Reinstated SSI Benefits, by Income Quintile: SIPP Data.. 23 5.2. Share of Reinstated SSI Benefits, by Income as a Share of Poverty Level.. 24 5.3a. Share of Reinstated SSI Benefits to Disabled, by Income Quintile. 25 5.3b. Share of Reinstated SSI Benefits to Elderly, by Income Quintile.... 25 5.4. Share of Reinstated Food Stamps Benefits, by Income Quintile... 26 5.5. Share of Reinstated Food Stamps Benefits, by Income as a Share of Poverty Level................ 27 5.6a. Share of Reinstated Food Stamps Benefits to Children, by Family Income Quintile................ 28 5.6b. Share of Reinstated Food Stamps Benefits to Elderly, by Family Income Quartile.................... 28 ix Tables 3.1. Eligibility of California’s Nonqualified Aliens for SSI and Food Stamps, Based on 1996–1998 Legislation............. 9 4.1. Income Losses for Immigrant Families Participating in SSI.... 13 4.2. Income Losses for Immigrant Families Participating in Food Stamps... 16 4.3. Income Sources for Immigrant Families Losing Food Stamps.... 18 5.1. Total Value of Reinstated SSI Benefits to California Immigrant Families..................... 22 5.2. Total Value of Reinstated Food Stamps Benefits to California Immigrant Families................... 24 A.1. Sample Sizes and Characteristics of the SIPP and CPS...... 38 A.2. Unweighted Subsample Sizes for SIPP and CPS Results...... 39 B.1a. Characteristics of Citizen and Immigrant Families and Their Welfare Participation in California: SIPP Data......... 45 B.1b. Characteristics of Citizen and Immigrant Families and Their Welfare Participation in California: CPS Data........... 46 B.2a. Sources of Income and Amounts Received by Immigrant Families Who Are Welfare Participants in California: SIPP Data...... 49 B.2b. Sources of Income and Amounts Received by Immigrant Families Who Are Welfare Participants in California: CPS Data....... 50 B.3a. Income Losses of Immigrant Families Affected by Elimination of Food Stamp and SSI Programs: SIPP Data, Best Case........ 53 B.3b. Income Losses of Immigrant Families Affected by Elimination of Food Stamp and SSI Programs: SIPP Data, Worst Case....... 54 x B.3c. Income Losses of Immigrant Families Affected by Elimination of Food Stamp and SSI Programs: CPS Data............. 55 B.4a. Income Status of Immigrant Families After Elimination of Food Stamp and SSI Programs: SIPP Data, Best Case....... 57 B.4b. Income Status of Immigrant Families After Elimination of Food Stamp and SSI Programs: SIPP Data, Worst Case........ 58 B.4c. Income Status of Immigrant Families After Elimination of Food Stamp and SSI Programs: CPS Data............ 59 B.5a. Shares of Reinstated Welfare Benefits Going to Various Categories of Immigrant Families in California: SIPP Data......... 62 B.5b. Shares of Reinstated Welfare Benefits Going to Various Categories of Immigrant Families in California: CPS Data......... 63 B.6a. Shares of Reinstated SSI Benefits Going to Various Categories of Immigrant Families in California: SIPP Data......... 65 B.6b. Shares of Reinstated SSI Benefits Going to Various Categories of Immigrant Families in California: CPS Data........... 66 B.7a. Shares of Reinstated Food Stamps Benefits Going to Various Categories of Immigrant Families in California: SIPP Data......... 69 B.7b. Shares of Reinstated Food Stamps Benefits Going to Various Categories of Immigrant Families in California: CPS Data......... 70 1 1. Introduction In recent debates over welfare reform, few issues have been as controversial as the denial of benefits to immigrants. The 1996 federal welfare reform law ruled that nearly all immigrants would be ineligible for federal welfare payments under the Food Stamp and Supplemental Security Income (SSI) programs. However, Congress reversed position a year later and restored payments to many immigrants. More payments were restored in 1998. Now, after two years of policy decisions and reversals, there is still little reliable evidence to assess whether the reforms for immigrants were appropriate or how the cuts would have affected the immigrant families that relied on public assistance. In the absence of such evidence, advocates on both sides of this debate relied heavily on anecdotes to support their positions. The policymakers who chose to deny immigrant benefits in 1996 had only limited information on how public assistance contributed to immigrants’ overall resources and how many immigrant families would lose their principal means of support. Similarly, when policymakers restored benefits, they did not know which immigrant families desperately needed the income support and which had incomes that placed them above poverty without welfare benefits. This study revisits the issue of welfare reforms affecting immigrants, focusing on California. Although the changes were nationwide, California was, in a very real sense, at the center of this policy debate. California has the most to gain from the restoration of federal benefits for two reasons. First, 40 percent of the immigrants affected by the 1996 reforms live in California. Second, when the federal government barred immigrant 2 eligibility, the California legislature aggressively sought to replace virtually all of the immigrant benefits using the state’s own funds. The 1997 restoration of federal SSI benefits mitigated the need for some of this legislation, although California restored Food Stamps to children and the elderly. The 1998 restoration of federal Food Stamps closely mirrors this strategy put in place by California a year earlier. Although the issue of immigrant eligibility seems settled for the moment, there were and continue to be reform choices beyond the extreme of the 1996 cutoff and the simplistic restorations. Moreover, the real effects of welfare reform have scarcely been felt yet. Over time, there may be further calls to reverse or at least moderate other substantial changes made in the 1996 federal welfare law and the 1997 California enactment of this reform. A careful reexamination of changes in immigrant families’ conditions may provide insight into alternative program designs that might better meet the goals of the original reform. In particular, this study asks, What were the economic circumstances for immigrant families in California, given the ban and then the restoration of public assistance? We focus on family resources, which steps outside the existing program rules, to take what is in our judgment a broader and more realistic look at the living conditions for immigrant welfare participants. We organize the analysis around the following five questions: Chapter 2 Why were immigrants targets of welfare reform? Chapter 3 How were the 1996 reforms reversed? Chapter 4 How much income would immigrant families have lost had the welfare reforms been adopted as they were originally conceived? Chapter 5 Which immigrant families gained from the restoration of their welfare benefits? Chapter 6 Does the case of immigrant benefits suggest options for reforming of the welfare system? 3 2. Why Did Welfare Reform Target Immigrants? One of the most dramatic elements of the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA) was the bar on welfare benefits for most legal immigrants. This restriction eliminated the eligibility of noncitizen immigrants for SSI and Food Stamps unless they demonstrated 40 quarters of work in the United States or met exemptions for military personnel and recently arrived refugees. In addition, all immigrants (except refugees) arriving after August 22, 1996, were barred from receiving benefits for these and other federally funded programs for at least five years after arrival. The primary motivation for the immigrant cuts was budgetary. For the Republican - controlled Congress, it was important that welfare reform save money. Yet many of the massive changes to the welfare system provided for no near-term savings. This made the restriction of benefits to immigrants a key budget-cutter. In the final version of the 1996 welfare reform law, the immigrant restrictions accounted for almost half of the total federal savings in the welfare reform law. There were, nevertheless, substantive reasons to target immigrant benefits in welfare reform. Immigrants arrived in much higher numbers in the 1980s, and these recent immigrants were relatively high users of welfare. This triggered debates about the net fiscal costs immigrants imposed on the United States and on high immigrant areas in 4 particular. Finally, immigrants and their sponsors appeared to be abusing welfare programs. We consider each of these reasons in turn. Between 1980 and 1994, 13 million immigrants entered the United States, doubling the total foreign-born population relative to 1970. By 1994, one in eleven U.S. residents was foreign-born, and 70 percent of these immigrants were noncitizens. Among noncitizen immigrants, 43 percent had not completed high school, and the median income for this group was only $10,930. As a result, 29 percent lived in poverty and 5 percent received public assistance. This put their rate of welfare receipt 70 percent higher than that of U.S. natives and 400 percent higher than that of naturalized citizens (Bureau of the Census, 1995). This high rate of immigrant welfare participation was a relatively new phenomenon. Before 1980, immigrant participation in welfare was similar to that of natives. After controlling for household characteristics, many economic studies of immigrant participation in welfare found little difference overall between immigrant and native households before 1980. However, studies based on the 1980 Census and later data show very different results. By 1990, recent immigrants were more likely than both natives and earlier cohorts of immigrants to participate in welfare (see Borjas, 1994). High immigration rates and high usage of welfare by immigrants triggered a series of controversial studies of the net fiscal costs of immigration. Most of these studies contended that immigrants received more in benefits, mainly education and public assistance, than they paid in taxes. The most recent research in this area suggests that immigrants are probably net contributors to the U.S. economy overall. However, for high immigration areas, immigrants do appear to impose a heavy fiscal burden. The National Research Council report on immigration concluded that in California, immigrants cost each native household $1,178 annually (Smith and Edmonston, 1997). Finally, immigrants were targeted in welfare reform because of perceived abuses of welfare programs, especially SSI. SSI is designed to provide income support for individuals who are unable to work because of age or disability and who lack sufficient work experience to receive Social Security. Most immigrants (62 percent in 1994) are sponsored by family members, who sign affidavits of support vouching that the immigrants will not become public charges; an immigrant who becomes a public charge can be deported in the first five years after entry. Based on affidavits of support, the SSI 5 program counts or “deems” sponsor income for three to five years after an immigrant’s entry into the United States. Although the deeming provisions are intended to discourage immigrant use of welfare, the provisions merely postpone entrance onto the rolls. Forty- six percent of noncitizen recipients applied for SSI within four years of entry, in other words, as soon as the deeming period expired (Ross, 1996). Anecdotal reports have suggested that some immigrant communities have come to view SSI for the elderly as a normal benefit of U.S. residency (Matloff, 1994). These behaviors are contrary to the policy aims of the program, but they are legal under the current rules. There are illegal activities as well. For example, there are a number of documented cases of fraudulent applications from noncitizens using translators to claim nonexistent disabilities (Ross, 1996). Overall, immigrants seemed to be straining the resources of the SSI program. During the period from 1986–1994, noncitizen cases doubled as a proportion of all SSI cases, and noncitizens now constitute one-third of the aged SSI population. By 1996, even nonconservatives, such as analysts with The Urban Institute, testified that “high and rising immigrant use of Supplemental Security Income . . . represent[s] a significant public policy issue that calls for legislative attention” (Fix et al., 1996). Together, these issues motivated a set of “statements of national policy concerning welfare and immigration” as part of the welfare reform legislation. Simply put, these statements of Congress reiterated the policy of self-sufficiency for immigrants, with the intent that immigrants not rely on public resources, “but rather rely on their own capabilities and the resources of their families” and their sponsors to meet their needs. This was at least the stated goal of the welfare reform provisions. 7 3. How Were the Reforms Reversed? When the welfare law passed in 1996, the immigrant restrictions were viewed as the most draconian elements of new legislation for two reasons. First, these restrictions went into effect earlier than other aspects of the reform. Immigrants were to lose benefits within one year, rather than within two to five years—the time limits for recipients of Aid to Families with Dependent Children (AFDC). Second, SSI benefits had been a primary means of support for elderly and disabled people who were not employable. Opposition to these measures came from several directions. President Clinton opposed the immigrant cuts even as he signed the welfare reform bill and made restoring these benefits part of his 1996 reelection platform. He particularly focused on restoring benefits to the disabled. In March 1997, immigrant rights groups in California and New York filed class-action lawsuits to block implementation of the immigrant provisions. By April, Florida filed its own lawsuit. Florida feared that its own taxpayers would end up picking up the costs of supporting these immigrants, just as California counties feared that immigrants would move from federally funded SSI benefits onto the locally funded General Assistance rolls. Finally, even well-respected newspapers such as the Wall Street Journal reported alleged cases of suicide among immigrants in California and elsewhere who faced losing benefits As a result of these pressures, Congress reversed position on SSI benefits as part of the Balanced Budget Act of 1997. The 1997 law restored SSI benefits to those immigrants receiving SSI as of August 22, 1996. It also allowed immigrants residing in the United 8 States on August 22, 1996, to be eligible for SSI if they become disabled in the future. New arrivals will not be eligible for SSI, nor will earlier immigrants who seek SSI in the future based on age. The act also extends refugee eligibility for SSI (but not other programs) from five years after arrival to seven years after arrival. The 1997 law made no changes regarding eligibility for Food Stamps. Later the same year, California chose to partially restore Food Stamps under a state- only program. Along with seven other states, California administered this program by purchasing federal Food Stamps coupons. 1 Under AB 1576, signed into law on August 18, 1997, California provided Food Stamps to legal immigrants age 65 or older and children under age 18 who lost federal eligibility based on PRWORA. This applied only to immigrants who were both legally in the United States and receiving Food Stamps before August 22, 1996. The benefits for these recipients will be the same as those provided under the federal Food Stamp program, with the same regulations in effect as well. California proved to be a policy leader, as Congress voted in 1998 to partially restore Food Stamps by reinstating eligibility for legal immigrant children and elderly persons who were legal residents at the time of the 1996 legislation. (Unlike the SSI restoration and the California legislation, the federal Food Stamps restoration does not require that benefits go only to previous recipients.) President Clinton signed these restorations into law on June 23, as part of the Agriculture Research, Extension and Education Reform Act of 1998. Table 3.1 outlines the eligibility changes over the past two years for immigrant recipients of Food Stamps and SSI in California. _________ 1Other states purchasing federal Food Stamps included Florida, Maryland, Nebraska, New Jersey, New York, Rhode Island, and Washington. Washington is the only state fully replacing Food Stamps for all legal immigrants who became ineligible. Massachusetts, Minnesota, and Texas have developed state food programs, with benefits substantially lower than the federal levels. 9 Table 3.1 Eligibility of California’s Nonqualified Aliens for SSI and Food Stamps, Based on 1996–1998 Legislation Eligibility by Category1996 Welfare Reform Law1997 Balanced Budget Act1997 California AB 15761998 Agriculture Research Act SSI Eligibility Legal immigrants resident as of 8/22/96 SSI recipients as of 8/22/967333 Immigrants who become disabled7333 Legal immigrants arriving after 8/22/967777 Food Stamps Eligibility Legal immigrants resident as of 8/22/96 Elderly or those who become elderly Food Stamps recipients as of 8/22/967733 Nonrecipients as of 8/22/967773 Children under age 18 Food Stamps recipients as of 8/22/967733 Nonrecipients as of 8/22/967773 Adults age 18–647777 Legal immigrants arriving after 8/22/967777 NOTE: 3 indicates made eligible; 7 made ineligible; 3 or 7 eligibility unchanged from prior law. 11 4. How Important Would the Income Loss Have Been for Immigrants? The specter of aged and disabled immigrants ending up on the streets when they lost their sole source of income provided much of the impetus for rescinding the SSI cuts. Although losing SSI clearly would have created some desperate cases, it is unclear from the anecdotal evidence whether such cases were representative of the entire immigrant SSI population. In the case of Food Stamps cuts, the anecdotes were far less dramatic, but this does not necessarily mean that cutting Food Stamps was any better or worse policy. In this chapter, we evaluate exactly how severe the SSI and Food Stamps cuts would have been for California immigrant families. We focus on two questions: What share of income would affected families have lost? How many more California families would have been in poverty? Assessing the Consequences of Denying Benefits To address these questions, we calculated the income losses that immigrant families faced under the 1996 welfare reform law, also known as PRWORA. Although usually nuclear families, some of the immigrant families are extended families, in that grandparents in the same household are counted as part of the family of their children and grandchildren. More important, these families need not match the case unit as defined by Food Stamps or SSI. (SSI cases usually refer to an individual or couple, whereas Food 12 Stamps considers all members of a household.) Appendix A provides further information on our use of families as the unit of observation. Our data are drawn from California observations in two national household datasets: the 1993–1994 Surveys of Income and Program Participation (SIPP) and the March 1993–1995 Current Population Surveys (CPS). The results from the two datasets are similar, except for a few key differences: The CPS shows less overall income for families, less welfare participation, and higher benefits among participants. A comparison of the advantages and disadvantages of each dataset, with definitions and methodology, is given in Appendix A. Detailed results from the SIPP and CPS data are included as Appendix B. We calculate benefit losses for two groups of California families: families of immigrants who received SSI (with or without Food Stamps) and families of immigrants who received Food Stamps but not SSI. The CPS and SIPP data can be considered as worst case and best case: The worst case presumes a complete loss of all family benefits if any family member is ineligible; the best case takes away benefits only from noncitizen members of the family for the months that they are in the family. The best case, using SIPP data, more accurately reflects the program rules. In addition, the SIPP data offer details not available in the CPS data. For these reasons, we focus more on the SIPP results, although all results are included in Appendix B. Immigrants Losing SSI In 1993–1995, one in five California families included at least one noncitizen immigrant. Immigrant families were twice as likely to live in poverty as citizen families. They were also more than twice as likely to participate in a major welfare program: SSI, Food Stamp, AFDC, or Medicaid. Out of all immigrant families in California, one in nine received benefits through the SSI program. However, not all immigrant SSI families would have lost benefits under PRWORA. Twenty-three percent of these immigrant families received SSI through either a citizen or an immigrant who met the exemptions allowed in the welfare law, particularly the exemption for refugees. We exclude these families when we calculate income losses. Table 4.1 shows the average losses immigrant families could have expected under the PRWORA as originally passed. The estimates for total annual income losses range from just over $5,000 to nearly $6,800. About $50 of this is actually lost Food Stamps 13 Table 4.1 Income Losses for Immigrant Families Participating in SSI Projected Loss of Benefit Income SIPP CPS Average annual loss of Food Stamps benefits $492 $474 AFDC ($290) $0 SSI $4,878 $6,323 Total benefits $5,080 $6,797 Annual income loss a 25th percentile $1,850 $3,174 50th percentile $4,612 $6,055 75th percentile $7,595 $8,002 aReported at three points in the income distribution, order- ing the losses from least to greatest. The xth percentile is the value that x percent of the distribution is below. The 50th per- centile is median income loss; it is the value of the loss such that half the families lose more than this amount and half lose less. benefits.1 Using the SIPP data, which allow more careful month-by-month calculations, we can also determine how much AFDC or Temporary Assistance for Needy Families (TANF) a family could get by switching a child from the SSI program into AFDC. Although this counteracts some of the lost income, it represents a gain of only $290 on average. The lower half of Table 4.1 gives more information about the distribution of losses. As we noted above, these losses can be properly understood only in the context of total family income. We divide families into four categories based on their projected income loss as a share of annual family income: (1) families losing less than 10 percent of annual income, (2) families losing 10 to 25 percent of annual income, (3) families losing 25 to 50 percent, and (4) families losing more than 50 percent of total family income. For both datasets, Figure 4.1 presents the percentage of families that fall into each of these four categories. Figure 4.1 illustrates two key facts in assessing the immigrant SSI policy. If SSI benefits had been eliminated as originally proposed, at least 20 percent and as many as 40 _________ 1In California, an individual cannot receive both Food Stamps and SSI, but as many as one in three immigrant SSI families also include another family member receiving Food Stamps. 14 Loss as share of annual income 10% or less Over 50 to 50% 10 to 25% 0 5 10 15 20 25 35 30 40 45 Percent of families SIPP CPS Figure 4.1—Benefit Losses for Families Receiving SSI percent of affected families would have lost more than half of their annual family income. This is consistent with the anecdotal evidence presented in the 1997 policy debate. Yet almost as many families would have lost less than 10 percent of their annual income. This side of the story did not come out in the debate. Of course, for a family in poverty, 10 percent of annual income could be crucial. In Figure 4.2, we show two income groups before and after the PRWORA benefit cuts. The two left hand bars show the change in the share of immigrant SSI families with income below the poverty line. The right hand bars show the change in the share of families with income above 185 percent of the poverty line (a common cutoff point for welfare benefits). For a family of four, 185 percent of the poverty level equals about $30,000. As in the previous figure, we see two distinct groups of families. On the left, we see that the cuts for SSI families would have increased the poverty rate for such families by 35 percent. With benefits, only 13 percent of immigrant SSI families live in poverty; without, this number rises to 48 percent. (For comparison, AFDC and Food Stamps combined in California paid only 72 percent of the poverty level.) On the other hand, 15 Income below poverty Income above 185% of poverty 0 5 10 15 20 25 30 35 40 45 50 Percent of SSI families Before (with SSI and Food Stamps) After (without SSI and Food Stamps) Figure 4.2—Change in Income Relative to Poverty Level Assuming Benefit Denial: SIPP Data with or without benefits, more than 30 percent of immigrant SSI families would not only be above poverty, but above 185 percent of poverty. How can 30 percent of SSI families have income above 185 percent of poverty, even without welfare benefits? Actually, according to the SIPP data, SSI benefits on average represented less than one-third of the annual income for immigrant SSI families. Figure 4.3 shows the breakdown of annual income for SSI families in SIPP. (CPS provides much less detail than SIPP on income sources over time.) Since 63 percent of recipients live in families with at least one wage earner—where earnings represent 79 percent of the family income for the median family—it is perhaps not surprising that over 40 percent of total income in SSI families came from earnings. In fact, among affected SSI families with earnings, one-fourth had earned income over $48,000 per year. These are the families in the “above 185 percent of poverty” category. Immigrants Losing Food Stamps Federal funding for Food Stamps benefits for immigrants ended in August 1997, although California restored partial funding. Cutting all Food Stamps benefits would 16    Earnings SSI 31% 43% 18% 6% 2DC Food Stamps Other including Social Security Figure 4.3—Income Sources for Immigrant SSI Families: SIPP Data have affected nearly one in five immigrant families in California. How much income would families have lost? The average loss of benefits for families participating in Food Stamps would fall between $1,300 and $1,700, as shown in Table 4.2. As with the SSI calculations, the differences between the SSI and the CPS versions are determined largely by whether benefits are recalculated monthly (SIPP) or simply eliminated for the entire year. In parallel with Table 4.1, the bottom half of Table 4.2 shows the distribution of benefits. Many families would lose less than $500 annually, although one in four would lose $1,800 or more (as high as $2,500 or more according to CPS). Table 4.2 Income Losses for Immigrant Families Participating in Food Stamps Projected Loss of Benefit Income SIPP CPS Average annual loss of Food Stamps benefits $1,295 $1,711 AFDC $0 $0 SSI $0 $0 Total benefits $1,295 $1,711 Annual income loss 25th percentile $496 $652 50th percentile $1,083 $1,396 75th percentile $1,876 $2,545 17 To put these losses in perspective, we can once again look at the losses as a share of total family income, as in Figure 4.4. The vast majority of families would lose less than 25 percent of their income, and most would lose less than 10 percent. The SIPP data show almost no families losing more than 25 percent of their income, although the CPS data show about 20 percent of families in this category. As noted above, the CPS calculation removes all annual benefits rather than doing a month-by-month determination. In addition, the CPS generally reports less overall income but more welfare income, so the potential losses represent a larger share. Unlike SSI, Food Stamps is not intended to be a sole source of income for recipients. In fact, almost all of the potentially affected Food Stamps families receive other public assistance. Table 4.3 shows the resources of these families in the absence of Food Stamps for immigrants. This table shows only the SIPP data, because SIPP provides much more information on income from welfare and other sources. Even after the cuts, 86 percent of these families would receive benefits through the AFDC/TANF program, through General Assistance or through Food Stamps to citizen members. A similar share of families have other unearned income, typically unemployment or Social Security benefits. Loss as share of annual income 10% or less Over 50 to 50% 10 to 25% 0 10 20 30 40 50 60 70 Percent of families SIPP CPS Figure 4.4—Benefit Losses for Families Receiving Food Stamps 18 Table 4.3 Income Sources for Immigrant Families Losing Food Stamps Income SourcePercent of Families Welfare benefits from major program86 AFDC/TANF 64 Food Stamps 58 SSI (a) General Assistance benefits 6 Other unearned income 87 Earnings82 Annual earnings 25th percentile $5,586 50th percentile $11,536 75th percentile $19,842 aCalculations based on immigrant families receiving Food Stamps but not SSI. In addition, 82 percent of the affected Food Stamps families have earnings. These are still poor families, however. The median income for families with earnings is only $11, 536. If we look at the change in poverty from cutting Food Stamps, we see a very different picture than we saw with SSI. Figure 4.5 shows the same categories as shown in Figure 4.2 for SSI cases. Here the poverty rate does not rise as dramatically, with the loss of Food Stamps pushing 8.2 percent more families into poverty. However, even with Food Stamps, more than half of the immigrant families were in poverty. Just over 10 percent had annual incomes over 185 percent of poverty. Summary of Key Findings Under the 1996 welfare reform law, most immigrants were to lose eligibility for SSI and Food Stamps. The potential losses can best be understood in the context of the family incomes of families that include an immigrant losing benefits. Calculating immigrant benefit losses in this framework, we found: •About a third of aged and disabled immigrants on SSI live well above poverty and would have lost only a small share of their income. Between 20 and 40 percent of immigrant families would have lost less than 10 percent of their annual income if SSI were cut. About one in three immigrant families receiving SSI had 19 Income below poverty Income above 185% of poverty 0 10 20 30 40 50 60 70 Percent of Food Stamps families Before (with Food Stamps) After (without Food Stamps) Figure 4.5—Change in Income Relative to Poverty Level Assuming Loss of Food Stamps: SIPP Data income over 185 percent, even without the SSI benefits. Many families had earnings exceeding $40,000 annually. •At the same time, another a third of the SSI immigrant population would have lost their primary source of support by the loss of SSI. Between 20 and 40 percent of immigrant families receiving SSI would have lost more than half their annual income. The poverty rate among SSI immigrants would have risen from 13 to 48 percent. •Nearly two-thirds of immigrant Food Stamps recipients would have lost less than 10 percent of their annual income with the denial of Food Stamps eligibility. The poverty rate among immigrant families losing Food Stamps would have risen from 54 percent to 63 percent. A negligible fraction (less than 5 percent) would lose more than a quarter of their total income with the denial of Food Stamps benefits. At $1,083, the median income loss for Food Stamps recipients would be less than one-fourth the loss suffered by SSI recipients. There are three programmatic differences between SSI and Food Stamps that are key to understanding the different policy effects. The first helps explain the decision to reinstate SSI. The SSI program was intended to be a primary source of income for people 20 unable to work: the disabled and the aged.2 Food Stamps merely supplement the income of poor families, where the majority of family income could come from work, from welfare, or from other benefits such as unemployment insurance. Thus, SSI recipients are much more dependent on the benefits. The other two differences work in the opposite direction, however. First, Food Stamps are issued to an entire household based on the entire household’s income. SSI, on the other hand, goes to individuals or couples with little outside income counted for eligibility, although a sponsor’s income theoretically may be counted for five years. Second, income eligibility is more routinely redetermined in the Food Stamp program than in the SSI program. Together, these two differences make it far more likely that an SSI recipient would live in a family with higher income. _________ 2Age and disability status also prevent some immigrant SSI recipients from meeting English language requirements for citizenship. 