Donate
Independent, objective, nonpartisan research

R 409CDR

Authors

R 409CDR

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_409CDR.pdf" ["wpmf_size"]=> string(6) "271489" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(64414) "Sanctions and Time Limits in California’s Welfare Program Caroline Danielson Deborah Reed, Mathematic Policy Research, Inc. with research support from Qian Li and Jay Liao April 2009 The Public Policy Institute of California is dedicated to informing and improving public policy in California through independent, objective, nonpartisan research on major economic, social, and political issues. The institute’s goal is to raise public awareness and to give elected representatives and other decisionmakers a more informed basis for developing policies and programs. The institute’s research focuses on the underlying forces shaping California's future, cutting across a wide range of public policy concerns, including economic development, education, environment and resources, governance, population, public finance, and social and health policy. PPIC is a private, nonprofit organization. It does not take or support positions on any ballot measures or on any local, state, or federal legislation, nor does it endorse, support, or oppose any political parties or candidates for public office. PPIC was established in 1994 with an endowment from William R. Hewlett. Mark Baldassare is President and Chief Executive Officer of PPIC. Thomas C. Sutton is Chair of the Board of Directors. Copyright © 2009 by Public Policy Institute of California All rights reserved San Francisco, CA Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source and the above copyright notice is included. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. Contents Summary Tables Acknowledgments Acronyms 1.  INTRODUCTION 2.  THE WELFARE CASELOAD AND FEDERAL WORK REGULATIONS 3.  FAMILY ECONOMIC CIRCUMSTANCES 4.  SANCTION AND TIME-LIMIT POLICIES 5.  EFFECTS OF SANCTIONS 6.  EFFECTS OF TIME LIMITS 7.  CONCLUSIONS Appendix. Data and Methods References v  iv  v  viii  1  3  6  7  8  10  12  14  15  Tables Table 5.1 – Effects of Work-Related Sanctions......................................................................... 8  Table 6.1 – Effects of Welfare Time Limits ............................................................................. 10  Summary California’s welfare program ‒ the California Work Opportunity and Responsibility to Kids (CalWORKs) program ‒ provides cash assistance to needy families while helping them gain self-sufficiency. Toward this end, most adults receiving CalWORKs are required to work; they may also (with some restrictions) combine work with education or training. If they do not work or do not seek employment and lack a valid exemption, CalWORKs adults risk losing a portion of their welfare grants. Federal rules require the state to have close to half of all adults on welfare working at least part-time, or engaged in a limited set of activities intended to lead to employment. Failure to meet this standard (the so-called “work participation rate”) can result in substantial fiscal penalties for the state. The most recent official statistics indicate that only about one-fifth (22.2%) of CalWORKs families required to comply with the federal standard actually did in 2006. In his 2007, 2008, and 2009 budget proposals, Governor Schwarzenegger suggested major changes to the sanction and time-limit policies in the CalWORKs program, seeking to boost the share of welfare adults who are working. Current state law allows cash assistance to continue to children whose parents have been removed from aid (“sanctioned”) for failing to meet work requirements. Similarly, current law limits adults to a maximum of 60 months of cash assistance, but their children’s eligibility is not time limited. The governor‘s proposals entailed eventually eliminating benefits to the entire family if parents are not working sufficient hours. To-date, the governor’s sanction and time-limit proposals have not been included in an enacted budget. This report examines the likely effects that increasing the severity of sanction and timelimit policies would have on the welfare caseload, the state’s work participation rate, and the economic circumstances of vulnerable families. Comparing California’s caseload and work participation rate to those in other states, we find that the state’s caseload would be substantially lower, and its work participation rate significantly higher, if the state adopted stricter sanction policies for adults who fail to meet work requirements. The effect on child poverty depends on whether a grant-elimination sanction simply removes welfare benefits from children or whether recipients respond to the policy in ways that ultimately increase family resources. Controlling for differing state characteristics, we find that states that implemented grant-elimination sanctions in the 1990s reduced child poverty somewhat, compared to states that imposed the less stringent sanction of reducing grant amounts. Our estimates imply that poverty among children in single-mother families in California would be slightly lower if the state adopted a gradual or immediate grantelimination sanction policy. Our findings for welfare time limits are somewhat different. We find that the state would be unlikely to see a change in the economic circumstances of families headed by single v mothers if grant-elimination time limits were adopted, but we are unable to determine whether the state’s work participation rate and caseload would be higher or lower under more severe time limits. We note three important limitations in our findings. First, our results suggest that moving to grant-elimination sanctions and time limits would not increase child poverty or worsen the economic circumstance of single-mother families on average, but it is possible that some highly vulnerable families would experience worse economic conditions under grantelimination policies. Second, our child poverty measures do not take into account the additional costs families face when moving from welfare to work (e.g., child care costs) although some families obtain subsidies or other assistance to offset those costs. Third, because the major welfare policy changes we examine occurred in the 1990s, our study covers a period of robust economic growth followed by a short recession and a period of slower growth. Using these data, it is difficult to predict what might happen to child poverty under grant-elimination policies during a prolonged recession. vi Acknowledgments In conceptualizing and preparing this report, we consulted numerous welfare policy experts and obtained data and invaluable input from fellow researchers. We would like to acknowledge the contributions of Phil Ansell, Mary Bellamy, Marianne Bitler, Todd Bland, John Cokley, Shantell Denny, Anastasia Dodson, Gene Falk, Mary Farrell, Scott Graves, Mark Greenberg, Jack Hailey, Cynthia Hinckley, Karen Kennedy, Jacob Klerman, Sara McCarthy, Casey McKeever, Charr Lee Metsker, Dean Misczynski, Max Neiman, David Neumark, Don Oellerich, Cathy Senderling, Paul Smilanick, Peverill Squire, Diane Tanaka, William Wascher, and the participants at sessions on TANF work participation at the annual Administration for Children & Families research and evaluation conferences in 2007 and 2008. However, we bear entire responsibility for the ultimate analysis and interpretation of the data used in this report. At PPIC, Gary Bjork and Ellen Hanak gave useful feedback that improved the final product. