Donate
Independent, objective, nonpartisan research

R 911CDR

Authors

R 911CDR

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(12) "R_911CDR.pdf" ["wpmf_size"]=> string(6) "872328" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(54056) "This report describes recent rapid changes to California’s CalFresh program—formerly known as the food stamp program—and assesses whether policies that the state is considering would improve the program— by increasing participation or lowering administrative costs. To the extent that CalFresh supports families when they experience economic difficulties and boosts local economies, policymakers may want to ensure that all eligible families have reasonable access to the program. The cost of the program is another important consideration. Although direct benefits are currently paid by the federal government—making it a relative bargain for the state—California’s costs to administer benefits nonetheless are higher compared to other states. Participation in safety net programs tends to grow when families lose employment or face work cutbacks in a poor economy. In California, CalFresh grew by 31 percent between December 2007, the official start of the last economic downturn, and its official end in June 2009. Participation has continued to grow as unemployment has remained high. Nevertheless, participation remains relatively low compared to other states. California and other states made substantial changes to program policies over the past decade, and policy- makers continue to consider further changes. Examining the policies of all 50 states, we conclude that the policy that has had the largest effect on program participation in California in recent years is the require- ment that most adults in food stamp households be fingerprinted. We find that eliminating fingerprinting would increase participation in California by about 7 percent (although we cannot rule out the possibility that a portion of the increase would come from fraud). Further, California is one of only two states with a requirement that most recipients report their status more frequently than semiannually, but we do not find this has depressed participation relative to other states. California’s costs of administering the program are also relatively high by national standards, and we find evidence that eliminating fingerprinting would not only increase participation but would also reduce the administrative costs of the program by about 13 percent. Quarterly reporting has not increased administrative costs relative to states with semiannual reporting. We conclude that California has already taken several important steps to bolster participation. It could additionally consider eliminating fingerprinting to boost participation and lower administrative costs. We find that moving to semiannual reporting would neither raise participation substantially nor reduce costs. AREERA Agricultural Research, Extension, and Education Reform Act of 1998 CalWORKs California Work Opportunity and Responsibility to Kids EBT Electronic Benefits Transfer IHSS In-Home Supportive Services SNAP Supplemental Nutrition Assistance Program TANF Temporary Assistance for Needy Families USDA United States Department of Agriculture Like every state, California has longstanding programs to assist poor and near-poor families. The food stamp program—recently renamed CalFresh—is now the second largest of these safety net programs, and as of May 2011, was serving about 3.7 million Californians.1 More than half are children. Its mechanism is simple—low-income families are credited with funds on a debit card that they then use to buy most groceries.2 The card is recharged with funds every month so long as the family remains eligible. Established nationwide in 1974, the food stamp program benefits not just low-income families with children, but also the low-income elderly. Households of one or more working-age adults can also participate, although qualifying for benefits is typically more difficult because work requirements can be more stringent. The food stamp program is relatively unusual in that the federal government pays for all benefits (which are not capped), while states—and, in California, also counties—share administrative costs.3 In 2009–2010 the total cost of CalFresh benefits plus the cost of administering the benefits was $5.62 billion in California, but the state and counties were responsible for $680 million—12 percent of total CalFresh spending, and 0.5 percent of the state’s 2009–2010 General Fund expenditures. Other state-federal funding models used in safety net programs require a higher state match, with states and counties typically contributing about half the cost of both benefits and program administration. If this model were applied to CalFresh, this would have meant total costs to the state and counties of about $2.84 billion in 2009–2010, not $680 million. Thus, for California, CalFresh is a relatively inexpensive social services program.4 Research indicates that the program also reduces food insecurity and boosts local economies (Hanson et al., 2002; Ratcliffe and McKernan, 2010).5 To be eligible for CalFresh, a household’s income after deductions cannot exceed federal poverty levels.6 For a single-member household—typically nearly a third of all CalFresh households—this means a monthly income after deductions of no more than $903. For families with two or three eligible members—who make up almost half of all CalFresh households—this translates to a monthly income of no more than $1,214 and $1,526, respectively.7 1 The largest such means-tested programs enroll children and adults in low-income families in health insurance. In California, these programs are called Medi-Cal and Healthy Families. 2 Usually families are unable to use their benefits to buy prepared foods, whether at a grocery store or a restaurant. Some localities, among them Los Angeles County—have obtained authority to enable homeless, elderly, or disabled participants to use their benefits at authorized restaurants. 3 Of special interest to the state is the state-funded program established to provide food assistance to legal immigrants ineligible for federal benefits, called California Food Assistance Program (CFAP). The state and counties bear both all administrative and all benefit costs for those in this program. However, most legal resident immigrants are eligible for federally-funded benefits, so this program is small: In spring 2011 about 39,500 immigrants received state-funded benefits, just over 1% of the number getting federally-funded benefits. 4 The federal food stamp program—officially called the Supplemental Nutrition Assistance Program (SNAP) is one of many discretionary federal programs that could see sharp reductions ordered by the new federal Joint Select Committee on Deficit Reduction. However, if the committee fails to come to agreement, triggering automatic federal spending cuts, SNAP would be exempt from those. 5 There is little solid evidence that participation in food stamps results in poor health outcomes (Baum, 2007; Gibson, 2003). The exception is women who stay on the program for several years or longer, who appear to be at increased risk of obesity (Smith, 2009). 6 Deductions include 20 percent for earned income, and others for expenses such as housing and child care. 7 These income thresholds are identical across all states, with the exception of Alaska and Hawaii, where they are higher. Thresholds are normally updated for inflation each year. However, they did not change between 2009 and 2011. Since 2008, asset limits, when applied, are also adjusted annually for inflation. The maximum monthly amount for a one-person household is $200; this maximum rises to $526 for a family of three.8 However, these maximums apply only to families with very little or no income. (Although in a minority, CalFresh households with no reported gross income nevertheless made up 19 percent of all CalFresh households in 2009.) For a family of three with one member employed at a low-wage ($11 an hour) job, the CalFresh amount would be $455—equivalent to an additional 27 percent in earnings.9 The recent growth of the CalFresh program is consistent with the recent economic downturn, but even so it has been striking. An additional 660,000 residents enrolled in food stamps between the start of the recession in December 2007 and its official end 18 months later—a 31 percent increase. This is more than double the increase in CalWORKs, the state’s main cash assistance program for families with children, over the same period. More recent statistics indicate that CalFresh participation grew by an additional 33 percent between June 2009 and May 2011.10 Nevertheless, California continues to have both lower participation and higher costs compared to other states. The gap in participation began to appear in about 2000 and continues, notwithstanding the more recent growth. In California, this growth was slowest among households of both eligible children and adults. The relative gap is apparent whether viewed as a percentage of the total population or as a percentage of those eligible. Using the first metric, about one in seven people in the other 49 states and the District of Columbia were receiving food assistance in early 2011, compared to one in ten Californians. Using the second metric, official estimates for 2008 (the latest that have been computed) indicate that about half of those eligible for CalFresh benefits actually receive them; in the rest of the country this ratio was above two- thirds (Leftin, 2010). Figure 1 details this gap more precisely: The percentage of state residents participating in food stamps in California and in the rest of the country was fairly similar in the 1990s, but a gap emerged in the early part of the following decade and became substantial by 2005.11 In California, participation increased only gradually until the start of the recent, severe recession in late 2007, while growth nationally began in 2001. 8 Benefit levels are updated for inflation each year. However, the 2009 American Recovery and Reinvestment Act (the ‚stimulus‛ bill) raised family benefits by an average of about 13 percent, and the automatic benefit inflation adjustment is suspended until 2013, when benefits will drop to where they would have been without the stimulus. 9 This benefit calculation uses the 20th percentile hourly wage among California workers in 2009. It assumes 35 hours of work each week, no unearned income (for example, from CalWORKs or child support payments), the fair market rent for a two-bedroom apartment in a low-cost county ($900), child care expenses of $500 a month, and the standard utility deduction. It also assumes that the household contains no elderly or disabled members, and no members ineligible for the program (for example, because of documentation status or CalWORKs sanction). 10 Participation does vary widely within California. As a percentage of the population, participation ranged from 2 percent to 22 percent across California’s counties in 2009. However, the percent using food stamps is highly correlated with the estimated percent of the county population in poverty—about 70 percent of the variation in participation is explained by poverty alone. Moreover, there is a one-to-one correlation between poverty at the county level and higher caseloads. 11 Trends in official estimates of the percentage of those eligible and receiving the benefit show similar evidence of an emerging gap (Leftin, 2010). Administrative costs for CalFresh have remained high compared to other states for several decades. Data from the U.S. Department of Agriculture’s (USDA) Food and Nutrition Service indicate that the combined monthly federal, state, and county administrative cost to administer food stamp benefits in California averaged $55 per household in 2009, second highest in the nation.12 Although these costs have been declining over the past decade in the state, from 1989 to 2009 they averaged nearly twice the per-household cost reported by all other states.13 And while the state and counties’ share of total program costs is low relative to other state safety net programs, relatively high administrative costs are still a concern during an economic downturn. 12 Total costs per case were even higher—$93 in California—but this total includes expenditures not closely related to administering benefits (e.g., nutrition education, employment and training programs, and automation development projects). The technical appendices describe the line items in states’ FNS-269FS reports that we categorize as ongoing costs of administering benefits. Logan and Klerman (2008) provide a further description and exploration of the underlying FNS-269FS reports that form the basis of our analysis of administrative costs. 13 There are several reasons why ongoing costs to administer the program might be higher in California than in other places. In the first place, California has a mix of cases that is arguably more expensive to administer (County Welfare Directors’ Association, 2010). Some of the cost difference is also likely attributable to higher wages in California (U.S. Bureau of the Census, n.d.). Still, policy changes that streamline the processing of applications and reduce ongoing paperwork might reduce the state’s administrative costs. 