21 5. Which Immigrants Gain from the Restoration of Benefits? In the previous chapter, we evaluated the effect of the 1996 welfare reform law on immigrant families receiving SSI and Food Stamps. Here we reverse directions. The thought exercise is this: Assuming that immigrants were ineligible for SSI and Food Stamps as originally planned under the PRWORA, what kinds of families will benefit from reinstating SSI or Food Stamps? We consider the consequences for California families of reinstating benefits under four alternative scenarios: • Restore all SSI benefits. • Restore SSI only to the disabled or only to the elderly. • Restore all Food Stamps. • Restore Food Stamps just to children and/or the elderly. To evaluate how these policy alternatives would affect the incomes of immigrant families, we start from the hypothetical situation where SSI and Food Stamps benefits are eliminated as intended in the 1996 reform legislation. We then consider the effect of restoring benefits on the total income of recipient families. As before, results from SIPP and CPS map to best and worst cases, where the best case removed benefits only from noncitizen family members only when they were present in the family and the worst case assumes that all benefits to a family are lost all year if any family member is ineligible. 22 SIPP thus assumes lower costs of restoring benefits because fewer benefits were eliminated to begin with. For each of the four policy scenarios, we examine the distribution of benefits to different categories of families, distinguished primarily by income or poverty level. We base this strategy on the idea that an important criterion for effective poverty policy is successfully targeting benefits toward the needy. The projected income gains to immigrants, of course, also represent the costs to the public treasury, so our analysis simultaneously estimates effects of the proposals on state and federal budgets. As of this year, the federal government has chosen to bear these costs, although California was prepared to cover the cost of replacing Food Stamps. In both cases, it is important to note that these results assume 1993–1995 participation rates. This will lead to overestimates for two reasons. First, California’s improving economy has led to declining caseloads, so fewer families receive benefits today, especially in the Food Stamp program. Second, we do not adjust the numbers to reflect growing citizenship rates. These cases, of course, will still qualify for benefits but would no longer be differentiable from the rest of the caseload. Restoring All SSI Benefits Table 5.1 provides the SIPP and CPS estimates for the total monetary value of reinstated SSI benefits for immigrant families in California. The two estimates are amazingly similar; both report an expected value of just over $1 billion annually. 1 These benefits were restored by the federal government as part of the Balanced Budget Act of 1997. Since California has a disproportionate share of immigrants on SSI, this represents Table 5.1 Total Value of Reinstated SSI Benefits to California Immigrant Families (millions of 1996$) Benfits Reinstated SIPP CPS All SSI $1,020 $1,097 SSI to disabled $272 $571 SSI to elderly $735 $579 _________ 1By definition, this is identical to the loss of income summed over all families described in the previous chapter. 23 a transfer of money from the rest of the United States to California. Benefits were restored only to immigrants legally in the United States in August 1996 who either were already receiving SSI or will qualify in the future based on disability. Therefore, the annual costs are likely to fall over time. Figures 5.1 and 5.2 show the breakdown of reinstated benefits across different income levels. Figure 5.1 divides benefits by the income quintile of immigrants’ families. Quintiles are determined by ranking all California families according to their income and dividing them into five equal-size groups. Thus, one-fifth of California families are in the lowest quintile, one-fifth in the “mid-low” quintile, and so on. A family with median income is in the middle quintile, and at least 80 percent of families have incomes below the income of a family that falls into the highest quintile. A poverty program typically serves families in the lowest or sometimes the next lowest or “mid-low” income quintile. In the case of reinstated SSI benefits, we do see 75 percent of benefit dollars going to families in the bottom two quintiles. However, we also see 20 percent of benefit dollars returned to families with incomes higher than 60 percent of all California families. In fact, $50 million of SSI benefits are being returned to families with incomes in the top- fifth of the entire California population. Immigrant families may be larger than other California families, in which case income quintiles would not accurately reflect individuals’ economic status. For this reason,    9% 66% 14% 6%5% Highest Mid-high Middle Mid-low Lowest  Figure 5.1—Share of Reinstated SSI Benefits, by Income Quintile: SIPP Data 24 Over 185% of poverty level From 100% to 185% From 66% to 100%Less than 66% of poverty level 47% 24% 19% 10% Figure 5.2—Share of Reinstated SSI Benefits, by Income as a Share of Poverty Level Figure 5.2 shows the benefits returned by the poverty status of the immigrant family. Poverty levels are based on family size as well as income and, in Figure 5.2, do not include SSI benefits. Two-thirds of benefit dollars go to families who would be in poverty without SSI. In fact, almost half of the reinstated benefit dollars go to families with incomes below two-thirds of the poverty level—families in extreme poverty. However, another 24 percent or nearly $250 million of the reinstated benefits go to families above 185 percent of poverty—before SSI benefits are even counted in their income. Restoring SSI Only to the Disabled or Only to the Elderly Neither SIPP nor CPS indicates whether a recipient qualified for SSI based on age or disability. However, there are age data, and SIPP also asks respondents to report disabilities. In CPS, we can identify only elderly and nonelderly recipients. With this information, we have roughly divided the reinstated benefits into those going to the nonelderly disabled and those going to the elderly. This breakdown is reported in Table 5.2. Table 5.2 Total Value of Reinstated Food Stamps Benefits to California Immigrant Families (millions of 1996$) Benfits Reinstated SIPP CPS All Food Stamps $561 $683 Food Stamps to children $134 $217 Food Stamps to elderly $15 $16 25 Figures 5.3a and 5.3b show the income quintile breakdown for the two groups. Although elderly immigrants are more likely to be in the lowest income quintile without SSI, they are also more likely to have income in the top two quintiles. If benefits were reinstated just to the elderly, 22 percent of dollars would go to families in the top 40 percent of the income scale. For the disabled, this share drops to 15 percent.     13% 12% 10%3% Highest Mid-high Middle Mid-low Lowest  Figure 5.3a—Share of Reinstated SSI Benefits to Disabled, by Income Quintile    7% 16% 4%6% Highest Mid-high Middle Mid-low Lowest  Figure 5.3b—Share of Reinstated SSI Benefits to Elderly, by Income Quintile 26 Restoring All Food Stamps President Clinton had called for the restoration of all Food Stamps benefits for immigrants, although the 1998 compromise restores them primarily to just children and the elderly. Had all benefits been restored, the Food Stamps would have totaled over $500 million for California families, based on 1993–1995 participation patterns. If these benefits were restored at the federal level, it would be a transfer of income to California, since California has more immigrants than any other state. These benefits dollars would be distributed as shown in Figure 5.4. This figure is parallel to Figure 5.1, dividing dollars according to where the recipients fall in the California income distribution. These benefits are calculated recognizing that SSI has already been reinstated; the distinction matters for determining the economic status of families before Food Stamps receipt. Eighty-seven percent of benefit dollars would be returned to families with incomes in the bottom 40 percent for California families. The remaining families usually have incomes around the median, with relative few in the upper income ranges. The families with higher incomes who receive Food Stamps are likely to be either large families or families who experience only short-term income losses. To control for family size, Figure 5.5 reports shares of reinstated benefits by poverty level. Three out of four benefit dollars would go to families in poverty. Another 20     30% 57% 4% 8%1% Highest Mid-high Middle Mid-low Lowest  Figure 5.4—Share of Reinstated Food Stamps Benefits, by Income Quintile 27 Over 185% of poverty level From 100% to 185% From 66% to 100%Less than 66% of poverty level 31% 5% 44% 20% Figure 5.5—Share of Reinstated Food Stamps Benefits, by Income as a Share of Poverty Level percent would go to families in near poverty. Only 5 percent of these benefits would go to families with income above 185 percent of poverty, one-fifth the share under SSI. Restoring Food Stamps to Children and the Elderly First in California and now nationally, Food Stamps have been restored to children and elderly immigrants who received Food Stamps before the welfare reform law. According to the survey data, these benefits account for between one-fourth and one-third of Food Stamps benefits to immigrants. Figures 5.6a and 5.6b provide the distribution of benefit dollars by income quintile for Food Stamps to children and the elderly. Compared to Food Stamps to all immigrants, Food Stamps to children will go overwhelmingly to families in the bottom two-fifths of the income distribution. Almost no benefit dollars would be returned to families at the top of the income distribution. These results are consistent between SIPP and CPS. The distribution for Food Stamps to the elderly shows one-fifth of dollars going to families in the mid-high income range but none to families in the middle quintile. The sample size for elderly immigrants receiving Food Stamps is extremely small, however, since the elderly would typically qualify for SSI instead of Food Stamps. 28  32% 62% 2% 4% Highest Mid-high Middle Mid-low Lowest  Figure 5.6a—Share of Reinstated Food Stamps Benefits to Children, by Family Income Quintile    14% 1% 20%Highest Mid-high Middle Mid-low Lowest  Figure 5.6b—Share of Reinstated Food Stamps Benefits to Elderly, by Family Income Quartile Summary of Key Findings Our first and last scenarios mirror the current policy. Our calculations consider California’s immigrant families who had received benefits from SSI and Food Stamps, starting from their baseline income in the absence of these two benefits. Reinstating first SSI and then Food Stamps for children and the elderly, we found: 29 • One in three SSI dollars will go to a family not in poverty; one in four to a family above 185 percent of poverty. The same result is evident in income quintile calculations where 20 percent of benefit dollars go to families with incomes higher than that of 60 percent of California families. This represents over $200 million in benefits to relatively well-off families, even before counting SSI income. • Elderly SSI recipients are more likely to be in the higher-income families. This presumably occurs because these recipients are living with their adult children. • The majority of reinstated SSI and Food Stamps benefits will go to families in poverty. $671 million out of about $1 billion in SSI benefits will go to families with incomes below the poverty level. $128 out of $149 million in state-only Food Stamps would go to these families. 31 6. Should We Be Rethinking Welfare Reform? The first element of the 1996 welfare reform to actually “bite” was the denial of SSI and Food Stamps to immigrants. In the wake of the law’s passage, the American public was barraged with emotional rhetoric depicting the fate of immigrants losing access to SSI and Food Stamp programs under the new law. Those wanting to restore benefits claimed that recipients would be destitute without public assistance. Advocates for denying benefits argued that immigrants create a burden on the welfare system and that many are well enough off without benefits. Who was right, and who was wrong? Our findings show that both are right, and both are wrong. Significant fractions of the immigrant welfare population match both descriptions. Supporting the reforms, our analysis of California’s immigrants reveals that immigrant families are more than twice as likely as citizen families to participate in a major welfare program. Further, reinstating welfare benefits to immigrants by simply reversing the 1996 reforms poorly targets “needy” immigrants families. A significant share of these reinstated benefits will go to families above the poverty level and to families without disabled members. Supporting the reinstatements, this study finds that immigrant families in the state are more than twice as likely as citizen families to live in poverty. Some immigrant families greatly rely on welfare benefits and it is not obvious how they would fare without public assistance. 32 Efficiency of Immigrant SSI Benefits SSI benefits are intended for low-income individuals who are unable to support themselves because of disability or age. However, instead of assisting only the state’s truly needy, we find that SSI benefits are often provided to immigrants whose total family incomes are well above the poverty level. This happens because the program provides assistance based on individual rather than family need. According to our estimates using 1993–1995 data, the cost of reinstating SSI benefits to the nearly 250,000 immigrant families in California who participate in the program is nearly $1 billion annually. 1 About one-third of these reinstated benefits are projected to go to immigrant families whose annual incomes exceed the poverty level. Nearly one-quarter goes to families with incomes 185 percent of the poverty level, a common income threshold for welfare programs to stop paying benefits. Even though the disabled were a main focus for the restoration of benefits, only half of the SSI benefits go to families that include a disabled member. By reinstating SSI benefits to all legal immigrants who were on the rolls in August 1996, lawmakers are essentially giving welfare dollars to thousands of people who do not necessarily need help. This is not to say that a modification of the federal welfare law was not in order. Some 55,00 immigrant families in the state, just under a quarter of the families who receive benefits during a year, stood to lose more than half of their income if SSI benefits were cut off. Two-thirds of SSI benefits go to families that would otherwise be in poverty. Efficiency of Immigrant Food Stamps Benefits The case for reinstating Food Stamps is less dramatic on both sides, because Food Stamps benefits are not a major source of income for recipients. Assuming 1993–1995 participation rates, our estimates indicate that fully restoring Food Stamps to all California immigrants would cost $561 million to $683 million annually and would provide income to approximately 430,000 families. Returning benefits just to children and the elderly would cost about $149 million. 2 _________ 1Unlike other welfare programs, SSI caseloads are virtually unchanged from the 1993–1995 period. 2Current Food Stamps caseloads are substantially below the 1993–1995 level used to calculate the estimates presented in this study, so these are overestimates of today’s costs. 33 For those arguing against reinstating Food Stamps, our findings support the view that the elimination of benefits was unlikely to impose substantial hardships—particularly when compared to elimination of SSI. About two-thirds of immigrant Food Stamps recipients would have lost less than 10 percent of their annual income with the denial of Food Stamps eligibility. Less than 2 percent would lose more than a quarter of their total income. Of those families who would forgo a significant fraction of their income, the vast majority receive welfare from other sources. Nearly 75 percent of immigrant families receiving Food Stamps also collect other forms of welfare (primarily AFDC plus some SSI), and, for these families, Food Stamps contribute on average slightly more than 10 percent of their total annual income. The remaining 25 percent of immigrant Food Stamps recipients rely on benefits to supplement their earnings, but Food Stamps make up only 5 percent of their family income. Expressing losses in dollar rather than relative amounts, well over three-quarters of the immigrant families collecting Food Stamps would have lost less than $2,000 in benefits per year. To place this amount in context, the enhancement of the Earned Income Tax Credit (EITC) that began in 1996 more than makes up for this loss for those who work—EITC pays up to $3,60 in benefits— meaning that working immigrant families who collected Food Stamps are better off now than they were in the early 1990s before the denial of Food Stamps benefits. On the other hand, our findings also support those favoring the restoration of Food Stamps to immigrants in that they demonstrate that the Food Stamp program is fairly well targeted to the poor. Reinstating benefits to all immigrants in California would give four in five dollars to families in poverty. Returning benefits just to immigrant children and the elderly, as legislated in California and now nationally, is even more efficient at assisting the poor, with 86 percent of restored benefits going to families with incomes below poverty. One in three of all reinstated Food Stamps dollars will go to families with disabled members, with one out of ten dollars going to families with disabled members who were either children or elderly. Were Immigrants a Special Case? Our analysis has focused on immigrants, since they were a principal target of the 1996 reform and the beneficiaries of the subsequent restoration of benefits. In fact, Michael Fix of The Urban Institute has described the denial of immigrant eligibility for welfare as a 34 switch from “immigration policy” or rules governing entry to the United States executed by the Immigration and Naturalization Service to “immigrant policy” or rules governing immigrants within the borders. Up until this change, however, the only practical difference between immigrants and natives for the purposes of welfare policy was the fact that most immigrants have family members who sponsor their immigration. A sponsor’s pledge to prevent the immigrant from becoming a public charge is the basis for deeming provisions in the determination of benefits. Deeming refers to counting sources of income other than the immigrant’s own as resources available to the immigrant. Income is counted to determine eligibility and to calculate benefits. Deeming of sponsor income occurs in both the SSI and the Food Stamp programs. According to rules of the SSI program, for example, the income of an immigrant’s sponsor is treated as unearned income for the immigrant, after subtracting an allocation for the sponsor and his or her dependents. A dollar increase in unearned income reduces SSI benefits by one dollar. Consequently, when sponsors are financially self sufficient, theoretically their income fully counts in lowering the immigrant’s potential SSI benefits, resulting in the immigrant losing eligibility if the sponsor is sufficiently well off. Before the 1996 welfare reform, deeming provisions held for three to five years after an immigrant’s arrival in the United States. As part of the reform, the deeming period was lengthened to the point when an immigrant becomes a “qualified alien” by accruing 40 quarters of work or becomes a U.S. citizen, and sponsors were required to have sufficient income to support an immigrant, defined as at least 125 percent of the poverty line. These changes do not apply to immigrants who became legal residents before January 1, 1998. Our data include no information about sponsor income and therefore may underestimate the resources available to immigrants. Some family members whose income we include may in fact be the immigrants’ sponsors, but sponsors need not reside with recipient relatives. Although we cannot identify sponsors, our findings do demonstrate that many SSI members reside in families with resources well beyond the poverty level. Consistent with these findings, Matloff (1994) reports that 42 percent of California’s elderly SSI recipients live with their children. Yet the difference between Food Stamps and SSI in their effectiveness at targeting needy families suggests a direction for rethinking the way income is counted for welfare. 35 The Food Stamp program typically treats all individuals living together (and sharing food preparation) as a single household for the purpose of determining eligibility and benefits. A similar strategy could be considered for SSI. Currently, SSI treats in-kind support and maintenance, such as when a recipient is living in the household of another, as equal to one-third or less of the SSI benefit level, regardless of income of other household members. If the income of other household members were counted, as in the Food Stamp program, benefits for many immigrants would be lowered and others would be disqualified. Yet immigrants living alone or in poor families would receive benefits through the program. During the debate over welfare and immigration reform, bills jointly sponsored by California’s two senators and three California representatives proposed restoring eligibility to any immigrant who otherwise qualified for SSI and “whose family is incapable of support, and who can demonstrate that he or she has no other sufficient means of support.” Defining the sources of income to be counted (deemed) in calculating SSI (or Food Stamps) benefits provides a natural framework for operationalizing the concept expressed in this bill. Such rules need not apply to immigrants alone. In fact, although the term “deeming” most commonly refers to a sponsor’s income, the same term applies whenever an additional person’s income is counted in determining welfare benefits. For example, the Social Security Handbook reports a number of types of deeming other than from sponsor to immigrant, such as from ineligible to eligible spouse and from parent to child. Each welfare program sets in own rules for whose income to count or deem. Unfortunately, despite the many dramatic changes in welfare reform, there was no reconsideration of these deeming rules as a strategy for addressing the policy concerns regarding immigrants’ receipt of welfare. As this analysis has shown, families of welfare recipients may have vastly different resources available. 3 We would like to ensure benefits to the neediest families, while asking self-sufficiency of less needy families. This lack of thoughtful program design was the most disturbing feature of the denial and subsequent reinstatement of benefits for immigrants. The welfare law itself, as well as the debate around reinstatement, pointed to other policy options that could have gone _________ 3Although the results in this study apply only to immigrants, they are likely to be the most extreme case. Previous research comparing recent immigrants to the California caseload overall showed higher incomes in immigrant welfare families (not controlling for family size) (MaCurdy and O’Brien-Strain, 1996). 36 beyond simply switching eligibility on and off. Welfare reform provided a rare opportunity to redesign programs to better meet the needs of our poorest families while promoting self-sufficiency and family responsibility. Categorically barring immigrant families from benefits did not meet these goals. Unfortunately, simply reversing the reforms did not meet these goals either. Denial of benefits to immigrants represented only the first in a series of tough challenges arising from the sweeping changes in the PRWORA. There are likely to be future cases requiring policymakers to rethink and refine the reforms enacted in 1996. If, instead, policymakers merely reverse direction each time, forgetting why they adopted the reforms in the first place, our welfare system will be no better than it was in 1996. 37 Appendix A Data and Methodology Our analysis is built on a profile of immigrant families that received welfare benefits during the period 1993 to 1995. There are two key features to our analytical approach. First, we use survey data to understand the economic circumstances of recipients to a depth not possible with administrative data. Second, we treat families as the relevant unit to count for benefits—a significant departure from the existing policy in most public assistance programs. In this appendix, we outline the source of the data and our definition of family and then explain how these elements are combined to assess the effects of the 1996 benefit cuts and the recent restorations. Data The empirical results presented in this study are derived from samples of California households in two different surveys conducted by the Bureau of the Census: the Survey of Income and Program Participation (SIPP) and the March Current Population Survey (CPS). Both surveys interview a nationally representative sample of households, collecting detailed demographic and income information. Compared to administrative data, these survey data create a more complete picture of the economic circumstances of households, families, and individuals. Specifically, we can include income counted for benefit determination as well as income sources typically excluded from these calculations and examine the economic status of recipients across the year, not just in months on aid. 38 The SIPP was the basis for our earlier report (MaCurdy and O’Brien-Strain, 1996).1 Although the structure of the SIPP is also ideal for the issues addressed here, we have expanded our analysis to include the CPS as well. We have two reasons for using the CPS data. First, the CPS provides a larger sample, especially when several years are pooled together. This may be particularly important when examining relatively small population groups such as immigrants receiving SSI. Second, the CPS is better known and more commonly used in the research community. Conducting the same analysis using both CPS and SIPP provides a link with earlier research and a comparison of the two datasets. The features of the two samples, as well as the advantages and disadvantages of each, are summarized in Table A.1. Data from the two samples are consistent for most of the key results, although as shown in Table A.2, the SIPP sample is very small for some of the detailed calculations. Table A.1 Sample Sizes and Characteristics of the SIPP and CPS SIPP 1993–1994 CPS 1993–1995 –Smaller sample 6,219 California families 1,022 immigrant families+Large sample 15,494 California families 4,654 immigrant families –Less known, more complex to use+Better known, more commonly used +Longitudinal panel, collected every 4 months for 40 months–Cross-section, covering previous year +Monthly data on income and participation–Annual data on income and participation +Reported income comes closer to independent estimates–Reported income falls short of independent estimatesa –Immigrant status identifiable only for persons over 15 years old+Immigrant status identifiable for all persons –Weighting problems specific to immigrant population found in 1993 and 1994 data NOTE: + indicates an advantage over the other dataset; – indicates a disadvantage. aCitro and Kalton (1993). _________ 1Further information about the SIPP sample can be found in Appendix A of that report. 39 Table A.2 Unweighted Subsample Sizes for SIPP and CPS Results SIPP 1993–1994CPS 1993–1995 Califoria families All California families 6,219 15,494 Citizen families 5,197 10,840 Immigrant families 1,022 4,654 Recent immigrant families 480 1,155 Immigrant welfare participant families Families participating in major programs 416 1,376 Families participating in AFDC 139 522 Families participating in SSI 99 287 Families participating in Food Stamps only 52 241 Immigrant recipients affected by benefit denials Receiving SSI 77 287 Receiving Food Stamps but not SSI 141 607 Receiving SSI or Food Stamps 218 894 However, where results differ between the two samples, these differences do not appear to be driven by the different sample sizes. In other words, the differences are not an artifact of the limited number of California immigrant observations in the SIPP. Instead, they result from the different survey strategies. Indeed, the same patterns of results occur in the national data and for all families in California (both cases with larger samples). Overall, there appear to be three fundamental differences between the two datasets. First, the monthly reporting from repeated surveys in SIPP identifies more income than CPS. For this reason, the poverty rate is higher in CPS than in SIPP. The SIPP also finds more welfare participation than CPS. Finally, CPS respondents who do report welfare participation report more benefits on average than welfare recipients in SIPP. Detailed results from each dataset are included in Appendix B. Similar findings regarding SIPP versus CPS have been documented in two studies by the National Academy of Sciences, one examining the SIPP (Citro and Kalton,1993) and a second examining the poverty measure (Citro and Michael, 1995). These reports also show the lower poverty rate in SIPP and note that there is more complete income reporting in SIPP than in CPS, coming closer to independent measures for most categories of income. This is especially true for transfer income, which is highlighted in 40 these studies. In fact, one recommendation in Citro and Michael is replacement of the CPS by the SIPP for calculation of the poverty rate. Focus on Families This analysis focuses on the resources of families receiving public assistance in California. It is perhaps surprising that families are rarely the unit of observation in welfare literature. Despite the rhetoric of family responsibility, researchers typically examine either individuals or cases, neither of which maps to policy goals. As policymakers, we hope to make families self-sufficient, but we do not expect self - sufficiency of every individual, especially children. Although administrative cases come closer to families, they need not match our concept of families. For example, a family consisting of a mother and her two children would be categorized as two different cases if one child is disabled and receiving SSI. In highlighting families and family income, we believe that we come closer to understanding the economic reality for people receiving public assistance. Of course, in the 1990s, defining a family is not as simple as it once was. For the purposes of this analysis, all people are assigned to families. A family is defined as one or more people living in the same household, related by birth, marriage, or adoption. Most of these families are conventional nuclear families: an individual, his or her spouse (if present in the home), and any children living at home. Some families are extended families, such as when a grandparent lives in the household with his or her child and grandchildren. Beyond these nuclear relationships, we do not combine relatives into the same families if there would be no expectation of shared resources even when they share a household. Thus, two adult sisters living in the same house represent two separate families. Individuals living alone or with nonrelatives are one-person families. This includes unmarried, cohabiting partners. This study examines the families of immigrants. We define immigrants to be foreign- born noncitizens. A family is considered an immigrant family if one or more members are immigrants. Immigrant families may include citizen members. However, we have no information on family sponsors, so an immigrant family may or may not include the sponsor. Therefore, in the case of sponsored immigrants, the resources of the family may 41 underestimate the total resources available—to the extent that sponsors fulfill their agreements to prevent immigrants from becoming public charges. Analytical Approach The core results in this report simulate the effects of denying and then restoring SSI and Food Stamps benefits to immigrant families. We calculate the amount (and share) of a family’s income that would be lost if an immigrant family no longer received SSI and Food Stamps. There are four steps to this calculation: 1. Identify affected individuals in affected families. A family is potentially affected if there is an immigrant family member and the family received SSI and/or Food Stamps. However, we exclude income received by family members who do not lose eligibility: citizen and refugee family members. 