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. vii Acronyms ACF AFDC CalWORKs CPS DSS DHHS DRA FNS MOE OFA PRWORA SDR SSP TANF TDR USDA Administration for Children & Families, DHHS Aid to Families with Dependent Children California Work Opportunity and Responsibility to Kids Current Population Survey California Department of Social Services U.S. Department of Health and Human Services Deficit Reduction Act of 2005 Food and Nutrition Service Maintenance of Effort Office of Family Assistance, ACF Personal Responsibility and Work Opportunity Reconciliation Act SSP-MOE Data Report (form ACF-204) Separate State Program Temporary Assistance for Needy Families TANF Data Report (form ACF-199) U.S. Department of Agriculture viii 1. Introduction The creation in 1996 of the Temporary Assistance for Needy Families (TANF) program represents a watershed in federal welfare policy. The legislation that created TANF blockgranted a program (Aid to Families with Dependent Children, or AFDC) that had long been uncapped, and it gave states extensive control over program eligibility and on-going requirements. Along with the capped block grant that states could use flexibly, the federal government stipulated penalties if states fail to show that enough welfare adults are meeting federal work requirements. The reauthorization of TANF in 2006 made work requirements considerably more stringent, increasing the likelihood of penalties against the states. Hence, California and other states continue to seek ways of increasing work among adults receiving welfare. In this report, we examine the effects of changing two policies over which states now have control: sanctions (the penalties families face if parents log too few hours in approved welfare-to-work activities) and time limits (restrictions on the length of time adults can receive aid).1 We focus on these policies because Governor Schwarzenegger has suggested modifications to the policies in his 2007, 2008, and 2009 budget proposals (California Department of Finance, 2007; California Department of Finance, 2008; California Department of Finance, 2009). In his most recent proposal, the governor recommended eliminating grants to families if parents who have exceeded their allowed 60 months of assistance are not working for an adequate number of hours.2 Earlier proposals recommended similar changes to the time limit policy and proposed eliminating family grants if sanctioned parents fail to reverse their sanction within a limited number of months. Currently, the state continues assistance to children regardless of whether their parents forfeited assistance due to a sanction or time limit. To date, the governor’s proposals have not been adopted as part of an enacted budget. Some of the impetus for increasing the severity of sanction and time-limit policies in California has been the expectation that doing so would increase the share of welfare adults meeting federal work requirements (the so-called work participation rate). We evaluate whether this would be the case and how policy changes would affect the number of families on welfare.3 1 Because we use other states’ experiences with similar but not identical sanction and time-limit policies to estimate the likely effect of changing these policies in California (and because the governor’s proposals have changed somewhat from year to year), we present the estimates as illustrative of the direction of the changes the state would experience if policymakers established more severe financial penalties for noncompliant or timed-out CalWORKs families. 2 In 2009, the governor also proposed eliminating a family’s welfare grant if adults not working for sufficient hours fail to attend an in-person, biannual self-sufficiency review. This proposal would have applied to sanctioned adults, but also to many other cases in which adults are not aided, but children are (Legislative Analyst’s Office, 2009). 3 Specifically, we examine changes in the number of families containing at least one “work eligible” adult—that is, an adult required to be included in the federal work participation rate calculation. The primary reason we draw attention to effects on the caseload is that a change in the work participation rate provides no information about caseload change: An increase, for instance, could occur in conjunction with a caseload increase, decrease, or no caseload change at all. Caseload declines resulting from modifications a state makes to its eligibility rules (e.g., modifying the time limit) cannot be claimed as a 1 However, we do not focus simply on expected effects on the state’s work participation rate. California’s TANF program—California Work Opportunity and Responsibility to Kids (CalWORKs)—is intended to be a safety net for families facing difficult economic circumstances. A major concern with increasing the severity of sanctions and time limits is that the economic circumstances of vulnerable families would suffer. Thus, we also examine the implications that changing these policies would have on child poverty, maternal employment, and household income in single-mother families.4 We construct statistical models relating this set of outcomes to state TANF policies in all fifty states, holding constant a broad group of economic, demographic, and political factors that may also differ across states. We focus on the period prior to the federal reauthorization of TANF in 2006. In drawing our conclusions, we compare the caseload, work participation rate and family economic circumstances in states that have California-like policies with the same outcomes in states that have more severe policies.5 We derive these outcomes from two sources: the TANF and Separate State Program (SSP) data reports that states collect and submit to the federal government in evidence of their work participation rates and other compliance requirements, and the Annual Social and Economic Supplement to the Current Population Survey (CPS), collected by the U.S. Bureau of the Census.6 We used the TANF and SSP data to determine the size of states’ welfare caseloads and work participation rates; we used the CPS data to measure poverty, income, and employment among vulnerable families. We obtained details on sanction and time-limit policies from the Urban Institute’s Welfare Rules Database. In the following sections, we discuss the shortfall in California’s work participation rate under new federal regulations, characteristics of single-mother families in California, and the likely effects that increasing the severity of sanction and time-limit policies would have on the welfare caseload, the state’s work participation rate, and the economic circumstances of single mothers and their children. We then conclude with a discussion of our findings. The appendix briefly summarizes the data and methods. Detailed descriptions of the data and methods are provided in the technical appendix. credit against the required work participation rate (such “caseload reduction credits” are described in further detail in the following section). 4 See also the syntheses presented in Blank (2002) and Grogger and Karoly (2005). These studies review the literature examining the effects of TANF policies on welfare use, poverty, employment and earnings, and other measures of family and child well-being. 5 In the case of the work participation rate and the caseload, we compare outcomes across states with more and less severe policies. In the case of family economic circumstances, we compare changes before and after implementation of more severe policies with changes in states that implemented California-like policies. 