0 5 10 15 Percent of population California U.S. outsideof California In the nation excluding California, the number of food stamp households began growing steadily in mid- 2000; by mid-2005, caseloads were about 56 percent higher than in 2000. By comparison, California’s caseloads grew 13 percent from mid-2000 to mid-2005. Californians’ participation grew at roughly the same rate as in the rest of the country after 2005, but the gap has not closed. A large part of the difference was due to slower increases among households containing both aided children and aided adults, rather than increases in households with children as the sole recipients of benefits or households with adult recipients only.14 Because eligibility for CalFresh is broadly aimed at low-income households, we examined three distinct household subgroups for further insight into caseload trends from 2000 to 2010 (Figure 2):  those with a mix of eligible adults and children;  those composed only of eligible children (because all adults in the household are ineligible);15  those with one or several eligible adults (aged 18 and older) and no children. We did not separately consider households containing only recipients aged 65 and older because this group made up only 1 to 2 percent of the caseload over the previous decade.16 The trend for the first of these three groups is clear: Mixed-age households actually shrank by about 10,000 from 2000 to 2005 while the numbers of the other two household types grew, as did all three groups nationally.17 Thus, 14 Data on characteristics of CalFresh households are publicly available after two years, so we can use data only through 2009 in this report. 15 In California, the most common reason that only children are aided is that parents are unauthorized immigrants. Ineligible adults may also be Supplemental Security Income (SSI) recipients, may be under a CalWORKs sanction, or may be relative caregivers. 16 The very small numbers of elderly recipients in California is largely the result of the state’s SSI cash-out policy, which we discuss below. In the U.S. outside of California, households containing only aided elderly made up 11 percent of the caseload in 2009. 390 380 513 167 229 272 115 163 328 0 200 400 600 800 1,000 1,200 200020052009 CalFresh cases (thousands) Only eligible adults Only eligible children Mix of eligible adults andchildren the place to look for further insight into diverging caseload trends is households with a mix of aided adults and aided children. A decade ago, CalFresh food assistance and CalWORKs—income support and work services for families with children—were closely linked in California. Although there was no explicit requirement that this be the case, the vast majority of all mixed-age CalFresh households also had assistance from CalWORKs—86 percent, or 336,000 of the 390,000 total (Figure 3). This had changed by 2005. The number of CalFresh households combining food and cash assistance dropped sharply from 336,000 to 243,000.18 Concurrently, the number of CalFresh households that had income from wages, income from other sources, or no income doubled or more. Not surprisingly, given the deterioration of the economy, all segments of the mixed-age caseload grew pre- to post-recession (2005 to 2009). Still, those combining food and cash assistance grew the least in percentage terms. In other words, the safety net for families with children changed substantially during the last decade. Relatively fewer California families with children are making use of the traditional safety net combination of food and cash assistance, and more are using CalFresh to supplement other sources of income—or using food stamps as their sole source of income. By 2009, just 58 percent of all mixed-age households were combining food and cash assistance, while 20 percent were using CalFresh to supplement earnings. Ten percent had income from other sources (most often unemployment insurance), and the remaining 12 percent had no cash income. 17 This does not imply that CalFresh households containing only aided children or only aided adults came to participate when they were in fact ineligible. In fact, calculations of program use among those eligible throughout the entire U.S. indicates that children who have immigrant parents tend to participate less frequently than those with citizen parents Leftin (2011). 18 Households not currently receiving CalWORKs benefits may nevertheless have recently received them. This is because California automatically extends food stamp eligibility for five months after a family leaves the CalWORKs program. 336 243 300 34 75 102 17 38 49 4 25 63 0 100 200 300 400 500 600 200020052009 CalFresh cases (thousands) No income Other income Earnings, no cashassistance Cash assistance A similar—but more pronounced—pattern also emerged in the other 49 states; the number of mixed-age households making use of both food stamps and cash assistance grew by only 4 percent over the entire period (2000–2009), while those combining food stamps with earnings, with other income, or that had no income, each doubled or more than doubled. In the U.S. outside of California in 2009, households on welfare made up just 28 percent of the mixed-age food stamps caseload, 40 percent had earnings, 20 percent had income from other sources, and the remaining households (12%) had no cash income. These patterns suggest that families’ use of safety net programs has changed and that reaching families eligible for CalFresh but that lack a current connection to CalWORKs has been a particular challenge. California’s demographics differ from other states, as does its economy and one longstanding CalFresh policy. Upon examination, however, none of these can explain its lower CalFresh participation rate during the last decade. Demographics. California is home to more noncitizens than any other state; these noncitizens may be reluctant to avail themselves of CalFresh assistance, and this could be depressing participation. Census data from 2008 show that among the 29 percent of all California households with incomes under 200 percent of the federal poverty level, 8 percent were noncitizen households—double the share among low-income households in the rest of the country. Although unauthorized immigrants are ineligible for the program, their citizen children are eligible if they meet other requirements. Ineligible parents may be unaware of their children’s eligibility or be reluctant to apply because of their own status. Unauthorized immigrants may also avoid the application process for their children because they think they themselves will be deported, or be denied citizenship in the future. Although the USDA’s Food and Nutrition Service issued instructions a decade ago that made clear that neither of these actions would be taken because of a food stamp application, concern may still linger. But other demographic characteristics might attenuate any effect that a large immigrant population has on participation. California’s low-income population is younger and has lower educational attainment compared to other states. These two factors tend to increase CalFresh participation compared to other parts of the country.19 Using a statistical model, we conclude that California’s distinctive population characteristics depress participation in food stamps by about 10 percent. This is far less than would be the case if we only considered the role that a larger immigrant population plays.20 In any case, population characteristics change relatively slowly, over the course of a few years or decades, and cannot account for the fast-moving trends since 2000. Economy. In contrast, the economy does change rapidly. When the economy worsens, more households become eligible for food stamps or for a larger amount of assistance. However, between 2000 and 2005 California’s economy—as measured by its unemployment rate—was in fact a little worse than in the rest of the nation. On average, California’s unemployment rate was about 0.75 percentage points higher than in the rest of the country, and the unemployment rate rose earlier and more steeply in California at the outset of 19 Specifically in 2008, 28 percent of low-income households in California had one or more children under age 5, and 71 percent had one or more children age 5–17. In the U.S. outside of California, 24 percent had at least one child under age 5 and 54 percent had older children. Further, in California 17 percent of low-income households contained only elderly members, while 22 percent of households in the rest of the U.S. were composed of only members age 65 and older. In California 33 percent of low-income households had a household head who had not finished high school; in the rest of the U.S., 26 percent of household heads lacked a high school degree. In all cases, these differences between California and the rest of the U.S. are statistically significant at the 1 percent level or better. 20 The covariates we include in the descriptive model are: the citizenship of household head, the citizenship of household members, the age of the household head, the age of other household members, household size, whether English is spoken at home, the household’s income as a percent of the federal poverty level, the labor force participation of household adults age 18–64, whether anyone in the household received SSI in the past year, whether anyone in the household received cash welfare in the past year, the educational attainment of the household head, and the marital status of household head. We document our approach, which relies on the research presented in Burstein et al. (2009) and Cody et al. (2007), in the technical appendices. the recent recession than it did in other places in the country. In sum, a better economy in California is not an explanation for lower participation. SSI cash-out. One policy difference between California and all other states predates the diverging trends of the last decade and so cannot explain them. California alone makes its 1.2 million aged, blind, and disabled Supplemental Security Income (SSI) recipients ineligible for CalFresh. Instituted in 1974 and known informally as ‚SSI cash out,‛ the policy originally added $10 to the monthly checks of all SSI recipients in place of food stamps; it was intended to reduce paperwork for SSI recipients and save administrative costs for the state. Changing this policy would have a complicated impact on the total number of CalFresh recipients in the state and the benefits that families receive. On the one hand, certain households—especially those containing only SSI recipients—would be eligible for more than $10 worth of food assistance each month if the policy were ended. On the other hand, they would be required to apply for the program. Some would fail to apply. Some larger households with one or a few SSI recipients would be ineligible if the SSI payment or payments (including the extra that California’s supplemental program provides) were then included in the entire household’s income. Simulations conducted by Mathematica Policy Research that take both factors into consideration indicate that more California households would be eligible for CalFresh if the policy were ended, but also that the total benefits received by Californians could drop slightly (Cunnyngham, 2010). The federal government began to give states increased freedom to adjust food stamp policies in the late 1990s. This new flexibility was principally aimed at giving states tools to increase working families’ access to food stamps, particularly those who had gone off cash welfare (while remaining eligible for food assistance), as occurred for large numbers in the wake of the 1996 federal overhaul of welfare (Super, 2004).21 In general, these policy changes were focused on eligibility restructuring and paperwork reduction. They could have the additional benefit of reducing administrative costs, by streamlining the oversight that social services offices conduct to ensure that they pay benefits accurately. The changes also might reduce ‚churning,‛ or the number of times families drop out and then return to the program because they fail to keep their eligibility current. California has made a number of such modifications and is considering others. Several occurred in the first half of the decade; several were put in place only very recently, between July 2009 and February 2011. To estimate comparative effects using nationwide data, we first compiled policy specifics for all other states, because they did not previously exist. We conducted a survey of food stamp program administrators in the 49 other states and the District of Columbia. Details are accurate through August 2010, and are discussed in greater detail in the technical appendices.22 Our models isolate any change in per capita food stamp participation or ongoing administrative costs per case that occurred after states introduced each policy.23 The models control for state differences, state trends, and national changes, and hold constant state population characteristics and other differences that affect eligibility, the likelihood of participating if eligible, or both.24 After including these controls, if a statistically detectable change followed the introduction of a new policy, we attribute it to the policy. EBT cards. The program’s previous name, food stamps, derives from the form in which families originally received benefits. In the early 1990s several states began converting these paper stamps to debit cards known as ‚Electronic Benefits Transfer‛ or EBT cards. In 1996, believing that the change would reduce fraud and increase the effectiveness of the program, Congress required all states to switch to EBT cards. By mid-2004, all of California’s counties were issuing benefits via EBT cards that are recharged electronically each month in the amount of the family’s benefit. We find that EBT had no statistically significant effects on participation or administrative costs. Online applications. Web-based program applications—enabling applicants to file and electronically sign an application from any location and at any time of the day—are a recent development in most states. In 21 We discuss major policy changes in this section, but not an exhaustive list. Other efforts that states have made are described in Rowe et al. (2010) and Food and Nutrition Service (2009). 