2 So some families who are potentially affected do not lose any benefits, and others lose benefits only for some family members. 2. Calculate the loss of SSI and Food Stamps. For noncitizen, nonrefugee family members, the cutoff of SSI and Food Stamps entailed losing all benefits from these programs. We add up benefits received by such family members over the course of a year and count this amount as the “projected loss” that is reported in the empirical results. 3. Determine eligibility for AFDC. Some immigrant families receive SSI for a disabled or blind child. In these families, the child may qualify for AFDC if SSI is denied. Since we can determine eligibility on a monthly basis in the SIPP data, we can determine AFDC eligibility after removing SSI. (We can also exclude only benefits for the specific months in which a now-ineligible person was the benefit recipient.) In SIPP calculations, we add back in the AFDC benefits. The monthly calculation is not possible with the CPS data. 4. Evaluate the family’s economic status after losing benefits. After calculating the projected income loss, we recalculate family income as a share of poverty and by income quintile. In doing so, we also report what other economic resources are still available to the family. These final estimates not only provide information about the income losses expected under denial of SSI and Food Stamps, but they also give us a distribution of family _________ 2We do not have direct information on refugee status. We assign refugee status to all immigrants from refugee-sending countries based on the year of arrival. In the SIPP, identifiable refugee-sending countries include Cuba, Czechoslovakia, Hungary, Poland, the U.S.S.R., and Vietnam. 42 income in the absence of these benefits. With this information, we can evaluate what kinds of families benefit from having SSI or Food Stamps restored, as recently passed at the federal and state levels. Qualifications in Interpreting Findings As we note in the discussion of our empirical analyses, our estimates of the total costs and values of losses and reimbursements are invariably on the high side for the policy reforms currently under consideration, because these estimates apply to the circumstances relevant in the 1993–1995 period when caseloads were larger. Moreover, our estimates rely on annual income, and not monthly income, as used to determine welfare eligibility. Thus, it is quite possible that a family could be well above the poverty level for the year as a whole but below for several months. We invite the reader to be the judge of how to evaluate this qualification. Although we believe that our estimates for totals are high for predicting the values of welfare benefits involved in the policy reforms affecting immigrants, the robustness of our findings on the shares of losses and gains going to the various immigrant groups leads us to be confident about the implications of our findings regarding the distributional consequences of alternative polices. 43 Appendix B Primary Tables and Description This appendix describes the construction of the primary tables that summarize our empirical findings. The discussion outlines the definitions of the variables presented in these tables, along with the samples used to calculate the estimates. Each table is reproduced for each of the two datasets. The SIPP results are shown first, then the CPS results. Data available in SIPP are not always reproducible in CPS. Therefore, for some rows of data, the CPS version of the table will be blank. In tables showing the simulation results for loss of benefits, there are two SIPP versions. The “best case” version considers eligibility and income loss on a monthly basis. The “worst case” version eliminates income for the entire year. The CPS results are roughly equivalent to the “worst case” version in SIPP. All incomes and benefits are converted to 1996 dollars using the June Consumer Price Indexes (CPIs) for 1993 to 1996. The SIPP results combine annual data for 1993 and 1994, and the CPS combines annual data for 1993 through 1995. There are certain conventions to keep in mind when reading the tables: 1.Panels: Panels in tables combine rows summarizing closely related information, and these panels are also replicated across tables. For simplicity, we may refer to panels to describe subsections of the tables. 2.Columns: Samples are common between rows in a given column, but are different across columns. 44 3.Indents: Indented rows indicate that the listed statistics are conditional on and only applicable to the group identified in the first nonindented row above the statistics. For example, an indented row below a nonindented participation rate means that the statistic is conditional on participating. Tables B.1a and B.1b: Characteristics of California Families Tables B.1a and B.1b present summary statistics on the characteristics of all California families and of immigrant families in particular. These characteristics include demographic characteristics, race/ethnicity, poverty status, and welfare participation. Sample: All California families. Columns: Tables B.1a and B.1b have four columns: All: All California families are combined for the first column. In the remaining three columns, families are divided by citizenship status. All families qualify as either citizen families or immigrant families. Citizens: The column labeled “Citizens” presents families in which all members are born or naturalized U.S. citizens. All Immigrants: These are families in which one or more members are foreign- born noncitizens. Recent Immigrants: These are a subset of “All Immigrants.” These are families in which one or more members are foreign-born noncitizens who arrived after 1985 (Table B.1a) or after 1990 (Table B.1b). Rows: The rows in these tables are divided into seven “panels.” The definitions of the variables in these panels are: Panel A: % of all families in California. The first row shows the share of all California families represented in the particular column. Panel B: Age of family head. This group of rows, along with the next two panels, presents demographic characteristics of families and/or family heads. In a family headed by a married couple, the head is always taken to be the male. The three rows in this panel divide family heads into different age groups (e.g., “% £25” refers to family heads who are age 25 or younger). 45 Table B.1a Characteristics of Citizen and Immigrant Families and Their Welfare Participation in California: SIPP Data Programs Family Characteristics All CitizensAll ImmigrantsRecent Immigrants A % of all families in California100.0 80.8 19.2 8.5 B Age of family head % £ 25 8.5 8.5 8.6 14.1 % 26–64 73.7 71.3 83.9 82.1 % 65+ 17.7 20.2 7.5 3.8 C Family characteristics % married 49.3 46.2 62.6 62.0 % female heads 28.9 31.2 18.8 17.3 % divorced 15.1 17.1 6.7 4.2 % never married 21.4 21.8 19.4 23.8 % with children 36.6 31.4 58.3 62.4 No. of children 2.0 1.9 2.3 2.2 % with child £ age 6 22.6 19.9 28.8 28.2 % with disability in family 19.6 20.0 18.0 14.1 % with child disability 21.1 21.2 20.3 16.3 D Education of family head % less than high school 21.7 16.1 45.3 41.7 % high school graduate 24.7 25.6 20.9 21.5 % some college + 53.6 58.3 33.8 36.7 E Race/ethnicity % white 63.5 73.9 19.8 18.2 % black 6.2 7.5 0.8 1.0 % Hispanic 20.9 12.1 57.8 53.9 % Asian 8.3 5.2 21.6 26.9 F Poverty status % < traditional poverty level 11.2 9.0 20.5 20.0 Income within poverty level % < 2/3 3.2 2.8 5.0 6.2 % 2/3–1 7.1 5.6 13.5 11.8 % 1–1.85 19.9 17.1 32.0 36.0 % 1.85+ 69.7 74.5 49.5 46.0 G Welfare participation % any 33.6 27.5 59.7 62.2 % any major program 24.3 20.3 41.4 44.4 % any major and minor 16.1 12.1 32.7 36.5 % only minor 9.3 7.2 18.2 17.8 46 Table B.1b Characteristics of Citizen and Immigrant Families and Their Welfare Participation in California: CPS Data Programs Family Characteristics All CitizensAll ImmigrantsRecent Immigrants A % of all families in California100.0 77.1 22.9 6.1 B Age of family head % £ 25 8.1 7.3 10.8 20.2 % 26–64 74.6 72.3 82.6 76.9 % 65+ 17.3 20.4 6.6 2.9 C Family characteristics % married 48.7 45.1 61.0 58.8 % female heads 29.0 31.6 20.2 20.1 % divorced 14.1 16.7 5.5 3.1 % never married 24.7 24.5 25.3 32.0 % with children 36.8 31.1 56.1 56.2 No. of children 2.0 1.8 2.2 2.1 % with child £ age 6 21.3 16.7 36.9 40.1 % with disability in family % with child disability D Education of family head % less than high school 20.0 11.2 49.0 49.9 % high school graduate 23.1 24.5 18.5 14.6 % some college + 57.0 64.3 32.5 35.6 E Race/ethnicity % white 60.7 73.8 16.5 16.2 % black 6.8 8.4 1.6 1.8 % Hispanic 23.6 12.0 62.8 55.9 % Asian 8.0 4.9 18.5 25.5 F Poverty status % < traditional poverty level 17.2 13.3 30.5 37.5 Income within poverty level % < 2/3 9.4 7.3 16.2 22.4 % 2/3–1 7.3 5.5 13.3 13.9 % 1–1.85 18.2 15.7 26.7 27.1 % 1.85+ 65.1 71.5 43.8 36.6 G Welfare participation % any 29.6 24.0 48.3 48.1 % any major program 17.5 13.9 29.6 31.6 % any major and minor 8.3 5.7 17.1 16.2 % only minor 12.1 10.1 18.7 16.5 47 Panel C: Family characteristics. This group of rows presents several family structure characteristics. The analysis classifies family heads as married, unmarried female, divorced, or never married. It also examines families with children and families with a member having physical disability. (Information on disability is not available in the CPS.) The indented rows indicate that the statistic is conditional on and only applicable to the category listed above (e.g., statistics for “No. of children” and “% with child £ age 6” are computed only for those families in the category “% with children”—the line above). Panel D: Education of family head. These rows show the distribution of educational attainment (highest grade completed) for heads making up each participation group. Panel E: Race/ethnicity. The four rows making up this group break down program participants into their racial/ethnic groups: whites, blacks, Hispanics, and Asians. Hispanics are defined to be white Hispanics, and whites are non-Hispanic. Panel F: Poverty status. The first row in this panel presents the percentage of each column below the traditional poverty level. The traditional poverty level does not included the cash value of Food Stamps in the calculation of total income. In contrast, our calculations always include the this cash value. The next four rows report how our income measure compares with each family’s poverty level (e.g., “% < 2/3” indicates that the family’s income is less than two-thirds of the poverty level, whereas “% 1.85+” designates that income exceeds 185 percent of the poverty level). We use 1.85 as a reference level because this value is used by many means-tested programs to end benefits. Panel G: Welfare participation. The final set of rows reports participation in public assistance programs. “Major programs” include the four major public assistance programs: AFDC, SSI, Food Stamp, and Medicaid (known in California as Medi-Cal). We identify any family who is enrolled in or collects any benefits from any major program to be a “welfare participant.” “Minor programs” include Special Supplemental Nutrition Program for Women, Infants and Children (WIC), GA, school-based food programs, housing assistance, and energy assistance. The first row in this group presents the percentage of families receiving benefits from any major or minor program. The second row refers to families participating in one or more of the major programs. “% any major and minor” includes families who participate in at least one major and one minor 48 program, and “% only minor” shows the percentage of families who participated only in minor assistance programs. Tables B.2a and B.2b: Income Sources for Immigrant Families Receiving Welfare Tables B.2a and B.2b report an array of statistics designed to summarize income sources and amounts received by immigrant families who received welfare in California. Sample: All immigrant families who participated in a major welfare program. Columns: Results in Tables B.2a and B.2b are divided into four columns based on program participation: Any major: This column presents results for all welfare-participant families (i.e., those who participated in any major program in at least one month during the sample period for the SIPP data or at any time during the three years of the CPS data). The remaining columns break down these program participants into three groups: AFDC: This column includes all families who collected benefits from the AFDC program some time in the observation period. SSI: This group includes all families who collected benefits from the SSI program some time in the observation period. Food Stamps only: This group includes families who received Food Stamps benefits but who collected no benefits from either AFDC or SSI—they could have been Medi-Cal participants. Families receiving both AFDC and SSI are included in both the SSI and the AFDC columns. Families receiving Medi-Cal only are included in the “any major” column but are not listed separately. Rows: The rows of Tables B.2a and B.2b group into three sections: Panel A: % of all immigrant families in California. This first row shows program participants as a share of all immigrant families. It not only shows the participation by program but it also relates the sample in this table to the information shown in Table A.1. For example, in the SIPP data, 41 percent of all immigrant families are reported to participate in any major program. This figure matches the number in the “% any major program” row for the “All Immigrants” column in Table B.1a. 49 Table B.2a Sources of Income and Amounts Received by Immigrant Families Who Are Welfare Participants in California: SIPP Data Programs Family Characteristics Any Major AFDC SSIFood Stamps Only A % of all immigrant families in California41.0 13.0 11.0 5.0 B Income (1996$) Annual total income 25th percentile 13,625 12,705 13,410 17,051 50th percentile 21,042 16,302 25,023 23,527 75th percentile 34,552 24,004 51,672 34,862 % of families with earnings 81.0 69.0 63.0 100.0 Annual earnings 25th percentile 9,086 2,713 13,924 10,671 50th percentile 18,986 7,676 31,073 18,064 75th percentile 33,404 18,030 54,402 31,787 % of income from earnings 25th percentile 36.0 13.0 44.0 51.0 50th percentile 67.0 34.0 79.0 76.0 75th percentile 91.0 60.0 91.0 90.0 % from all cash welfare 24.0 50.0 40.0 6.0 % from AFDC+SSI+Food Stamps 23.0 49.0 39.0 5.0 % from AFDC 11.0 35.0 6.0 0.0 % from SSI/State Supplementary Payment 8.0 4.0 31.0 0.0 % from Food Stamps 4.0 11.0 2.0 5.0 % from other income 26.0 24.0 18.0 25.0 C Welfare utilization (1996$) % receiving benefits ³ 6 months 53.0 94.0 84.0 62.0 % of income from cash benefits 25th percentile 0.0 15.0 6.0 1.0 50th percentile 5.0 47.0 17.0 2.0 75th percentile 37.0 92.0 86.0 7.0 Annual cash benefits 25th percentile 148 4,816 2,399 272 50th percentile 1,850 8,990 6,642 606 75th percentile 7,463 12,870 7,978 1,513 % AFDC participants 32.0 100.0 21.0 0.0 % of income from AFDC 35.0 35.0 31.0 . Annual AFDC benefits 6,179 6,179 7,203 . % SSI participants 26.0 17.0 100.0 0.0 % of income from SSI 31.0 21.0 31.0 . Annual SSI benefits 4,851 4,791 4,851 . % Food Stamps participants 45.0 91.0 34.0 100.0 % of income from Food Stamps 9.0 12.0 7.0 5.0 Annual Food Stamps benefits 1,727 2,209 1,649 1,007 % General Assistance (GA) 5.0 9.0 9.0 10.0 % of income from GA 10.0 5.0 3.0 11.0 Annual GA benefits 1,522 1,201 787 2,055 % other welfare 90.0 98.0 80.0 94.0 % Medicaid participants 98.0 100.0 100.0 86.0 50 Table B.2b Sources of Income and Amounts Received by Immigrant Families Who Are Welfare Participants in California: CPS Data Programs Family Characteristics Any Major AFDC SSIFood Stamps Only A % of all immigrant families in California29.6 10.7 6.5 14.0 B Income (1996$) Annual total income 25th percentile 10,295 9,998 8,610 6,700 50th percentile 16,472 13,699 15,794 14,134 75th percentile 28,660 23,334 33,508 20,126 % of families with earnings 67.6 54.8 46.9 80.0 Annual earnings 25th percentile 8,648 3,809 6,524 8,236 50th percentile 16,961 9,765 20,590 13,482 75th percentile 29,856 21,745 43,888 19,255 % of income from earnings 25th percentile 59.5 29.2 30.7 72.8 50th percentile 84.0 59.8 67.5 88.1 75th percentile 98.3 77.7 83.0 96.5 % from all cash welfare % from AFDC+SSI+Food Stamps 36.3 62.9 52.0 23.3 % from AFDC 17.6 47.0 6.2 0.0 % from SSI/State Supplementary Payment 9.8 4.2 42.7 0.0 % from Food Stamps 8.9 11.8 3.1 23.3 % from other income 13.8 7.6 21.5 12.3 C Welfare utilization (1996$) % receiving benefits ³ 6 months % of income from cash benefits 25th percentile 0.0 27.3 15.6 2.7 50th percentile 16.6 70.7 47.7 8.5 75th percentile 77.0 100.0 100.0 25.5 Annual cash benefits 25th percentile 0 4,520 3,867 362 50th percentile 2,539 8,437 7,412 951 75th percentile 7,867 12,082 11,413 1,853 % AFDC participants 37.4 100.0 20.9 0.0 % of income from AFDC 47.0 47.0 29.7 . Annual AFDC benefits 6,787 6,787 6,533 . % SSI participants 22.8 12.8 100.0 0.0 % of income from SSI 42.7 32.8 42.7 . Annual SSI benefits 6,510 6,548 6,510 . % Food Stamps participants 49.1 76.8 25.0 100.0 % of income from Food Stamps 18.2 15.4 12.5 23.3 Annual Food Stamps benefits 1,773 2,047 2,103 1,301 % GA . . . . % of income from GA . . . . Annual GA benefits . . . . % other welfare 58.9 71.3 54.4 57.1 % Medicaid participants 70.8 60.2 84.9 35.2 51 Panel B: Income (1996$). Annual total income: The first group of rows measures the total annual income received by participant families, showing the 25th, 50th, and 75th percentiles of the total annual income distribution. % of families with earnings: This row shows the percentage of immigrant welfare families who had earnings and are members of the sample designated by the column heading. Using SIPP for an example, 81 percent of all welfare participant immigrant families worked at some time during the 1993–1994 period. Sixty-nine percent of those who received AFDC also had earnings. The indented rows under this subheading are conditional on families having earnings. They first report the 25th, 50th, and 75th percentiles of the distribution of annual earnings obtained by working families making up the relevant sample. The next set of rows shows the 25th, 50th, and 75th percentiles of the distribution of total income shares provided by family earnings. % from all cash welfare: The remaining rows in this panel show the average percentage of family annual income from sources other than earnings, including income from cash benefits received from all public assistance programs, from cumulated AFDC + SSI + Food Stamps benefits, from the individual major programs, and from sources of “other” income (including Social Security, Unemployment Insurance, pensions and so on). The percentage from all cash welfare (or AFDC + SSI + Food Stamps for CPS), from other income and from earnings together total 100 percent. Panel C: Welfare utilization (1996$). The first row of this panel shows the percentage of families in the relevant sample who participated in any combination of major programs for six months or more. This information is available only in the SIPP. The next rows report the 25th, 50th, and 75th percentiles of the distribution of shares of total annual income contributed by cash benefits from all public assistance programs, and the next group gives the corresponding percentiles for the annual cash benefits. For each public assistance program with information in SIPP, the remaining rows show the participation rate of the column sample in each program. For programs with income information, indented rows show the average share of annual income attributable to benefits from that program and the average annual payments conditional on participating in the program. For example, considering all immigrant welfare participants (the first column of Table B.2a), 26 percent collected SSI benefits; these benefits represented 31 52 percent of SSI recipient families’ income on average; and the average annual payment to SSI families was $4,851. Income information for GA is not available in CPS. Tables B.3a, B.3b, and B.3c: Income Losses of Immigrant Families Tables B.3a, B.3b, and B.3c present the projected effects of the denial of SSI and Food Stamps benefits to legal immigrants. It is the first table to include the two alternative SIPP calculations: the best case (Table B.3a) and the worst case (Table B.3b). As noted, the essential difference between these two calculations is the method used to compute lost benefits. In all cases, only noncitizen nonrefugees are considered ineligible for benefits. In the CPS data and the SIPP worst case, anyone ineligible loses all benefits received annually. In the SIPP case, the benefits are recalculated using the appropriate formulas for monthly AFDC benefits as well as for SSI and Food Stamps. This results in lower benefit losses but also allows AFDC benefits to replace SSI benefits. Sample: Immigrant families who participated in SSI or Food Stamps. Columns: The columns in these tables clarify the different effects on immigrants depending on which program or programs they participated in: Food Stamps/no SSI: This column indicates families who received Food Stamps benefits but no SSI benefits. These families may also receive benefits from other welfare programs such as AFDC. SSI: This column indicates families who collected benefits from the SSI program. They may or may not have also received Food Stamps or other welfare benefits. Food Stamps &/or SSI: This is the summation of the other two categories. This column counts together families receiving benefits from either or both programs. Rows: These tables have four panels each: Panel A: % of recipient immigrant families potentially affected. The families who potentially lose benefits are only a subset of all immigrant families receiving welfare. This first row relates these affected families to the samples in Tables B.1 and B.2. Panel B: Sources of eligibility. The first two rows in this panel show the share of potentially affected families that include still-eligible family members—either refugees or citizens. (Federal legislation provides exemptions for refugees, which is not the same as a 53 Table B.3a Income Losses of Immigrant Families Affected by Elimination of Food Stamp and SSI Programs: SIPP Data, Best Case Programs Family CharacteristicsFood Stamps/ No SSI SSIFood Stamps and/or SSI A % of recipient immigrant families potentially affected36.0 25.6 61.6 B Sources of eligibility % of families with refugees 11.9 14.6 13.0 % of families with citizens 83.1 56.8 72.1 % of potential group actually losing benefits 93.9 77.3 87.0 C Projected loss of benefits and income (1996$) Average loss of annual Food Stamps benefits 1,295 492 998 Average recovery of annual benefits from AFDC 0 (290) (107) Average loss of annual SSI benefits 0 4,878 1,803 Average loss of total annual benefits 1,295 5,080 2,694 Annual income loss 25th percentile 496 1,850 786 50th percentile 1,083 4,612 1,509 75th percentile 1,876 7,595 3,157 % of families with annual income losses ³ 10 % of total income 35.5 62.0 45.3 ³ 25 % of total income 1.6 46.9 18.3 ³ 50 % of total income 0.0 22.4 8.3 D Change in poverty status Before eligibility denial % above 185% of poverty level 14.4 37.2 22.8 With children 12.3 21.0 15.5 Single 3.7 3.5 3.6 Family head age 65+ 0.0 3.8 1.4 % below poverty level 54.5 13.0 39.1 With children 52.9 7.5 36.1 Single 20.5 1.7 13.6 Family head age 65+ 0.0 4.0 1.5 % below 2/3 of poverty level 8.8 2.9 6.6 With children 8.8 1.7 6.2 Single 2.6 1.7 2.2 Family head age 65+ 0.0 1.2 0.4 After eligibility denial % above 185% of poverty level 11.5 31.8 19.0 With children 10.1 18.4 13.2 Single 2.5 2.2 2.4 Family head age 65+ 0.0 0.9 0.3 % below poverty level 62.7 47.6 57.1 With children 60.9 14.9 43.9 Single 27.1 6.4 19.5 Family head age 65+ 0.0 28.9 10.7 % below 2/3 of poverty level 22.3 28.6 24.6 With children 22.3 6.1 16.3 Single 8.7 1.7 6.1 Family head age 65+ 0.0 20.0 7.4 54 Table B.3b Income Losses of Immigrant Families Affected by Elimination of Food Stamp and SSI Programs: SIPP Data, Worst Case Programs Family CharacteristicsFood Stamps/ No SSI SSIFood Stamps and/or SSI A % of recipient immigrant families potentially affected36.0 25.6 61.6 B Sources of eligibility.. . % of families with refugees 11.9 14.6 13.0 % of families with citizens 83.1 56.8 72.1 % of potential group actually losing benefits 93.9 77.3 87.0 C Projected loss of benefits and income (1996$) Average loss of annual Food Stamps benefits 1,717 611 1,308 Average recovery of annual benefits from AFDC 0 0 0 Average loss of annual SSI benefits 0 4,988 1,844 Average loss of total annual benefits 1,717 5,599 3,152 Annual income loss . . . 25th percentile 593 2,086 867 50th percentile 1,509 5,512 2,072 75th percentile 2,544 7,867 4,004 % of families with annual income losses . . . ³ 10 % of total income 47.3 67.4 54.8 ³ 25 % of total income 10.9 50.1 25.4 ³ 50 % of total income 0.0 28.5 10.5 D Change in poverty status.. . Before eligibility denial . . . % above 185% of poverty level 14.4 37.2 22.8 With children 12.3 21.0 15.5 Single 3.7 3.5 3.6 Family head age 65+ 0.0 3.8 1.4 % below poverty level 54.5 13.0 39.1 With children 52.9 7.5 36.1 Single 20.5 1.7 13.6 Family head age 65+ 0.0 4.0 1.5 % below 2/3 of poverty level 8.8 2.9 6.6 With children 8.8 1.7 6.2 Single 2.6 1.7 2.2 Family head age 65+ 0.0 1.2 0.4 After eligibility denial . . . % above 185% of poverty level 11.5 31.8 19.0 With children 10.1 18.4 13.2 Single 2.5 2.2 2.4 Family head age 65+ 0.0 0.9 0.3 % below poverty level 64.6 50.3 59.3 With children 62.7 17.6 46.1 Single 27.1 7.7 20.0 Family head age 65+ 0.0 28.9 10.7 % below 2/3 of poverty level 28.7 35.0 31.1 With children 28.7 12.3 22.7 Single 12.7 6.4 10.3 Family head age 65+ 0.0 21.5 7.9 55 Table B.3c Income Losses of Immigrant Families Affected by Elimination of Food Stamp and SSI Programs: CPS Data Programs Family CharacteristicsFood Stamps/ No SSI SSIFood Stamps and/or SSI A % of recipient immigrant families potentially affected41.9 22.1 63.9 B Sources of eligibility % of families with refugees 4.3 2.9 3.8 % of families with citizens 74.1 51.9 66.5 % of potential group actually losing benefits 96.5 90.5 94.4 C Projected loss of benefits and income (1996$) Average loss of annual Food Stamps benefits 1,711 474 1,302 Average recovery of annual benefits from AFDC . . . Average loss of annual SSI benefits 0 6,323 2,091 Average loss of total annual benefits 1,711 6,797 3,393 Annual income loss 25th percentile 652 3,174 976 50th percentile 1,396 6,055 2,088 75th percentile 2,545 8,002 4,063 % of families with annual income losses ³ 10 % of total income 55.9 80.1 63.9 ³ 25 % of total income 21.4 58.5 33.7 ³ 50 % of total income 7.5 39.6 18.1 D Change in poverty status Before eligibility denial % above 185% of poverty level 7.7 25.2 13.5 With children 5.1 12.7 7.7 Single 1.8 4.1 2.6 Family head age 65+ 0.0 3.5 1.2 % below poverty level 64.5 28.7 52.7 With children 52.7 11.7 38.6 Single 22.5 4.0 16.1 Family head age 65+ 0.7 6.8 2.8 % below 2/3 of poverty level 33.1 10.2 25.5 With children 25.0 4.9 18.1 Single 10.2 2.2 7.5 Family head age 65+ 0.5 2.7 1.2 After eligibility denial % above 185% of poverty level 7.4 19.9 11.5 With children 4.7 11.3 7.0 Single 1.5 3.9 2.3 Family head age 65+ 0.0 1.3 0.4 % below poverty level 70.1 61.4 67.2 With children 57.2 19.8 44.3 Single 23.6 6.0 17.5 Family head age 65+ 0.7 23.9 8.7 % below 2/3 of poverty level 48.8 48.0 48.6 With children 39.2 14.8 30.8 Single 16.4 5.2 12.5 Family head age 65+ 0.6 19.1 7.0 56 family immigrating from a refugee-sending country, but SIPP does not designate refugee status.) Refugee-sending countries identifiable in the SIPP are: Cuba, Czechoslovakia, Hungary, Poland, the U.S.S.R., and Vietnam. The final row lists the share of families who would actually lose benefits after accounting for all eligible family members. Panel C: Projected loss of benefits and income (1996$). These rows summarize the annual income losses projected if all affected families identified in the last row of panel B lose Food Stamps and SSI. As noted above, Table B.3a shows the income loss when benefits are recalculated on a monthly basis in the SIPP. Tables B.3b and B.3c remove all benefits received in a year by ineligible persons. The first set presents the average losses for Food Stamps, AFDC, SSI, and total annual cash benefits. The second set of rows reports the 25th, 50th, and 75th percentiles of the distribution of the annual dollar loss in benefits across affected families. The third set lists the share of families losing more than 10, 25, or 50 percent of their total annual family income. Panel D: Income status. This panel is divided into two parts: The top portion reports family income status before eligibility denial from these programs, and the bottom projects family income status after eligibility denial. For each portion, the results give the percentage of affected families who had incomes above 185 percent of the poverty level, below the poverty level, and below two-thirds of the poverty level. Within these subgroups, the table reports the percentage of families who fall into the poverty level category and who have children, are single parents, or have a family head over age 65. For example, before eligibility denial, 21 percent of affected SSI families were families with children and with incomes over 185 percent of the poverty level, whereas 4 percent were families headed by an elderly person living below the poverty line. Tables B.4a, B.4b, and B.4c: Income Status After Elimination of Food Stamp and SSI Programs These tables consider the other resources available to immigrant families if they were denied Food Stamps and SSI. As in Tables B.3a, B.3b, and B.3c, they include the two alternative SIPP calculations: the best case (Table B.4a) and the worst case (Table B.4b). Sample: Immigrant families who lose SSI or Food Stamps. Columns: The columns in this table are the same as those in Tables B.3a, B.3b, and B.3c. 57 Table B.4a Income Status of Immigrant Families After Elimination of Food Stamp and SSI Programs: SIPP Data, Best Case Programs Family CharacteristicsFood Stamps/ No SSI SSIFood Stamps and/or SSI A Welfare utilization (1996$) % receiving benefits 86.1 50.8 73.1 % of income from cash benefits 25th percentile 4.5 0.0 0.0 50th percentile 42.5 0.6 18.4 75th percentile 100.0 52.4 98.3 Annual cash benefits 25th percentile 1,269 0 0 50th percentile 7,391 166 3,790 75th percentile 12,325 6,076 11,135 % AFDC participants 64.3 37.0 54.2 % of income from AFDC 50.8 36.1 47.1 Annual AFDC benefits 6,192 5,068 5,908 % SSI participants 0.0 24.3 9.0 % of income from SSI . 6.9 6.9 Annual SSI benefits . 1,387 1,387 % Food Stamps participants 57.9 20.0 43.9 % of income from Food Stamps 5.4 4.1 5.2 Annual Food Stamps benefits 730 654 718 % GA 6.1 8.9 7.1 % of income from GA 14.2 3.9 9.4 Annual GA benefits 2,232 616 1,482 % other welfare 97.0 82.7 91.7 % Medicaid participants 94.9 100.0 96.8 B Income (1996$) % of families with earnings 82.0 57.7 73.0 Annual earnings 25th percentile 5,586 11,782 6,336 50th percentile 11,536 26,850 13,924 75th percentile 19,842 47,995 30,520 % of income from earnings 25th percentile 38.8 49.7 44.3 50th percentile 69.5 84.7 75.3 75th percentile 88.7 91.5 90.5 % of families with other income 87.1 80.3 84.6 Annual other income 25th percentile 896 1,493 1,101 75th percentile 15,668 10,256 12,736 % of income from other income 25th percentile 5.4 7.9 6.0 75th percentile 74.7 100.0 90.6 58 Table B.4b Income Status of Immigrant Families After Elimination of Food Stamp and SSI Programs: SIPP Data, Worst Case Programs Family CharacteristicsFood Stamps/ No SSI SSIFood Stamps and/or SSI A Welfare utilization (1996$) % receiving benefits 76.3 40.5 63.1 % of income from cash benefits . . . 25th percentile 1.4 0.0 0.0 50th percentile 42.5 0.0 17.9 75th percentile 100.0 45.1 98.2 Annual cash benefits . . . 25th percentile 303 0 0 50th percentile 6,928 0 3,189 75th percentile 11,960 3,772 10,364 % AFDC participants 64.3 22.1 48.7 % of income from AFDC 54.0 60.5 55.1 Annual AFDC benefits 6,192 7,196 6,361 % SSI participants 0.0 15.3 5.7 % of income from SSI . 8.8 8.8 Annual SSI benefits . 1,478 1,478 % Food Stamps participants 1.0 1.2 1.1 % of income from Food Stamps 0.8 3.5 1.9 Annual Food Stamps benefits 103 951 458 % GA 6.1 8.9 7.1 % of income from GA 15.1 4.5 10.2 Annual GA benefits 2,232 616 1,482 % other welfare 97.0 82.7 91.7 % Medicaid participants 94.9 100.0 96.8 B Income (1996$).. . % of families with earnings 82.0 57.7 73.0 Annual earnings . . . 