6 The TANF and SSP data reports cover federal fiscal years 1998-2005 (October 1997-September 2005), and we use calendar years 1990-2005 from the CPS. As allowed by federal rules before the 2006 reauthorization of the TANF program, states operated SSPs to move whole segments of the caseload out of the work participation calculation. By 2005 California and thirty-one other states had created such SSPs. These were often created for two-parent families (Office of Family Assistance, 2007a). Such families were typically required to meet state program requirements, which included work requirements. The SSP data reports include the same information about work participation as the TANF data reports. 2 2. The Welfare Caseload and Federal Work Regulations The federal legislation creating TANF in 1996 required that states engage 50 percent of most welfare adults for 20 or 30 hours a week in work or in related activities (job search or education related to employment, work experience, and so on)—or face fiscal penalties. States were also required to engage 90 percent of those families with both parents on welfare in work activities. These are known, respectively, as the “All Families” and the “Two Parent” work participation rate requirements in the federal legislation.1 Prior to the reauthorization of TANF in early 2006, most states easily met these requirements because they were extended credits for caseload reductions that reduced the rate they were required to meet.2 Between 2004 and 2006, California’s All Families rate credit averaged 45.5 points (Office of Family Assistance, 2006; Office of Family Assistance, 2007a; Office of Family Assistance, 2007b).3 The size of the state’s caseload reduction credit implies that the CalWORKs caseload fell dramatically in the late 1990s and early 2000s, and indeed it did: The caseload dropped an average of 10 percentage points each year between 1998 and 2000, and then continued to decline at a slower pace for most of the rest of the period between 2001 and 2005, although it rose by 3 percent in 2004 (California Department of Social Services, n.d.).4 Changes in the legislation governing TANF, effective fiscal year 2007, have made it more difficult for states to meet federal work participation requirements because of changes in the ways states must calculate their credits. In California, the Department of Social Services estimated in 2007 that the caseload reduction credit under the new rules would drop to about 4 percentage points in coming years, implying that California’s effective All Families work 1 Single parents of children under age 6 fulfill the federal requirement by working at least 20 hours each week, while most single parents of older children, and one adult in two-parent families with children of any age, must work at least 30 hours a week. States face limitations on the length of time adults can remain in some work-related activities, as well as on the fraction of adults that can be engaged in some of these activities, and still have the adults’ effort count. Rules governing the calculation of the two-parent rate differ somewhat. For details on current program rules and restrictions on countable work activities, see Department of Health and Human Services (2008). California has its own work rules for CalWORKs adults, which differ somewhat from the federal rules. They are summarized at www.dss.cahwnet.gov/CDSSWEB/PG141.htm. 2 The so-called caseload reduction credit is a one percentage point reduction in the required work participation rate standard for every percentage point drop in the welfare caseload as compared to the baseline year. The original baseline year was 1995 and the current baseline year is 2005. Before the 2006 reauthorization, states could also exclude from the rate calculation those families receiving state-funded SSP assistance as well as families with children, but not parents, receiving welfare because the latter had been sanctioned or had reached a time limit. 3 Because two-parent CalWORKs families were in the state’s SSP, California had no Two Parent rate to meet until October 2006. While we focus in this report on the All Families rate, we include two-parent families in the rate calculation because their work effort must now be counted in both rate calculations. 4 Danielson (2008) describes changes in the education, citizenship, and family characteristics of welfare adults who were required to work in California and in the rest of the nation between 1999 and 2005. 3 participation rate requirement would be about 46 percent.5 It appears that the weak economy may reduce this credit to zero in the immediate future: As of October 2008, the CalWORKs caseload had exceeded its 2005 average (California Department of Social Services, n.d.). The penalty for failing to meet the All Families rate is a reduction of 5 percent in the state’s federal TANF block grant. This penalty can be increased by up to 2 percentage points each year, to a maximum of 21 percent of the state’s federal block grant. The state must make up this penalty out of its own funds and, in addition, must increase its required Maintenance of Effort (MOE) welfare expenditures.6 To put these penalties in perspective, California’s TANF block grant is $3.7 billion for the current and upcoming fiscal years, and the state’s minimum CalWORKs and related program spending is $2.7 billion. The penalties imply that California may be required to make up a $149 million shortfall in its TANF block grant, which could grow over time, and must also increase its minimum required state expenditures by $180 million each year it is out of compliance (Legislative Analyst’s Office, 2007; Parrott et al., 2007).7 A state can avoid penalties (but it must still increase its MOE expenditures) by entering into a compliance plan with the U.S. Department of Health and Human Services and achieving an agreed-upon increase in its work participation rate. In 2005, the final year before the reauthorization legislation passed, California achieved its adjusted All Families rate, but it would not have done so absent its caseload reduction credit. With their caseload reduction credits, 49 states and the District of Columbia met their adjusted standards. Absent these credits, only nine states would have been in compliance; three more were within 5 percentage points of meeting the rate absent the credit (Office of Family Assistance, 2006).8 Thus, most states, including California, face a substantial shortfall under the new federal requirements. 5 This caseload reduction credit is the so-called “natural reduction” credit and is distinct from the “excess MOE” credit that California has considered. The final federal rule published in February 2008 limits flexibility to claim such a credit. See www.acf.hhs.gov/programs/ofa. Both are discussed further in Legislative Analyst’s Office (2008). 6 MOE is the minimum a state must spend of its own funds in order to qualify for its entire TANF block grant. The level is pegged at 75 percent of its fiscal year 1994 AFDC expenditure, or 80 percent if it fails to meet federal requirements. 7 Before penalties are calculated, the state’s block grant is first adjusted downward to reflect amounts transferred to the child care or social services block grants and amounts spent through tribal TANF programs. If the state is in compliance with the All Families rate but not the Two Parent rate, the penalty is reduced to reflect the share of the caseload in two-parent families. In California, this is approximately 8 percent. 