22 In addition, see Danielson et al. (2011). 23 The participation models we report do not include state-funded food stamp recipients (in California, CFAP recipients). However, the technical appendices report models that do include state-funded food stamp recipients, and the results do not change our conclusions about the effects of fingerprinting or any other policy on participation. 24 We carried out our analysis of food stamp participation using the percentage of the entire population receiving food stamps rather than estimates of the percentage of the eligible population receiving food stamps. This is mainly because estimates of the eligible population exclude those made eligible by changes in asset policies. We provide a detailed discussion in the technical appendices. California, a pilot program began in October 2007, and all counties have been able to accept online applications since December 2010. We find that enabling households to submit an application online does not increase participation measurably, but we estimate that it lowers ongoing administrative costs by 19 percent.25 Assets. Determining a household’s food stamp eligibility in the past usually required a detailed assessment of a household’s assets. Households with more than $2,000 in liquid assets ($3,000 if the household contains an elderly or disabled member) were ineligible for the program, regardless of their income. Two ways in which this has changed are the treatment of the value of the vehicles that a household may own and the consideration of other household assets (for example, bank accounts, retirement accounts, and the like).26 Since 2001, the federal government has given states the option to evaluate vehicle assets using rules from other assistance programs (TANF, child care, or foster care). This option has had the effect of allowing states to exclude one or more household vehicles from the food stamp asset test. In 2004 California began exempting the value of all vehicles. And since 1996 the federal government has allowed states to exempt households from other asset tests. (This is termed ‚categorical eligibility‛). In July 2009, California counties began exempting households with children from the asset test altogether. The state expanded this exemption to all households in February 2011. All households still must qualify based on income. We find that eliminating the asset test increases participation by an estimated 5.6 percent.27 (However, the earlier switch to excluding cars when considering a household’s eligibility for food stamps does not appear to have boosted participation.)28 Households containing a mix of aided children and adults are substantially affected by this policy, increasing by an estimated 6.3 percent in the wake of its adoption. (Other types of households are affected as well.) While it might seem that eliminating asset tests would reduce the complexity of eligibility determinations and so streamline the application process for social services offices, we find no evidence that exempting vehicles from the asset test or eliminating the asset test altogether lowers the costs of administering the program. Fingerprinting. California made one policy change in December 2001 that added a step to the application process—fingerprinting.29 This was instituted to prevent recipients receiving benefits from multiple social services offices. California is one of five states to use fingerprinting in recent years, but CalFresh is not the only program within the state that uses fingerprinting: CalWORKs adults, General Assistance participants, and In-Home Supportive Services (IHSS) caregivers must also be fingerprinted.30 25 We exclude capital and other start-up costs for automation projects in our analysis of factors driving administrative costs. See the technical appendices for details. 26 Households have always obtained exemptions for a house that they live in, vehicles used for a business or to live in. Most other available assets were long limited to $2,000 ($3,000 if the household contains an elderly or disabled member). 27 For further details of this policy, see California Department of Social Services (2009a) and California Department of Social Services (2011). 28 Evidence from other research indicates that such policies increased participation among some types of households while simultaneously reducing participation among other types of households, leaving the size of the overall caseload unchanged (Klerman and Danielson, 2011). In particular, that research found that vehicle policies was partly responsible for shifting the composition of the caseload away from those combining cash and food assistance and toward those getting food assistance in the absence of cash assistance. 29 California uses an electronic imaging process rather than conventional fingerprinting. 30 The most recent legislative effort to eliminate fingerprinting is AB 6 (Fuentes), introduced in December 2010. As amended in August 2011, it would change only the CalFresh fingerprinting requirement. Our model indicates that fingerprinting reduces the share of California’s CalFresh population by about 7 percent. The effect is concentrated in mixed-age households, which dropped by 9.6 percent after fingerprinting was introduced. In fact, modeling the policy effect on households with no aided adults (often, although not always, because of adults’ immigration status), we find no statistically significant reduction.31 Any deterrent effect of fingerprinting on the CalFresh program may have been softened as a result of changes to other application requirements made in 2008 and 2009 in California. This is because interviews to determine eligibility can now be conducted over the phone, and adults must come in to the social services office to be fingerprinted within 12 months of applying for benefits (California Department of Social Services, 2008; California Department of Social Services, 2009b). We estimate that the introduction of fingerprinting also raises administrative costs by about 13 percent. Thus, while eliminating fingerprinting would increase state administrative costs because more households would enroll in CalFresh, administrative costs per case would decline, offsetting the greater cost of a higher caseload. The additional administrative costs of requiring fingerprints might be justified if the policy effectively detects or deters fraud, but the evidence for this is inconclusive. First, we considered whether the number of positive fraud findings increased in a state after fingerprinting was implemented and whether the rate of positive fraud findings (positive findings as a fraction of total investigations) improved. The first would occur if a state that adopted digital fingerprinting invests additional resources in fraud investigations that the new technology enables them to detect. The second of these would occur if fingerprinting enabled a state to detect fraud more efficiently. We found no evidence that fingerprinting increases findings of fraud, or that it increases the rate of positive fraud findings. This is in line with the previous research that has found that fingerprinting detects fraud relatively rarely (State Auditor, 2003; Sticha et al., 1999).32 Second, we considered whether fingerprinting deters applicants. Because we lack information about those who do not apply for food assistance, we cannot easily determine how much fingerprinting deters fraudulent applications, as opposed to cutting legitimate applications among those who simply find the requirement burdensome.33 Other research on this question provides little light on this issue because the research studied only those who actually applied for assistance. Case studies of small groups of applicants and recipients who refused to comply with the fingerprinting requirement in food and cash assistance programs were conducted more than a decade ago in Los Angeles County, Arizona, and Texas; these studies found that a minority of those who refused to comply were cases of probable or verified duplicate aid. In Los Angeles County, investigators found probable or verified fraud in 69 percent of the 137 cases they investigated, but only 6 percent represented verifiable duplicate-aid fraud (State Auditor, 2003; Sticha, 1999).34 In sum, the evidence from our and other research indicates that fingerprinting works primarily by deterring applications rather than catching fraudulent applicants—and it is uncertain who, exactly, is being deterred. 31 The estimate for the children-only caseload is also much smaller—roughly 40 percent of the size of the estimate for mixed-age households. 32 The technical appendices to this report provide details. The overall prevalence of detected fraud is low—nationwide, about 1% of households participating in the program—and has been flat over the past decade across both states that have fingerprinting and those that do not. 33 Legitimate applicants could also be turned away if administrative errors in evaluating finger images are made. The evaluation conducted in Lost Angeles County in the mid-1990s found that such errors do occur (State Auditor, 2003). 34 In a further 10 percent of cases investigators suspected duplicate-aid fraud but could not verify it. An additional 7 percent could not be located at all. In the remaining cases in which inspectors suspected or found fraud, clients had concealed income, misrepresented household composition, or the like. Quarterly and semiannual reporting by recipients. Believing that social services offices can require less paperwork of families and still issue benefits accurately, the federal government has encouraged all states to move to less frequent, semiannual reports of family income. Because California still tracks families’ incomes using quarterly reporting it has had to request several waivers from the federal government to continue this policy. Moreover, the USDA is still strongly urging the state to halve the number of required reports as rapidly as possible (USDA, 2011).35 We find that making this switch would not increase participation in CalFresh. The plausible reason lies with the second main paperwork policy. Periodically, social services offices conduct a comprehensive reassessment of households’ eligibility and benefit amounts, known as certification. California already certifies most households annually, and a switch to semiannual reporting would not affect this practice. (The most common certification period in other states in 2009 was six months.) Our estimates suggest that, given California’s longstanding practice of using relatively long certification periods, the additional effort required to submit quarterly, instead of semiannual, interim reports each year is relatively minor.36 While it might seem obvious that requiring two rather than four interim reports each year would reduce administrative costs, we do also not find empirical support for this. Because the administrative cost data are at a summary level, it is difficult to explore this finding further. It is possible that reducing the number of recipient reports does lower costs, but that social services offices use the resources that are freed up to improve other aspects of program operations. 35 The state and counties will incur costs to reprogram data systems to accept semiannual rather than quarterly reports, and, possibly, to delink the CalWORKs quarterly reports from the CalFresh semiannual reports. See Wagner (2010). The most recent legislative attempt to switch to semiannual reporting is AB 6 (Fuentes), introduced in December 2010, and which was sent to the governor in September 2011. 