25th percentile 5,586 11,782 6,336 50th percentile 11,536 26,850 13,924 75th percentile 19,842 47,995 30,520 % of income from earnings . . . 25th percentile 38.8 52.2 44.4 50th percentile 70.5 84.7 76.0 75th percentile 90.6 92.9 91.6 % of families with other income 87.1 80.3 84.6 Annual other income . . . 25th percentile 896 1,493 1,101 75th percentile 15,668 10,256 12,736 % of income from other income . . . 25th percentile 5.4 8.1 6.1 75th percentile 74.7 100.0 94.3 59 Table B.4c Income Status of Immigrant Families After Elimination of Food Stamp and SSI Programs: CPS Data Programs Family CharacteristicsFood Stamps/ No SSI SSIFood Stamps and/or SSI A Welfare utilization (1996$) % receiving benefits 58.2 28.5 48.4 % of income from cash benefits 25th percentile 0.0 0.0 0.0 50th percentile 22.2 0.0 7.6 75th percentile 95.1 31.3 80.2 Annual cash benefits 25th percentile 0 0 0 50th percentile 2,520 0 0 75th percentile 7,697 1,305 6,393 % AFDC participants 58.2 22.6 46.4 % of income from AFDC 65.1 57.4 63.8 Annual AFDC benefits 7,197 6,538 7,091 % SSI participants 0.0 3.9 1.3 % of income from SSI . 25.7 25.7 Annual SSI benefits . 5,952 5,952 % Food Stamps participants 0 0 0 % of income from Food Stamps . . . Annual Food Stamps benefits . . . % GA . . . % of income from GA . . . Annual GA benefits . . . % other welfare 66.7 53.8 62.5 % Medicaid participants 46.0 83.9 58.5 B Income (1996$) % of families with earnings 62.6 44.6 56.6 Annual earnings 25th percentile 5,219 5,971 5,290 50th percentile 10,873 19,561 12,354 75th percentile 18,531 43,888 21,794 % of income from earnings 25th percentile 52.3 39.4 50.8 50th percentile 81.9 80.6 81.8 75th percentile 100.0 99.5 100.0 % of families with other income 36.7 62.9 45.4 Annual other income 25th percentile 927 1,371 1,023 75th percentile 5,250 8,303 6,997 % of income from other income 25th percentile 6.5 12.8 7.6 75th percentile 44.6 100.0 82.3 60 Food Stamps/no SSI: This column indicates families who received Food Stamps benefits but no SSI benefits. SSI: This group includes families who collected benefits from the SSI program. Food Stamps &/or SSI: This is the summation of the other two categories. Rows: These tables have two panels each. Panel A: Welfare utilization (1996$). The rows in this panel are identical to those in Table B.2, panel C, except for the first row. In this table, the first row lists the share of affected families who would still receive public assistance even without SSI or Food Stamps. The income shares are adjusted to reflect the benefits lost due to the ineligibility of some or all family members. Note that the last row of this panel lists Medicaid participation. This calculation does not reflect the possibility of losing Medicaid through the loss of categorical eligibility through SSI. Panel B: Income (1996$): % of families with earnings: As in Table B.2, this row shows the percentage of immigrant families who lost benefits (as designated by the column heading) but who had earnings. The indented rows under this subheading are conditional on families having earnings. They first report the 25th, 50th, and 75th percentiles of the distribution of annual earnings obtained by working families making up the relevant sample. The next set of rows shows the 25th, 50th, and 75th percentiles of the distribution of total income shares provided by family earnings after elimination of Food Stamps and SSI. % of families with other income: These rows are structured in a similar fashion to the preceding rows. Here, the first row shows the percentage of immigrant welfare families who lost benefits but had “other” income such as Social Security, pensions, or Unemployment Insurance. The indented rows under this subheading are conditional on families having other income. They first report the 25th and 75th percentiles of the distribution of annual other income obtained by families in the column category who had other income. The next set of rows shows the 25th and 75th percentiles of the distribution of total income shares provided by other income after elimination of Food Stamps and SSI. 61 Tables B.5a and B.5b: Share of Reinstated Welfare Benefits Tables B.5a and B.5b count shares of dollars instead of shares of families. They consider the total benefits lost in the three scenarios of Table B.3 and the distribution of those dollars to various categories of families. Because of the similarity of results, only the “best” case for SIPP is included in this and the remaining tables in this appendix. Sample: Projected benefits lost by immigrant families denied SSI or Food Stamps. Columns: This table splits the reinstated dollars into the three family categories used for the last two tables: Food Stamps/no SSI: This indicates benefits theoretically reinstated to families who previously received Food Stamps benefits but no SSI benefits. SSI: This indicates benefits theoretically reinstated to families who participated in SSI, with or without Food Stamps. Food Stamps &/or SSI: This column is the summation of the other two categories. This column counts benefits reinstated to all immigrant families who would have lost benefits from either or both programs. Rows: These tables have five panels each. The first lists the total benefits reinstated, and the remaining four show the distribution of these benefits to different groups. Panel A: Total benefit loss/reinstated benefits (millions of 1996$). This row lists the costs of reinstating benefits for both programs to the three categories of families. Thus, under the SIPP best case, $1,060 million would be reinstated to SSI families if both their SSI and their Food Stamps benefits were restored. The total cost of restoring both programs would be $1,520 million. Panel B: Share of reinstated benefits by income quintile. This panel shows how restoration of the benefits listed in panel A would be distributed by income quintiles of the recipient families. The quintiles order all California families (immigrant and citizen) by their family income and split them into five equal-sized groups. The lowest income quintile, therefore, includes the 20 percent of California families with the lowest income. If the $460 million were split evenly across all California families, 20 percent would go to each quintile. However, 6 percent of the benefits restored to Food Stamps families who did not receive SSI would go to families in the bottom income quintile. Panel C: Share of reinstated benefits by poverty level. This panel shows how restoration of the benefits listed in panel A would be distributed based on the poverty level 62 Table B.5a Shares of Reinstated Welfare Benefits Going to Various Categories of Immigrant Families in California: SIPP Data Programs Family CharacteristicsFood Stamps/ No SSI SSIFood Stamps and/or SSI A Total benefit loss/reinstated benefits (millions of 1996$)460 1,060 1,520 B Share of reinstated benefits by income quintiles Lowest 66.4 63.4 64.3 Mid-low 21.1 11.7 14.6 Middle 8.0 6.3 6.8 Mid-high 4.1 13.5 10.6 Highest 0.4 5.1 3.7 C Share of reinstated benefits by poverty level < 0.66 35.7 47.8 44.1 0.66–1 43.6 18.7 26.2 1–1.85 15.3 10.8 12.2 > 1.85 5.5 22.7 17.5 D Share allocation by poverty level Income below 2/3 poverty level 35.7 47.8 44.1 With children < age 18 35.7 9.0 17.1 Single 12.7 0.7 4.3 Family head age 65+ . 35.5 24.8 Single . 12.5 8.7 Income below poverty level 79.2 66.5 70.3 With children < age 18 78.6 18.2 36.5 Single 33.3 5.6 14.0 Family head age 65+ . 43.5 30.3 Single . 15.4 10.7 Income from 1–1.85 of poverty level 15.3 10.8 12.2 With children < age 18 14.5 6.2 8.7 Single 4.2 0.9 1.9 Family head age 65+ 0.4 3.5 2.6 Single . 0.5 0.3 Income above 1.85 of poverty level 5.5 22.7 17.5 With children < age 18 5.5 15.7 12.6 Single 0.7 1.5 1.2 Family head age 65+ . 0.3 0.2 Single . . . E Shares received by AFDC recipients All AFDC families 77.2 32.0 45.7 Single 36.2 6.9 15.8 63 Table B.5b Shares of Reinstated Welfare Benefits Going to Various Categories of Immigrant Families in California: CPS Data Programs Family CharacteristicsFood Stamps/ No SSI SSIFood Stamps and/or SSI A Total benefit loss/reinstated benefits (millions of 1996$)601 1,179 1,780 B Share of reinstated benefits by income quintiles Lowest 56.5 62.4 60.4 Mid-low 31.6 14.1 20.0 Middle 8.8 6.4 7.2 Mid-high 2.8 11.4 8.5 Highest 0.4 5.8 4.0 C Share of reinstated benefits by poverty level < 0.66 56.7 60.6 59.3 0.66–1 21.3 9.1 13.2 1–1.85 17.5 12.5 14.2 > 1.85 4.4 17.8 13.3 D Share allocation by poverty level Income below 2/3 poverty level 56.7 60.6 59.3 With children < age 18 53.3 21.5 32.3 Single 19.6 6.8 11.2 Family head age 65+ 0.5 26.4 17.7 Single 0.4 11.9 8.0 Income below poverty level 78.0 69.7 72.5 With children < age 18 73.5 27.0 42.7 Single 26.8 8.2 14.4 Family head age 65+ 0.5 28.5 19.1 Single 0.4 12.8 8.6 Income from 1–1.85 of poverty level 17.5 12.5 14.2 With children < age 18 16.4 7.3 10.4 Single 4.6 0.3 1.8 Family head age 65+ 0.4 4.0 2.8 Single 0.2 0.8 0.6 Income above 1.85 of poverty level 4.4 17.8 13.3 With children < age 18 3.5 8.4 6.8 Single 0.9 2.9 2.2 Family head age 65+ 0.0 0.9 0.6 Single 0.0 0.0 0.0 E Shares received by AFDC recipients All AFDC families 68.3 24.3 39.2 Single 27.3 6.4 13.4 64 of the families in the absence of these benefits. For example, out of the $460 million potentially restored to families who previously received Food Stamps but not SSI, 35.7 percent would go to families with income below two-thirds of the poverty level. Panel D: Share allocation by poverty level. This panel provides the same information as panel C above, but in greater detail. For example, both panels C and D show in the second column that 22.7 percent of SSI and Food Stamps benefits restored to SSI families goes to families over 185 percent of the poverty level (without these benefits). Panel D shows that 15.7 percent of the $1,060 million would go to families over 185 percent of the poverty level who have children under age 18. As in all the tables, indented rows indicate that the statistics are conditional on belonging to the previous group. Panel E: Shares received by AFDC recipients. Many of the SSI and Food Stamps families also received AFDC. The final panel in this table shows what share of reinstated benefit dollars would go to families on AFDC and to single parent families on AFDC. Tables B.6a and B.6b: Shares of Reinstated SSI Benefits Going to Various Categories of Immigrant Families in California Tables B.6a and B.6b calculate the value of SSI benefits restored to California families when the 1997 budget deal undid these feature of the PRWORA. These tables detail the distribution of these benefits to various categories of California families. Like Table B.5, these tables count shares of dollars instead of shares of families. Sample: Projected benefits lost by immigrant families potentially denied SSI. Columns: These tables show the reinstated SSI benefit dollars split in two ways. First, they show the distribution of benefits by income level based on whether or not Food Stamps benefits are also restored. Second, they break the total SSI benefits into benefits restored to the disabled and benefits restored to the elderly. Accordingly, there are four relevant categories when evaluating the distribution of SSI program dollars: SSI alone: This category counts all SSI benefits reinstated to immigrant families, assuming that no Food Stamps benefits are restored. SSI after Food Stamps: This column counts all SSI benefits reinstated for immigrant families, but assumes that Food Stamps benefits have already been restored. The distinction between this column and the first column matters only for the distribution of benefits by income level. 65 Table B.6a Shares of Reinstated SSI Benefits Going to Various Categories of Immigrant Families in California: SIPP Data Programs Family CharacteristicsSSI AloneSSI After Food StampsSSI DisabledSSI Elderly A Total value of reinstated benefits (millions of 1996$)1,020 1,020 272 735 B Share of reinstated benefits by income quintiles Lowest 65.0 61.2 62.8 67.3 Mid-low 9.4 13.2 12.7 7.2 Middle 5.9 5.9 9.7 3.5 Mid-high 14.4 14.4 11.9 15.7 Highest 5.3 5.3 2.9 6.2 C Share of reinstated benefits by poverty level < 0.66 47.2 45.0 48.6 44.1 0.66–1 18.6 20.8 25.4 20.2 1–1.85 10.4 10.4 5.2 11.5 > 1.85 23.8 23.8 20.8 24.2 D Share allocation by poverty level Income below 2/3 poverty level 47.2 45.0 48.6 44.1 With children < age 18 6.8 4.7 25.5 0.4 Single 0.3 0.3 1.0 0.4 Family head age 65+ 37.0 37.0 10.4 . Single 13.0 13.0 0.0 . Income below poverty level 65.8 65.8 74.0 64.3 With children < age 18 15.6 15.6 43.2 5.9 Single 6.1 6.1 14.1 3.6 Family head age 65+ 45.8 45.8 10.4 59.8 Single 16.5 16.5 0.0 22.8 Income from 1–1.85 of poverty level 10.4 10.4 5.2 11.5 With children < age 18 5.7 5.7 3.6 6.3 Single 1.2 1.2 0.0 1.6 Family head age 65+ 4.0 4.0 0.6 4.8 Single 0.6 0.6 0.0 0.9 Income above 1.85 of poverty level 23.8 23.8 20.8 24.2 With children < age 18 16.7 16.7 11.2 18.9 Single 1.5 1.5 0.0 2.0 Family head age 65+ 0.3 0.3 0.0 . Single . . . . E Shares received by disabled recipients All families with disabled members 50.2 50.2 100.0 31.0 With children < age 18 3.2 3.2 57.9 6.3 Family head age 65+ 26.6 26.6 10.9 25.6 66 Table B.6b Shares of Reinstated SSI Benefits Going to Various Categories of Immigrant Families in California: CPS Data Programs Family CharacteristicsSSI AloneSSI After Food StampsSSI DisabledSSI Elderly A Total value of reinstated benefits (millions of 1996$)1,097 1,097 571 579 B Share of reinstated benefits by income quintiles Lowest 63.6 62.5 68.0 60.9 Mid-low 12.2 12.3 9.0 15.9 Middle 5.9 6.8 4.1 7.0 Mid-high 12.1 12.1 8.0 15.0 Highest 6.2 6.2 10.8 1.2 C Share of reinstated benefits by poverty level < 0.66 61.0 59.6 63.2 60.8 0.66–1 8.2 9.6 7.8 7.8 1–1.85 11.9 11.9 7.1 17.3 > 1.85 18.9 18.9 22.0 14.1 D Share allocation by poverty level Income below 2/3 poverty level 61.0 59.6 63.2 60.8 With children < age 18 19.8 18.3 3.5 34.0 Single 6.1 6.1 0.4 11.2 Family head age 65+ 28.3 28.3 54.3 3.3 Single 12.7 12.7 24.4 0.7 Income below poverty level 69.2 69.2 71.0 68.6 With children < age 18 24.2 24.2 7.0 38.9 Single 7.5 7.5 1.2 13.0 Family head age 65+ 30.5 30.5 58.6 3.3 Single 13.6 13.6 26.2 0.7 Income from 1–1.85 of poverty level 11.9 11.9 7.1 17.3 With children < age 18 6.5 6.5 0.9 11.5 Single 0.4 0.4 0.0 0.7 Family head age 65+ 3.9 3.9 6.8 2.4 Single 0.8 0.8 1.5 0.0 Income above 1.85 of poverty level 18.9 18.9 22.0 14.1 With children < age 18 8.9 8.9 11.1 5.9 Single 3.1 3.1 3.4 2.5 Family head age 65+ 1.0 1.0 2.0 0.0 Single 0.0 0.0 0.1 0.0 E Shares received by disabled recipients All families with disabled members . . . . With children < age 18 . . . . Family head age 65+ . . . . 67 SSI disabled: This category counts SSI benefits restored to disabled recipients. The disabled are self-identified in the SIPP; the CPS does not ask disability status. SSI elderly: This column counts SSI benefits restored to elderly recipients. The disabled and elderly columns do not necessarily add to the full reinstated benefits. Some cases in SIPP include neither an elderly recipient nor one who reports a disability. In the CPS, families without elderly recipients are assumed to include disabled recipients. Rows: Tables B.6a and B.6b have five panels. The first lists the total benefits reinstated, and the remaining four show the distribution of these benefits to different groups. Panel A: Total value of reinstated benefits (millions of 1996$). This row lists the costs of reinstating benefits for SSI recipients in California. Since eligibility for SSI or Food Stamps is not dependent on the other, the SSI benefits reinstated are the same whether or not Food Stamps are also restored. According to the SIPP version of the table, restoring SSI in California accounts for $1,020 million. Panel B: Share of reinstated benefits by income quintile. This panel shows how restoration of the benefits listed in panel A is to be distributed by income quintiles of the recipient families. As in Table B.5, the quintiles order all California families (immigrant and citizen) by their family income and split them into five equal-sized groups. The lowest income quintile, therefore, includes the 20 percent of California families with the lowest income. These families receive 65 percent of reinstated benefits. If Food Stamps had been restored first, some of these families would no longer be in the bottom income quintile, so the share of reinstated SSI going to families at the bottom of the income distribution would fall to 61.2 percent. Panel C: Share of reinstated benefits by poverty level. This panel shows how restoration of the benefits listed in panel A is to be distributed based on the poverty level of the families in the absence of these benefits. For example, out of the $1,020 million in restored SSI benefits, 47.2 percent would go to families with incomes below two-thirds of the poverty level. If SSI benefits were restored after Food Stamps, 45 percent of the SSI benefits would go to families below two-thirds of poverty. Panel D: Share allocation by poverty level. This panel provides the same information as panel C above, but in greater detail. For example, both panels C and D show in the second column that 23.8 percent of restored SSI benefits go to families over 68 185 percent of the poverty level (without these benefits). Panel D shows that 16.7 percent of the $1,020 million would go to families over 185 percent of the poverty level who have children under age 18. As in all the tables, indented rows indicate that the statistics are conditional on belonging to the previous group. Panel E: Shares received by disabled recipients. SIPP reports on disability. This last panel shows the share of restored SSI benefits that goes to families with disabled members. Just over half of SSI benefits would go to families with a disabled member. The order of benefit restoration has no effect on the share of benefits going to disabled recipients. This detail on disabled family members is not available in the CPS data. Tables B.7a and B.7b: Shares of Reinstated Food Stamps Benefits Going to Various Categories of Immigrant Families in California In parallel with Tables B.6a and B.6b, Tables B.7a and B.7b calculate the value of Food Stamps benefits potentially restored to California families. These tables detail the distribution of benefits to various categories of California families under four different scenarios. Like the last two tables, Tables B.7a and B.7b count shares of dollars instead of shares of families. Sample: Projected benefits lost by immigrant families potentially denied Food Stamps. Columns: These tables show the value of reinstating Food Stamps benefit dollars to all or some of the immigrants denied Food Stamps under the new federal rules. The first two columns show the effect of restoring all Food Stamps. The last two columns show the benefits to the two groups who will continue to receive benefits in California: children and the elderly. To detail the four columns: Food Stamps alone: This column counts all Food Stamps benefits reinstated to immigrant families, assuming that no SSI benefits have been restored. Food Stamps after SSI: This counts all Food Stamps benefits potentially restored to immigrant families, but assumes that SSI benefits have already been restored. The distinction between this column and the first column matters only for the distribution of benefits by income level. 69 Table B.7a Shares of Reinstated Food Stamps Benefits Going to Various Categories of Immigrant Families in California: SIPP Data Programs Family CharacteristicsFood Stamps AloneFood Stamps After SSIFood Stamps to Children (After SSI Disabled)Food Stamps to Elderly (After SSI Disabled) A Total value of reinstated benefits (millions of 1996$)561 561 134 15 B Share of reinstated benefits by income quintiles Lowest 62.1 56.8 61.1 64.8 Mid-low 24.9 30.2 32.4 14.2 Middle 8.5 8.5 4.4 0.0 Mid-high 4.0 4.0 2.1 19.6 Highest 0.5 0.5 0.0 1.4 C Share of reinstated benefits by poverty level < 0.66 36.8 30.5 37.4 35.1 0.66–1 41.5 44.4 49.2 42.4 1–1.85 16.4 19.8 11.2 12.4 > 1.85 5.3 5.3 2.1 10.0 D Share allocation by poverty level Income below 2/3 poverty level 36.8 30.5 37.4 35.1 With children < age 18 36.8 30.5 37.4 35.1 Single 11.7 11.7 19.3 8.5 Family head age 65+ 0.0 0.0 0.0 . Single 0.0 0.0 0.0 . Income below poverty level 78.3 74.9 86.7 77.6 With children < age 18 77.7 74.4 86.7 76.2 Single 30.6 28.6 39.0 36.9 Family head age 65+ 0.6 0.0 0.0 . Single 0.6 0.0 0.0 . Income from 1–1.85 of poverty level 16.4 19.8 11.2 12.4 With children < age 18 15.5 18.8 11.2 12.4 Single 4.1 6.1 5.0 1.4 Family head age 65+ 0.7 1.3 0.0 3.7 Single 0.3 0.9 0.0 1.4 Income above 1.85 of poverty level 5.3 5.3 2.1 10.0 With children < age 18 5.0 5.0 2.1 8.6 Single 0.7 0.7 0.3 0.0 Family head age 65+ 0.0 0.0 0.0 . Single . 0.0 . . E Shares received by disabled recipients All families with disabled members 35.3 35.3 40.7 30.4 With children < age 18 7.4 7.4 40.7 30.4 Family head age 65+ 3.0 3.0 0.0 2.2 70 Table B.7b Shares of Reinstated Food Stamps Benefits Going to Various Categories of Immigrant Families in California: CPS Data Programs Family CharacteristicsFood Stamps AloneFood Stamps After SSIFood Stamps to Children (After SSI Disabled)Food Stamps to Elderly (After SSI Disabled) A Total value of reinstated benefits (millions of 1996$)683 683 217 16 B Share of reinstated benefits by income quintiles Lowest 55.2 54.1 66.5 67.9 Mid-low 32.5 31.4 26.3 24.3 Middle 9.3 11.5 6.8 7.8 Mid-high 2.7 2.7 0.4 0.0 Highest 0.4 0.4 0.0 0.0 C Share of reinstated benefits by poverty level < 0.66 56.6 52.9 69.6 56.0 0.66–1 21.3 22.3 20.5 15.9 1–1.85 17.8 20.6 8.8 18.2 > 1.85 4.2 4.2 1.1 10.0 D Share allocation by poverty level Income below 2/3 poverty level 56.6 52.9 69.6 56.0 With children < age 18 52.3 49.6 68.5 37.6 Single 19.3 18.9 32.1 28.9 Family head age 65+ 0.6 0.4 0.0 3.1 Single 0.6 0.3 0.0 3.1 Income below poverty level 77.9 75.2 90.1 71.9 With children < age 18 72.4 70.4 89.0 53.5 Single 25.6 25.4 39.0 29.9 Family head age 65+ 0.7 0.4 0.0 3.1 Single 0.6 0.4 0.0 3.1 Income from 1–1.85 of poverty level 17.8 20.6 8.8 18.2 With children < age 18 16.6 18.6 8.4 6.5 Single 4.0 4.2 2.2 3.5 Family head age 65+ 0.9 1.2 0.3 10.9 Single 0.3 0.5 0.1 8.4 Income above 1.85 of poverty level 4.2 4.2 1.1 10.0 With children < age 18 3.3 3.3 0.7 0.0 Single 0.8 0.8 0.2 0.0 Family head age 65+ 0.0 0.0 0.0 0.0 Single 0.0 0.0 0.0 0.0 E Shares received by disabled recipients All families with disabled members . . . . With children < age 18 . . . . Family head age 65+ . . . . 71 Food Stamps to children (after SSI disabled): This category counts Food Stamps payments restored to children under age 18. We specify “after SSI disabled” benefits are restored, since without SSI some disabled SSI children might have moved onto the Food Stamps rolls. These calculations assume no new cases due to such switching. Food Stamps to elderly (after SSI disabled): This counts Food Stamps benefits restored to recipients age 65 and older. Once again, we specify “after SSI disabled” benefits are restored, since without SSI some recipients might have moved onto the Food Stamps rolls. These calculations assume no new cases due to such switching. After accounting for SSI, there are very few elderly Food Stamps recipients. Rows: Tables B.7a and B.7b have five panels each. The first lists the total benefits reinstated, and the remaining four show the distribution of these benefits to different groups. Panel A: Total value of reinstated benefits (millions of 1996$). This row lists the costs of reinstating benefits for Food Stamps recipients in California. Since eligibility for SSI or Food Stamps is not dependent on the other, the Food Stamps benefits reinstated are the same whether or not SSI has also been restored. According to the SIPP version of the table, restoring Food Stamps to all California immigrant families would cost $561 million. Panel B: Share of reinstated benefits by income quintile. This panel shows how restoration of the benefits listed in panel A are to be distributed by income quintiles of the recipient families. As in Tables B.6a and B.6b, the quintiles order all California families (immigrant and citizen) by their family income and split them into five equal-sized groups. The lowest income quintile, therefore, includes the 20 percent of California families with the lowest income. Restoring benefits to children gives 61.1 percent of reinstated benefits to this lowest income group. Panel C: Share of reinstated benefits by poverty level. This panel shows how restoration of the benefits listed in panel A are to be distributed based on the poverty level of the families in the absence of these benefits. For example, out of the $561 million in total restored Food Stamps benefits, 36.8 percent would go to families with income below two-thirds of the poverty level. Given the restoration of SSI, only 30.5 percent of the Food Stamps benefits would go to families below two-thirds of poverty. 72 Panel D: Share allocation by poverty level. As in the last two tables, this panel provides the same information as panel C above, but in greater detail. For example, both panels C and D show in the third column that 86.7 percent of restored Food Stamps benefits to children go to families below poverty (without these benefits). Panel D shows that 39 percent of the $134 million would go to single parent families in poverty. As in all the tables, indented rows indicate that the statistics are conditional on belonging to the previous group. Panel E: Share Received by Disabled Recipients. SIPP reports on disability. This last panel shows the share of restored Food Stamps benefits that goes to families that report at least one disabled member. Although disability is not a factor in Food Stamps eligibility, 35.3 percent of all Food Stamps benefits would go to immigrant families with a disabled family member. The order of benefit restoration has no effect on the share of benefits going to disabled recipients. This detail on disabled family members is not available in the CPS data. 73 Bibliography Borjas, George J., “The Economics of Immigration,” Journal of Economic Literature, Vol. 32, 1994, pp. 1667–1717. Citro, Constance F., and Graham Kalton (eds.), The Future of the Survey of Income and Program Participation, National Academy Press, Washington, D.C., 1993. Citro, Constance, F., and Robert T. Michael (eds.), Measuring Poverty: A New Approach, National Resarch Council, National Academy Press, Washington, D.C., 1995. Fix, Michael, and Jeffrey S. Passel, Immigration and Immigrants: Setting the Record Straight, The Urban Institute, Washington, D.C., 1994. Fix, Michael, and Jeffrey S. Passel, “ Setting the Record Straight: What Are the Costs to the Public,” Public Welfare, Vol. 52, No. 2, 1995, pp. 6–20. Fix, Michael, Jeffrey S. Passel, and Wendy Zimmerman, “The Use of SSI and Other Welfare Programs by Immigrants,” Testimony before the House Ways and Means Committee, May 23, 1996. Downloaded 11/24/1997 from The Urban Institute: http://www.urban.org/TESTIMON/fix.htm. Fix, Michael, and Karen Tumlin, “Welfare Reform and the Devolution of Immigrant Policy,” New Federalism: Issues and Options for States, Series A, No. A-15, The Urban Institute, Washington, D.C., October 1997. MaCurdy, Thomas, and Margaret O’Brien-Strain, Who Will Be Affected by Welfare Reform in California? Public Policy Institute of California, San Francisco, California, 1996. Matloff, Norman, “From ‘Jiu Ji Jin’ to ‘Fu Li Jin’: Some Chinese Immigrants Mistakenly See Welfare as a Fringe Benefit,”The New Democrat, November 1994. Migration News, Vol. 3, No. 10, October 1996. Ross, Jane L., Supplemental Security Income: Noncitizens Have Been a Major Source of Caseload Growth, GAS/T-HEHS-96-88, Government Accounting Office, Washington, D.C., February 6, 1996. 74 Smith, James, and Barry Edmonston (eds.), The New Americans: Economic, Demographic and Fiscal Effects of Immigration, National Academy Press, Washington, D.C., 1997. U.S. Department of Commerce, Bureau of the Census, “The Foreign Born Population: 1994,” Current Population Reports, August 1995. U.S. Department of Justice, Immigration and Naturalization Service, INS Fact Book: Summary of Recent Immigration Data, August 1995. 75 About the Authors THOMAS E. MACURDY Thomas MaCurdy, an adjunct fellow of the Public Policy Institute of California, is a professor of economics at Stanford University and a senior fellow at the Hoover Institution. His principal research explores topics in the areas of income transfer programs, human resources, and labor markets. His recent studies investigate the consequences of governmental policies underlying welfare programs, unemployment compensation, Social Security, Medicare, Medicaid, and various forms of public assistance for low-income populations. These studies analyze the work disincentives and the determinants of participation induced by policies, along with the characteristics and behavioral patterns of beneficiaries. MaCurdy received an A.B. from the University of Washington and a Ph.D. in economics from the University of Chicago. MARGARET O’BRIEN-STRAIN Margaret O'Brien-Strain is a research fellow at the Public Policy Institute of California. Before coming to PPIC, she spent three years at Mathematics Policy Research, Inc., where she worked on studies of child care, welfare, and adult literacy. A recipient of a National Science Foundation fellowship, she is interested in a broad range of public assistance questions and has written about the minimum wage, the Earned Income Tax Credit, immigrants’ use of welfare, the combined effects of public assistance programs, and welfare reform in California. O’Brien-Strain received a B.A. in economics from Swarthmore College and will soon complete a Ph.D. in economics at Stanford." } ["___content":protected]=> string(104) "

R 1298TMR