8 Although federal fiscal year 2007 was the first year rates were calculated under the new rules, we use the 2005, not the 2006, statistics to represent the status quo ante because the new legislation was enacted in February 2006, less than half way into fiscal year 2006. Indiana, the one state that failed to meet the federal requirement, was within 3 percentage points of meeting its adjusted rate. Two of the nine states that achieved a 50 percent or higher rate had a waiver to work participation rules in effect. In the absence of the waiver, both states’ rates would have been well below 50 percent, and one of the states (Tennessee) would not have met its adjusted rate. 4 However, California’s work participation rate was particularly low, ranking 36th among the fifty states in 2005 (Office of Family Assistance, 2007b).9 The difference does not appear to be attributable to California’s distinctive demographic characteristics: Adjusting the state’s share of immigrants and single mothers, educational attainment, and race/ethnic mix to match that in the rest of the nation pushes the rate even lower.10 In sections 5 and 6, we assess the potential for proposed policy changes to increase the work participation rate by shifting welfare adults in and out of compliance as well as on and off the caseload. After adjusting for demographic, economic, and political differences across states, we examine whether states with stricter sanction and time-limit policies have higher work participation rates and lower caseloads. 9 These rankings were computed under the old work participation rules, implying that they exclude all cases in SSPs (most of which were two-parent cases). It does not take into account waivers to federal rules in effect in two states in 2005. Using those waiver-adjusted work participation rates, California ranked 38th in the nation. California’s official rate for 2005 was 25.9 percent, and its rate for 2006 (the most recent available) was 22.2 percent (Office of Family Assistance, 2007a, 2007b). 10 Our simulation uses the same model as used for the sanction and time limit estimates reported in sections 5 and 6. The number of adults in the denominator of the rate is lower in the simulation, but the simulated number meeting work requirements is even lower. The result is a lower simulated work participation rate. Further details of the variables used in the simulation are provided in the technical appendix, available at http://www.ppic.org/content/pubs/other/409CDR_appendix.pdf. 5 3. Family Economic Circumstances Although raising California’s work participation rate has been of concern to state policymakers, getting adults on welfare to work is not the sole objective of the CalWORKs program. CalWORKs was crafted to assist low-income families in becoming self-sufficient while providing a safety net for children (Zellman et al., 1999). In this section, we review the circumstances of families liable to be affected by welfare policies. We define these families as single women between the ages of 16 and 46 who have less than a four-year college degree and who have children living with them. Most of these families do not receive a welfare check in any particular year, but far more will obtain welfare assistance over their lifetimes. Thus, their actions are likely to be affected by changes in welfare policies, regardless of whether they are current recipients of CalWORKs benefits. For instance, they may work more or less if sanction or time-limit policies are more or less strict, or they may take other action to change the resources available to their families (such as moving in with friends or family or obtaining more education). In contrast, mothers with more resources (those with more education and married mothers) are much less likely to substantially alter their behavior in the wake of a welfare policy change.1 Between 2003 and 2005, the endpoint in our period of analysis, an average of 8.4 percent of families in California were headed by a single mother who lacked a four-year college degree. These families had less than two children (1.7 on average), and close to half (47.6%) had a child under age 6. Slightly over two-fifths of mothers (43.3%) had some college experience, while about three out of ten had a high school diploma or equivalent (29.3%) and about the same share had not completed high school (27.4%). These families had higher incomes in the mid-2000s than they did in the early 1990s before the advent of CalWORKs: In 2003-2005, an average of 34.9 percent had incomes below the federal poverty line, down from 44.1 percent in 1990-1992.2 Single mothers were also more likely to be working: In 2003-2005, 72.9 percent had worked at some point during the reporting year, up from 61.2 percent in 1990-1992.3 Earnings among the group of single mothers who had worked at all during the year averaged $23,132 in 2003-2005, compared to $15,436 in 1990-1992.4 Clearly, at least by some measures these vulnerable families were better off in the mid-2000s than they were in the early 1990s. Although welfare reform policies may have contributed to these improvements, it is not clear whether the specific sanction and time-limit policies that California adopted are important for family well-being. We explore this issue in the sections that follow. 1 Although some have argued welfare policies can affect the incidence of single motherhood by providing an alternative to marriage, the existing research finds little or no effect of welfare policies on marriage or fertility among single women. However, research is still ongoing in this area (Grogger and Karoly, 2005). While married couples are eligible for welfare, the vast majority of welfare families (approximately 95 percent nationwide) include only one parent in the assistance unit. 2 Examining poverty from the child’s point of view, 40.9 percent of children living with single mothers who did not have a college degree lived in poverty in 2003-2005. 3 CPS respondents were also asked whether they worked in the previous week. Using this metric, an average of 62.3 percent were working in 2003-2005 while 48.4 percent were working in 1990-1992. 4 Dollar values are adjusted for inflation to 2005 levels. 6 4. Sanction and Time-Limit Policies Sanctions are the reduction or elimination of a family’s aid payment if parents fail to meet work-related requirements, while time limits restrict the length of time adults can receive welfare assistance. When state lawmakers fashioned the CalWORKs program in 1996 and 1997, they established policies intended to further the state’s “work-first” approach for adults on welfare while preserving a safety net for children (Zellman et al., 1999).1 Consequently, CalWORKs sanction and time-limit policies focus penalties on adults. California continued the sanction policy established under AFDC, the federal cash assistance program preceding TANF. That policy calls for a reduction in the family’s aid payment equal to the sanctioned adult’s share for as long as that adult remains out of compliance with requirements.2 Most states established more severe penalties.3 Likewise, California is one of four states that have not applied the federal 60-month time limit on aid to families on welfare. In California, adults have a 60-month time limit on welfare benefits, but children can continue to receive welfare benefits after parents have reached time limits, as long as they remain otherwise eligible.4 California uses state funds to continue assistance to these children. The governor’s 2008 budget proposal included a recommendation to eliminate family grants over a period of 12 months if sanctioned parents fail to reverse their sanction (the 2007 proposal suggested a grace period of 3 months). Likewise, the 2007, 2008, and 2009 budget proposals recommended eliminating family grants once a parent has reached the time limit unless that parent meets federal work requirements (California Department of Finance, 2007; California Department of Finance, 2008; California Department of Finance, 2009).5 1 We emphasize that the CalWORKs program is far more than its sanction and time limit policies. CalWORKs incorporates a range of policies designed to promote work among adult welfare recipients, ranging from assistance with job search to subsidized child care for those who become employed. 