36 In our statistical models, we compare states that introduce policies of semiannual interim reporting joined with semiannual or annual certification periods to states with policies of quarterly interim reporting joined with semiannual or annual certification periods. We find no statistically significant difference between the two. As a sensitivity check, we also model certification and reporting policies separately, and again find no statistically significant difference between states that adopt quarterly reporting and states that adopt semiannual reporting. The technical appendices provide details. The number of low-income Californians making use of food stamps has grown sharply over the past several years. In the absence of any action on the part of policymakers, we would expect CalFresh participation to remain high for several years to come. This is because the state’s unemployment rate remains high, roughly twice what it was in the fall of 2007. Previous research on the link between the state of the economy and food stamp caseloads indicates that that demand for food assistance will grow for several years even after the economy has begun to improve substantially (Klerman and Danielson, 2011). The CalFresh caseload should drop once an economic recovery is securely in place. Still, in both good economic times and bad, California has had a large number of residents eligible for CalFresh who do not take advantage of the food assistance it offers. Could policy changes help more eligible families make use of the program? While policymakers have taken several steps—including, recently, switching from income- and assets-based eligibility tests to solely income- based eligibility tests—that have boosted participation, they could additionally consider completely eliminating the requirement that most adult applicants be fingerprinted. The state could instead rely on existing data systems and applicant social security numbers to track duplicate aid.37 Our evidence indicates that the group of households most sensitive to the fingerprinting policy are mixed-age households, those whose participation lagged in the early 2000s. We also find evidence that the increased cost of administering a larger caseload we would expect from ceasing fingerprinting would be offset by a reduction in cost per case. The widespread use of fingerprinting in safety net programs sets California apart from most other states. There is no federal requirement that CalFresh adults be fingerprinted and the policy appears to deter some eligible families from applying. However, we are unable to rule out the possibility that the incidence of benefits being paid from multiple social services offices would increase if the state were to eliminate fingerprinting. California’s fingerprinting policy will ultimately reflect policymakers’ ongoing efforts to balance two program goals: reasonable access and program integrity. Our evidence also indicates that families with children are distinctly less likely to be combining food and cash assistance in traditional safety net fashion than they used to do. Instead, they are making use of CalFresh as a supplement to wage earnings or other sources of income—or food stamps are their sole income. Looking ahead, policymakers should consider paying special attention to trends among these mixed-age households. They are a low-income population traditionally of concern to many policymakers, but their participation has lagged. Our evidence indicates that these households are generally responsive to policy changes aimed at increasing access among those eligible. Thus, continued efforts to fine-tune CalFresh to reach this population have the potential to make a real difference in the lives of these children and their parents. 37 The state’s Medi-Cal Eligibility Determination System (MEDS) has been used to detect social security number-based duplicate aid fraud across counties, within counties, and in Nevada, Oregon, and Arizona (State Auditor, 2003). . Burstein, N. R., S. Patrabansh, William L. Hamilton, and Sarah Y. Siegel. 2009. Understanding the Determinants of Supplemental Nutrition Assistance Program Participation: Final Report. Alexandria, VA: U.S. Department of Agriculture Food and Nutrition Service. Breunig, Robert, Indraneel Dasgupta, Craig Gundersen, and Prasanta Pattanaik. 2001. Explaining the Food Stamp Cash-Out Puzzle. Food Assistance and Nutrition Research Report No. 12. Washington, DC: U.S. Department of Agriculture Economic Research Service. Available at www.ers.usda.gov/publications/fanrr12/fanrr12.pdf. California Department of Social Services. 2008. ‚Food Stamp Waiver of Face-to-Face Interview and Non-Assistance Food Stamp Household Recertification Form, FS 27.‛ All-County Letter No. 08-32, July 14. Available at www.dss.cahwnet.gov/lettersnotices/entres/getinfo/acl08/08-32.pdf. California Department of Social Services. 2009a. ‚Modified Categorical Eligibility for the Food Stamp Program : Questions and Answers.‛ All-County Letter No. 09-24, May 27. Available at www.dss.cahwnet.gov/lettersnotices/entres/getinfo/acl/2009/09-24.pdf. California Department of Social Services. 2009b. ‚Food Stamp Program (FSP) Waiver of Face-to-Face Interview.‛ All-County Letter No. 09-62, October 23. Available at www.dss.cahwnet.gov/lettersnotices/entres/getinfo/acl/2009/09-62.pdf. California Department of Social Services. 2011. ‚Expansion of Modified Categorical Eligibility for CalFresh to All Non- Assistance CalFresh (NACF) Households.‛ All-County Letter No. 11-11, January 27. Available at www.dss.cahwnet.gov/lettersnotices/entres/getinfo/acl/2011/11-11.pdf. California State Auditor. 2003. Statewide Fingerprint Imaging System: The State Must Weigh Factors Other Than Need and Cost- Effectiveness When Determining Future Funding for the System. January 2003, 2001-015. Sacramento, CA. Available at www.bsa.ca.gov/pdfs/reports/2001-015.pdf. California Welfare Directors’ Association. 2010. California Counties Work to Improve Food Stamp Access. Sacramento, CA. Available at www.cwda.org/downloads/publications/foodstamps/Food-Stamp-Fact-Sheet-with-Addendum-April- 2010.pdf. Center on Budget and Policy Priorities. 2008. States’ Vehicle Asset Policies in the Food Stamp Program. Revised July 1. Available at www.cbpp.org/cms/?fa=view&id=554. Cody, Scott, Allen Schirm, Elizabeth Stuart, Laura Castner, and Alan Zaslavsky. 2007. Sources of Variation in State-Level Food Stamp Participation Rates. Final report (submitted to U.S. Department of Agriculture). Available at www.mathematica-mpr.com/Publications/PDFs/variation-fspr.pdf. Cunnyngham, Karen. 2010. ‚Estimated Effects on the Supplemental Nutrition Assistance Program of Eliminating California’s SSI Cash-Out Policy.‛ Final report (submitted to California Food Policy Advocates). Available at www.mathematica-mpr.com/publications/PDFs/nutrition/SNAP_CA_cashout.pdf. Danielson, Caroline, and Jacob A. Klerman. 2008. ‚Did Welfare Reform Cause the Caseload Decline?‛ Social Service Review 82 (4): 703–30. Danielson, Caroline, Jacob Alex Klerman, Margaret Andrews, and Daniel Krimm. 2011. "Reporting Requirement and Asset Test Policies in the Supplemental Nutrition Assistance Program." Unpublished manuscript (March). Fernández, Alexis. 2010. California Food Stamps Characteristics Report. Oakland, CA: California Food Policy Advocates. Available at www.cfpa.net/foodstamps/foodstamp_characteristics.pdf. Food and Nutrition Service. 1999. ‚Letter to Regions on Categorical Eligibility, July 14, 1999.‛ Washington DC: United States Department of Agriculture. Available at www.fns.usda.gov/snap/rules/Memo/1999/raletter2.htm. Food and Nutrition Service. 2001. ‚Food Stamp Provisions of the Agriculture, Rural Development, Food and Drug Administration, and Related Agencies Appropriations Act, 2001, January 4, 2001.‛ Washington DC: United States Department of Agriculture. Available at www.fns.usda.gov/snap/rules/Memo/2001/TANFVehicleRules.htm. Food and Nutrition Service. 2009a. State Options Report, Eighth Edition. Alexandria, VA: U.S. Department of Agriculture, Food and Nutrition Service. Available at www.fns.usda.gov/snap/rules/Memo/Support/State_Options/8-State_Options.pdf. Food and Nutrition Service. 2009b. ‚Improving Access to SNAP Through Broad-Based Categorical Eligibility.‛ Memorandum to Regional Administrators, September 30. Available at www.fns.usda.gov/snap/rules/Memo/2009/093009.pdf. Hoynes, H. W., and D. W. Schanzenbach. 2009. ‚Consumption Responses to In-Kind Transfers: Evidence from the Introduction of the Food Stamp Program.‛ American Economic Journal - Applied Economics 1 (4). Kabbani, N. S. and P. E. Wilde. 2003. ‚Short Recertification Periods in the U.S. Food Stamp Program.‛ Journal of Human Resources 38 (Supplement): 1112–38. Klerman, Jacob A., and Caroline Danielson. 2011. ‚The Transformation of the Supplemental Nutrition Assistance Program.‛ Journal of Policy Analysis and Management 30 (4): 863–88. Leftin, Joshua. 2010. Trends in Supplemental Nutrition Assistance Program Participation Rates: 2001 to 2008. Alexandria, VA: U.S. Department of Agriculture, Food and Nutrition Service. Available at http://fns.usda.gov/fns/. Leftin, Joshua, Andrew Gothro, Esa Eslami, Daisy Ewell, and Katherine Bencio. 2010. Technical Documentation for the Fiscal Year 2009 SNAP QC Database and QC Minimodel. Washington DC: Mathematica Policy Research. Available at http://hostm142.mathematica-mpr.com/fns/. Logan, Christopher, and Jacob Alex Klerman. 2008. Feasibility of Assessing Causes of State Variation in Food Stamp Program Administrative Costs: Final Report. Alexandria, VA: U.S. Department of Agriculture, Food and Nutrition Service. Mabli, J., L. Castner, J. Ohls, M. K. Fox, M. K. Crepinsek, and E. Condon. 2010. ‚Food Expenditures and Diet Quality among Low-Income Households and Individuals.‛ Final report (submitted to Food and Nutrition Service, U.S. Department of Agriculture). Available at www.mathematica-mpr.com/publications/PDFs/nutrition/FoodExpendDietQuality.pdf. Mabli, J., E. S. Martin, and L. Castner. 2009. ‚Effects of Economic Conditions and Program Policy on State Food Stamp Program Caseloads, 2000 to 2006.‛ Contractor and Cooperator Report No. 26, Mathematica Policy Research. Available at http://ddr.nal.usda.gov/bitstream/10113/35893/1/CAT31037490.pdf. Ratcliffe, Caroline, Signe-Mary McKernan, and Kenneth Finegold. 2008a. Effect of State Food Stamp and TANF Policies on Food Stamp Program Participation. Contractor and Cooperator Report No. 36, U.S. Department of Agriculture, Economic Research Service. Ratcliffe, Caroline, Signe Mary McKernan, and Kenneth Finegold. 2008b. ‚Effects of Food Stamp and TANF Policies on Food Stamp Receipt.‛ Social Service Review 82 (2): 291–334. Ratcliffe, Caroline and Signe-Mary McKernan. 2010. How Much Does SNAP Reduce Food Insecurity? Washington DC: Urban Institute. Available at www.urban.org/UploadedPDF/412065_reduce_food_insecurity.pdf. Rowe, Gretchen, Carolyn T. O’Brien, Sam Hall, Nancy M. Pindus, Lauren Eyster, Robin Koralek, and Alexandra Stanczyk. 2010 Enhancing Supplemental Nutrition Assistance Program (SNAP) Certification: SNAP Modernization Efforts: Final Report - Volumes I & II. Washington DC: Urban Institute. Available at www.urban.org/publications/412179.html. Smith, Patricia K. 2009. Obesity among Poor Americans: Is Public Assistance the Problem? Nashville, TN: Vanderbilt University Press. Schexnayder, Deanna T., Jerome A. Olson, Daniel P. O’Shea, Davy Norris, Daniel Schroeder, and Christopher T. King. 1997. Lone Star Image System Evaluation: Final Report. Center for the Study of Human Resources, Lyndon B. Johnson School of Public Affairs, University of Texas at Austin. Available at www.utexas.edu/research/cshr/pubs/pdf/LSIS_Final_Report.pdf. Sticha, Paul J., David Thomas, Chris Zamberlan, and Monica A. Gribben.1999. ‚Use of Biometric Identification Technology to Reduce Fraud in the Food Stamp Program: Final Report.‛ Submitted to U.S. Department of Agriculture Food and Nutrition Service. Available at www.fns.usda.gov/ora/menu/published/SNAP/FILES/ProgramIntegrity/biomeval.pdf. Super, D. A. 2004. ‚The Quiet ‘Welfare’ Revolution: Resurrecting the Food Stamp Program in the Wake of the 1996 Welfare Law.‛ New York University Law Review 79: 1271–1397. United States Department of Health and Human Services. 1996. Review of the Ongoing Los Angeles County Finger Imaging Demonstration Project. Washington DC: US. Department of Health and Human Services, Office of Inspector General. Available at http://oig.hhs.gov/oas/reports/region9/99500054.pdf. United States Department of Agriculture. 2011. ‚SNAP-10-6/Waiver 2030030 CA.‛ Memo, Food and Nutrition Service, Western Region, April 6. Caroline Danielson is a policy fellow at the Public Policy Institute of California. Her research focuses on social safety net programs, including nutrition assistance, welfare, child welfare, and child care subsidies. Before coming to PPIC she was a principal analyst at the University of California's Welfare Policy Research Project, a RAND graduate fellow, and an assistant professor of politics at the State University of New York, Potsdam. She holds a Ph.D. in political science from the University of Michigan. Jacob Klerman is a principal associate in Social and Economic Policy at Abt Associates and an Abt Senior Fellow. He has more than two decades of experience in social policy research and is an expert in nonexperimental evaluation of social programs. He was formerly a senior economist at the RAND Corporation, director of RAND’s Center for the Study of Social Welfare Policy, and a professor in the Pardee RAND Graduate School of Policy Studies. We received invaluable advice and data from a wide range of researchers, policy experts, and program officials during the course of preparing this report. They include Margaret Andrews, Kim Belshé, Todd Bland, Anastasia Dodson, Katie Fitzpatrick, Helen Lee, Charr Lee Metsker, David Neumark, Nadine Nichols, Mike Papin, Karen Peko, Hope Rios, Paul Smilanick, Shelly Ver Ploeg, and David Young. Richard Greene provided strong editing assistance. We gratefully acknowledge all of their contributions and the input of others whom we may have neglected to name here. However, we bear the entire responsibility for any errors and for the ultimate analysis and interpretation of the data used in this report." } ["___content":protected]=> string(102) "

R 911CDR