" ["_permalink":protected]=> string(123) "https://www.ppic.org/publication/reform-reversed-the-restoration-of-welfare-benefits-to-immigrants-in-california/r_1298tmr/" ["_next":protected]=> array(0) { } ["_prev":protected]=> array(0) { } ["_css_class":protected]=> NULL ["id"]=> int(8077) ["ID"]=> int(8077) ["post_author"]=> string(1) "1" ["post_content"]=> string(0) "" ["post_date"]=> string(19) "2017-05-20 02:34:43" ["post_excerpt"]=> string(0) "" ["post_parent"]=> int(3160) ["post_status"]=> string(7) "inherit" ["post_title"]=> string(9) "R 1298TMR" ["post_type"]=> string(10) "attachment" ["slug"]=> string(9) "r_1298tmr" ["__type":protected]=> NULL ["_wp_attached_file"]=> string(13) "R_1298TMR.pdf" ["wpmf_size"]=> string(6) "216583" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(134260) "iii Foreword Designing an efficient and effective social safety net is one of the most intractable aspects of pubic policy. For example, simple rules for eligibility are easy to enforce but highly inefficient because they tend to provide benefits to some individuals and families not really in need. More complex rules that isolate only those most deserving are difficult to administer and often create expensive bureaucratic nightmares. The 1996 welfare reform legislation included many sweeping changes, including time limits and restricted eligibility for benefits. In the wake of this wave of reform, however, it became clear that some of the most draconian decisions were not politically feasible. Subsequent legislation has already undone some of the 1996 reforms. In this report, Thomas MaCurdy and Margaret O’Brien-Strain examine one of the most controversial elements of the legislation—the denial and subsequent reinstatement of benefits for noncitizen immigrants under the federal Supplemental Security Income (SSI) and Food Stamp programs. The authors argue that both the original decision and its reversal did not meet the most important goals of welfare reform, which were to meet the needs of the poorest families while promoting self-sufficiency and family responsibility. Taking away all SSI and Food Stamps benefits would have plunged two-thirds of immigrant recipients into poverty. However, in simply reversing the reforms and reinstating the benefits, legislators missed a second opportunity to design a program that efficiently targets the benefits to those truly in need. Under the reinstated program, one in four SSI dollars goes to families with incomes substantially above the poverty line. And iv 20 percent of the total benefits paid out to immigrants under SSI goes to families with incomes higher than 60 percent of all California families. The authors presented the work reported here to Congressional staff, federal agency staff, and numerous members of nongovernmental agencies in June 1997. Although the decision to reinstate benefits is now behind us, many other reforms are just beginning to take effect. Thus, there will be numerous opportunities to rethink, perhaps more carefully, welfare program design over the coming years. This work is presented with that thought in mind, and also as a reminder that targeting of the social safety net is every bit as important as the social insurance that the net provides. This is especially true in an environment where government programs are being judged not only for their effectiveness but also for the efficiency of their design. David W. Lyon President and CEO Public Policy Institute of California v Contents Foreword................... iii Figures............. vi Tables.............. ix 1. INTRODUCTION........... 1 2. WHY DID WELFARE REFORM TARGET IMMIGRANTS?.. 3 3. HOW WERE THE REFORMS REVERSED?...... 7 4. HOW IMPORTANT WOULD THE INCOME LOSS HAVE BEEN FOR IMMIGRANTS?............ 11 Assessing the Consequences of Denying Benefits..... 11 Immigrants Losing SSI........... 12 Immigrants Losing Food Stamps........... 15 Summary of Key Findings............. 18 5. WHICH IMMIGRANTS GAIN FROM THE RESTORATION OF BENEFITS?................. 21 Restoring All SSI Benefits............. 22 Restoring SSI Only to the Disabled or Only to the Elderly..... 24 Restoring All Food Stamps.................. 26 Restoring Food Stamps to Children and the Elderly...... 27 Summary of Key Findings................... 28 6. SHOULD WE BE RETHINKING WELFARE REFORM?....... 31 Efficiency of Immigrant SSI Benefits............. 32 Efficiency of Immigrant Food Stamps Benefits....... 32 Were Immigrants a Special Case?............... 33 vi Appendix A. Data and Methodology............... 37 B. Primary Tables and Description........... 43 Bibliography.............. 73 About the Authors............. 75 vii Figures 4.1. Benefit Losses for Families Receiving SSI........... 14 4.2. Change in Income Relative to Poverty Level Assuming Benefit Denial: SIPP Data............... 15 4.3. Income Sources for Immigrant SSI Families: SIPP Data..... 16 4.4. Benefit Losses for Families Receiving Food Stamps..... 17 4.5. Change in Income Relative to Poverty Level Assuming Loss of Food Stamps: SIPP Data.............. 19 5.1. Share of Reinstated SSI Benefits, by Income Quintile: SIPP Data.. 23 5.2. Share of Reinstated SSI Benefits, by Income as a Share of Poverty Level.. 24 5.3a. Share of Reinstated SSI Benefits to Disabled, by Income Quintile. 25 5.3b. Share of Reinstated SSI Benefits to Elderly, by Income Quintile.... 25 5.4. Share of Reinstated Food Stamps Benefits, by Income Quintile... 26 5.5. Share of Reinstated Food Stamps Benefits, by Income as a Share of Poverty Level................ 27 5.6a. Share of Reinstated Food Stamps Benefits to Children, by Family Income Quintile................ 28 5.6b. Share of Reinstated Food Stamps Benefits to Elderly, by Family Income Quartile.................... 28 ix Tables 3.1. Eligibility of California’s Nonqualified Aliens for SSI and Food Stamps, Based on 1996–1998 Legislation............. 9 4.1. Income Losses for Immigrant Families Participating in SSI.... 13 4.2. Income Losses for Immigrant Families Participating in Food Stamps... 16 4.3. Income Sources for Immigrant Families Losing Food Stamps.... 18 5.1. Total Value of Reinstated SSI Benefits to California Immigrant Families..................... 22 5.2. Total Value of Reinstated Food Stamps Benefits to California Immigrant Families................... 24 A.1. Sample Sizes and Characteristics of the SIPP and CPS...... 38 A.2. Unweighted Subsample Sizes for SIPP and CPS Results...... 39 B.1a. Characteristics of Citizen and Immigrant Families and Their Welfare Participation in California: SIPP Data......... 45 B.1b. Characteristics of Citizen and Immigrant Families and Their Welfare Participation in California: CPS Data........... 46 B.2a. Sources of Income and Amounts Received by Immigrant Families Who Are Welfare Participants in California: SIPP Data...... 49 B.2b. Sources of Income and Amounts Received by Immigrant Families Who Are Welfare Participants in California: CPS Data....... 50 B.3a. Income Losses of Immigrant Families Affected by Elimination of Food Stamp and SSI Programs: SIPP Data, Best Case........ 53 B.3b. Income Losses of Immigrant Families Affected by Elimination of Food Stamp and SSI Programs: SIPP Data, Worst Case....... 54 x B.3c. Income Losses of Immigrant Families Affected by Elimination of Food Stamp and SSI Programs: CPS Data............. 55 B.4a. Income Status of Immigrant Families After Elimination of Food Stamp and SSI Programs: SIPP Data, Best Case....... 57 B.4b. Income Status of Immigrant Families After Elimination of Food Stamp and SSI Programs: SIPP Data, Worst Case........ 58 B.4c. Income Status of Immigrant Families After Elimination of Food Stamp and SSI Programs: CPS Data............ 59 B.5a. Shares of Reinstated Welfare Benefits Going to Various Categories of Immigrant Families in California: SIPP Data......... 62 B.5b. Shares of Reinstated Welfare Benefits Going to Various Categories of Immigrant Families in California: CPS Data......... 63 B.6a. Shares of Reinstated SSI Benefits Going to Various Categories of Immigrant Families in California: SIPP Data......... 65 B.6b. Shares of Reinstated SSI Benefits Going to Various Categories of Immigrant Families in California: CPS Data........... 66 B.7a. Shares of Reinstated Food Stamps Benefits Going to Various Categories of Immigrant Families in California: SIPP Data......... 69 B.7b. Shares of Reinstated Food Stamps Benefits Going to Various Categories of Immigrant Families in California: CPS Data......... 70 1 1. Introduction In recent debates over welfare reform, few issues have been as controversial as the denial of benefits to immigrants. The 1996 federal welfare reform law ruled that nearly all immigrants would be ineligible for federal welfare payments under the Food Stamp and Supplemental Security Income (SSI) programs. However, Congress reversed position a year later and restored payments to many immigrants. More payments were restored in 1998. Now, after two years of policy decisions and reversals, there is still little reliable evidence to assess whether the reforms for immigrants were appropriate or how the cuts would have affected the immigrant families that relied on public assistance. In the absence of such evidence, advocates on both sides of this debate relied heavily on anecdotes to support their positions. The policymakers who chose to deny immigrant benefits in 1996 had only limited information on how public assistance contributed to immigrants’ overall resources and how many immigrant families would lose their principal means of support. Similarly, when policymakers restored benefits, they did not know which immigrant families desperately needed the income support and which had incomes that placed them above poverty without welfare benefits. This study revisits the issue of welfare reforms affecting immigrants, focusing on California. Although the changes were nationwide, California was, in a very real sense, at the center of this policy debate. California has the most to gain from the restoration of federal benefits for two reasons. First, 40 percent of the immigrants affected by the 1996 reforms live in California. Second, when the federal government barred immigrant 2 eligibility, the California legislature aggressively sought to replace virtually all of the immigrant benefits using the state’s own funds. The 1997 restoration of federal SSI benefits mitigated the need for some of this legislation, although California restored Food Stamps to children and the elderly. The 1998 restoration of federal Food Stamps closely mirrors this strategy put in place by California a year earlier. Although the issue of immigrant eligibility seems settled for the moment, there were and continue to be reform choices beyond the extreme of the 1996 cutoff and the simplistic restorations. Moreover, the real effects of welfare reform have scarcely been felt yet. Over time, there may be further calls to reverse or at least moderate other substantial changes made in the 1996 federal welfare law and the 1997 California enactment of this reform. A careful reexamination of changes in immigrant families’ conditions may provide insight into alternative program designs that might better meet the goals of the original reform. In particular, this study asks, What were the economic circumstances for immigrant families in California, given the ban and then the restoration of public assistance? We focus on family resources, which steps outside the existing program rules, to take what is in our judgment a broader and more realistic look at the living conditions for immigrant welfare participants. We organize the analysis around the following five questions: Chapter 2 Why were immigrants targets of welfare reform? Chapter 3 How were the 1996 reforms reversed? Chapter 4 How much income would immigrant families have lost had the welfare reforms been adopted as they were originally conceived? Chapter 5 Which immigrant families gained from the restoration of their welfare benefits? Chapter 6 Does the case of immigrant benefits suggest options for reforming of the welfare system? 3 2. Why Did Welfare Reform Target Immigrants? One of the most dramatic elements of the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA) was the bar on welfare benefits for most legal immigrants. This restriction eliminated the eligibility of noncitizen immigrants for SSI and Food Stamps unless they demonstrated 40 quarters of work in the United States or met exemptions for military personnel and recently arrived refugees. In addition, all immigrants (except refugees) arriving after August 22, 1996, were barred from receiving benefits for these and other federally funded programs for at least five years after arrival. The primary motivation for the immigrant cuts was budgetary. For the Republican - controlled Congress, it was important that welfare reform save money. Yet many of the massive changes to the welfare system provided for no near-term savings. This made the restriction of benefits to immigrants a key budget-cutter. In the final version of the 1996 welfare reform law, the immigrant restrictions accounted for almost half of the total federal savings in the welfare reform law. There were, nevertheless, substantive reasons to target immigrant benefits in welfare reform. Immigrants arrived in much higher numbers in the 1980s, and these recent immigrants were relatively high users of welfare. This triggered debates about the net fiscal costs immigrants imposed on the United States and on high immigrant areas in 4 particular. Finally, immigrants and their sponsors appeared to be abusing welfare programs. We consider each of these reasons in turn. Between 1980 and 1994, 13 million immigrants entered the United States, doubling the total foreign-born population relative to 1970. By 1994, one in eleven U.S. residents was foreign-born, and 70 percent of these immigrants were noncitizens. Among noncitizen immigrants, 43 percent had not completed high school, and the median income for this group was only $10,930. As a result, 29 percent lived in poverty and 5 percent received public assistance. This put their rate of welfare receipt 70 percent higher than that of U.S. natives and 400 percent higher than that of naturalized citizens (Bureau of the Census, 1995). This high rate of immigrant welfare participation was a relatively new phenomenon. Before 1980, immigrant participation in welfare was similar to that of natives. After controlling for household characteristics, many economic studies of immigrant participation in welfare found little difference overall between immigrant and native households before 1980. However, studies based on the 1980 Census and later data show very different results. By 1990, recent immigrants were more likely than both natives and earlier cohorts of immigrants to participate in welfare (see Borjas, 1994). High immigration rates and high usage of welfare by immigrants triggered a series of controversial studies of the net fiscal costs of immigration. Most of these studies contended that immigrants received more in benefits, mainly education and public assistance, than they paid in taxes. The most recent research in this area suggests that immigrants are probably net contributors to the U.S. economy overall. However, for high immigration areas, immigrants do appear to impose a heavy fiscal burden. The National Research Council report on immigration concluded that in California, immigrants cost each native household $1,178 annually (Smith and Edmonston, 1997). Finally, immigrants were targeted in welfare reform because of perceived abuses of welfare programs, especially SSI. SSI is designed to provide income support for individuals who are unable to work because of age or disability and who lack sufficient work experience to receive Social Security. Most immigrants (62 percent in 1994) are sponsored by family members, who sign affidavits of support vouching that the immigrants will not become public charges; an immigrant who becomes a public charge can be deported in the first five years after entry. Based on affidavits of support, the SSI 5 program counts or “deems” sponsor income for three to five years after an immigrant’s entry into the United States. Although the deeming provisions are intended to discourage immigrant use of welfare, the provisions merely postpone entrance onto the rolls. Forty- six percent of noncitizen recipients applied for SSI within four years of entry, in other words, as soon as the deeming period expired (Ross, 1996). Anecdotal reports have suggested that some immigrant communities have come to view SSI for the elderly as a normal benefit of U.S. residency (Matloff, 1994). These behaviors are contrary to the policy aims of the program, but they are legal under the current rules. There are illegal activities as well. For example, there are a number of documented cases of fraudulent applications from noncitizens using translators to claim nonexistent disabilities (Ross, 1996). Overall, immigrants seemed to be straining the resources of the SSI program. During the period from 1986–1994, noncitizen cases doubled as a proportion of all SSI cases, and noncitizens now constitute one-third of the aged SSI population. By 1996, even nonconservatives, such as analysts with The Urban Institute, testified that “high and rising immigrant use of Supplemental Security Income . . . represent[s] a significant public policy issue that calls for legislative attention” (Fix et al., 1996). Together, these issues motivated a set of “statements of national policy concerning welfare and immigration” as part of the welfare reform legislation. Simply put, these statements of Congress reiterated the policy of self-sufficiency for immigrants, with the intent that immigrants not rely on public resources, “but rather rely on their own capabilities and the resources of their families” and their sponsors to meet their needs. This was at least the stated goal of the welfare reform provisions. 7 3. How Were the Reforms Reversed? When the welfare law passed in 1996, the immigrant restrictions were viewed as the most draconian elements of new legislation for two reasons. First, these restrictions went into effect earlier than other aspects of the reform. Immigrants were to lose benefits within one year, rather than within two to five years—the time limits for recipients of Aid to Families with Dependent Children (AFDC). Second, SSI benefits had been a primary means of support for elderly and disabled people who were not employable. Opposition to these measures came from several directions. President Clinton opposed the immigrant cuts even as he signed the welfare reform bill and made restoring these benefits part of his 1996 reelection platform. He particularly focused on restoring benefits to the disabled. In March 1997, immigrant rights groups in California and New York filed class-action lawsuits to block implementation of the immigrant provisions. By April, Florida filed its own lawsuit. Florida feared that its own taxpayers would end up picking up the costs of supporting these immigrants, just as California counties feared that immigrants would move from federally funded SSI benefits onto the locally funded General Assistance rolls. Finally, even well-respected newspapers such as the Wall Street Journal reported alleged cases of suicide among immigrants in California and elsewhere who faced losing benefits As a result of these pressures, Congress reversed position on SSI benefits as part of the Balanced Budget Act of 1997. The 1997 law restored SSI benefits to those immigrants receiving SSI as of August 22, 1996. It also allowed immigrants residing in the United 8 States on August 22, 1996, to be eligible for SSI if they become disabled in the future. New arrivals will not be eligible for SSI, nor will earlier immigrants who seek SSI in the future based on age. The act also extends refugee eligibility for SSI (but not other programs) from five years after arrival to seven years after arrival. The 1997 law made no changes regarding eligibility for Food Stamps. Later the same year, California chose to partially restore Food Stamps under a state- only program. Along with seven other states, California administered this program by purchasing federal Food Stamps coupons. 1 Under AB 1576, signed into law on August 18, 1997, California provided Food Stamps to legal immigrants age 65 or older and children under age 18 who lost federal eligibility based on PRWORA. This applied only to immigrants who were both legally in the United States and receiving Food Stamps before August 22, 1996. The benefits for these recipients will be the same as those provided under the federal Food Stamp program, with the same regulations in effect as well. California proved to be a policy leader, as Congress voted in 1998 to partially restore Food Stamps by reinstating eligibility for legal immigrant children and elderly persons who were legal residents at the time of the 1996 legislation. (Unlike the SSI restoration and the California legislation, the federal Food Stamps restoration does not require that benefits go only to previous recipients.) President Clinton signed these restorations into law on June 23, as part of the Agriculture Research, Extension and Education Reform Act of 1998. Table 3.1 outlines the eligibility changes over the past two years for immigrant recipients of Food Stamps and SSI in California. _________ 1Other states purchasing federal Food Stamps included Florida, Maryland, Nebraska, New Jersey, New York, Rhode Island, and Washington. Washington is the only state fully replacing Food Stamps for all legal immigrants who became ineligible. Massachusetts, Minnesota, and Texas have developed state food programs, with benefits substantially lower than the federal levels. 9 Table 3.1 Eligibility of California’s Nonqualified Aliens for SSI and Food Stamps, Based on 1996–1998 Legislation Eligibility by Category1996 Welfare Reform Law1997 Balanced Budget Act1997 California AB 15761998 Agriculture Research Act SSI Eligibility Legal immigrants resident as of 8/22/96 SSI recipients as of 8/22/967333 Immigrants who become disabled7333 Legal immigrants arriving after 8/22/967777 Food Stamps Eligibility Legal immigrants resident as of 8/22/96 Elderly or those who become elderly Food Stamps recipients as of 8/22/967733 Nonrecipients as of 8/22/967773 Children under age 18 Food Stamps recipients as of 8/22/967733 Nonrecipients as of 8/22/967773 Adults age 18–647777 Legal immigrants arriving after 8/22/967777 NOTE: 3 indicates made eligible; 7 made ineligible; 3 or 7 eligibility unchanged from prior law. 11 4. How Important Would the Income Loss Have Been for Immigrants? The specter of aged and disabled immigrants ending up on the streets when they lost their sole source of income provided much of the impetus for rescinding the SSI cuts. Although losing SSI clearly would have created some desperate cases, it is unclear from the anecdotal evidence whether such cases were representative of the entire immigrant SSI population. In the case of Food Stamps cuts, the anecdotes were far less dramatic, but this does not necessarily mean that cutting Food Stamps was any better or worse policy. In this chapter, we evaluate exactly how severe the SSI and Food Stamps cuts would have been for California immigrant families. We focus on two questions: What share of income would affected families have lost? How many more California families would have been in poverty? Assessing the Consequences of Denying Benefits To address these questions, we calculated the income losses that immigrant families faced under the 1996 welfare reform law, also known as PRWORA. Although usually nuclear families, some of the immigrant families are extended families, in that grandparents in the same household are counted as part of the family of their children and grandchildren. More important, these families need not match the case unit as defined by Food Stamps or SSI. (SSI cases usually refer to an individual or couple, whereas Food 12 Stamps considers all members of a household.) Appendix A provides further information on our use of families as the unit of observation. Our data are drawn from California observations in two national household datasets: the 1993–1994 Surveys of Income and Program Participation (SIPP) and the March 1993–1995 Current Population Surveys (CPS). The results from the two datasets are similar, except for a few key differences: The CPS shows less overall income for families, less welfare participation, and higher benefits among participants. A comparison of the advantages and disadvantages of each dataset, with definitions and methodology, is given in Appendix A. Detailed results from the SIPP and CPS data are included as Appendix B. We calculate benefit losses for two groups of California families: families of immigrants who received SSI (with or without Food Stamps) and families of immigrants who received Food Stamps but not SSI. The CPS and SIPP data can be considered as worst case and best case: The worst case presumes a complete loss of all family benefits if any family member is ineligible; the best case takes away benefits only from noncitizen members of the family for the months that they are in the family. The best case, using SIPP data, more accurately reflects the program rules. In addition, the SIPP data offer details not available in the CPS data. For these reasons, we focus more on the SIPP results, although all results are included in Appendix B. Immigrants Losing SSI In 1993–1995, one in five California families included at least one noncitizen immigrant. Immigrant families were twice as likely to live in poverty as citizen families. They were also more than twice as likely to participate in a major welfare program: SSI, Food Stamp, AFDC, or Medicaid. Out of all immigrant families in California, one in nine received benefits through the SSI program. However, not all immigrant SSI families would have lost benefits under PRWORA. Twenty-three percent of these immigrant families received SSI through either a citizen or an immigrant who met the exemptions allowed in the welfare law, particularly the exemption for refugees. We exclude these families when we calculate income losses. Table 4.1 shows the average losses immigrant families could have expected under the PRWORA as originally passed. The estimates for total annual income losses range from just over $5,000 to nearly $6,800. About $50 of this is actually lost Food Stamps 13 Table 4.1 Income Losses for Immigrant Families Participating in SSI Projected Loss of Benefit Income SIPP CPS Average annual loss of Food Stamps benefits $492 $474 AFDC ($290) $0 SSI $4,878 $6,323 Total benefits $5,080 $6,797 Annual income loss a 25th percentile $1,850 $3,174 50th percentile $4,612 $6,055 75th percentile $7,595 $8,002 aReported at three points in the income distribution, order- ing the losses from least to greatest. The xth percentile is the value that x percent of the distribution is below. The 50th per- centile is median income loss; it is the value of the loss such that half the families lose more than this amount and half lose less. benefits.1 Using the SIPP data, which allow more careful month-by-month calculations, we can also determine how much AFDC or Temporary Assistance for Needy Families (TANF) a family could get by switching a child from the SSI program into AFDC. Although this counteracts some of the lost income, it represents a gain of only $290 on average. The lower half of Table 4.1 gives more information about the distribution of losses. As we noted above, these losses can be properly understood only in the context of total family income. We divide families into four categories based on their projected income loss as a share of annual family income: (1) families losing less than 10 percent of annual income, (2) families losing 10 to 25 percent of annual income, (3) families losing 25 to 50 percent, and (4) families losing more than 50 percent of total family income. For both datasets, Figure 4.1 presents the percentage of families that fall into each of these four categories. Figure 4.1 illustrates two key facts in assessing the immigrant SSI policy. If SSI benefits had been eliminated as originally proposed, at least 20 percent and as many as 40 _________ 1In California, an individual cannot receive both Food Stamps and SSI, but as many as one in three immigrant SSI families also include another family member receiving Food Stamps. 14 Loss as share of annual income 10% or less Over 50 to 50% 10 to 25% 0 5 10 15 20 25 35 30 40 45 Percent of families SIPP CPS Figure 4.1—Benefit Losses for Families Receiving SSI percent of affected families would have lost more than half of their annual family income. This is consistent with the anecdotal evidence presented in the 1997 policy debate. Yet almost as many families would have lost less than 10 percent of their annual income. This side of the story did not come out in the debate. Of course, for a family in poverty, 10 percent of annual income could be crucial. In Figure 4.2, we show two income groups before and after the PRWORA benefit cuts. The two left hand bars show the change in the share of immigrant SSI families with income below the poverty line. The right hand bars show the change in the share of families with income above 185 percent of the poverty line (a common cutoff point for welfare benefits). For a family of four, 185 percent of the poverty level equals about $30,000. As in the previous figure, we see two distinct groups of families. On the left, we see that the cuts for SSI families would have increased the poverty rate for such families by 35 percent. With benefits, only 13 percent of immigrant SSI families live in poverty; without, this number rises to 48 percent. (For comparison, AFDC and Food Stamps combined in California paid only 72 percent of the poverty level.) On the other hand, 15 Income below poverty Income above 185% of poverty 0 5 10 15 20 25 30 35 40 45 50 Percent of SSI families Before (with SSI and Food Stamps) After (without SSI and Food Stamps) Figure 4.2—Change in Income Relative to Poverty Level Assuming Benefit Denial: SIPP Data with or without benefits, more than 30 percent of immigrant SSI families would not only be above poverty, but above 185 percent of poverty. How can 30 percent of SSI families have income above 185 percent of poverty, even without welfare benefits? Actually, according to the SIPP data, SSI benefits on average represented less than one-third of the annual income for immigrant SSI families. Figure 4.3 shows the breakdown of annual income for SSI families in SIPP. (CPS provides much less detail than SIPP on income sources over time.) Since 63 percent of recipients live in families with at least one wage earner—where earnings represent 79 percent of the family income for the median family—it is perhaps not surprising that over 40 percent of total income in SSI families came from earnings. In fact, among affected SSI families with earnings, one-fourth had earned income over $48,000 per year. These are the families in the “above 185 percent of poverty” category. Immigrants Losing Food Stamps Federal funding for Food Stamps benefits for immigrants ended in August 1997, although California restored partial funding. Cutting all Food Stamps benefits would 16    Earnings SSI 31% 43% 18% 6% 2DC Food Stamps Other including Social Security Figure 4.3—Income Sources for Immigrant SSI Families: SIPP Data have affected nearly one in five immigrant families in California. How much income would families have lost? The average loss of benefits for families participating in Food Stamps would fall between $1,300 and $1,700, as shown in Table 4.2. As with the SSI calculations, the differences between the SSI and the CPS versions are determined largely by whether benefits are recalculated monthly (SIPP) or simply eliminated for the entire year. In parallel with Table 4.1, the bottom half of Table 4.2 shows the distribution of benefits. Many families would lose less than $500 annually, although one in four would lose $1,800 or more (as high as $2,500 or more according to CPS). Table 4.2 Income Losses for Immigrant Families Participating in Food Stamps Projected Loss of Benefit Income SIPP CPS Average annual loss of Food Stamps benefits $1,295 $1,711 AFDC $0 $0 SSI $0 $0 Total benefits $1,295 $1,711 Annual income loss 25th percentile $496 $652 50th percentile $1,083 $1,396 75th percentile $1,876 $2,545 17 To put these losses in perspective, we can once again look at the losses as a share of total family income, as in Figure 4.4. The vast majority of families would lose less than 25 percent of their income, and most would lose less than 10 percent. The SIPP data show almost no families losing more than 25 percent of their income, although the CPS data show about 20 percent of families in this category. As noted above, the CPS calculation removes all annual benefits rather than doing a month-by-month determination. In addition, the CPS generally reports less overall income but more welfare income, so the potential losses represent a larger share. Unlike SSI, Food Stamps is not intended to be a sole source of income for recipients. In fact, almost all of the potentially affected Food Stamps families receive other public assistance. Table 4.3 shows the resources of these families in the absence of Food Stamps for immigrants. This table shows only the SIPP data, because SIPP provides much more information on income from welfare and other sources. Even after the cuts, 86 percent of these families would receive benefits through the AFDC/TANF program, through General Assistance or through Food Stamps to citizen members. A similar share of families have other unearned income, typically unemployment or Social Security benefits. Loss as share of annual income 10% or less Over 50 to 50% 10 to 25% 0 10 20 30 40 50 60 70 Percent of families SIPP CPS Figure 4.4—Benefit Losses for Families Receiving Food Stamps 18 Table 4.3 Income Sources for Immigrant Families Losing Food Stamps Income SourcePercent of Families Welfare benefits from major program86 AFDC/TANF 64 Food Stamps 58 SSI (a) General Assistance benefits 6 Other unearned income 87 Earnings82 Annual earnings 25th percentile $5,586 50th percentile $11,536 75th percentile $19,842 aCalculations based on immigrant families receiving Food Stamps but not SSI. In addition, 82 percent of the affected Food Stamps families have earnings. These are still poor families, however. The median income for families with earnings is only $11, 536. If we look at the change in poverty from cutting Food Stamps, we see a very different picture than we saw with SSI. Figure 4.5 shows the same categories as shown in Figure 4.2 for SSI cases. Here the poverty rate does not rise as dramatically, with the loss of Food Stamps pushing 8.2 percent more families into poverty. However, even with Food Stamps, more than half of the immigrant families were in poverty. Just over 10 percent had annual incomes over 185 percent of poverty. Summary of Key Findings Under the 1996 welfare reform law, most immigrants were to lose eligibility for SSI and Food Stamps. The potential losses can best be understood in the context of the family incomes of families that include an immigrant losing benefits. Calculating immigrant benefit losses in this framework, we found: •About a third of aged and disabled immigrants on SSI live well above poverty and would have lost only a small share of their income. Between 20 and 40 percent of immigrant families would have lost less than 10 percent of their annual income if SSI were cut. About one in three immigrant families receiving SSI had 19 Income below poverty Income above 185% of poverty 0 10 20 30 40 50 60 70 Percent of Food Stamps families Before (with Food Stamps) After (without Food Stamps) Figure 4.5—Change in Income Relative to Poverty Level Assuming Loss of Food Stamps: SIPP Data income over 185 percent, even without the SSI benefits. Many families had earnings exceeding $40,000 annually. •At the same time, another a third of the SSI immigrant population would have lost their primary source of support by the loss of SSI. Between 20 and 40 percent of immigrant families receiving SSI would have lost more than half their annual income. The poverty rate among SSI immigrants would have risen from 13 to 48 percent. •Nearly two-thirds of immigrant Food Stamps recipients would have lost less than 10 percent of their annual income with the denial of Food Stamps eligibility. The poverty rate among immigrant families losing Food Stamps would have risen from 54 percent to 63 percent. A negligible fraction (less than 5 percent) would lose more than a quarter of their total income with the denial of Food Stamps benefits. At $1,083, the median income loss for Food Stamps recipients would be less than one-fourth the loss suffered by SSI recipients. There are three programmatic differences between SSI and Food Stamps that are key to understanding the different policy effects. The first helps explain the decision to reinstate SSI. The SSI program was intended to be a primary source of income for people 20 unable to work: the disabled and the aged.2 Food Stamps merely supplement the income of poor families, where the majority of family income could come from work, from welfare, or from other benefits such as unemployment insurance. Thus, SSI recipients are much more dependent on the benefits. The other two differences work in the opposite direction, however. First, Food Stamps are issued to an entire household based on the entire household’s income. SSI, on the other hand, goes to individuals or couples with little outside income counted for eligibility, although a sponsor’s income theoretically may be counted for five years. Second, income eligibility is more routinely redetermined in the Food Stamp program than in the SSI program. Together, these two differences make it far more likely that an SSI recipient would live in a family with higher income. _________ 2Age and disability status also prevent some immigrant SSI recipients from meeting English language requirements for citizenship. 