2 Adult welfare recipients were subject to so-called “durational sanctions” until 2007, which meant that those who had been sanctioned more than once had to spend a minimum number of months in sanction status before their full grant would be restored. Pursuant to AB 1808 (2006), this is no longer the case. 3 For a more detailed typology of the sanction and time-limit policies across states, see Table A.2 of the technical appendix, available at http://www.ppic.org/content/pubs/other/409CDR_appendix.pdf. 4 Children continue to be eligible if their total family income remains below an income threshold, which varies by family size. The threshold is lower after adult(s) reach their time limit. 5 The intent of these changes, at least in part, is to increase California’s work participation rate (California Department of Finance, 2007, 2008). In the wake of the federal reauthorization of TANF in 2006, California must include in the work participation rate calculation sanctioned adults whose children continue receiving assistance. Sanctioned adults who have been sanctioned for three months or less in the previous twelve months can be excluded. This means that roughly half of California’s sanctioned caseload can be excluded from the calculation (Danielson, 2006). Federal rules in the wake of TANF reauthorization require California to count time-limited parents whose children continue to receive aid when calculating the work participation rate. That is, these parents will be added to the denominator of the calculation and may also be included in the numerator (if they are working an adequate number of hours). 7 5. Effects of Sanctions How would the welfare caseload, the work participation rate, child poverty, and single mothers’ employment differ under a more severe sanction policy? In this section, we examine the implications of moving from California’s current grant-reduction sanction to a policy of gradual or immediate grant elimination.26 We use statistical methods to account for a range of other conditions in states—strength of the economy, political climate, and demographic characteristics—that, apart from policies, may also cause states’ work participation rates to differ. In sum, we find that increasing the financial penalties associated with noncompliance with work rules would reduce California’s welfare caseload, substantially increase its work participation rate, and slightly reduce poverty among children living with single mothers (Table 5.1). Table 5.1. Effects of work-related sanctions Gradual grant elimination (%) Immediate grant elimination (%) Caseload Work participation rate Child poverty Mother’s employment Mother’s earnings Household income -27+ 50* -4* 1 6 0.03 -52* 94* -5* 4 3 4* SOURCES: Authors’ calculations from TANF and SSP-MOE data reports (1998-2005), the Annual Social and Economic Supplement to the CPS (1990-2005), and the variables described in technical appendix Table A.3. NOTES: Significance is in comparison to grant-reduction sanction. Table entries derived from model estimates in technical appendix Table A.4, column 1, and Table A.8, columns 1 and 2. Caseload refers to the portion of the caseload required to be counted in the federal rate calculation. Child poverty, mothers’ employment, and mothers’ earnings are among families headed by single mothers ages 16-46 who have less than a four-year college degree. For non-linear models (poverty and employment), we simulate the effects at the means of explanatory variables for California in 2003-2005. Household income is cash income from all sources for all residents of housing units that contain such a family. + Significantly different at the 10% level or lower. * Significantly different at the 5% level or lower. 26 The gradual grant elimination category we use includes several variants of this type of sanction. Some states eliminated the grant at the instance of a second or higher sanction. Some eliminated the grant after a sanction lasting more than a certain number of months. This number of months ranged from 1 to 12, with a median of 4 months. Several states imposed a grant reduction sanction if the family had received aid for fewer than 24 months, and a grant elimination sanction if it had received aid for more months. 8 Adjusting for other differences, states with gradual or immediate grant-elimination sanctions have adult-headed caseloads that are lower by about a quarter to a half as compared to states with grant-reduction sanctions. 27 The work participation rate is sharply higher, by 50 percent in the case of a gradual sanction and by over 90 percent in the case of an immediate sanction. In other words, if the state had had a gradual sanction in place, California’s work participation rate would have been 34 percent in 2005 rather than 23 percent. Under an immediate sanction, it would have been 44 percent. These differences are large; however, a portion of the change (about half in the case of a gradual sanction) is due to a federal rule change in 2006. This rule change requires more of California’s sanctioned adults to be counted in the state’s work participation rate, thus lowering the rate even though the state has not altered its sanction policy. Our estimates suggest that moving to some form of a grant-elimination sanction will result in a smaller caseload and a higher work participation rate. A serious concern with such a change is that it would be harmful to low-income families. If the policy simply removes all benefits from families who fail to meet work requirements, or discourages such families from ever applying for benefits, it could increase poverty among at-risk families. On the other hand, it is possible that a grant elimination sanction would encourage work, either because sanctioned adults would be able to replace lost assistance with work or because the more stringent sanction would motivate more parents on assistance to comply with work requirements while on welfare, or both.28 We explore these issues by investigating the effect of more severe sanction policies on the economic circumstances of single-mother families.29 We find that, relative to California-like policies, grant-elimination sanctions reduce child poverty by a small amount. About 41 percent of California children in single-mother families were poor in 2003-2005. With grant elimination sanctions, the percentage would have been about 39 or 40 percent (a reduction of approximately 4 to 5 percent, as shown in Table 5.1). Single-mother employment and earnings are higher, but the effects are not statistically significant.30 Implementation of an immediate (but not gradual) grant-elimination sanction increases income in the households in which these families live by a small amount (4 percent). Overall, the results suggest that grant-elimination sanctions do not negatively affect the cash resources of single-mother families.31 27 Previous studies examining the effects of welfare sanctions on caseload find mixed evidence of a negative effect of gradual full-family sanctions and a negative effect of immediate full-family sanctions (Danielson and Klerman, 2008; Grogger and Karoly, 2005). These studies were not able to exclude childonly cases, which are little affected by work-related TANF policies. 