" ["_permalink":protected]=> string(121) "https://www.ppic.org/publication/californias-food-stamp-program-participation-and-cost-challenges-for-the-state/r_911cdr/" ["_next":protected]=> array(0) { } ["_prev":protected]=> array(0) { } ["_css_class":protected]=> NULL ["id"]=> int(8762) ["ID"]=> int(8762) ["post_author"]=> string(1) "1" ["post_content"]=> string(0) "" ["post_date"]=> string(19) "2017-05-20 02:40:35" ["post_excerpt"]=> string(0) "" ["post_parent"]=> int(4086) ["post_status"]=> string(7) "inherit" ["post_title"]=> string(8) "R 911CDR" ["post_type"]=> string(10) "attachment" ["slug"]=> string(8) "r_911cdr" ["__type":protected]=> NULL ["_wp_attached_file"]=> string(12) "R_911CDR.pdf" ["wpmf_size"]=> string(6) "872328" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(54056) "This report describes recent rapid changes to California’s CalFresh program—formerly known as the food stamp program—and assesses whether policies that the state is considering would improve the program— by increasing participation or lowering administrative costs. To the extent that CalFresh supports families when they experience economic difficulties and boosts local economies, policymakers may want to ensure that all eligible families have reasonable access to the program. The cost of the program is another important consideration. Although direct benefits are currently paid by the federal government—making it a relative bargain for the state—California’s costs to administer benefits nonetheless are higher compared to other states. Participation in safety net programs tends to grow when families lose employment or face work cutbacks in a poor economy. In California, CalFresh grew by 31 percent between December 2007, the official start of the last economic downturn, and its official end in June 2009. Participation has continued to grow as unemployment has remained high. Nevertheless, participation remains relatively low compared to other states. California and other states made substantial changes to program policies over the past decade, and policy- makers continue to consider further changes. Examining the policies of all 50 states, we conclude that the policy that has had the largest effect on program participation in California in recent years is the require- ment that most adults in food stamp households be fingerprinted. We find that eliminating fingerprinting would increase participation in California by about 7 percent (although we cannot rule out the possibility that a portion of the increase would come from fraud). Further, California is one of only two states with a requirement that most recipients report their status more frequently than semiannually, but we do not find this has depressed participation relative to other states. California’s costs of administering the program are also relatively high by national standards, and we find evidence that eliminating fingerprinting would not only increase participation but would also reduce the administrative costs of the program by about 13 percent. Quarterly reporting has not increased administrative costs relative to states with semiannual reporting. We conclude that California has already taken several important steps to bolster participation. It could additionally consider eliminating fingerprinting to boost participation and lower administrative costs. We find that moving to semiannual reporting would neither raise participation substantially nor reduce costs. AREERA Agricultural Research, Extension, and Education Reform Act of 1998 CalWORKs California Work Opportunity and Responsibility to Kids EBT Electronic Benefits Transfer IHSS In-Home Supportive Services SNAP Supplemental Nutrition Assistance Program TANF Temporary Assistance for Needy Families USDA United States Department of Agriculture Like every state, California has longstanding programs to assist poor and near-poor families. The food stamp program—recently renamed CalFresh—is now the second largest of these safety net programs, and as of May 2011, was serving about 3.7 million Californians.1 More than half are children. Its mechanism is simple—low-income families are credited with funds on a debit card that they then use to buy most groceries.2 The card is recharged with funds every month so long as the family remains eligible. Established nationwide in 1974, the food stamp program benefits not just low-income families with children, but also the low-income elderly. Households of one or more working-age adults can also participate, although qualifying for benefits is typically more difficult because work requirements can be more stringent. The food stamp program is relatively unusual in that the federal government pays for all benefits (which are not capped), while states—and, in California, also counties—share administrative costs.3 In 2009–2010 the total cost of CalFresh benefits plus the cost of administering the benefits was $5.62 billion in California, but the state and counties were responsible for $680 million—12 percent of total CalFresh spending, and 0.5 percent of the state’s 2009–2010 General Fund expenditures. Other state-federal funding models used in safety net programs require a higher state match, with states and counties typically contributing about half the cost of both benefits and program administration. If this model were applied to CalFresh, this would have meant total costs to the state and counties of about $2.84 billion in 2009–2010, not $680 million. Thus, for California, CalFresh is a relatively inexpensive social services program.4 Research indicates that the program also reduces food insecurity and boosts local economies (Hanson et al., 2002; Ratcliffe and McKernan, 2010).5 To be eligible for CalFresh, a household’s income after deductions cannot exceed federal poverty levels.6 For a single-member household—typically nearly a third of all CalFresh households—this means a monthly income after deductions of no more than $903. For families with two or three eligible members—who make up almost half of all CalFresh households—this translates to a monthly income of no more than $1,214 and $1,526, respectively.7 1 The largest such means-tested programs enroll children and adults in low-income families in health insurance. In California, these programs are called Medi-Cal and Healthy Families. 2 Usually families are unable to use their benefits to buy prepared foods, whether at a grocery store or a restaurant. Some localities, among them Los Angeles County—have obtained authority to enable homeless, elderly, or disabled participants to use their benefits at authorized restaurants. 3 Of special interest to the state is the state-funded program established to provide food assistance to legal immigrants ineligible for federal benefits, called California Food Assistance Program (CFAP). The state and counties bear both all administrative and all benefit costs for those in this program. However, most legal resident immigrants are eligible for federally-funded benefits, so this program is small: In spring 2011 about 39,500 immigrants received state-funded benefits, just over 1% of the number getting federally-funded benefits. 4 The federal food stamp program—officially called the Supplemental Nutrition Assistance Program (SNAP) is one of many discretionary federal programs that could see sharp reductions ordered by the new federal Joint Select Committee on Deficit Reduction. However, if the committee fails to come to agreement, triggering automatic federal spending cuts, SNAP would be exempt from those. 5 There is little solid evidence that participation in food stamps results in poor health outcomes (Baum, 2007; Gibson, 2003). The exception is women who stay on the program for several years or longer, who appear to be at increased risk of obesity (Smith, 2009). 6 Deductions include 20 percent for earned income, and others for expenses such as housing and child care. 7 These income thresholds are identical across all states, with the exception of Alaska and Hawaii, where they are higher. Thresholds are normally updated for inflation each year. However, they did not change between 2009 and 2011. Since 2008, asset limits, when applied, are also adjusted annually for inflation. The maximum monthly amount for a one-person household is $200; this maximum rises to $526 for a family of three.8 However, these maximums apply only to families with very little or no income. (Although in a minority, CalFresh households with no reported gross income nevertheless made up 19 percent of all CalFresh households in 2009.) For a family of three with one member employed at a low-wage ($11 an hour) job, the CalFresh amount would be $455—equivalent to an additional 27 percent in earnings.9 The recent growth of the CalFresh program is consistent with the recent economic downturn, but even so it has been striking. An additional 660,000 residents enrolled in food stamps between the start of the recession in December 2007 and its official end 18 months later—a 31 percent increase. This is more than double the increase in CalWORKs, the state’s main cash assistance program for families with children, over the same period. More recent statistics indicate that CalFresh participation grew by an additional 33 percent between June 2009 and May 2011.10 Nevertheless, California continues to have both lower participation and higher costs compared to other states. The gap in participation began to appear in about 2000 and continues, notwithstanding the more recent growth. In California, this growth was slowest among households of both eligible children and adults. The relative gap is apparent whether viewed as a percentage of the total population or as a percentage of those eligible. Using the first metric, about one in seven people in the other 49 states and the District of Columbia were receiving food assistance in early 2011, compared to one in ten Californians. Using the second metric, official estimates for 2008 (the latest that have been computed) indicate that about half of those eligible for CalFresh benefits actually receive them; in the rest of the country this ratio was above two- thirds (Leftin, 2010). Figure 1 details this gap more precisely: The percentage of state residents participating in food stamps in California and in the rest of the country was fairly similar in the 1990s, but a gap emerged in the early part of the following decade and became substantial by 2005.11 In California, participation increased only gradually until the start of the recent, severe recession in late 2007, while growth nationally began in 2001. 8 Benefit levels are updated for inflation each year. However, the 2009 American Recovery and Reinvestment Act (the ‚stimulus‛ bill) raised family benefits by an average of about 13 percent, and the automatic benefit inflation adjustment is suspended until 2013, when benefits will drop to where they would have been without the stimulus. 9 This benefit calculation uses the 20th percentile hourly wage among California workers in 2009. It assumes 35 hours of work each week, no unearned income (for example, from CalWORKs or child support payments), the fair market rent for a two-bedroom apartment in a low-cost county ($900), child care expenses of $500 a month, and the standard utility deduction. It also assumes that the household contains no elderly or disabled members, and no members ineligible for the program (for example, because of documentation status or CalWORKs sanction). 10 Participation does vary widely within California. As a percentage of the population, participation ranged from 2 percent to 22 percent across California’s counties in 2009. However, the percent using food stamps is highly correlated with the estimated percent of the county population in poverty—about 70 percent of the variation in participation is explained by poverty alone. Moreover, there is a one-to-one correlation between poverty at the county level and higher caseloads. 11 Trends in official estimates of the percentage of those eligible and receiving the benefit show similar evidence of an emerging gap (Leftin, 2010). Administrative costs for CalFresh have remained high compared to other states for several decades. Data from the U.S. Department of Agriculture’s (USDA) Food and Nutrition Service indicate that the combined monthly federal, state, and county administrative cost to administer food stamp benefits in California averaged $55 per household in 2009, second highest in the nation.12 Although these costs have been declining over the past decade in the state, from 1989 to 2009 they averaged nearly twice the per-household cost reported by all other states.13 And while the state and counties’ share of total program costs is low relative to other state safety net programs, relatively high administrative costs are still a concern during an economic downturn. 12 Total costs per case were even higher—$93 in California—but this total includes expenditures not closely related to administering benefits (e.g., nutrition education, employment and training programs, and automation development projects). The technical appendices describe the line items in states’ FNS-269FS reports that we categorize as ongoing costs of administering benefits. Logan and Klerman (2008) provide a further description and exploration of the underlying FNS-269FS reports that form the basis of our analysis of administrative costs. 13 There are several reasons why ongoing costs to administer the program might be higher in California than in other places. In the first place, California has a mix of cases that is arguably more expensive to administer (County Welfare Directors’ Association, 2010). Some of the cost difference is also likely attributable to higher wages in California (U.S. Bureau of the Census, n.d.). Still, policy changes that streamline the processing of applications and reduce ongoing paperwork might reduce the state’s administrative costs. 