21 5. Which Immigrants Gain from the Restoration of Benefits? In the previous chapter, we evaluated the effect of the 1996 welfare reform law on immigrant families receiving SSI and Food Stamps. Here we reverse directions. The thought exercise is this: Assuming that immigrants were ineligible for SSI and Food Stamps as originally planned under the PRWORA, what kinds of families will benefit from reinstating SSI or Food Stamps? We consider the consequences for California families of reinstating benefits under four alternative scenarios: • Restore all SSI benefits. • Restore SSI only to the disabled or only to the elderly. • Restore all Food Stamps. • Restore Food Stamps just to children and/or the elderly. To evaluate how these policy alternatives would affect the incomes of immigrant families, we start from the hypothetical situation where SSI and Food Stamps benefits are eliminated as intended in the 1996 reform legislation. We then consider the effect of restoring benefits on the total income of recipient families. As before, results from SIPP and CPS map to best and worst cases, where the best case removed benefits only from noncitizen family members only when they were present in the family and the worst case assumes that all benefits to a family are lost all year if any family member is ineligible. 22 SIPP thus assumes lower costs of restoring benefits because fewer benefits were eliminated to begin with. For each of the four policy scenarios, we examine the distribution of benefits to different categories of families, distinguished primarily by income or poverty level. We base this strategy on the idea that an important criterion for effective poverty policy is successfully targeting benefits toward the needy. The projected income gains to immigrants, of course, also represent the costs to the public treasury, so our analysis simultaneously estimates effects of the proposals on state and federal budgets. As of this year, the federal government has chosen to bear these costs, although California was prepared to cover the cost of replacing Food Stamps. In both cases, it is important to note that these results assume 1993–1995 participation rates. This will lead to overestimates for two reasons. First, California’s improving economy has led to declining caseloads, so fewer families receive benefits today, especially in the Food Stamp program. Second, we do not adjust the numbers to reflect growing citizenship rates. These cases, of course, will still qualify for benefits but would no longer be differentiable from the rest of the caseload. Restoring All SSI Benefits Table 5.1 provides the SIPP and CPS estimates for the total monetary value of reinstated SSI benefits for immigrant families in California. The two estimates are amazingly similar; both report an expected value of just over $1 billion annually. 1 These benefits were restored by the federal government as part of the Balanced Budget Act of 1997. Since California has a disproportionate share of immigrants on SSI, this represents Table 5.1 Total Value of Reinstated SSI Benefits to California Immigrant Families (millions of 1996$) Benfits Reinstated SIPP CPS All SSI $1,020 $1,097 SSI to disabled $272 $571 SSI to elderly $735 $579 _________ 1By definition, this is identical to the loss of income summed over all families described in the previous chapter. 23 a transfer of money from the rest of the United States to California. Benefits were restored only to immigrants legally in the United States in August 1996 who either were already receiving SSI or will qualify in the future based on disability. Therefore, the annual costs are likely to fall over time. Figures 5.1 and 5.2 show the breakdown of reinstated benefits across different income levels. Figure 5.1 divides benefits by the income quintile of immigrants’ families. Quintiles are determined by ranking all California families according to their income and dividing them into five equal-size groups. Thus, one-fifth of California families are in the lowest quintile, one-fifth in the “mid-low” quintile, and so on. A family with median income is in the middle quintile, and at least 80 percent of families have incomes below the income of a family that falls into the highest quintile. A poverty program typically serves families in the lowest or sometimes the next lowest or “mid-low” income quintile. In the case of reinstated SSI benefits, we do see 75 percent of benefit dollars going to families in the bottom two quintiles. However, we also see 20 percent of benefit dollars returned to families with incomes higher than 60 percent of all California families. In fact, $50 million of SSI benefits are being returned to families with incomes in the top- fifth of the entire California population. Immigrant families may be larger than other California families, in which case income quintiles would not accurately reflect individuals’ economic status. For this reason,    9% 66% 14% 6%5% Highest Mid-high Middle Mid-low Lowest  Figure 5.1—Share of Reinstated SSI Benefits, by Income Quintile: SIPP Data 24 Over 185% of poverty level From 100% to 185% From 66% to 100%Less than 66% of poverty level 47% 24% 19% 10% Figure 5.2—Share of Reinstated SSI Benefits, by Income as a Share of Poverty Level Figure 5.2 shows the benefits returned by the poverty status of the immigrant family. Poverty levels are based on family size as well as income and, in Figure 5.2, do not include SSI benefits. Two-thirds of benefit dollars go to families who would be in poverty without SSI. In fact, almost half of the reinstated benefit dollars go to families with incomes below two-thirds of the poverty level—families in extreme poverty. However, another 24 percent or nearly $250 million of the reinstated benefits go to families above 185 percent of poverty—before SSI benefits are even counted in their income. Restoring SSI Only to the Disabled or Only to the Elderly Neither SIPP nor CPS indicates whether a recipient qualified for SSI based on age or disability. However, there are age data, and SIPP also asks respondents to report disabilities. In CPS, we can identify only elderly and nonelderly recipients. With this information, we have roughly divided the reinstated benefits into those going to the nonelderly disabled and those going to the elderly. This breakdown is reported in Table 5.2. Table 5.2 Total Value of Reinstated Food Stamps Benefits to California Immigrant Families (millions of 1996$) Benfits Reinstated SIPP CPS All Food Stamps $561 $683 Food Stamps to children $134 $217 Food Stamps to elderly $15 $16 25 Figures 5.3a and 5.3b show the income quintile breakdown for the two groups. Although elderly immigrants are more likely to be in the lowest income quintile without SSI, they are also more likely to have income in the top two quintiles. If benefits were reinstated just to the elderly, 22 percent of dollars would go to families in the top 40 percent of the income scale. For the disabled, this share drops to 15 percent.     13% 12% 10%3% Highest Mid-high Middle Mid-low Lowest  Figure 5.3a—Share of Reinstated SSI Benefits to Disabled, by Income Quintile    7% 16% 4%6% Highest Mid-high Middle Mid-low Lowest  Figure 5.3b—Share of Reinstated SSI Benefits to Elderly, by Income Quintile 26 Restoring All Food Stamps President Clinton had called for the restoration of all Food Stamps benefits for immigrants, although the 1998 compromise restores them primarily to just children and the elderly. Had all benefits been restored, the Food Stamps would have totaled over $500 million for California families, based on 1993–1995 participation patterns. If these benefits were restored at the federal level, it would be a transfer of income to California, since California has more immigrants than any other state. These benefits dollars would be distributed as shown in Figure 5.4. This figure is parallel to Figure 5.1, dividing dollars according to where the recipients fall in the California income distribution. These benefits are calculated recognizing that SSI has already been reinstated; the distinction matters for determining the economic status of families before Food Stamps receipt. Eighty-seven percent of benefit dollars would be returned to families with incomes in the bottom 40 percent for California families. The remaining families usually have incomes around the median, with relative few in the upper income ranges. The families with higher incomes who receive Food Stamps are likely to be either large families or families who experience only short-term income losses. To control for family size, Figure 5.5 reports shares of reinstated benefits by poverty level. Three out of four benefit dollars would go to families in poverty. Another 20     30% 57% 4% 8%1% Highest Mid-high Middle Mid-low Lowest  Figure 5.4—Share of Reinstated Food Stamps Benefits, by Income Quintile 27 Over 185% of poverty level From 100% to 185% From 66% to 100%Less than 66% of poverty level 31% 5% 44% 20% Figure 5.5—Share of Reinstated Food Stamps Benefits, by Income as a Share of Poverty Level percent would go to families in near poverty. Only 5 percent of these benefits would go to families with income above 185 percent of poverty, one-fifth the share under SSI. Restoring Food Stamps to Children and the Elderly First in California and now nationally, Food Stamps have been restored to children and elderly immigrants who received Food Stamps before the welfare reform law. According to the survey data, these benefits account for between one-fourth and one-third of Food Stamps benefits to immigrants. Figures 5.6a and 5.6b provide the distribution of benefit dollars by income quintile for Food Stamps to children and the elderly. Compared to Food Stamps to all immigrants, Food Stamps to children will go overwhelmingly to families in the bottom two-fifths of the income distribution. Almost no benefit dollars would be returned to families at the top of the income distribution. These results are consistent between SIPP and CPS. The distribution for Food Stamps to the elderly shows one-fifth of dollars going to families in the mid-high income range but none to families in the middle quintile. The sample size for elderly immigrants receiving Food Stamps is extremely small, however, since the elderly would typically qualify for SSI instead of Food Stamps. 28  32% 62% 2% 4% Highest Mid-high Middle Mid-low Lowest  Figure 5.6a—Share of Reinstated Food Stamps Benefits to Children, by Family Income Quintile    14% 1% 20%Highest Mid-high Middle Mid-low Lowest  Figure 5.6b—Share of Reinstated Food Stamps Benefits to Elderly, by Family Income Quartile Summary of Key Findings Our first and last scenarios mirror the current policy. Our calculations consider California’s immigrant families who had received benefits from SSI and Food Stamps, starting from their baseline income in the absence of these two benefits. Reinstating first SSI and then Food Stamps for children and the elderly, we found: 29 • One in three SSI dollars will go to a family not in poverty; one in four to a family above 185 percent of poverty. The same result is evident in income quintile calculations where 20 percent of benefit dollars go to families with incomes higher than that of 60 percent of California families. This represents over $200 million in benefits to relatively well-off families, even before counting SSI income. • Elderly SSI recipients are more likely to be in the higher-income families. This presumably occurs because these recipients are living with their adult children. • The majority of reinstated SSI and Food Stamps benefits will go to families in poverty. $671 million out of about $1 billion in SSI benefits will go to families with incomes below the poverty level. $128 out of $149 million in state-only Food Stamps would go to these families. 31 6. Should We Be Rethinking Welfare Reform? The first element of the 1996 welfare reform to actually “bite” was the denial of SSI and Food Stamps to immigrants. In the wake of the law’s passage, the American public was barraged with emotional rhetoric depicting the fate of immigrants losing access to SSI and Food Stamp programs under the new law. Those wanting to restore benefits claimed that recipients would be destitute without public assistance. Advocates for denying benefits argued that immigrants create a burden on the welfare system and that many are well enough off without benefits. Who was right, and who was wrong? Our findings show that both are right, and both are wrong. Significant fractions of the immigrant welfare population match both descriptions. Supporting the reforms, our analysis of California’s immigrants reveals that immigrant families are more than twice as likely as citizen families to participate in a major welfare program. Further, reinstating welfare benefits to immigrants by simply reversing the 1996 reforms poorly targets “needy” immigrants families. A significant share of these reinstated benefits will go to families above the poverty level and to families without disabled members. Supporting the reinstatements, this study finds that immigrant families in the state are more than twice as likely as citizen families to live in poverty. Some immigrant families greatly rely on welfare benefits and it is not obvious how they would fare without public assistance. 32 Efficiency of Immigrant SSI Benefits SSI benefits are intended for low-income individuals who are unable to support themselves because of disability or age. However, instead of assisting only the state’s truly needy, we find that SSI benefits are often provided to immigrants whose total family incomes are well above the poverty level. This happens because the program provides assistance based on individual rather than family need. According to our estimates using 1993–1995 data, the cost of reinstating SSI benefits to the nearly 250,000 immigrant families in California who participate in the program is nearly $1 billion annually. 1 About one-third of these reinstated benefits are projected to go to immigrant families whose annual incomes exceed the poverty level. Nearly one-quarter goes to families with incomes 185 percent of the poverty level, a common income threshold for welfare programs to stop paying benefits. Even though the disabled were a main focus for the restoration of benefits, only half of the SSI benefits go to families that include a disabled member. By reinstating SSI benefits to all legal immigrants who were on the rolls in August 1996, lawmakers are essentially giving welfare dollars to thousands of people who do not necessarily need help. This is not to say that a modification of the federal welfare law was not in order. Some 55,00 immigrant families in the state, just under a quarter of the families who receive benefits during a year, stood to lose more than half of their income if SSI benefits were cut off. Two-thirds of SSI benefits go to families that would otherwise be in poverty. Efficiency of Immigrant Food Stamps Benefits The case for reinstating Food Stamps is less dramatic on both sides, because Food Stamps benefits are not a major source of income for recipients. Assuming 1993–1995 participation rates, our estimates indicate that fully restoring Food Stamps to all California immigrants would cost $561 million to $683 million annually and would provide income to approximately 430,000 families. Returning benefits just to children and the elderly would cost about $149 million. 2 _________ 1Unlike other welfare programs, SSI caseloads are virtually unchanged from the 1993–1995 period. 2Current Food Stamps caseloads are substantially below the 1993–1995 level used to calculate the estimates presented in this study, so these are overestimates of today’s costs. 33 For those arguing against reinstating Food Stamps, our findings support the view that the elimination of benefits was unlikely to impose substantial hardships—particularly when compared to elimination of SSI. About two-thirds of immigrant Food Stamps recipients would have lost less than 10 percent of their annual income with the denial of Food Stamps eligibility. Less than 2 percent would lose more than a quarter of their total income. Of those families who would forgo a significant fraction of their income, the vast majority receive welfare from other sources. Nearly 75 percent of immigrant families receiving Food Stamps also collect other forms of welfare (primarily AFDC plus some SSI), and, for these families, Food Stamps contribute on average slightly more than 10 percent of their total annual income. The remaining 25 percent of immigrant Food Stamps recipients rely on benefits to supplement their earnings, but Food Stamps make up only 5 percent of their family income. Expressing losses in dollar rather than relative amounts, well over three-quarters of the immigrant families collecting Food Stamps would have lost less than $2,000 in benefits per year. To place this amount in context, the enhancement of the Earned Income Tax Credit (EITC) that began in 1996 more than makes up for this loss for those who work—EITC pays up to $3,60 in benefits— meaning that working immigrant families who collected Food Stamps are better off now than they were in the early 1990s before the denial of Food Stamps benefits. On the other hand, our findings also support those favoring the restoration of Food Stamps to immigrants in that they demonstrate that the Food Stamp program is fairly well targeted to the poor. Reinstating benefits to all immigrants in California would give four in five dollars to families in poverty. Returning benefits just to immigrant children and the elderly, as legislated in California and now nationally, is even more efficient at assisting the poor, with 86 percent of restored benefits going to families with incomes below poverty. One in three of all reinstated Food Stamps dollars will go to families with disabled members, with one out of ten dollars going to families with disabled members who were either children or elderly. Were Immigrants a Special Case? Our analysis has focused on immigrants, since they were a principal target of the 1996 reform and the beneficiaries of the subsequent restoration of benefits. In fact, Michael Fix of The Urban Institute has described the denial of immigrant eligibility for welfare as a 34 switch from “immigration policy” or rules governing entry to the United States executed by the Immigration and Naturalization Service to “immigrant policy” or rules governing immigrants within the borders. Up until this change, however, the only practical difference between immigrants and natives for the purposes of welfare policy was the fact that most immigrants have family members who sponsor their immigration. A sponsor’s pledge to prevent the immigrant from becoming a public charge is the basis for deeming provisions in the determination of benefits. Deeming refers to counting sources of income other than the immigrant’s own as resources available to the immigrant. Income is counted to determine eligibility and to calculate benefits. Deeming of sponsor income occurs in both the SSI and the Food Stamp programs. According to rules of the SSI program, for example, the income of an immigrant’s sponsor is treated as unearned income for the immigrant, after subtracting an allocation for the sponsor and his or her dependents. A dollar increase in unearned income reduces SSI benefits by one dollar. Consequently, when sponsors are financially self sufficient, theoretically their income fully counts in lowering the immigrant’s potential SSI benefits, resulting in the immigrant losing eligibility if the sponsor is sufficiently well off. Before the 1996 welfare reform, deeming provisions held for three to five years after an immigrant’s arrival in the United States. As part of the reform, the deeming period was lengthened to the point when an immigrant becomes a “qualified alien” by accruing 40 quarters of work or becomes a U.S. citizen, and sponsors were required to have sufficient income to support an immigrant, defined as at least 125 percent of the poverty line. These changes do not apply to immigrants who became legal residents before January 1, 1998. Our data include no information about sponsor income and therefore may underestimate the resources available to immigrants. Some family members whose income we include may in fact be the immigrants’ sponsors, but sponsors need not reside with recipient relatives. Although we cannot identify sponsors, our findings do demonstrate that many SSI members reside in families with resources well beyond the poverty level. Consistent with these findings, Matloff (1994) reports that 42 percent of California’s elderly SSI recipients live with their children. Yet the difference between Food Stamps and SSI in their effectiveness at targeting needy families suggests a direction for rethinking the way income is counted for welfare. 35 The Food Stamp program typically treats all individuals living together (and sharing food preparation) as a single household for the purpose of determining eligibility and benefits. A similar strategy could be considered for SSI. Currently, SSI treats in-kind support and maintenance, such as when a recipient is living in the household of another, as equal to one-third or less of the SSI benefit level, regardless of income of other household members. If the income of other household members were counted, as in the Food Stamp program, benefits for many immigrants would be lowered and others would be disqualified. Yet immigrants living alone or in poor families would receive benefits through the program. During the debate over welfare and immigration reform, bills jointly sponsored by California’s two senators and three California representatives proposed restoring eligibility to any immigrant who otherwise qualified for SSI and “whose family is incapable of support, and who can demonstrate that he or she has no other sufficient means of support.” Defining the sources of income to be counted (deemed) in calculating SSI (or Food Stamps) benefits provides a natural framework for operationalizing the concept expressed in this bill. Such rules need not apply to immigrants alone. In fact, although the term “deeming” most commonly refers to a sponsor’s income, the same term applies whenever an additional person’s income is counted in determining welfare benefits. For example, the Social Security Handbook reports a number of types of deeming other than from sponsor to immigrant, such as from ineligible to eligible spouse and from parent to child. Each welfare program sets in own rules for whose income to count or deem. Unfortunately, despite the many dramatic changes in welfare reform, there was no reconsideration of these deeming rules as a strategy for addressing the policy concerns regarding immigrants’ receipt of welfare. As this analysis has shown, families of welfare recipients may have vastly different resources available. 3 We would like to ensure benefits to the neediest families, while asking self-sufficiency of less needy families. This lack of thoughtful program design was the most disturbing feature of the denial and subsequent reinstatement of benefits for immigrants. The welfare law itself, as well as the debate around reinstatement, pointed to other policy options that could have gone _________ 3Although the results in this study apply only to immigrants, they are likely to be the most extreme case. Previous research comparing recent immigrants to the California caseload overall showed higher incomes in immigrant welfare families (not controlling for family size) (MaCurdy and O’Brien-Strain, 1996). 36 beyond simply switching eligibility on and off. Welfare reform provided a rare opportunity to redesign programs to better meet the needs of our poorest families while promoting self-sufficiency and family responsibility. Categorically barring immigrant families from benefits did not meet these goals. Unfortunately, simply reversing the reforms did not meet these goals either. Denial of benefits to immigrants represented only the first in a series of tough challenges arising from the sweeping changes in the PRWORA. There are likely to be future cases requiring policymakers to rethink and refine the reforms enacted in 1996. If, instead, policymakers merely reverse direction each time, forgetting why they adopted the reforms in the first place, our welfare system will be no better than it was in 1996. 37 Appendix A Data and Methodology Our analysis is built on a profile of immigrant families that received welfare benefits during the period 1993 to 1995. There are two key features to our analytical approach. First, we use survey data to understand the economic circumstances of recipients to a depth not possible with administrative data. Second, we treat families as the relevant unit to count for benefits—a significant departure from the existing policy in most public assistance programs. In this appendix, we outline the source of the data and our definition of family and then explain how these elements are combined to assess the effects of the 1996 benefit cuts and the recent restorations. Data The empirical results presented in this study are derived from samples of California households in two different surveys conducted by the Bureau of the Census: the Survey of Income and Program Participation (SIPP) and the March Current Population Survey (CPS). Both surveys interview a nationally representative sample of households, collecting detailed demographic and income information. Compared to administrative data, these survey data create a more complete picture of the economic circumstances of households, families, and individuals. Specifically, we can include income counted for benefit determination as well as income sources typically excluded from these calculations and examine the economic status of recipients across the year, not just in months on aid. 38 The SIPP was the basis for our earlier report (MaCurdy and O’Brien-Strain, 1996).1 Although the structure of the SIPP is also ideal for the issues addressed here, we have expanded our analysis to include the CPS as well. We have two reasons for using the CPS data. First, the CPS provides a larger sample, especially when several years are pooled together. This may be particularly important when examining relatively small population groups such as immigrants receiving SSI. Second, the CPS is better known and more commonly used in the research community. Conducting the same analysis using both CPS and SIPP provides a link with earlier research and a comparison of the two datasets. The features of the two samples, as well as the advantages and disadvantages of each, are summarized in Table A.1. Data from the two samples are consistent for most of the key results, although as shown in Table A.2, the SIPP sample is very small for some of the detailed calculations. Table A.1 Sample Sizes and Characteristics of the SIPP and CPS SIPP 1993–1994 CPS 1993–1995 –Smaller sample 6,219 California families 1,022 immigrant families+Large sample 15,494 California families 4,654 immigrant families –Less known, more complex to use+Better known, more commonly used +Longitudinal panel, collected every 4 months for 40 months–Cross-section, covering previous year +Monthly data on income and participation–Annual data on income and participation +Reported income comes closer to independent estimates–Reported income falls short of independent estimatesa –Immigrant status identifiable only for persons over 15 years old+Immigrant status identifiable for all persons –Weighting problems specific to immigrant population found in 1993 and 1994 data NOTE: + indicates an advantage over the other dataset; – indicates a disadvantage. aCitro and Kalton (1993). _________ 1Further information about the SIPP sample can be found in Appendix A of that report. 