28 To be consistent with our finding that the smaller caseload is smaller when more severe sanctions are in place, in the latter scenario it must be the case that some on welfare who work more leave assistance. 29 If we restrict the sample to mothers with less than a high school degree, a larger fraction of whom received income from public assistance, estimates remain substantially the same, but sample sizes are smaller and coefficients are estimated less precisely. 30 The table reports results using any maternal work during the past year as the outcome. We found similar results when we examined any work in the last week. Grant elimination sanctions did not have a statistically significant effect on non-work factors associated with poverty: number of children per mother, number of adults in the household, and the probability of a single woman being a mother. 31 The previous research does not specifically examine the effects of more stringent sanctions on singlemother employment and earnings. One study finds that greater financial penalties associated with a sanction reduce deep poverty among single mothers by a small amount but do not have an effect on overall single-mother poverty (McKernan and Ratcliffe, 2006). 9 6. Effects of Time Limits How are the set of outcomes we consider—welfare caseload, work participation rate, child poverty, and single mothers’ employment—different if a grant-elimination time limit is in place rather than a grant-reduction policy? In this section, we examine the implications of moving from California’s current policy, a grant-reduction time limit, to two variants of a grantelimination time limit. The first policy is simply one of halting assistance to families in which an adult has reached the time limit. The second extends additional months of assistance to the family as long as the time-limited adult is working sufficient hours to meet the requirement.32 In sum, due to the data limitations discussed below, we are unable to estimate the effect of changing California’s time-limit policy on the work participation rate and the caseload of those required to be counted in that rate. However, we find that increasing the financial penalty associated with reaching a time limit is not likely to change the economic circumstances of vulnerable families (Table 6.1). Table 6.1. Effects of welfare time limits Grant elimination (%) Grant elimination with extension if meeting work requirement (%) Child poverty Mothers’ employment Mother’s earnings Household income 3 -1 1 -2 1 0 5 2 SOURCES: Authors’ calculations from the Annual Social and Economic Supplement to the CPS (19902005) and the variables described in technical appendix Table A.3. NOTE: Significance is in comparison to grant-reduction time limit (reached). Table entries derived from model estimates in technical appendix Table A.8, columns 1 and 2. Child poverty, mothers’ employment, and mothers’ earnings are among families headed by single mothers ages 16-46 who have less than a four-year college degree. For non-linear models (poverty and employment), we simulate the effects at the means of explanatory variables for California in 2003-2005. Household income is cash income from all sources for all residents of housing units that contain such a family. + Significantly different at the 10% level or lower. * Significantly different at the 5% level or lower. An important rule change in 2006 requires states that continue children on welfare once parents have reached a time limit to include those timed-out parents in their work participation rate calculation. Before 2006, states with grant-reduction policies like California’s excluded 32 Adults may qualify for a time limit exemption or extension for other reasons. Table A.3 of the Technical Appendix lists the reasons that we include in the set of TANF program characteristics that we hold constant in our regression models. 10 time-limited adults from their rate calculations. This rule change implies that we lack a reasonable baseline policy to estimate differences across states with more and less severe time limits using data from before the federal reauthorization of TANF. Thus, we are unable to estimate whether California’s work participation rate and caseload would be higher or lower under a grant-elimination time limit. It is important to note that time limits set into motion potentially complex behavioral changes among welfare recipients and potential recipients, and it is possible that a more severe time limit might either reduce or increase the state’s work participation rate.33 For example, a grant-elimination time limit might increase the incentive of welfare adults to work. However, it is also possible that families considering whether to apply for CalWORKs might defer their applications until they face more extreme hardship in order to avoid exhausting their months of eligibility. Families facing relatively greater hardships may be less able to find and keep employment. Such a change in the types of families on welfare could lower the state’s work participation rate. As with grant-elimination sanctions, a major concern with grant-elimination time limits is that they would harm low-income families. Clearly, the loss of benefits to children without any compensating actions on the part of adults would worsen their economic situation. At the same time, it is conceivable that a full-family time limit might change work incentives or other incentives enough to improve the family’s economic situation. Examining the economic circumstances of single-mother families (in which the mother does not have a college degree), we find that relative to states in which families had reached grant-reduction time limits, states in which families had reached grant-elimination time limits had somewhat higher child poverty rates, but the difference is not statistically significant.34 Grant-elimination time limits, with or without an extension for those complying with work rules, have mixed effects on single mothers’ employment and earnings and on household income, but the differences are never statistically significant.35 In other words, based on the experience of other states (holding constant many other factors), we do not find robust evidence that moving from a grantreduction to a grant-elimination time limit would substantially increase poverty among children in single-mother families.36 33 We note that we estimate the work participation rate to be lower in states with grant-elimination time limits as compared to states with no time limit. We also estimate the caseload to be smaller. However, the differences are not statistically significant. Previous research has found that time limits (as compared to no time limit) reduce single mothers’ use of welfare and the welfare caseload (Danielson and Klerman, 2008; Grogger, 2001; Grogger, 2003; Mazzolari, 2007). 34 Our model controls, among other things, for the length of the time limit and the set of time-limit exemptions and extensions that a state permits. Table A.3 in the technical appendix (available at http://www.ppic.org/content/pubs/other/409CDR_appendix.pdf) lists the full set of control variables. 