0 5 10 15 Percent of population California U.S. outsideof California In the nation excluding California, the number of food stamp households began growing steadily in mid- 2000; by mid-2005, caseloads were about 56 percent higher than in 2000. By comparison, California’s caseloads grew 13 percent from mid-2000 to mid-2005. Californians’ participation grew at roughly the same rate as in the rest of the country after 2005, but the gap has not closed. A large part of the difference was due to slower increases among households containing both aided children and aided adults, rather than increases in households with children as the sole recipients of benefits or households with adult recipients only.14 Because eligibility for CalFresh is broadly aimed at low-income households, we examined three distinct household subgroups for further insight into caseload trends from 2000 to 2010 (Figure 2):  those with a mix of eligible adults and children;  those composed only of eligible children (because all adults in the household are ineligible);15  those with one or several eligible adults (aged 18 and older) and no children. We did not separately consider households containing only recipients aged 65 and older because this group made up only 1 to 2 percent of the caseload over the previous decade.16 The trend for the first of these three groups is clear: Mixed-age households actually shrank by about 10,000 from 2000 to 2005 while the numbers of the other two household types grew, as did all three groups nationally.17 Thus, 14 Data on characteristics of CalFresh households are publicly available after two years, so we can use data only through 2009 in this report. 15 In California, the most common reason that only children are aided is that parents are unauthorized immigrants. Ineligible adults may also be Supplemental Security Income (SSI) recipients, may be under a CalWORKs sanction, or may be relative caregivers. 16 The very small numbers of elderly recipients in California is largely the result of the state’s SSI cash-out policy, which we discuss below. In the U.S. outside of California, households containing only aided elderly made up 11 percent of the caseload in 2009. 390 380 513 167 229 272 115 163 328 0 200 400 600 800 1,000 1,200 200020052009 CalFresh cases (thousands) Only eligible adults Only eligible children Mix of eligible adults andchildren the place to look for further insight into diverging caseload trends is households with a mix of aided adults and aided children. A decade ago, CalFresh food assistance and CalWORKs—income support and work services for families with children—were closely linked in California. Although there was no explicit requirement that this be the case, the vast majority of all mixed-age CalFresh households also had assistance from CalWORKs—86 percent, or 336,000 of the 390,000 total (Figure 3). This had changed by 2005. The number of CalFresh households combining food and cash assistance dropped sharply from 336,000 to 243,000.18 Concurrently, the number of CalFresh households that had income from wages, income from other sources, or no income doubled or more. Not surprisingly, given the deterioration of the economy, all segments of the mixed-age caseload grew pre- to post-recession (2005 to 2009). Still, those combining food and cash assistance grew the least in percentage terms. In other words, the safety net for families with children changed substantially during the last decade. Relatively fewer California families with children are making use of the traditional safety net combination of food and cash assistance, and more are using CalFresh to supplement other sources of income—or using food stamps as their sole source of income. By 2009, just 58 percent of all mixed-age households were combining food and cash assistance, while 20 percent were using CalFresh to supplement earnings. Ten percent had income from other sources (most often unemployment insurance), and the remaining 12 percent had no cash income. 17 This does not imply that CalFresh households containing only aided children or only aided adults came to participate when they were in fact ineligible. In fact, calculations of program use among those eligible throughout the entire U.S. indicates that children who have immigrant parents tend to participate less frequently than those with citizen parents Leftin (2011). 18 Households not currently receiving CalWORKs benefits may nevertheless have recently received them. This is because California automatically extends food stamp eligibility for five months after a family leaves the CalWORKs program. 336 243 300 34 75 102 17 38 49 4 25 63 0 100 200 300 400 500 600 200020052009 CalFresh cases (thousands) No income Other income Earnings, no cashassistance Cash assistance A similar—but more pronounced—pattern also emerged in the other 49 states; the number of mixed-age households making use of both food stamps and cash assistance grew by only 4 percent over the entire period (2000–2009), while those combining food stamps with earnings, with other income, or that had no income, each doubled or more than doubled. In the U.S. outside of California in 2009, households on welfare made up just 28 percent of the mixed-age food stamps caseload, 40 percent had earnings, 20 percent had income from other sources, and the remaining households (12%) had no cash income. These patterns suggest that families’ use of safety net programs has changed and that reaching families eligible for CalFresh but that lack a current connection to CalWORKs has been a particular challenge. California’s demographics differ from other states, as does its economy and one longstanding CalFresh policy. Upon examination, however, none of these can explain its lower CalFresh participation rate during the last decade. Demographics. California is home to more noncitizens than any other state; these noncitizens may be reluctant to avail themselves of CalFresh assistance, and this could be depressing participation. Census data from 2008 show that among the 29 percent of all California households with incomes under 200 percent of the federal poverty level, 8 percent were noncitizen households—double the share among low-income households in the rest of the country. Although unauthorized immigrants are ineligible for the program, their citizen children are eligible if they meet other requirements. Ineligible parents may be unaware of their children’s eligibility or be reluctant to apply because of their own status. Unauthorized immigrants may also avoid the application process for their children because they think they themselves will be deported, or be denied citizenship in the future. Although the USDA’s Food and Nutrition Service issued instructions a decade ago that made clear that neither of these actions would be taken because of a food stamp application, concern may still linger. But other demographic characteristics might attenuate any effect that a large immigrant population has on participation. California’s low-income population is younger and has lower educational attainment compared to other states. These two factors tend to increase CalFresh participation compared to other parts of the country.19 Using a statistical model, we conclude that California’s distinctive population characteristics depress participation in food stamps by about 10 percent. This is far less than would be the case if we only considered the role that a larger immigrant population plays.20 In any case, population characteristics change relatively slowly, over the course of a few years or decades, and cannot account for the fast-moving trends since 2000. Economy. In contrast, the economy does change rapidly. When the economy worsens, more households become eligible for food stamps or for a larger amount of assistance. However, between 2000 and 2005 California’s economy—as measured by its unemployment rate—was in fact a little worse than in the rest of the nation. On average, California’s unemployment rate was about 0.75 percentage points higher than in the rest of the country, and the unemployment rate rose earlier and more steeply in California at the outset of 19 Specifically in 2008, 28 percent of low-income households in California had one or more children under age 5, and 71 percent had one or more children age 5–17. In the U.S. outside of California, 24 percent had at least one child under age 5 and 54 percent had older children. Further, in California 17 percent of low-income households contained only elderly members, while 22 percent of households in the rest of the U.S. were composed of only members age 65 and older. In California 33 percent of low-income households had a household head who had not finished high school; in the rest of the U.S., 26 percent of household heads lacked a high school degree. In all cases, these differences between California and the rest of the U.S. are statistically significant at the 1 percent level or better. 20 The covariates we include in the descriptive model are: the citizenship of household head, the citizenship of household members, the age of the household head, the age of other household members, household size, whether English is spoken at home, the household’s income as a percent of the federal poverty level, the labor force participation of household adults age 18–64, whether anyone in the household received SSI in the past year, whether anyone in the household received cash welfare in the past year, the educational attainment of the household head, and the marital status of household head. We document our approach, which relies on the research presented in Burstein et al. (2009) and Cody et al. (2007), in the technical appendices. the recent recession than it did in other places in the country. In sum, a better economy in California is not an explanation for lower participation. SSI cash-out. One policy difference between California and all other states predates the diverging trends of the last decade and so cannot explain them. California alone makes its 1.2 million aged, blind, and disabled Supplemental Security Income (SSI) recipients ineligible for CalFresh. Instituted in 1974 and known informally as ‚SSI cash out,‛ the policy originally added $10 to the monthly checks of all SSI recipients in place of food stamps; it was intended to reduce paperwork for SSI recipients and save administrative costs for the state. Changing this policy would have a complicated impact on the total number of CalFresh recipients in the state and the benefits that families receive. On the one hand, certain households—especially those containing only SSI recipients—would be eligible for more than $10 worth of food assistance each month if the policy were ended. On the other hand, they would be required to apply for the program. Some would fail to apply. Some larger households with one or a few SSI recipients would be ineligible if the SSI payment or payments (including the extra that California’s supplemental program provides) were then included in the entire household’s income. Simulations conducted by Mathematica Policy Research that take both factors into consideration indicate that more California households would be eligible for CalFresh if the policy were ended, but also that the total benefits received by Californians could drop slightly (Cunnyngham, 2010). The federal government began to give states increased freedom to adjust food stamp policies in the late 1990s. This new flexibility was principally aimed at giving states tools to increase working families’ access to food stamps, particularly those who had gone off cash welfare (while remaining eligible for food assistance), as occurred for large numbers in the wake of the 1996 federal overhaul of welfare (Super, 2004).21 In general, these policy changes were focused on eligibility restructuring and paperwork reduction. They could have the additional benefit of reducing administrative costs, by streamlining the oversight that social services offices conduct to ensure that they pay benefits accurately. The changes also might reduce ‚churning,‛ or the number of times families drop out and then return to the program because they fail to keep their eligibility current. California has made a number of such modifications and is considering others. Several occurred in the first half of the decade; several were put in place only very recently, between July 2009 and February 2011. To estimate comparative effects using nationwide data, we first compiled policy specifics for all other states, because they did not previously exist. We conducted a survey of food stamp program administrators in the 49 other states and the District of Columbia. Details are accurate through August 2010, and are discussed in greater detail in the technical appendices.22 Our models isolate any change in per capita food stamp participation or ongoing administrative costs per case that occurred after states introduced each policy.23 The models control for state differences, state trends, and national changes, and hold constant state population characteristics and other differences that affect eligibility, the likelihood of participating if eligible, or both.24 After including these controls, if a statistically detectable change followed the introduction of a new policy, we attribute it to the policy. EBT cards. The program’s previous name, food stamps, derives from the form in which families originally received benefits. In the early 1990s several states began converting these paper stamps to debit cards known as ‚Electronic Benefits Transfer‛ or EBT cards. In 1996, believing that the change would reduce fraud and increase the effectiveness of the program, Congress required all states to switch to EBT cards. By mid-2004, all of California’s counties were issuing benefits via EBT cards that are recharged electronically each month in the amount of the family’s benefit. We find that EBT had no statistically significant effects on participation or administrative costs. Online applications. Web-based program applications—enabling applicants to file and electronically sign an application from any location and at any time of the day—are a recent development in most states. In 21 We discuss major policy changes in this section, but not an exhaustive list. Other efforts that states have made are described in Rowe et al. (2010) and Food and Nutrition Service (2009). 