39 Table A.2 Unweighted Subsample Sizes for SIPP and CPS Results SIPP 1993–1994CPS 1993–1995 Califoria families All California families 6,219 15,494 Citizen families 5,197 10,840 Immigrant families 1,022 4,654 Recent immigrant families 480 1,155 Immigrant welfare participant families Families participating in major programs 416 1,376 Families participating in AFDC 139 522 Families participating in SSI 99 287 Families participating in Food Stamps only 52 241 Immigrant recipients affected by benefit denials Receiving SSI 77 287 Receiving Food Stamps but not SSI 141 607 Receiving SSI or Food Stamps 218 894 However, where results differ between the two samples, these differences do not appear to be driven by the different sample sizes. In other words, the differences are not an artifact of the limited number of California immigrant observations in the SIPP. Instead, they result from the different survey strategies. Indeed, the same patterns of results occur in the national data and for all families in California (both cases with larger samples). Overall, there appear to be three fundamental differences between the two datasets. First, the monthly reporting from repeated surveys in SIPP identifies more income than CPS. For this reason, the poverty rate is higher in CPS than in SIPP. The SIPP also finds more welfare participation than CPS. Finally, CPS respondents who do report welfare participation report more benefits on average than welfare recipients in SIPP. Detailed results from each dataset are included in Appendix B. Similar findings regarding SIPP versus CPS have been documented in two studies by the National Academy of Sciences, one examining the SIPP (Citro and Kalton,1993) and a second examining the poverty measure (Citro and Michael, 1995). These reports also show the lower poverty rate in SIPP and note that there is more complete income reporting in SIPP than in CPS, coming closer to independent measures for most categories of income. This is especially true for transfer income, which is highlighted in 40 these studies. In fact, one recommendation in Citro and Michael is replacement of the CPS by the SIPP for calculation of the poverty rate. Focus on Families This analysis focuses on the resources of families receiving public assistance in California. It is perhaps surprising that families are rarely the unit of observation in welfare literature. Despite the rhetoric of family responsibility, researchers typically examine either individuals or cases, neither of which maps to policy goals. As policymakers, we hope to make families self-sufficient, but we do not expect self - sufficiency of every individual, especially children. Although administrative cases come closer to families, they need not match our concept of families. For example, a family consisting of a mother and her two children would be categorized as two different cases if one child is disabled and receiving SSI. In highlighting families and family income, we believe that we come closer to understanding the economic reality for people receiving public assistance. Of course, in the 1990s, defining a family is not as simple as it once was. For the purposes of this analysis, all people are assigned to families. A family is defined as one or more people living in the same household, related by birth, marriage, or adoption. Most of these families are conventional nuclear families: an individual, his or her spouse (if present in the home), and any children living at home. Some families are extended families, such as when a grandparent lives in the household with his or her child and grandchildren. Beyond these nuclear relationships, we do not combine relatives into the same families if there would be no expectation of shared resources even when they share a household. Thus, two adult sisters living in the same house represent two separate families. Individuals living alone or with nonrelatives are one-person families. This includes unmarried, cohabiting partners. This study examines the families of immigrants. We define immigrants to be foreign- born noncitizens. A family is considered an immigrant family if one or more members are immigrants. Immigrant families may include citizen members. However, we have no information on family sponsors, so an immigrant family may or may not include the sponsor. Therefore, in the case of sponsored immigrants, the resources of the family may 41 underestimate the total resources available—to the extent that sponsors fulfill their agreements to prevent immigrants from becoming public charges. Analytical Approach The core results in this report simulate the effects of denying and then restoring SSI and Food Stamps benefits to immigrant families. We calculate the amount (and share) of a family’s income that would be lost if an immigrant family no longer received SSI and Food Stamps. There are four steps to this calculation: 1. Identify affected individuals in affected families. A family is potentially affected if there is an immigrant family member and the family received SSI and/or Food Stamps. However, we exclude income received by family members who do not lose eligibility: citizen and refugee family members. 2 So some families who are potentially affected do not lose any benefits, and others lose benefits only for some family members. 2. Calculate the loss of SSI and Food Stamps. For noncitizen, nonrefugee family members, the cutoff of SSI and Food Stamps entailed losing all benefits from these programs. We add up benefits received by such family members over the course of a year and count this amount as the “projected loss” that is reported in the empirical results. 3. Determine eligibility for AFDC. Some immigrant families receive SSI for a disabled or blind child. In these families, the child may qualify for AFDC if SSI is denied. Since we can determine eligibility on a monthly basis in the SIPP data, we can determine AFDC eligibility after removing SSI. (We can also exclude only benefits for the specific months in which a now-ineligible person was the benefit recipient.) In SIPP calculations, we add back in the AFDC benefits. The monthly calculation is not possible with the CPS data. 4. Evaluate the family’s economic status after losing benefits. After calculating the projected income loss, we recalculate family income as a share of poverty and by income quintile. In doing so, we also report what other economic resources are still available to the family. These final estimates not only provide information about the income losses expected under denial of SSI and Food Stamps, but they also give us a distribution of family _________ 2We do not have direct information on refugee status. We assign refugee status to all immigrants from refugee-sending countries based on the year of arrival. In the SIPP, identifiable refugee-sending countries include Cuba, Czechoslovakia, Hungary, Poland, the U.S.S.R., and Vietnam. 42 income in the absence of these benefits. With this information, we can evaluate what kinds of families benefit from having SSI or Food Stamps restored, as recently passed at the federal and state levels. Qualifications in Interpreting Findings As we note in the discussion of our empirical analyses, our estimates of the total costs and values of losses and reimbursements are invariably on the high side for the policy reforms currently under consideration, because these estimates apply to the circumstances relevant in the 1993–1995 period when caseloads were larger. Moreover, our estimates rely on annual income, and not monthly income, as used to determine welfare eligibility. Thus, it is quite possible that a family could be well above the poverty level for the year as a whole but below for several months. We invite the reader to be the judge of how to evaluate this qualification. Although we believe that our estimates for totals are high for predicting the values of welfare benefits involved in the policy reforms affecting immigrants, the robustness of our findings on the shares of losses and gains going to the various immigrant groups leads us to be confident about the implications of our findings regarding the distributional consequences of alternative polices. 43 Appendix B Primary Tables and Description This appendix describes the construction of the primary tables that summarize our empirical findings. The discussion outlines the definitions of the variables presented in these tables, along with the samples used to calculate the estimates. Each table is reproduced for each of the two datasets. The SIPP results are shown first, then the CPS results. Data available in SIPP are not always reproducible in CPS. Therefore, for some rows of data, the CPS version of the table will be blank. In tables showing the simulation results for loss of benefits, there are two SIPP versions. The “best case” version considers eligibility and income loss on a monthly basis. The “worst case” version eliminates income for the entire year. The CPS results are roughly equivalent to the “worst case” version in SIPP. All incomes and benefits are converted to 1996 dollars using the June Consumer Price Indexes (CPIs) for 1993 to 1996. The SIPP results combine annual data for 1993 and 1994, and the CPS combines annual data for 1993 through 1995. There are certain conventions to keep in mind when reading the tables: 1.Panels: Panels in tables combine rows summarizing closely related information, and these panels are also replicated across tables. For simplicity, we may refer to panels to describe subsections of the tables. 2.Columns: Samples are common between rows in a given column, but are different across columns. 44 3.Indents: Indented rows indicate that the listed statistics are conditional on and only applicable to the group identified in the first nonindented row above the statistics. For example, an indented row below a nonindented participation rate means that the statistic is conditional on participating. Tables B.1a and B.1b: Characteristics of California Families Tables B.1a and B.1b present summary statistics on the characteristics of all California families and of immigrant families in particular. These characteristics include demographic characteristics, race/ethnicity, poverty status, and welfare participation. Sample: All California families. Columns: Tables B.1a and B.1b have four columns: All: All California families are combined for the first column. In the remaining three columns, families are divided by citizenship status. All families qualify as either citizen families or immigrant families. Citizens: The column labeled “Citizens” presents families in which all members are born or naturalized U.S. citizens. All Immigrants: These are families in which one or more members are foreign- born noncitizens. Recent Immigrants: These are a subset of “All Immigrants.” These are families in which one or more members are foreign-born noncitizens who arrived after 1985 (Table B.1a) or after 1990 (Table B.1b). Rows: The rows in these tables are divided into seven “panels.” The definitions of the variables in these panels are: Panel A: % of all families in California. The first row shows the share of all California families represented in the particular column. Panel B: Age of family head. This group of rows, along with the next two panels, presents demographic characteristics of families and/or family heads. In a family headed by a married couple, the head is always taken to be the male. The three rows in this panel divide family heads into different age groups (e.g., “% £25” refers to family heads who are age 25 or younger). 45 Table B.1a Characteristics of Citizen and Immigrant Families and Their Welfare Participation in California: SIPP Data Programs Family Characteristics All CitizensAll ImmigrantsRecent Immigrants A % of all families in California100.0 80.8 19.2 8.5 B Age of family head % £ 25 8.5 8.5 8.6 14.1 % 26–64 73.7 71.3 83.9 82.1 % 65+ 17.7 20.2 7.5 3.8 C Family characteristics % married 49.3 46.2 62.6 62.0 % female heads 28.9 31.2 18.8 17.3 % divorced 15.1 17.1 6.7 4.2 % never married 21.4 21.8 19.4 23.8 % with children 36.6 31.4 58.3 62.4 No. of children 2.0 1.9 2.3 2.2 % with child £ age 6 22.6 19.9 28.8 28.2 % with disability in family 19.6 20.0 18.0 14.1 % with child disability 21.1 21.2 20.3 16.3 D Education of family head % less than high school 21.7 16.1 45.3 41.7 % high school graduate 24.7 25.6 20.9 21.5 % some college + 53.6 58.3 33.8 36.7 E Race/ethnicity % white 63.5 73.9 19.8 18.2 % black 6.2 7.5 0.8 1.0 % Hispanic 20.9 12.1 57.8 53.9 % Asian 8.3 5.2 21.6 26.9 F Poverty status % < traditional poverty level 11.2 9.0 20.5 20.0 Income within poverty level % < 2/3 3.2 2.8 5.0 6.2 % 2/3–1 7.1 5.6 13.5 11.8 % 1–1.85 19.9 17.1 32.0 36.0 % 1.85+ 69.7 74.5 49.5 46.0 G Welfare participation % any 33.6 27.5 59.7 62.2 % any major program 24.3 20.3 41.4 44.4 % any major and minor 16.1 12.1 32.7 36.5 % only minor 9.3 7.2 18.2 17.8 46 Table B.1b Characteristics of Citizen and Immigrant Families and Their Welfare Participation in California: CPS Data Programs Family Characteristics All CitizensAll ImmigrantsRecent Immigrants A % of all families in California100.0 77.1 22.9 6.1 B Age of family head % £ 25 8.1 7.3 10.8 20.2 % 26–64 74.6 72.3 82.6 76.9 % 65+ 17.3 20.4 6.6 2.9 C Family characteristics % married 48.7 45.1 61.0 58.8 % female heads 29.0 31.6 20.2 20.1 % divorced 14.1 16.7 5.5 3.1 % never married 24.7 24.5 25.3 32.0 % with children 36.8 31.1 56.1 56.2 No. of children 2.0 1.8 2.2 2.1 % with child £ age 6 21.3 16.7 36.9 40.1 % with disability in family % with child disability D Education of family head % less than high school 20.0 11.2 49.0 49.9 % high school graduate 23.1 24.5 18.5 14.6 % some college + 57.0 64.3 32.5 35.6 E Race/ethnicity % white 60.7 73.8 16.5 16.2 % black 6.8 8.4 1.6 1.8 % Hispanic 23.6 12.0 62.8 55.9 % Asian 8.0 4.9 18.5 25.5 F Poverty status % < traditional poverty level 17.2 13.3 30.5 37.5 Income within poverty level % < 2/3 9.4 7.3 16.2 22.4 % 2/3–1 7.3 5.5 13.3 13.9 % 1–1.85 18.2 15.7 26.7 27.1 % 1.85+ 65.1 71.5 43.8 36.6 G Welfare participation % any 29.6 24.0 48.3 48.1 % any major program 17.5 13.9 29.6 31.6 % any major and minor 8.3 5.7 17.1 16.2 % only minor 12.1 10.1 18.7 16.5 47 Panel C: Family characteristics. This group of rows presents several family structure characteristics. The analysis classifies family heads as married, unmarried female, divorced, or never married. It also examines families with children and families with a member having physical disability. (Information on disability is not available in the CPS.) The indented rows indicate that the statistic is conditional on and only applicable to the category listed above (e.g., statistics for “No. of children” and “% with child £ age 6” are computed only for those families in the category “% with children”—the line above). Panel D: Education of family head. These rows show the distribution of educational attainment (highest grade completed) for heads making up each participation group. Panel E: Race/ethnicity. The four rows making up this group break down program participants into their racial/ethnic groups: whites, blacks, Hispanics, and Asians. Hispanics are defined to be white Hispanics, and whites are non-Hispanic. Panel F: Poverty status. The first row in this panel presents the percentage of each column below the traditional poverty level. The traditional poverty level does not included the cash value of Food Stamps in the calculation of total income. In contrast, our calculations always include the this cash value. The next four rows report how our income measure compares with each family’s poverty level (e.g., “% < 2/3” indicates that the family’s income is less than two-thirds of the poverty level, whereas “% 1.85+” designates that income exceeds 185 percent of the poverty level). We use 1.85 as a reference level because this value is used by many means-tested programs to end benefits. Panel G: Welfare participation. The final set of rows reports participation in public assistance programs. “Major programs” include the four major public assistance programs: AFDC, SSI, Food Stamp, and Medicaid (known in California as Medi-Cal). We identify any family who is enrolled in or collects any benefits from any major program to be a “welfare participant.” “Minor programs” include Special Supplemental Nutrition Program for Women, Infants and Children (WIC), GA, school-based food programs, housing assistance, and energy assistance. The first row in this group presents the percentage of families receiving benefits from any major or minor program. The second row refers to families participating in one or more of the major programs. “% any major and minor” includes families who participate in at least one major and one minor 48 program, and “% only minor” shows the percentage of families who participated only in minor assistance programs. Tables B.2a and B.2b: Income Sources for Immigrant Families Receiving Welfare Tables B.2a and B.2b report an array of statistics designed to summarize income sources and amounts received by immigrant families who received welfare in California. Sample: All immigrant families who participated in a major welfare program. Columns: Results in Tables B.2a and B.2b are divided into four columns based on program participation: Any major: This column presents results for all welfare-participant families (i.e., those who participated in any major program in at least one month during the sample period for the SIPP data or at any time during the three years of the CPS data). The remaining columns break down these program participants into three groups: AFDC: This column includes all families who collected benefits from the AFDC program some time in the observation period. SSI: This group includes all families who collected benefits from the SSI program some time in the observation period. Food Stamps only: This group includes families who received Food Stamps benefits but who collected no benefits from either AFDC or SSI—they could have been Medi-Cal participants. Families receiving both AFDC and SSI are included in both the SSI and the AFDC columns. Families receiving Medi-Cal only are included in the “any major” column but are not listed separately. Rows: The rows of Tables B.2a and B.2b group into three sections: Panel A: % of all immigrant families in California. This first row shows program participants as a share of all immigrant families. It not only shows the participation by program but it also relates the sample in this table to the information shown in Table A.1. For example, in the SIPP data, 41 percent of all immigrant families are reported to participate in any major program. This figure matches the number in the “% any major program” row for the “All Immigrants” column in Table B.1a. 49 Table B.2a Sources of Income and Amounts Received by Immigrant Families Who Are Welfare Participants in California: SIPP Data Programs Family Characteristics Any Major AFDC SSIFood Stamps Only A % of all immigrant families in California41.0 13.0 11.0 5.0 B Income (1996$) Annual total income 25th percentile 13,625 12,705 13,410 17,051 50th percentile 21,042 16,302 25,023 23,527 75th percentile 34,552 24,004 51,672 34,862 % of families with earnings 81.0 69.0 63.0 100.0 Annual earnings 25th percentile 9,086 2,713 13,924 10,671 50th percentile 18,986 7,676 31,073 18,064 75th percentile 33,404 18,030 54,402 31,787 % of income from earnings 25th percentile 36.0 13.0 44.0 51.0 50th percentile 67.0 34.0 79.0 76.0 75th percentile 91.0 60.0 91.0 90.0 % from all cash welfare 24.0 50.0 40.0 6.0 % from AFDC+SSI+Food Stamps 23.0 49.0 39.0 5.0 % from AFDC 11.0 35.0 6.0 0.0 % from SSI/State Supplementary Payment 8.0 4.0 31.0 0.0 % from Food Stamps 4.0 11.0 2.0 5.0 % from other income 26.0 24.0 18.0 25.0 C Welfare utilization (1996$) % receiving benefits ³ 6 months 53.0 94.0 84.0 62.0 % of income from cash benefits 25th percentile 0.0 15.0 6.0 1.0 50th percentile 5.0 47.0 17.0 2.0 75th percentile 37.0 92.0 86.0 7.0 Annual cash benefits 25th percentile 148 4,816 2,399 272 50th percentile 1,850 8,990 6,642 606 75th percentile 7,463 12,870 7,978 1,513 % AFDC participants 32.0 100.0 21.0 0.0 % of income from AFDC 35.0 35.0 31.0 . Annual AFDC benefits 6,179 6,179 7,203 . % SSI participants 26.0 17.0 100.0 0.0 % of income from SSI 31.0 21.0 31.0 . Annual SSI benefits 4,851 4,791 4,851 . % Food Stamps participants 45.0 91.0 34.0 100.0 % of income from Food Stamps 9.0 12.0 7.0 5.0 Annual Food Stamps benefits 1,727 2,209 1,649 1,007 % General Assistance (GA) 5.0 9.0 9.0 10.0 % of income from GA 10.0 5.0 3.0 11.0 Annual GA benefits 1,522 1,201 787 2,055 % other welfare 90.0 98.0 80.0 94.0 % Medicaid participants 98.0 100.0 100.0 86.0 50 Table B.2b Sources of Income and Amounts Received by Immigrant Families Who Are Welfare Participants in California: CPS Data Programs Family Characteristics Any Major AFDC SSIFood Stamps Only A % of all immigrant families in California29.6 10.7 6.5 14.0 B Income (1996$) Annual total income 25th percentile 10,295 9,998 8,610 6,700 50th percentile 16,472 13,699 15,794 14,134 75th percentile 28,660 23,334 33,508 20,126 % of families with earnings 67.6 54.8 46.9 80.0 Annual earnings 25th percentile 8,648 3,809 6,524 8,236 50th percentile 16,961 9,765 20,590 13,482 75th percentile 29,856 21,745 43,888 19,255 % of income from earnings 25th percentile 59.5 29.2 30.7 72.8 50th percentile 84.0 59.8 67.5 88.1 75th percentile 98.3 77.7 83.0 96.5 % from all cash welfare % from AFDC+SSI+Food Stamps 36.3 62.9 52.0 23.3 % from AFDC 17.6 47.0 6.2 0.0 % from SSI/State Supplementary Payment 9.8 4.2 42.7 0.0 % from Food Stamps 8.9 11.8 3.1 23.3 % from other income 13.8 7.6 21.5 12.3 C Welfare utilization (1996$) % receiving benefits ³ 6 months % of income from cash benefits 25th percentile 0.0 27.3 15.6 2.7 50th percentile 16.6 70.7 47.7 8.5 75th percentile 77.0 100.0 100.0 25.5 Annual cash benefits 25th percentile 0 4,520 3,867 362 50th percentile 2,539 8,437 7,412 951 75th percentile 7,867 12,082 11,413 1,853 % AFDC participants 37.4 100.0 20.9 0.0 % of income from AFDC 47.0 47.0 29.7 . Annual AFDC benefits 6,787 6,787 6,533 . % SSI participants 22.8 12.8 100.0 0.0 % of income from SSI 42.7 32.8 42.7 . Annual SSI benefits 6,510 6,548 6,510 . % Food Stamps participants 49.1 76.8 25.0 100.0 % of income from Food Stamps 18.2 15.4 12.5 23.3 Annual Food Stamps benefits 1,773 2,047 2,103 1,301 % GA . . . . % of income from GA . . . . Annual GA benefits . . . . % other welfare 58.9 71.3 54.4 57.1 % Medicaid participants 70.8 60.2 84.9 35.2 51 Panel B: Income (1996$). Annual total income: The first group of rows measures the total annual income received by participant families, showing the 25th, 50th, and 75th percentiles of the total annual income distribution. % of families with earnings: This row shows the percentage of immigrant welfare families who had earnings and are members of the sample designated by the column heading. Using SIPP for an example, 81 percent of all welfare participant immigrant families worked at some time during the 1993–1994 period. Sixty-nine percent of those who received AFDC also had earnings. The indented rows under this subheading are conditional on families having earnings. They first report the 25th, 50th, and 75th percentiles of the distribution of annual earnings obtained by working families making up the relevant sample. The next set of rows shows the 25th, 50th, and 75th percentiles of the distribution of total income shares provided by family earnings. % from all cash welfare: The remaining rows in this panel show the average percentage of family annual income from sources other than earnings, including income from cash benefits received from all public assistance programs, from cumulated AFDC + SSI + Food Stamps benefits, from the individual major programs, and from sources of “other” income (including Social Security, Unemployment Insurance, pensions and so on). The percentage from all cash welfare (or AFDC + SSI + Food Stamps for CPS), from other income and from earnings together total 100 percent. Panel C: Welfare utilization (1996$). The first row of this panel shows the percentage of families in the relevant sample who participated in any combination of major programs for six months or more. This information is available only in the SIPP. The next rows report the 25th, 50th, and 75th percentiles of the distribution of shares of total annual income contributed by cash benefits from all public assistance programs, and the next group gives the corresponding percentiles for the annual cash benefits. For each public assistance program with information in SIPP, the remaining rows show the participation rate of the column sample in each program. For programs with income information, indented rows show the average share of annual income attributable to benefits from that program and the average annual payments conditional on participating in the program. For example, considering all immigrant welfare participants (the first column of Table B.2a), 26 percent collected SSI benefits; these benefits represented 31 52 percent of SSI recipient families’ income on average; and the average annual payment to SSI families was $4,851. Income information for GA is not available in CPS. Tables B.3a, B.3b, and B.3c: Income Losses of Immigrant Families Tables B.3a, B.3b, and B.3c present the projected effects of the denial of SSI and Food Stamps benefits to legal immigrants. It is the first table to include the two alternative SIPP calculations: the best case (Table B.3a) and the worst case (Table B.3b). As noted, the essential difference between these two calculations is the method used to compute lost benefits. In all cases, only noncitizen nonrefugees are considered ineligible for benefits. In the CPS data and the SIPP worst case, anyone ineligible loses all benefits received annually. In the SIPP case, the benefits are recalculated using the appropriate formulas for monthly AFDC benefits as well as for SSI and Food Stamps. This results in lower benefit losses but also allows AFDC benefits to replace SSI benefits. Sample: Immigrant families who participated in SSI or Food Stamps. Columns: The columns in these tables clarify the different effects on immigrants depending on which program or programs they participated in: Food Stamps/no SSI: This column indicates families who received Food Stamps benefits but no SSI benefits. These families may also receive benefits from other welfare programs such as AFDC. SSI: This column indicates families who collected benefits from the SSI program. They may or may not have also received Food Stamps or other welfare benefits. Food Stamps &/or SSI: This is the summation of the other two categories. This column counts together families receiving benefits from either or both programs. Rows: These tables have four panels each: Panel A: % of recipient immigrant families potentially affected. The families who potentially lose benefits are only a subset of all immigrant families receiving welfare. This first row relates these affected families to the samples in Tables B.1 and B.2. Panel B: Sources of eligibility. The first two rows in this panel show the share of potentially affected families that include still-eligible family members—either refugees or citizens. (Federal legislation provides exemptions for refugees, which is not the same as a 53 Table B.3a Income Losses of Immigrant Families Affected by Elimination of Food Stamp and SSI Programs: SIPP Data, Best Case Programs Family CharacteristicsFood Stamps/ No SSI SSIFood Stamps and/or SSI A % of recipient immigrant families potentially affected36.0 25.6 61.6 B Sources of eligibility % of families with refugees 11.9 14.6 13.0 % of families with citizens 83.1 56.8 72.1 % of potential group actually losing benefits 93.9 77.3 87.0 C Projected loss of benefits and income (1996$) Average loss of annual Food Stamps benefits 1,295 492 998 Average recovery of annual benefits from AFDC 0 (290) (107) Average loss of annual SSI benefits 0 4,878 1,803 Average loss of total annual benefits 1,295 5,080 2,694 Annual income loss 25th percentile 496 1,850 786 50th percentile 1,083 4,612 1,509 75th percentile 1,876 7,595 3,157 % of families with annual income losses ³ 10 % of total income 35.5 62.0 45.3 ³ 25 % of total income 1.6 46.9 18.3 ³ 50 % of total income 0.0 22.4 8.3 D Change in poverty status Before eligibility denial % above 185% of poverty level 14.4 37.2 22.8 With children 12.3 21.0 15.5 Single 3.7 3.5 3.6 Family head age 65+ 0.0 3.8 1.4 % below poverty level 54.5 13.0 39.1 With children 52.9 7.5 36.1 Single 20.5 1.7 13.6 Family head age 65+ 0.0 4.0 1.5 % below 2/3 of poverty level 8.8 2.9 6.6 With children 8.8 1.7 6.2 Single 2.6 1.7 2.2 Family head age 65+ 0.0 1.2 0.4 After eligibility denial % above 185% of poverty level 11.5 31.8 19.0 With children 10.1 18.4 13.2 Single 2.5 2.2 2.4 Family head age 65+ 0.0 0.9 0.3 % below poverty level 62.7 47.6 57.1 With children 60.9 14.9 43.9 Single 27.1 6.4 19.5 Family head age 65+ 0.0 28.9 10.7 % below 2/3 of poverty level 22.3 28.6 24.6 With children 22.3 6.1 16.3 Single 8.7 1.7 6.1 Family head age 65+ 0.0 20.0 7.4 54 Table B.3b Income Losses of Immigrant Families Affected by Elimination of Food Stamp and SSI Programs: SIPP Data, Worst Case Programs Family CharacteristicsFood Stamps/ No SSI SSIFood Stamps and/or SSI A % of recipient immigrant families potentially affected36.0 25.6 61.6 B Sources of eligibility.. . % of families with refugees 11.9 14.6 13.0 % of families with citizens 83.1 56.8 72.1 % of potential group actually losing benefits 93.9 77.3 87.0 C Projected loss of benefits and income (1996$) Average loss of annual Food Stamps benefits 1,717 611 1,308 Average recovery of annual benefits from AFDC 0 0 0 Average loss of annual SSI benefits 0 4,988 1,844 Average loss of total annual benefits 1,717 5,599 3,152 Annual income loss . . . 25th percentile 593 2,086 867 50th percentile 1,509 5,512 2,072 75th percentile 2,544 7,867 4,004 % of families with annual income losses . . . ³ 10 % of total income 47.3 67.4 54.8 ³ 25 % of total income 10.9 50.1 25.4 ³ 50 % of total income 0.0 28.5 10.5 D Change in poverty status.. . Before eligibility denial . . . % above 185% of poverty level 14.4 37.2 22.8 With children 12.3 21.0 15.5 Single 3.7 3.5 3.6 Family head age 65+ 0.0 3.8 1.4 % below poverty level 54.5 13.0 39.1 With children 52.9 7.5 36.1 Single 20.5 1.7 13.6 Family head age 65+ 0.0 4.0 1.5 % below 2/3 of poverty level 8.8 2.9 6.6 With children 8.8 1.7 6.2 Single 2.6 1.7 2.2 Family head age 65+ 0.0 1.2 0.4 After eligibility denial . . . % above 185% of poverty level 11.5 31.8 19.0 With children 10.1 18.4 13.2 Single 2.5 2.2 2.4 Family head age 65+ 0.0 0.9 0.3 % below poverty level 64.6 50.3 59.3 With children 62.7 17.6 46.1 Single 27.1 7.7 20.0 Family head age 65+ 0.0 28.9 10.7 % below 2/3 of poverty level 28.7 35.0 31.1 With children 28.7 12.3 22.7 Single 12.7 6.4 10.3 Family head age 65+ 0.0 21.5 7.9 55 Table B.3c Income Losses of Immigrant Families Affected by Elimination of Food Stamp and SSI Programs: CPS Data Programs Family CharacteristicsFood Stamps/ No SSI SSIFood Stamps and/or SSI A % of recipient immigrant families potentially affected41.9 22.1 63.9 B Sources of eligibility % of families with refugees 4.3 2.9 3.8 % of families with citizens 74.1 51.9 66.5 % of potential group actually losing benefits 96.5 90.5 94.4 C Projected loss of benefits and income (1996$) Average loss of annual Food Stamps benefits 1,711 474 1,302 Average recovery of annual benefits from AFDC . . . Average loss of annual SSI benefits 0 6,323 2,091 Average loss of total annual benefits 1,711 6,797 3,393 Annual income loss 25th percentile 652 3,174 976 50th percentile 1,396 6,055 2,088 75th percentile 2,545 8,002 4,063 % of families with annual income losses ³ 10 % of total income 55.9 80.1 63.9 ³ 25 % of total income 21.4 58.5 33.7 ³ 50 % of total income 7.5 39.6 18.1 D Change in poverty status Before eligibility denial % above 185% of poverty level 7.7 25.2 13.5 With children 5.1 12.7 7.7 Single 1.8 4.1 2.6 Family head age 65+ 0.0 3.5 1.2 % below poverty level 64.5 28.7 52.7 With children 52.7 11.7 38.6 Single 22.5 4.0 16.1 Family head age 65+ 0.7 6.8 2.8 % below 2/3 of poverty level 33.1 10.2 25.5 With children 25.0 4.9 18.1 Single 10.2 2.2 7.5 Family head age 65+ 0.5 2.7 1.2 After eligibility denial % above 185% of poverty level 7.4 19.9 11.5 With children 4.7 11.3 7.0 Single 1.5 3.9 2.3 Family head age 65+ 0.0 1.3 0.4 % below poverty level 70.1 61.4 67.2 With children 57.2 19.8 44.3 Single 23.6 6.0 17.5 Family head age 65+ 0.7 23.9 8.7 % below 2/3 of poverty level 48.8 48.0 48.6 With children 39.2 14.8 30.8 Single 16.4 5.2 12.5 Family head age 65+ 0.6 19.1 7.0 56 family immigrating from a refugee-sending country, but SIPP does not designate refugee status.) Refugee-sending countries identifiable in the SIPP are: Cuba, Czechoslovakia, Hungary, Poland, the U.S.S.R., and Vietnam. The final row lists the share of families who would actually lose benefits after accounting for all eligible family members. Panel C: Projected loss of benefits and income (1996$). These rows summarize the annual income losses projected if all affected families identified in the last row of panel B lose Food Stamps and SSI. As noted above, Table B.3a shows the income loss when benefits are recalculated on a monthly basis in the SIPP. Tables B.3b and B.3c remove all benefits received in a year by ineligible persons. The first set presents the average losses for Food Stamps, AFDC, SSI, and total annual cash benefits. The second set of rows reports the 25th, 50th, and 75th percentiles of the distribution of the annual dollar loss in benefits across affected families. The third set lists the share of families losing more than 10, 25, or 50 percent of their total annual family income. Panel D: Income status. This panel is divided into two parts: The top portion reports family income status before eligibility denial from these programs, and the bottom projects family income status after eligibility denial. For each portion, the results give the percentage of affected families who had incomes above 185 percent of the poverty level, below the poverty level, and below two-thirds of the poverty level. Within these subgroups, the table reports the percentage of families who fall into the poverty level category and who have children, are single parents, or have a family head over age 65. For example, before eligibility denial, 21 percent of affected SSI families were families with children and with incomes over 185 percent of the poverty level, whereas 4 percent were families headed by an elderly person living below the poverty line. Tables B.4a, B.4b, and B.4c: Income Status After Elimination of Food Stamp and SSI Programs These tables consider the other resources available to immigrant families if they were denied Food Stamps and SSI. As in Tables B.3a, B.3b, and B.3c, they include the two alternative SIPP calculations: the best case (Table B.4a) and the worst case (Table B.4b). Sample: Immigrant families who lose SSI or Food Stamps. Columns: The columns in this table are the same as those in Tables B.3a, B.3b, and B.3c. 57 Table B.4a Income Status of Immigrant Families After Elimination of Food Stamp and SSI Programs: SIPP Data, Best Case Programs Family CharacteristicsFood Stamps/ No SSI SSIFood Stamps and/or SSI A Welfare utilization (1996$) % receiving benefits 86.1 50.8 73.1 % of income from cash benefits 25th percentile 4.5 0.0 0.0 50th percentile 42.5 0.6 18.4 75th percentile 100.0 52.4 98.3 Annual cash benefits 25th percentile 1,269 0 0 50th percentile 7,391 166 3,790 75th percentile 12,325 6,076 11,135 % AFDC participants 64.3 37.0 54.2 % of income from AFDC 50.8 36.1 47.1 Annual AFDC benefits 6,192 5,068 5,908 % SSI participants 0.0 24.3 9.0 % of income from SSI . 