22 In addition, see Danielson et al. (2011). 23 The participation models we report do not include state-funded food stamp recipients (in California, CFAP recipients). However, the technical appendices report models that do include state-funded food stamp recipients, and the results do not change our conclusions about the effects of fingerprinting or any other policy on participation. 24 We carried out our analysis of food stamp participation using the percentage of the entire population receiving food stamps rather than estimates of the percentage of the eligible population receiving food stamps. This is mainly because estimates of the eligible population exclude those made eligible by changes in asset policies. We provide a detailed discussion in the technical appendices. California, a pilot program began in October 2007, and all counties have been able to accept online applications since December 2010. We find that enabling households to submit an application online does not increase participation measurably, but we estimate that it lowers ongoing administrative costs by 19 percent.25 Assets. Determining a household’s food stamp eligibility in the past usually required a detailed assessment of a household’s assets. Households with more than $2,000 in liquid assets ($3,000 if the household contains an elderly or disabled member) were ineligible for the program, regardless of their income. Two ways in which this has changed are the treatment of the value of the vehicles that a household may own and the consideration of other household assets (for example, bank accounts, retirement accounts, and the like).26 Since 2001, the federal government has given states the option to evaluate vehicle assets using rules from other assistance programs (TANF, child care, or foster care). This option has had the effect of allowing states to exclude one or more household vehicles from the food stamp asset test. In 2004 California began exempting the value of all vehicles. And since 1996 the federal government has allowed states to exempt households from other asset tests. (This is termed ‚categorical eligibility‛). In July 2009, California counties began exempting households with children from the asset test altogether. The state expanded this exemption to all households in February 2011. All households still must qualify based on income. We find that eliminating the asset test increases participation by an estimated 5.6 percent.27 (However, the earlier switch to excluding cars when considering a household’s eligibility for food stamps does not appear to have boosted participation.)28 Households containing a mix of aided children and adults are substantially affected by this policy, increasing by an estimated 6.3 percent in the wake of its adoption. (Other types of households are affected as well.) While it might seem that eliminating asset tests would reduce the complexity of eligibility determinations and so streamline the application process for social services offices, we find no evidence that exempting vehicles from the asset test or eliminating the asset test altogether lowers the costs of administering the program. Fingerprinting. California made one policy change in December 2001 that added a step to the application process—fingerprinting.29 This was instituted to prevent recipients receiving benefits from multiple social services offices. California is one of five states to use fingerprinting in recent years, but CalFresh is not the only program within the state that uses fingerprinting: CalWORKs adults, General Assistance participants, and In-Home Supportive Services (IHSS) caregivers must also be fingerprinted.30 25 We exclude capital and other start-up costs for automation projects in our analysis of factors driving administrative costs. See the technical appendices for details. 26 Households have always obtained exemptions for a house that they live in, vehicles used for a business or to live in. Most other available assets were long limited to $2,000 ($3,000 if the household contains an elderly or disabled member). 27 For further details of this policy, see California Department of Social Services (2009a) and California Department of Social Services (2011). 28 Evidence from other research indicates that such policies increased participation among some types of households while simultaneously reducing participation among other types of households, leaving the size of the overall caseload unchanged (Klerman and Danielson, 2011). In particular, that research found that vehicle policies was partly responsible for shifting the composition of the caseload away from those combining cash and food assistance and toward those getting food assistance in the absence of cash assistance. 29 California uses an electronic imaging process rather than conventional fingerprinting. 30 The most recent legislative effort to eliminate fingerprinting is AB 6 (Fuentes), introduced in December 2010. As amended in August 2011, it would change only the CalFresh fingerprinting requirement. Our model indicates that fingerprinting reduces the share of California’s CalFresh population by about 7 percent. The effect is concentrated in mixed-age households, which dropped by 9.6 percent after fingerprinting was introduced. In fact, modeling the policy effect on households with no aided adults (often, although not always, because of adults’ immigration status), we find no statistically significant reduction.31 Any deterrent effect of fingerprinting on the CalFresh program may have been softened as a result of changes to other application requirements made in 2008 and 2009 in California. This is because interviews to determine eligibility can now be conducted over the phone, and adults must come in to the social services office to be fingerprinted within 12 months of applying for benefits (California Department of Social Services, 2008; California Department of Social Services, 2009b). We estimate that the introduction of fingerprinting also raises administrative costs by about 13 percent. Thus, while eliminating fingerprinting would increase state administrative costs because more households would enroll in CalFresh, administrative costs per case would decline, offsetting the greater cost of a higher caseload. The additional administrative costs of requiring fingerprints might be justified if the policy effectively detects or deters fraud, but the evidence for this is inconclusive. First, we considered whether the number of positive fraud findings increased in a state after fingerprinting was implemented and whether the rate of positive fraud findings (positive findings as a fraction of total investigations) improved. The first would occur if a state that adopted digital fingerprinting invests additional resources in fraud investigations that the new technology enables them to detect. The second of these would occur if fingerprinting enabled a state to detect fraud more efficiently. We found no evidence that fingerprinting increases findings of fraud, or that it increases the rate of positive fraud findings. This is in line with the previous research that has found that fingerprinting detects fraud relatively rarely (State Auditor, 2003; Sticha et al., 1999).32 Second, we considered whether fingerprinting deters applicants. Because we lack information about those who do not apply for food assistance, we cannot easily determine how much fingerprinting deters fraudulent applications, as opposed to cutting legitimate applications among those who simply find the requirement burdensome.33 Other research on this question provides little light on this issue because the research studied only those who actually applied for assistance. Case studies of small groups of applicants and recipients who refused to comply with the fingerprinting requirement in food and cash assistance programs were conducted more than a decade ago in Los Angeles County, Arizona, and Texas; these studies found that a minority of those who refused to comply were cases of probable or verified duplicate aid. In Los Angeles County, investigators found probable or verified fraud in 69 percent of the 137 cases they investigated, but only 6 percent represented verifiable duplicate-aid fraud (State Auditor, 2003; Sticha, 1999).34 In sum, the evidence from our and other research indicates that fingerprinting works primarily by deterring applications rather than catching fraudulent applicants—and it is uncertain who, exactly, is being deterred. 31 The estimate for the children-only caseload is also much smaller—roughly 40 percent of the size of the estimate for mixed-age households. 32 The technical appendices to this report provide details. The overall prevalence of detected fraud is low—nationwide, about 1% of households participating in the program—and has been flat over the past decade across both states that have fingerprinting and those that do not. 33 Legitimate applicants could also be turned away if administrative errors in evaluating finger images are made. The evaluation conducted in Lost Angeles County in the mid-1990s found that such errors do occur (State Auditor, 2003). 34 In a further 10 percent of cases investigators suspected duplicate-aid fraud but could not verify it. An additional 7 percent could not be located at all. In the remaining cases in which inspectors suspected or found fraud, clients had concealed income, misrepresented household composition, or the like. Quarterly and semiannual reporting by recipients. Believing that social services offices can require less paperwork of families and still issue benefits accurately, the federal government has encouraged all states to move to less frequent, semiannual reports of family income. Because California still tracks families’ incomes using quarterly reporting it has had to request several waivers from the federal government to continue this policy. Moreover, the USDA is still strongly urging the state to halve the number of required reports as rapidly as possible (USDA, 2011).35 We find that making this switch would not increase participation in CalFresh. The plausible reason lies with the second main paperwork policy. Periodically, social services offices conduct a comprehensive reassessment of households’ eligibility and benefit amounts, known as certification. California already certifies most households annually, and a switch to semiannual reporting would not affect this practice. (The most common certification period in other states in 2009 was six months.) Our estimates suggest that, given California’s longstanding practice of using relatively long certification periods, the additional effort required to submit quarterly, instead of semiannual, interim reports each year is relatively minor.36 While it might seem obvious that requiring two rather than four interim reports each year would reduce administrative costs, we do also not find empirical support for this. Because the administrative cost data are at a summary level, it is difficult to explore this finding further. It is possible that reducing the number of recipient reports does lower costs, but that social services offices use the resources that are freed up to improve other aspects of program operations. 35 The state and counties will incur costs to reprogram data systems to accept semiannual rather than quarterly reports, and, possibly, to delink the CalWORKs quarterly reports from the CalFresh semiannual reports. See Wagner (2010). The most recent legislative attempt to switch to semiannual reporting is AB 6 (Fuentes), introduced in December 2010, and which was sent to the governor in September 2011. 36 In our statistical models, we compare states that introduce policies of semiannual interim reporting joined with semiannual or annual certification periods to states with policies of quarterly interim reporting joined with semiannual or annual certification periods. We find no statistically significant difference between the two. As a sensitivity check, we also model certification and reporting policies separately, and again find no statistically significant difference between states that adopt quarterly reporting and states that adopt semiannual reporting. The technical appendices provide details. The number of low-income Californians making use of food stamps has grown sharply over the past several years. In the absence of any action on the part of policymakers, we would expect CalFresh participation to remain high for several years to come. This is because the state’s unemployment rate remains high, roughly twice what it was in the fall of 2007. Previous research on the link between the state of the economy and food stamp caseloads indicates that that demand for food assistance will grow for several years even after the economy has begun to improve substantially (Klerman and Danielson, 2011). The CalFresh caseload should drop once an economic recovery is securely in place. Still, in both good economic times and bad, California has had a large number of residents eligible for CalFresh who do not take advantage of the food assistance it offers. Could policy changes help more eligible families make use of the program? While policymakers have taken several steps—including, recently, switching from income- and assets-based eligibility tests to solely income- based eligibility tests—that have boosted participation, they could additionally consider completely eliminating the requirement that most adult applicants be fingerprinted. The state could instead rely on existing data systems and applicant social security numbers to track duplicate aid.37 Our evidence indicates that the group of households most sensitive to the fingerprinting policy are mixed-age households, those whose participation lagged in the early 2000s. We also find evidence that the increased cost of administering a larger caseload we would expect from ceasing fingerprinting would be offset by a reduction in cost per case. The widespread use of fingerprinting in safety net programs sets California apart from most other states. There is no federal requirement that CalFresh adults be fingerprinted and the policy appears to deter some eligible families from applying. However, we are unable to rule out the possibility that the incidence of benefits being paid from multiple social services offices would increase if the state were to eliminate fingerprinting. California’s fingerprinting policy will ultimately reflect policymakers’ ongoing efforts to balance two program goals: reasonable access and program integrity. Our evidence also indicates that families with children are distinctly less likely to be combining food and cash assistance in traditional safety net fashion than they used to do. Instead, they are making use of CalFresh as a supplement to wage earnings or other sources of income—or food stamps are their sole income. Looking ahead, policymakers should consider paying special attention to trends among these mixed-age households. They are a low-income population traditionally of concern to many policymakers, but their participation has lagged. Our evidence indicates that these households are generally responsive to policy changes aimed at increasing access among those eligible. Thus, continued efforts to fine-tune CalFresh to reach this population have the potential to make a real difference in the lives of these children and their parents. 37 The state’s Medi-Cal Eligibility Determination System (MEDS) has been used to detect social security number-based duplicate aid fraud across counties, within counties, and in Nevada, Oregon, and Arizona (State Auditor, 2003). . Burstein, N. R., S. Patrabansh, William L. Hamilton, and Sarah Y. Siegel. 2009. Understanding the Determinants of Supplemental Nutrition Assistance Program Participation: Final Report. Alexandria, VA: U.S. Department of Agriculture Food and Nutrition Service. Breunig, Robert, Indraneel Dasgupta, Craig Gundersen, and Prasanta Pattanaik. 2001. Explaining the Food Stamp Cash-Out Puzzle. Food Assistance and Nutrition Research Report No. 12. Washington, DC: U.S. Department of Agriculture Economic Research Service. Available at www.ers.usda.gov/publications/fanrr12/fanrr12.pdf. California Department of Social Services. 2008. ‚Food Stamp Waiver of Face-to-Face Interview and Non-Assistance Food Stamp Household Recertification Form, FS 27.‛ All-County Letter No. 08-32, July 14. Available at www.dss.cahwnet.gov/lettersnotices/entres/getinfo/acl08/08-32.pdf. California Department of Social Services. 2009a. ‚Modified Categorical Eligibility for the Food Stamp Program : Questions and Answers.‛ All-County Letter No. 09-24, May 27. Available at www.dss.cahwnet.gov/lettersnotices/entres/getinfo/acl/2009/09-24.pdf. California Department of Social Services. 2009b. ‚Food Stamp Program (FSP) Waiver of Face-to-Face Interview.‛ All-County Letter No. 09-62, October 23. Available at www.dss.cahwnet.gov/lettersnotices/entres/getinfo/acl/2009/09-62.pdf. California Department of Social Services. 2011. ‚Expansion of Modified Categorical Eligibility for CalFresh to All Non- Assistance CalFresh (NACF) Households.‛ All-County Letter No. 11-11, January 27. Available at www.dss.cahwnet.gov/lettersnotices/entres/getinfo/acl/2011/11-11.pdf. California State Auditor. 2003. Statewide Fingerprint Imaging System: The State Must Weigh Factors Other Than Need and Cost- Effectiveness When Determining Future Funding for the System. January 2003, 2001-015. Sacramento, CA. Available at www.bsa.ca.gov/pdfs/reports/2001-015.pdf. California Welfare Directors’ Association. 2010. California Counties Work to Improve Food Stamp Access. Sacramento, CA. Available at www.cwda.org/downloads/publications/foodstamps/Food-Stamp-Fact-Sheet-with-Addendum-April- 2010.pdf. Center on Budget and Policy Priorities. 2008. States’ Vehicle Asset Policies in the Food Stamp Program. Revised July 1. Available at www.cbpp.org/cms/?fa=view&id=554. Cody, Scott, Allen Schirm, Elizabeth Stuart, Laura Castner, and Alan Zaslavsky. 2007. Sources of Variation in State-Level Food Stamp Participation Rates. Final report (submitted to U.S. Department of Agriculture). Available at www.mathematica-mpr.com/Publications/PDFs/variation-fspr.pdf. Cunnyngham, Karen. 2010. ‚Estimated Effects on the Supplemental Nutrition Assistance Program of Eliminating California’s SSI Cash-Out Policy.‛ Final report (submitted to California Food Policy Advocates). Available at www.mathematica-mpr.com/publications/PDFs/nutrition/SNAP_CA_cashout.pdf. Danielson, Caroline, and Jacob A. Klerman. 2008. ‚Did Welfare Reform Cause the Caseload Decline?‛ Social Service Review 82 (4): 703–30. Danielson, Caroline, Jacob Alex Klerman, Margaret Andrews, and Daniel Krimm. 2011. "Reporting Requirement and Asset Test Policies in the Supplemental Nutrition Assistance Program." Unpublished manuscript (March). Fernández, Alexis. 2010. California Food Stamps Characteristics Report. Oakland, CA: California Food Policy Advocates. Available at www.cfpa.net/foodstamps/foodstamp_characteristics.pdf. Food and Nutrition Service. 1999. ‚Letter to Regions on Categorical Eligibility, July 14, 1999.‛ Washington DC: United States Department of Agriculture. Available at www.fns.usda.gov/snap/rules/Memo/1999/raletter2.htm. Food and Nutrition Service. 2001. ‚Food Stamp Provisions of the Agriculture, Rural Development, Food and Drug Administration, and Related Agencies Appropriations Act, 2001, January 4, 2001.‛ Washington DC: United States Department of Agriculture. Available at www.fns.usda.gov/snap/rules/Memo/2001/TANFVehicleRules.htm. Food and Nutrition Service. 2009a. State Options Report, Eighth Edition. Alexandria, VA: U.S. Department of Agriculture, Food and Nutrition Service. Available at www.fns.usda.gov/snap/rules/Memo/Support/State_Options/8-State_Options.pdf. Food and Nutrition Service. 2009b. ‚Improving Access to SNAP Through Broad-Based Categorical Eligibility.‛ Memorandum to Regional Administrators, September 30. Available at www.fns.usda.gov/snap/rules/Memo/2009/093009.pdf. Hoynes, H. W., and D. W. Schanzenbach. 2009. ‚Consumption Responses to In-Kind Transfers: Evidence from the Introduction of the Food Stamp Program.‛ American Economic Journal - Applied Economics 1 (4). Kabbani, N. S. and P. E. Wilde. 2003. ‚Short Recertification Periods in the U.S. Food Stamp Program.‛ Journal of Human Resources 38 (Supplement): 1112–38. Klerman, Jacob A., and Caroline Danielson. 2011. ‚The Transformation of the Supplemental Nutrition Assistance Program.‛ Journal of Policy Analysis and Management 30 (4): 863–88. Leftin, Joshua. 2010. Trends in Supplemental Nutrition Assistance Program Participation Rates: 2001 to 2008. Alexandria, VA: U.S. Department of Agriculture, Food and Nutrition Service. Available at http://fns.usda.gov/fns/. Leftin, Joshua, Andrew Gothro, Esa Eslami, Daisy Ewell, and Katherine Bencio. 2010. Technical Documentation for the Fiscal Year 2009 SNAP QC Database and QC Minimodel. Washington DC: Mathematica Policy Research. Available at http://hostm142.mathematica-mpr.com/fns/. Logan, Christopher, and Jacob Alex Klerman. 2008. Feasibility of Assessing Causes of State Variation in Food Stamp Program Administrative Costs: Final Report. Alexandria, VA: U.S. Department of Agriculture, Food and Nutrition Service. Mabli, J., L. Castner, J. Ohls, M. K. Fox, M. K. Crepinsek, and E. Condon. 2010. ‚Food Expenditures and Diet Quality among Low-Income Households and Individuals.‛ Final report (submitted to Food and Nutrition Service, U.S. Department of Agriculture). Available at www.mathematica-mpr.com/publications/PDFs/nutrition/FoodExpendDietQuality.pdf. Mabli, J., E. S. Martin, and L. Castner. 2009. ‚Effects of Economic Conditions and Program Policy on State Food Stamp Program Caseloads, 2000 to 2006.‛ Contractor and Cooperator Report No. 26, Mathematica Policy Research. Available at http://ddr.nal.usda.gov/bitstream/10113/35893/1/CAT31037490.pdf. Ratcliffe, Caroline, Signe-Mary McKernan, and Kenneth Finegold. 2008a. Effect of State Food Stamp and TANF Policies on Food Stamp Program Participation. Contractor and Cooperator Report No. 36, U.S. Department of Agriculture, Economic Research Service. Ratcliffe, Caroline, Signe Mary McKernan, and Kenneth Finegold. 2008b. ‚Effects of Food Stamp and TANF Policies on Food Stamp Receipt.‛ Social Service Review 82 (2): 291–334. Ratcliffe, Caroline and Signe-Mary McKernan. 2010. How Much Does SNAP Reduce Food Insecurity? Washington DC: Urban Institute. Available at www.urban.org/UploadedPDF/412065_reduce_food_insecurity.pdf. Rowe, Gretchen, Carolyn T. O’Brien, Sam Hall, Nancy M. Pindus, Lauren Eyster, Robin Koralek, and Alexandra Stanczyk. 2010 Enhancing Supplemental Nutrition Assistance Program (SNAP) Certification: SNAP Modernization Efforts: Final Report - Volumes I & II. Washington DC: Urban Institute. Available at www.urban.org/publications/412179.html. Smith, Patricia K. 2009. Obesity among Poor Americans: Is Public Assistance the Problem? Nashville, TN: Vanderbilt University Press. Schexnayder, Deanna T., Jerome A. Olson, Daniel P. O’Shea, Davy Norris, Daniel Schroeder, and Christopher T. King. 1997. Lone Star Image System Evaluation: Final Report. Center for the Study of Human Resources, Lyndon B. Johnson School of Public Affairs, University of Texas at Austin. Available at www.utexas.edu/research/cshr/pubs/pdf/LSIS_Final_Report.pdf. Sticha, Paul J., David Thomas, Chris Zamberlan, and Monica A. Gribben.1999. ‚Use of Biometric Identification Technology to Reduce Fraud in the Food Stamp Program: Final Report.‛ Submitted to U.S. Department of Agriculture Food and Nutrition Service. Available at www.fns.usda.gov/ora/menu/published/SNAP/FILES/ProgramIntegrity/biomeval.pdf. Super, D. A. 2004. ‚The Quiet ‘Welfare’ Revolution: Resurrecting the Food Stamp Program in the Wake of the 1996 Welfare Law.‛ New York University Law Review 79: 1271–1397. United States Department of Health and Human Services. 1996. Review of the Ongoing Los Angeles County Finger Imaging Demonstration Project. Washington DC: US. Department of Health and Human Services, Office of Inspector General. Available at http://oig.hhs.gov/oas/reports/region9/99500054.pdf. United States Department of Agriculture. 2011. ‚SNAP-10-6/Waiver 2030030 CA.‛ Memo, Food and Nutrition Service, Western Region, April 6. Caroline Danielson is a policy fellow at the Public Policy Institute of California. Her research focuses on social safety net programs, including nutrition assistance, welfare, child welfare, and child care subsidies. Before coming to PPIC she was a principal analyst at the University of California's Welfare Policy Research Project, a RAND graduate fellow, and an assistant professor of politics at the State University of New York, Potsdam. She holds a Ph.D. in political science from the University of Michigan. Jacob Klerman is a principal associate in Social and Economic Policy at Abt Associates and an Abt Senior Fellow. He has more than two decades of experience in social policy research and is an expert in nonexperimental evaluation of social programs. He was formerly a senior economist at the RAND Corporation, director of RAND’s Center for the Study of Social Welfare Policy, and a professor in the Pardee RAND Graduate School of Policy Studies. We received invaluable advice and data from a wide range of researchers, policy experts, and program officials during the course of preparing this report. They include Margaret Andrews, Kim Belshé, Todd Bland, Anastasia Dodson, Katie Fitzpatrick, Helen Lee, Charr Lee Metsker, David Neumark, Nadine Nichols, Mike Papin, Karen Peko, Hope Rios, Paul Smilanick, Shelly Ver Ploeg, and David Young. Richard Greene provided strong editing assistance. We gratefully acknowledge all of their contributions and the input of others whom we may have neglected to name here. However, we bear the entire responsibility for any errors and for the ultimate analysis and interpretation of the data used in this report." ["post_date_gmt"]=> string(19) "2017-05-20 09:40:35" ["comment_status"]=> string(4) "open" ["ping_status"]=> string(6) "closed" ["post_password"]=> string(0) "" ["post_name"]=> string(8) "r_911cdr" ["to_ping"]=> string(0) "" ["pinged"]=> string(0) "" ["post_modified"]=> string(19) "2017-05-20 02:40:35" ["post_modified_gmt"]=> string(19) "2017-05-20 09:40:35" ["post_content_filtered"]=> string(0) "" ["guid"]=> string(50) "http://148.62.4.17/wp-content/uploads/R_911CDR.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) }