6.9 6.9 Annual SSI benefits . 1,387 1,387 % Food Stamps participants 57.9 20.0 43.9 % of income from Food Stamps 5.4 4.1 5.2 Annual Food Stamps benefits 730 654 718 % GA 6.1 8.9 7.1 % of income from GA 14.2 3.9 9.4 Annual GA benefits 2,232 616 1,482 % other welfare 97.0 82.7 91.7 % Medicaid participants 94.9 100.0 96.8 B Income (1996$) % of families with earnings 82.0 57.7 73.0 Annual earnings 25th percentile 5,586 11,782 6,336 50th percentile 11,536 26,850 13,924 75th percentile 19,842 47,995 30,520 % of income from earnings 25th percentile 38.8 49.7 44.3 50th percentile 69.5 84.7 75.3 75th percentile 88.7 91.5 90.5 % of families with other income 87.1 80.3 84.6 Annual other income 25th percentile 896 1,493 1,101 75th percentile 15,668 10,256 12,736 % of income from other income 25th percentile 5.4 7.9 6.0 75th percentile 74.7 100.0 90.6 58 Table B.4b Income Status of Immigrant Families After Elimination of Food Stamp and SSI Programs: SIPP Data, Worst Case Programs Family CharacteristicsFood Stamps/ No SSI SSIFood Stamps and/or SSI A Welfare utilization (1996$) % receiving benefits 76.3 40.5 63.1 % of income from cash benefits . . . 25th percentile 1.4 0.0 0.0 50th percentile 42.5 0.0 17.9 75th percentile 100.0 45.1 98.2 Annual cash benefits . . . 25th percentile 303 0 0 50th percentile 6,928 0 3,189 75th percentile 11,960 3,772 10,364 % AFDC participants 64.3 22.1 48.7 % of income from AFDC 54.0 60.5 55.1 Annual AFDC benefits 6,192 7,196 6,361 % SSI participants 0.0 15.3 5.7 % of income from SSI . 8.8 8.8 Annual SSI benefits . 1,478 1,478 % Food Stamps participants 1.0 1.2 1.1 % of income from Food Stamps 0.8 3.5 1.9 Annual Food Stamps benefits 103 951 458 % GA 6.1 8.9 7.1 % of income from GA 15.1 4.5 10.2 Annual GA benefits 2,232 616 1,482 % other welfare 97.0 82.7 91.7 % Medicaid participants 94.9 100.0 96.8 B Income (1996$).. . % of families with earnings 82.0 57.7 73.0 Annual earnings . . . 25th percentile 5,586 11,782 6,336 50th percentile 11,536 26,850 13,924 75th percentile 19,842 47,995 30,520 % of income from earnings . . . 25th percentile 38.8 52.2 44.4 50th percentile 70.5 84.7 76.0 75th percentile 90.6 92.9 91.6 % of families with other income 87.1 80.3 84.6 Annual other income . . . 25th percentile 896 1,493 1,101 75th percentile 15,668 10,256 12,736 % of income from other income . . . 25th percentile 5.4 8.1 6.1 75th percentile 74.7 100.0 94.3 59 Table B.4c Income Status of Immigrant Families After Elimination of Food Stamp and SSI Programs: CPS Data Programs Family CharacteristicsFood Stamps/ No SSI SSIFood Stamps and/or SSI A Welfare utilization (1996$) % receiving benefits 58.2 28.5 48.4 % of income from cash benefits 25th percentile 0.0 0.0 0.0 50th percentile 22.2 0.0 7.6 75th percentile 95.1 31.3 80.2 Annual cash benefits 25th percentile 0 0 0 50th percentile 2,520 0 0 75th percentile 7,697 1,305 6,393 % AFDC participants 58.2 22.6 46.4 % of income from AFDC 65.1 57.4 63.8 Annual AFDC benefits 7,197 6,538 7,091 % SSI participants 0.0 3.9 1.3 % of income from SSI . 25.7 25.7 Annual SSI benefits . 5,952 5,952 % Food Stamps participants 0 0 0 % of income from Food Stamps . . . Annual Food Stamps benefits . . . % GA . . . % of income from GA . . . Annual GA benefits . . . % other welfare 66.7 53.8 62.5 % Medicaid participants 46.0 83.9 58.5 B Income (1996$) % of families with earnings 62.6 44.6 56.6 Annual earnings 25th percentile 5,219 5,971 5,290 50th percentile 10,873 19,561 12,354 75th percentile 18,531 43,888 21,794 % of income from earnings 25th percentile 52.3 39.4 50.8 50th percentile 81.9 80.6 81.8 75th percentile 100.0 99.5 100.0 % of families with other income 36.7 62.9 45.4 Annual other income 25th percentile 927 1,371 1,023 75th percentile 5,250 8,303 6,997 % of income from other income 25th percentile 6.5 12.8 7.6 75th percentile 44.6 100.0 82.3 60 Food Stamps/no SSI: This column indicates families who received Food Stamps benefits but no SSI benefits. SSI: This group includes families who collected benefits from the SSI program. Food Stamps &/or SSI: This is the summation of the other two categories. Rows: These tables have two panels each. Panel A: Welfare utilization (1996$). The rows in this panel are identical to those in Table B.2, panel C, except for the first row. In this table, the first row lists the share of affected families who would still receive public assistance even without SSI or Food Stamps. The income shares are adjusted to reflect the benefits lost due to the ineligibility of some or all family members. Note that the last row of this panel lists Medicaid participation. This calculation does not reflect the possibility of losing Medicaid through the loss of categorical eligibility through SSI. Panel B: Income (1996$): % of families with earnings: As in Table B.2, this row shows the percentage of immigrant families who lost benefits (as designated by the column heading) but who had earnings. The indented rows under this subheading are conditional on families having earnings. They first report the 25th, 50th, and 75th percentiles of the distribution of annual earnings obtained by working families making up the relevant sample. The next set of rows shows the 25th, 50th, and 75th percentiles of the distribution of total income shares provided by family earnings after elimination of Food Stamps and SSI. % of families with other income: These rows are structured in a similar fashion to the preceding rows. Here, the first row shows the percentage of immigrant welfare families who lost benefits but had “other” income such as Social Security, pensions, or Unemployment Insurance. The indented rows under this subheading are conditional on families having other income. They first report the 25th and 75th percentiles of the distribution of annual other income obtained by families in the column category who had other income. The next set of rows shows the 25th and 75th percentiles of the distribution of total income shares provided by other income after elimination of Food Stamps and SSI. 61 Tables B.5a and B.5b: Share of Reinstated Welfare Benefits Tables B.5a and B.5b count shares of dollars instead of shares of families. They consider the total benefits lost in the three scenarios of Table B.3 and the distribution of those dollars to various categories of families. Because of the similarity of results, only the “best” case for SIPP is included in this and the remaining tables in this appendix. Sample: Projected benefits lost by immigrant families denied SSI or Food Stamps. Columns: This table splits the reinstated dollars into the three family categories used for the last two tables: Food Stamps/no SSI: This indicates benefits theoretically reinstated to families who previously received Food Stamps benefits but no SSI benefits. SSI: This indicates benefits theoretically reinstated to families who participated in SSI, with or without Food Stamps. Food Stamps &/or SSI: This column is the summation of the other two categories. This column counts benefits reinstated to all immigrant families who would have lost benefits from either or both programs. Rows: These tables have five panels each. The first lists the total benefits reinstated, and the remaining four show the distribution of these benefits to different groups. Panel A: Total benefit loss/reinstated benefits (millions of 1996$). This row lists the costs of reinstating benefits for both programs to the three categories of families. Thus, under the SIPP best case, $1,060 million would be reinstated to SSI families if both their SSI and their Food Stamps benefits were restored. The total cost of restoring both programs would be $1,520 million. Panel B: Share of reinstated benefits by income quintile. This panel shows how restoration of the benefits listed in panel A would be distributed by income quintiles of the recipient families. The quintiles order all California families (immigrant and citizen) by their family income and split them into five equal-sized groups. The lowest income quintile, therefore, includes the 20 percent of California families with the lowest income. If the $460 million were split evenly across all California families, 20 percent would go to each quintile. However, 6 percent of the benefits restored to Food Stamps families who did not receive SSI would go to families in the bottom income quintile. Panel C: Share of reinstated benefits by poverty level. This panel shows how restoration of the benefits listed in panel A would be distributed based on the poverty level 62 Table B.5a Shares of Reinstated Welfare Benefits Going to Various Categories of Immigrant Families in California: SIPP Data Programs Family CharacteristicsFood Stamps/ No SSI SSIFood Stamps and/or SSI A Total benefit loss/reinstated benefits (millions of 1996$)460 1,060 1,520 B Share of reinstated benefits by income quintiles Lowest 66.4 63.4 64.3 Mid-low 21.1 11.7 14.6 Middle 8.0 6.3 6.8 Mid-high 4.1 13.5 10.6 Highest 0.4 5.1 3.7 C Share of reinstated benefits by poverty level < 0.66 35.7 47.8 44.1 0.66–1 43.6 18.7 26.2 1–1.85 15.3 10.8 12.2 > 1.85 5.5 22.7 17.5 D Share allocation by poverty level Income below 2/3 poverty level 35.7 47.8 44.1 With children < age 18 35.7 9.0 17.1 Single 12.7 0.7 4.3 Family head age 65+ . 35.5 24.8 Single . 12.5 8.7 Income below poverty level 79.2 66.5 70.3 With children < age 18 78.6 18.2 36.5 Single 33.3 5.6 14.0 Family head age 65+ . 43.5 30.3 Single . 15.4 10.7 Income from 1–1.85 of poverty level 15.3 10.8 12.2 With children < age 18 14.5 6.2 8.7 Single 4.2 0.9 1.9 Family head age 65+ 0.4 3.5 2.6 Single . 0.5 0.3 Income above 1.85 of poverty level 5.5 22.7 17.5 With children < age 18 5.5 15.7 12.6 Single 0.7 1.5 1.2 Family head age 65+ . 0.3 0.2 Single . . . E Shares received by AFDC recipients All AFDC families 77.2 32.0 45.7 Single 36.2 6.9 15.8 63 Table B.5b Shares of Reinstated Welfare Benefits Going to Various Categories of Immigrant Families in California: CPS Data Programs Family CharacteristicsFood Stamps/ No SSI SSIFood Stamps and/or SSI A Total benefit loss/reinstated benefits (millions of 1996$)601 1,179 1,780 B Share of reinstated benefits by income quintiles Lowest 56.5 62.4 60.4 Mid-low 31.6 14.1 20.0 Middle 8.8 6.4 7.2 Mid-high 2.8 11.4 8.5 Highest 0.4 5.8 4.0 C Share of reinstated benefits by poverty level < 0.66 56.7 60.6 59.3 0.66–1 21.3 9.1 13.2 1–1.85 17.5 12.5 14.2 > 1.85 4.4 17.8 13.3 D Share allocation by poverty level Income below 2/3 poverty level 56.7 60.6 59.3 With children < age 18 53.3 21.5 32.3 Single 19.6 6.8 11.2 Family head age 65+ 0.5 26.4 17.7 Single 0.4 11.9 8.0 Income below poverty level 78.0 69.7 72.5 With children < age 18 73.5 27.0 42.7 Single 26.8 8.2 14.4 Family head age 65+ 0.5 28.5 19.1 Single 0.4 12.8 8.6 Income from 1–1.85 of poverty level 17.5 12.5 14.2 With children < age 18 16.4 7.3 10.4 Single 4.6 0.3 1.8 Family head age 65+ 0.4 4.0 2.8 Single 0.2 0.8 0.6 Income above 1.85 of poverty level 4.4 17.8 13.3 With children < age 18 3.5 8.4 6.8 Single 0.9 2.9 2.2 Family head age 65+ 0.0 0.9 0.6 Single 0.0 0.0 0.0 E Shares received by AFDC recipients All AFDC families 68.3 24.3 39.2 Single 27.3 6.4 13.4 64 of the families in the absence of these benefits. For example, out of the $460 million potentially restored to families who previously received Food Stamps but not SSI, 35.7 percent would go to families with income below two-thirds of the poverty level. Panel D: Share allocation by poverty level. This panel provides the same information as panel C above, but in greater detail. For example, both panels C and D show in the second column that 22.7 percent of SSI and Food Stamps benefits restored to SSI families goes to families over 185 percent of the poverty level (without these benefits). Panel D shows that 15.7 percent of the $1,060 million would go to families over 185 percent of the poverty level who have children under age 18. As in all the tables, indented rows indicate that the statistics are conditional on belonging to the previous group. Panel E: Shares received by AFDC recipients. Many of the SSI and Food Stamps families also received AFDC. The final panel in this table shows what share of reinstated benefit dollars would go to families on AFDC and to single parent families on AFDC. Tables B.6a and B.6b: Shares of Reinstated SSI Benefits Going to Various Categories of Immigrant Families in California Tables B.6a and B.6b calculate the value of SSI benefits restored to California families when the 1997 budget deal undid these feature of the PRWORA. These tables detail the distribution of these benefits to various categories of California families. Like Table B.5, these tables count shares of dollars instead of shares of families. Sample: Projected benefits lost by immigrant families potentially denied SSI. Columns: These tables show the reinstated SSI benefit dollars split in two ways. First, they show the distribution of benefits by income level based on whether or not Food Stamps benefits are also restored. Second, they break the total SSI benefits into benefits restored to the disabled and benefits restored to the elderly. Accordingly, there are four relevant categories when evaluating the distribution of SSI program dollars: SSI alone: This category counts all SSI benefits reinstated to immigrant families, assuming that no Food Stamps benefits are restored. SSI after Food Stamps: This column counts all SSI benefits reinstated for immigrant families, but assumes that Food Stamps benefits have already been restored. The distinction between this column and the first column matters only for the distribution of benefits by income level. 65 Table B.6a Shares of Reinstated SSI Benefits Going to Various Categories of Immigrant Families in California: SIPP Data Programs Family CharacteristicsSSI AloneSSI After Food StampsSSI DisabledSSI Elderly A Total value of reinstated benefits (millions of 1996$)1,020 1,020 272 735 B Share of reinstated benefits by income quintiles Lowest 65.0 61.2 62.8 67.3 Mid-low 9.4 13.2 12.7 7.2 Middle 5.9 5.9 9.7 3.5 Mid-high 14.4 14.4 11.9 15.7 Highest 5.3 5.3 2.9 6.2 C Share of reinstated benefits by poverty level < 0.66 47.2 45.0 48.6 44.1 0.66–1 18.6 20.8 25.4 20.2 1–1.85 10.4 10.4 5.2 11.5 > 1.85 23.8 23.8 20.8 24.2 D Share allocation by poverty level Income below 2/3 poverty level 47.2 45.0 48.6 44.1 With children < age 18 6.8 4.7 25.5 0.4 Single 0.3 0.3 1.0 0.4 Family head age 65+ 37.0 37.0 10.4 . Single 13.0 13.0 0.0 . Income below poverty level 65.8 65.8 74.0 64.3 With children < age 18 15.6 15.6 43.2 5.9 Single 6.1 6.1 14.1 3.6 Family head age 65+ 45.8 45.8 10.4 59.8 Single 16.5 16.5 0.0 22.8 Income from 1–1.85 of poverty level 10.4 10.4 5.2 11.5 With children < age 18 5.7 5.7 3.6 6.3 Single 1.2 1.2 0.0 1.6 Family head age 65+ 4.0 4.0 0.6 4.8 Single 0.6 0.6 0.0 0.9 Income above 1.85 of poverty level 23.8 23.8 20.8 24.2 With children < age 18 16.7 16.7 11.2 18.9 Single 1.5 1.5 0.0 2.0 Family head age 65+ 0.3 0.3 0.0 . Single . . . . E Shares received by disabled recipients All families with disabled members 50.2 50.2 100.0 31.0 With children < age 18 3.2 3.2 57.9 6.3 Family head age 65+ 26.6 26.6 10.9 25.6 66 Table B.6b Shares of Reinstated SSI Benefits Going to Various Categories of Immigrant Families in California: CPS Data Programs Family CharacteristicsSSI AloneSSI After Food StampsSSI DisabledSSI Elderly A Total value of reinstated benefits (millions of 1996$)1,097 1,097 571 579 B Share of reinstated benefits by income quintiles Lowest 63.6 62.5 68.0 60.9 Mid-low 12.2 12.3 9.0 15.9 Middle 5.9 6.8 4.1 7.0 Mid-high 12.1 12.1 8.0 15.0 Highest 6.2 6.2 10.8 1.2 C Share of reinstated benefits by poverty level < 0.66 61.0 59.6 63.2 60.8 0.66–1 8.2 9.6 7.8 7.8 1–1.85 11.9 11.9 7.1 17.3 > 1.85 18.9 18.9 22.0 14.1 D Share allocation by poverty level Income below 2/3 poverty level 61.0 59.6 63.2 60.8 With children < age 18 19.8 18.3 3.5 34.0 Single 6.1 6.1 0.4 11.2 Family head age 65+ 28.3 28.3 54.3 3.3 Single 12.7 12.7 24.4 0.7 Income below poverty level 69.2 69.2 71.0 68.6 With children < age 18 24.2 24.2 7.0 38.9 Single 7.5 7.5 1.2 13.0 Family head age 65+ 30.5 30.5 58.6 3.3 Single 13.6 13.6 26.2 0.7 Income from 1–1.85 of poverty level 11.9 11.9 7.1 17.3 With children < age 18 6.5 6.5 0.9 11.5 Single 0.4 0.4 0.0 0.7 Family head age 65+ 3.9 3.9 6.8 2.4 Single 0.8 0.8 1.5 0.0 Income above 1.85 of poverty level 18.9 18.9 22.0 14.1 With children < age 18 8.9 8.9 11.1 5.9 Single 3.1 3.1 3.4 2.5 Family head age 65+ 1.0 1.0 2.0 0.0 Single 0.0 0.0 0.1 0.0 E Shares received by disabled recipients All families with disabled members . . . . With children < age 18 . . . . Family head age 65+ . . . . 67 SSI disabled: This category counts SSI benefits restored to disabled recipients. The disabled are self-identified in the SIPP; the CPS does not ask disability status. SSI elderly: This column counts SSI benefits restored to elderly recipients. The disabled and elderly columns do not necessarily add to the full reinstated benefits. Some cases in SIPP include neither an elderly recipient nor one who reports a disability. In the CPS, families without elderly recipients are assumed to include disabled recipients. Rows: Tables B.6a and B.6b have five panels. The first lists the total benefits reinstated, and the remaining four show the distribution of these benefits to different groups. Panel A: Total value of reinstated benefits (millions of 1996$). This row lists the costs of reinstating benefits for SSI recipients in California. Since eligibility for SSI or Food Stamps is not dependent on the other, the SSI benefits reinstated are the same whether or not Food Stamps are also restored. According to the SIPP version of the table, restoring SSI in California accounts for $1,020 million. Panel B: Share of reinstated benefits by income quintile. This panel shows how restoration of the benefits listed in panel A is to be distributed by income quintiles of the recipient families. As in Table B.5, the quintiles order all California families (immigrant and citizen) by their family income and split them into five equal-sized groups. The lowest income quintile, therefore, includes the 20 percent of California families with the lowest income. These families receive 65 percent of reinstated benefits. If Food Stamps had been restored first, some of these families would no longer be in the bottom income quintile, so the share of reinstated SSI going to families at the bottom of the income distribution would fall to 61.2 percent. Panel C: Share of reinstated benefits by poverty level. This panel shows how restoration of the benefits listed in panel A is to be distributed based on the poverty level of the families in the absence of these benefits. For example, out of the $1,020 million in restored SSI benefits, 47.2 percent would go to families with incomes below two-thirds of the poverty level. If SSI benefits were restored after Food Stamps, 45 percent of the SSI benefits would go to families below two-thirds of poverty. Panel D: Share allocation by poverty level. This panel provides the same information as panel C above, but in greater detail. For example, both panels C and D show in the second column that 23.8 percent of restored SSI benefits go to families over 68 185 percent of the poverty level (without these benefits). Panel D shows that 16.7 percent of the $1,020 million would go to families over 185 percent of the poverty level who have children under age 18. As in all the tables, indented rows indicate that the statistics are conditional on belonging to the previous group. Panel E: Shares received by disabled recipients. SIPP reports on disability. This last panel shows the share of restored SSI benefits that goes to families with disabled members. Just over half of SSI benefits would go to families with a disabled member. The order of benefit restoration has no effect on the share of benefits going to disabled recipients. This detail on disabled family members is not available in the CPS data. Tables B.7a and B.7b: Shares of Reinstated Food Stamps Benefits Going to Various Categories of Immigrant Families in California In parallel with Tables B.6a and B.6b, Tables B.7a and B.7b calculate the value of Food Stamps benefits potentially restored to California families. These tables detail the distribution of benefits to various categories of California families under four different scenarios. Like the last two tables, Tables B.7a and B.7b count shares of dollars instead of shares of families. Sample: Projected benefits lost by immigrant families potentially denied Food Stamps. Columns: These tables show the value of reinstating Food Stamps benefit dollars to all or some of the immigrants denied Food Stamps under the new federal rules. The first two columns show the effect of restoring all Food Stamps. The last two columns show the benefits to the two groups who will continue to receive benefits in California: children and the elderly. To detail the four columns: Food Stamps alone: This column counts all Food Stamps benefits reinstated to immigrant families, assuming that no SSI benefits have been restored. Food Stamps after SSI: This counts all Food Stamps benefits potentially restored to immigrant families, but assumes that SSI benefits have already been restored. The distinction between this column and the first column matters only for the distribution of benefits by income level. 69 Table B.7a Shares of Reinstated Food Stamps Benefits Going to Various Categories of Immigrant Families in California: SIPP Data Programs Family CharacteristicsFood Stamps AloneFood Stamps After SSIFood Stamps to Children (After SSI Disabled)Food Stamps to Elderly (After SSI Disabled) A Total value of reinstated benefits (millions of 1996$)561 561 134 15 B Share of reinstated benefits by income quintiles Lowest 62.1 56.8 61.1 64.8 Mid-low 24.9 30.2 32.4 14.2 Middle 8.5 8.5 4.4 0.0 Mid-high 4.0 4.0 2.1 19.6 Highest 0.5 0.5 0.0 1.4 C Share of reinstated benefits by poverty level < 0.66 36.8 30.5 37.4 35.1 0.66–1 41.5 44.4 49.2 42.4 1–1.85 16.4 19.8 11.2 12.4 > 1.85 5.3 5.3 2.1 10.0 D Share allocation by poverty level Income below 2/3 poverty level 36.8 30.5 37.4 35.1 With children < age 18 36.8 30.5 37.4 35.1 Single 11.7 11.7 19.3 8.5 Family head age 65+ 0.0 0.0 0.0 . Single 0.0 0.0 0.0 . Income below poverty level 78.3 74.9 86.7 77.6 With children < age 18 77.7 74.4 86.7 76.2 Single 30.6 28.6 39.0 36.9 Family head age 65+ 0.6 0.0 0.0 . Single 0.6 0.0 0.0 . Income from 1–1.85 of poverty level 16.4 19.8 11.2 12.4 With children < age 18 15.5 18.8 11.2 12.4 Single 4.1 6.1 5.0 1.4 Family head age 65+ 0.7 1.3 0.0 3.7 Single 0.3 0.9 0.0 1.4 Income above 1.85 of poverty level 5.3 5.3 2.1 10.0 With children < age 18 5.0 5.0 2.1 8.6 Single 0.7 0.7 0.3 0.0 Family head age 65+ 0.0 0.0 0.0 . Single . 0.0 . . E Shares received by disabled recipients All families with disabled members 35.3 35.3 40.7 30.4 With children < age 18 7.4 7.4 40.7 30.4 Family head age 65+ 3.0 3.0 0.0 2.2 70 Table B.7b Shares of Reinstated Food Stamps Benefits Going to Various Categories of Immigrant Families in California: CPS Data Programs Family CharacteristicsFood Stamps AloneFood Stamps After SSIFood Stamps to Children (After SSI Disabled)Food Stamps to Elderly (After SSI Disabled) A Total value of reinstated benefits (millions of 1996$)683 683 217 16 B Share of reinstated benefits by income quintiles Lowest 55.2 54.1 66.5 67.9 Mid-low 32.5 31.4 26.3 24.3 Middle 9.3 11.5 6.8 7.8 Mid-high 2.7 2.7 0.4 0.0 Highest 0.4 0.4 0.0 0.0 C Share of reinstated benefits by poverty level < 0.66 56.6 52.9 69.6 56.0 0.66–1 21.3 22.3 20.5 15.9 1–1.85 17.8 20.6 8.8 18.2 > 1.85 4.2 4.2 1.1 10.0 D Share allocation by poverty level Income below 2/3 poverty level 56.6 52.9 69.6 56.0 With children < age 18 52.3 49.6 68.5 37.6 Single 19.3 18.9 32.1 28.9 Family head age 65+ 0.6 0.4 0.0 3.1 Single 0.6 0.3 0.0 3.1 Income below poverty level 77.9 75.2 90.1 71.9 With children < age 18 72.4 70.4 89.0 53.5 Single 25.6 25.4 39.0 29.9 Family head age 65+ 0.7 0.4 0.0 3.1 Single 0.6 0.4 0.0 3.1 Income from 1–1.85 of poverty level 17.8 20.6 8.8 18.2 With children < age 18 16.6 18.6 8.4 6.5 Single 4.0 4.2 2.2 3.5 Family head age 65+ 0.9 1.2 0.3 10.9 Single 0.3 0.5 0.1 8.4 Income above 1.85 of poverty level 4.2 4.2 1.1 10.0 With children < age 18 3.3 3.3 0.7 0.0 Single 0.8 0.8 0.2 0.0 Family head age 65+ 0.0 0.0 0.0 0.0 Single 0.0 0.0 0.0 0.0 E Shares received by disabled recipients All families with disabled members . . . . With children < age 18 . . . . Family head age 65+ . . . . 71 Food Stamps to children (after SSI disabled): This category counts Food Stamps payments restored to children under age 18. We specify “after SSI disabled” benefits are restored, since without SSI some disabled SSI children might have moved onto the Food Stamps rolls. These calculations assume no new cases due to such switching. Food Stamps to elderly (after SSI disabled): This counts Food Stamps benefits restored to recipients age 65 and older. Once again, we specify “after SSI disabled” benefits are restored, since without SSI some recipients might have moved onto the Food Stamps rolls. These calculations assume no new cases due to such switching. After accounting for SSI, there are very few elderly Food Stamps recipients. Rows: Tables B.7a and B.7b have five panels each. The first lists the total benefits reinstated, and the remaining four show the distribution of these benefits to different groups. Panel A: Total value of reinstated benefits (millions of 1996$). This row lists the costs of reinstating benefits for Food Stamps recipients in California. Since eligibility for SSI or Food Stamps is not dependent on the other, the Food Stamps benefits reinstated are the same whether or not SSI has also been restored. According to the SIPP version of the table, restoring Food Stamps to all California immigrant families would cost $561 million. Panel B: Share of reinstated benefits by income quintile. This panel shows how restoration of the benefits listed in panel A are to be distributed by income quintiles of the recipient families. As in Tables B.6a and B.6b, the quintiles order all California families (immigrant and citizen) by their family income and split them into five equal-sized groups. The lowest income quintile, therefore, includes the 20 percent of California families with the lowest income. Restoring benefits to children gives 61.1 percent of reinstated benefits to this lowest income group. Panel C: Share of reinstated benefits by poverty level. This panel shows how restoration of the benefits listed in panel A are to be distributed based on the poverty level of the families in the absence of these benefits. For example, out of the $561 million in total restored Food Stamps benefits, 36.8 percent would go to families with income below two-thirds of the poverty level. Given the restoration of SSI, only 30.5 percent of the Food Stamps benefits would go to families below two-thirds of poverty. 72 Panel D: Share allocation by poverty level. As in the last two tables, this panel provides the same information as panel C above, but in greater detail. For example, both panels C and D show in the third column that 86.7 percent of restored Food Stamps benefits to children go to families below poverty (without these benefits). Panel D shows that 39 percent of the $134 million would go to single parent families in poverty. As in all the tables, indented rows indicate that the statistics are conditional on belonging to the previous group. Panel E: Share Received by Disabled Recipients. SIPP reports on disability. This last panel shows the share of restored Food Stamps benefits that goes to families that report at least one disabled member. Although disability is not a factor in Food Stamps eligibility, 35.3 percent of all Food Stamps benefits would go to immigrant families with a disabled family member. The order of benefit restoration has no effect on the share of benefits going to disabled recipients. This detail on disabled family members is not available in the CPS data. 73 Bibliography Borjas, George J., “The Economics of Immigration,” Journal of Economic Literature, Vol. 32, 1994, pp. 1667–1717. Citro, Constance F., and Graham Kalton (eds.), The Future of the Survey of Income and Program Participation, National Academy Press, Washington, D.C., 1993. Citro, Constance, F., and Robert T. Michael (eds.), Measuring Poverty: A New Approach, National Resarch Council, National Academy Press, Washington, D.C., 1995. Fix, Michael, and Jeffrey S. Passel, Immigration and Immigrants: Setting the Record Straight, The Urban Institute, Washington, D.C., 1994. Fix, Michael, and Jeffrey S. Passel, “ Setting the Record Straight: What Are the Costs to the Public,” Public Welfare, Vol. 52, No. 2, 1995, pp. 6–20. Fix, Michael, Jeffrey S. Passel, and Wendy Zimmerman, “The Use of SSI and Other Welfare Programs by Immigrants,” Testimony before the House Ways and Means Committee, May 23, 1996. Downloaded 11/24/1997 from The Urban Institute: http://www.urban.org/TESTIMON/fix.htm. Fix, Michael, and Karen Tumlin, “Welfare Reform and the Devolution of Immigrant Policy,” New Federalism: Issues and Options for States, Series A, No. A-15, The Urban Institute, Washington, D.C., October 1997. MaCurdy, Thomas, and Margaret O’Brien-Strain, Who Will Be Affected by Welfare Reform in California? Public Policy Institute of California, San Francisco, California, 1996. Matloff, Norman, “From ‘Jiu Ji Jin’ to ‘Fu Li Jin’: Some Chinese Immigrants Mistakenly See Welfare as a Fringe Benefit,”The New Democrat, November 1994. Migration News, Vol. 3, No. 10, October 1996. Ross, Jane L., Supplemental Security Income: Noncitizens Have Been a Major Source of Caseload Growth, GAS/T-HEHS-96-88, Government Accounting Office, Washington, D.C., February 6, 1996. 74 Smith, James, and Barry Edmonston (eds.), The New Americans: Economic, Demographic and Fiscal Effects of Immigration, National Academy Press, Washington, D.C., 1997. U.S. Department of Commerce, Bureau of the Census, “The Foreign Born Population: 1994,” Current Population Reports, August 1995. U.S. Department of Justice, Immigration and Naturalization Service, INS Fact Book: Summary of Recent Immigration Data, August 1995. 75 About the Authors THOMAS E. MACURDY Thomas MaCurdy, an adjunct fellow of the Public Policy Institute of California, is a professor of economics at Stanford University and a senior fellow at the Hoover Institution. His principal research explores topics in the areas of income transfer programs, human resources, and labor markets. His recent studies investigate the consequences of governmental policies underlying welfare programs, unemployment compensation, Social Security, Medicare, Medicaid, and various forms of public assistance for low-income populations. These studies analyze the work disincentives and the determinants of participation induced by policies, along with the characteristics and behavioral patterns of beneficiaries. MaCurdy received an A.B. from the University of Washington and a Ph.D. in economics from the University of Chicago. MARGARET O’BRIEN-STRAIN Margaret O'Brien-Strain is a research fellow at the Public Policy Institute of California. Before coming to PPIC, she spent three years at Mathematics Policy Research, Inc., where she worked on studies of child care, welfare, and adult literacy. A recipient of a National Science Foundation fellowship, she is interested in a broad range of public assistance questions and has written about the minimum wage, the Earned Income Tax Credit, immigrants’ use of welfare, the combined effects of public assistance programs, and welfare reform in California. O’Brien-Strain received a B.A. in economics from Swarthmore College and will soon complete a Ph.D. in economics at Stanford." ["post_date_gmt"]=> string(19) "2017-05-20 09:34:43" ["comment_status"]=> string(4) "open" ["ping_status"]=> string(6) "closed" ["post_password"]=> string(0) "" ["post_name"]=> string(9) "r_1298tmr" ["to_ping"]=> string(0) "" ["pinged"]=> string(0) "" ["post_modified"]=> string(19) "2017-05-20 02:34:43" ["post_modified_gmt"]=> string(19) "2017-05-20 09:34:43" ["post_content_filtered"]=> string(0) "" ["guid"]=> string(51) "http://148.62.4.17/wp-content/uploads/R_1298TMR.pdf" ["menu_order"]=> int(0) ["post_mime_type"]=> string(15) "application/pdf" ["comment_count"]=> string(1) "0" ["filter"]=> string(3) "raw" ["status"]=> string(7) "inherit" ["attachment_authors"]=> bool(false) }