35 The table shows results for any maternal work during the past year. We found similar results when we examined any work in the past week. 36 Note that California does impose a time limit, so that differences described in the text are in comparison to that time limit. Analyzing data that end in the early 2000s, the previous literature finds mixed evidence that a grant-elimination time limit in comparison to no time limit increases employment among single mothers but no evidence that a grant-elimination time limit alters income, the number of weeks worked, or annual earnings. See Grogger and Karoly (2005) for a summary. Making that same comparison between no time limit and a grant-elimination time limit, our estimates imply no statistically significant change in income, employment, or poverty. 11 7. Conclusions Increasing the severity of CalWORKs sanction and time limit policies has been considered in three successive budget cycles. In this report, we examine the likely effects of modifying these policies. We find that states with stricter sanction policies have substantially higher work participation rates and lower caseloads than states with a California-like sanction policy. Neither of the sanction policies that the governor proposed in 2007 and 2008 would likely make up the expected shortfall in California’s work participation rate; however, the increase in the participation rate in the wake of either one could be dramatic. One important concern is that adopting grant-elimination sanctions will worsen the economic circumstances of children in vulnerable families. We find that implementing a stricter sanction policy actually lowers poverty rates slightly among children living in single-mother families. The evidence we muster indicates states that adopted grant-elimination time limits did not see increased child poverty as a result; we also do not find evidence that grant-elimination time limits increased single mothers’ employment. There are three important limitations of our findings. First, our results suggest that moving to a grant-elimination sanction or time-limit would not increase child poverty or worsen the economic circumstance of single-mother families on average, but it is possible that some highly vulnerable families would experience worse economic conditions under grantelimination policies.37 Such families may be discouraged from applying for welfare, or they may lose benefits under stricter policies and not be able to increase earnings or other sources of income to make up the loss. Second, the child poverty measure does not take into account the additional costs families face when moving from welfare to work (e.g., less parental time with children and work-related expenses). Incorporating these costs could reduce the measured benefits of policies that promote work.38 Third, because the welfare reform period began in the 1990s, our study covers a period of major economic growth followed by a short recession and a period of slower growth.39 Using these data, it is difficult to predict what will happen to child poverty under grant-elimination policies during a prolonged recession in which labor market opportunities are limited.40 37 Bitler, Gelbach, and Hoynes (2006) show that the impact of welfare reform policies varied substantially across families. 38 Families receiving CalWORKs assistance, those that have recently left CalWORKS, and some other lowincome families, are eligible for services that help to offset the additional expenses of work. The most important of these is subsidized child care. 39 During the 1990s, the robust economy and strong labor market were important factors in encouraging single mothers to work, although policies also played an important role (Blank, 2002). 40 Herbst (2008) provides preliminary evidence that some TANF policies have larger effects on work among single mothers when the economy is stronger. The time period examined in this research is similar to the time period we use, implying that it also cannot answer the question of the role of policy during a sustained downturn. In theory, a state’s work participation rate is less sensitive than low-wage employment to economic conditions because welfare adults can count a number of different activities 12 Our findings are at odds with the belief that the economic conditions of children are substantially improved by maintaining their benefits when parents are sanctioned for not working or for reaching the time limit. However, they are similar to the findings of studies examining the introduction of TANF and earlier TANF-like state reform efforts. The 1996 welfare overhaul introduced stronger work requirements and enforced them with financial penalties levied on those who did not meet the requirements. At the time, many argued that the reform would increase poverty among vulnerable children. Drawing on multiple studies, the general consensus is that the 1996 reform did not increase child poverty. Indeed, several studies conclude that the introduction of TANF helped reduce child poverty by encouraging maternal employment.41 Clearly, full-family sanctions and time limits remove welfare benefits from some children. But their economic circumstances will depend on how their parents respond to these incentives to work. toward their required hours. At the same time, employment is the most common activity that welfare adults pursue (Danielson, 2006), and it is also a route that is less expensive for states than supporting subsidized employment, on-the-job training, and related efforts. 41 For summaries of the national studies, see Grogger and Karoly (2005), Blank (2002), and Meyer and Holtz-Eakin (2001). See also Sawhill and Haskins (2007), O’Neill and Korenman (2004), and Haskins and Primus (2002). McKernan and Ratcliffe (2006) find no overall effect of welfare reform on child poverty and mixed effects for specific welfare reform policies. For evidence of welfare program effects on other child outcomes, see Gennetian et al., (2002) and Morris et al. (2005). 13 Appendix: Data and Methods This appendix provides a brief description of the data and methods used in this study. A detailed technical appendix is available on the PPIC website http://www.ppic.org/content/pubs/other/409CDR_appendix.pdf). Here, we first describe our approach to analyzing the work participation rate and welfare caseload, and then our approach to arrive at estimates of changes in employment, poverty, and income among singlemother families. To measure the work participation rate and the number of cases containing an adult required to be counted in the rate calculation, we use official statistics from the Office of Family Assistance reports, 1998-2005(OFA, n.d.). We construct the participation rate from OFA information on the size of the TANF and SSP caseloads meeting federal work participation requirements and the size of the caseloads not meeting these requirements. We use the natural log of the work participation rate and the natural log of the caseload divided by the population of women ages 16-46 in the state and year. We divide by the population statistic in order to adjust for that source of variation in the size of welfare caseload: State populations vary widely. We obtained the denominator for the caseload outcome from U.S. Bureau of the Census population estimates. We estimate the effect of state TANF policies using the variation in policies across states. We control for national changes across years, state-level measures of economic conditions, demographic characteristics, political variables, and a detailed set of welfare policies. For family economic conditions, we use the Annual Social and Economic Supplement of the Current Population Survey. These data, collected by the U.S. Census Bureau, are used for official poverty statistics. We focus our analysis on the main population thought to be affected by welfare policy: single mothers ages 16 to 46 who do not have a bachelor’s degree. We investigate four main outcomes: poverty, maternal employment, annual earnings, and household income. We estimate the effect of state TANF policies using the variation in policies across states and over time. We use a “difference-in-difference-in-differences” approach. We control for state and year fixed effects, individual-level demographic characteristics (such as age, education, age of youngest child, and race/ethnicity) and state-level factors (policies, economic conditions, and political variables). We identify the effect on family economic conditions of the change in state TANF policies using data from 1990-1996 and 1999-2005 (we exclude 1997 and 1998 because during those years many states had only partially implemented their TANF programs). We use single women in the same age range and with the same level of education as a “control group” ‒ we assume that any effect of the TANF policy variables on this group is spurious and that the additional effect on single mothers is causal. In robustness checks, we also use married mothers in the same age range and with the same education as a control group. Using this control group, we find policy effects that are similar in magnitude although they are not statistically significant. 14 References Assistant Secretary for Planning and Evaluation, “Temporary Assistance for Needy Families (TANF): Data and Documentation,” available at aspe.hhs.gov/ftp/hsp/tanf-data/index.htm. Berry, William D., Evan J. Ringquist, Richard C. Fording, and Russell L. Hanson, “Measuring Citizen and Government Ideology in the American States, 1960-93,” American Journal of Political Science, 1998, 42, pp. 327-48. Bitler, Marianne, Jonah Gelbach, and Hilary Hoynes, “What Mean Impacts Miss: Distributional Effects of Welfare Reform Experiments,” American Economic Review, 2006, 4 (September), pp. 988-1012. Blank, Rebecca M., “Evaluating Welfare Reform in the United States,” Journal of Economic Literature, 2002, Vol. 40, No. 4, pp. 1105-66. California Department of Finance, Governor’s Budget Summary 2007-08, January 2007, available at 200708.archives.ebudget.ca.gov/pdf/BudgetSummary/FullBudgetSummary.pdf. California Department of Finance, Governor’s Budget Summary 2008-09, January 2008, available at 200809.archives.ebudget.ca.gov/pdf/BudgetSummary/FullBudgetSummary.pdf. California Department of Finance, Governor’s Budget Summary 2009-10, January 2009, available at www.ebudget.ca.gov/pdf/BudgetSummary/FullBudgetSummary.pdf. California Department of Social Services, CalWORKs Benefits Model, 2003, available at www.dss.cahwnet.gov/research/res/pdf/BenefitModel.pdf. California Department of Social Services, “CA 237 CW - CalWORKs Cash Grant Caseload Movement Report,” n.d., available at www.dss.cahwnet.gov/research/res/. Citro, Constance F., and Robert T. Michael, eds, Measuring Poverty: A New Approach, Washington, D.C., National Academy Press, 1995. Crouse, Gil, “State Implementation of Major Changes to Welfare Policies, 1992-1998,“ Assistant Secretary for Planning and Evaluation, DHSS, Washington, DC, 1999, available at aspe.hhs.gov/HSP/Waiver-Policies99/policy_CEA.htm. Danielson, Caroline, and Jacob Alex Klerman, “Did Welfare Reform Cause the Caseload to Decline?” Social Service Review, 2008, Vol. 82, No. 4, pp. 703-730. Danielson, Caroline, Patterns of Work Participation in CalWORKs, Occasional Paper, Public Policy Institute of California, San Francisco, California, December 2006. Danielson, Caroline, The CalWORKs Caseload in 1999 and 2005, Working Paper, Public Policy Institute of California, San Francisco, California, September 2008. Department of Health and Human Services, “Reauthorization of the Temporary Assistance for Needy Families Program; Interim Final Rule,” Federal Register, Vol. 71, no. 125, June 29, 2006. 15 Department of Health and Human Services, “Estimated Additional Work Participation Needed Under the Deficit Reduction Act of 2005,” 2007, available at peerta.acf.hhs.gov/pdf/Public_WPR.pdf. Department of Health and Human Services, “Reauthorization of the Temporary Assistance for Needy Families Program; Final Rule,” Federal Register, Vol. 73, No. 24, February 5, 2008, available at www.acf.hhs.gov/programs/ofa/lawreg/finalrule/tanf_final_rule.pdf. Gennetian, Lisa A., Greg J. Duncan, Virginia W. Knox, Wanda G. Vargas, Elizabeth Clark-Kauffman, Andrew S. London, How Welfare and Work Policies for Parents Affect Adolescents: A Synthesis of Research, New York: MDRC, May 2002, available at www.mdrc.org. Grogger, Jeffrey, “The Effects of Time Limits, the EITC, and Other Policy Changes on Welfare Use, Work, and Income Among Female-Headed Families,” Review of Economics and Statistics, Vol. 85, No. 2 May 2003,pp. 394-408. Grogger, Jeffrey, “Time Limits and Welfare Use,” Journal of Human Resources, Vol. 39, No. 2 Spring 2004, pp. 405-424. Grogger, Jeffrey, and Lynn A. Karoly, Welfare Reform: Effects of a Decade of Change, Cambridge, Massachusetts: Harvard University Press, 2005. Haskins, Ron, and Wendell Primus, “Welfare Reform and Poverty,” Brookings Institution, July 2001, available at http://www.brookings.edu/papers/2001/07poverty_haskins.aspx. Herbst, Chris M., “Do Social Policy Reforms Have Different Impacts on Employment and Welfare Use as Economic Conditions Change?” Journal of Policy Analysis and Management, Vol. 27, No. 4, 2008, pp.867-894. Klarner, Carl, “The Measurement of the Partisan Balance of State Government,” State Politics and Policy Quarterly, Vol. 3, No. 3, 2003. Klerman, Jacob Alex, Jeanne S. Ringel, and Beth Roth, “Underreporting of Medicaid and Welfare in the Current Population Survey,” Santa Monica, California, RAND WR-169-3, March 2005, available at www.rand.org/pubs/working_papers/2005/RAND_WR1693.pdf. Legislative Analyst’s Office, “Analysis of the 2007-08 Budget Bill,” Legislative Analyst’s Office, Sacramento, California, February 2007, available at www.lao.ca.gov. Legislative Analyst’s Office, “Analysis of the 2008-09 Budget Bill,” Legislative Analyst’s Office, Sacramento, California, February 2008, available at www.lao.ca.gov. Mazzolari, Francesca, “Welfare Use when Approaching the Time Limit.” Journal of Human Resources, Vol. 42, No. 3, 2007, pp. 596–618. McKernan, Signe-Mary, and Caroline Ratcliffe, The Effect of Specific Welfare Policies on Poverty, The Urban Institute (mimeo.), Washington D.C., April 2006. 16 Meyer, Bruce D., and Douglas Holtz-Eakin, eds., Making Work Pay: The Earned Income Tax Credit and Its Impact on America’s Families, New York: Russell Sage Foundation, 2001. Morris, Pamela A., Lisa A. Gennetian, and Greg J. Duncan, Effects of Welfare and Employment Policies on Young Children: New Findings on Policy Experiments Conducted in the Early 1990s, New York: MDRC, April 2005, available at www.mdrc.org. Neumark, David, and William Wascher. “Minimum Wages, the Earned Income Tax Credit, and Employment: Evidence from the Post-Welfare Reform Era,” Cambridge, Massachusetts: NBER Working Paper No. 12915, February 2007. Office of Family Assistance, “TANF and SSP-MOE Data Reports,” n.d., available at www.acf.hhs.gov/programs/ofa/tanfrpts/index.htm. Office of Family Assistance, 2006, “Temporary Assistance for Needy Families Program Information Memorandum No. TANF-ACF-IM-2006-1,” January 17, 2006, available at www.acf.hhs.gov/programs/ofa/particip/2004/IM04RATE.htm. Office of Family Assistance, 2007a, “Temporary Assistance for Needy Families Program Information Memorandum No. TANF-ACF-IM-2007-5,” August 16, 2007, available at www.acf.hhs.gov/programs/ofa/particip/2006/IM06RATE_REV_CN20070814.htm. Office of Family Assistance, 2007b, “Temporary Assistance for Needy Families Program Information Memorandum No. TANF-ACF-IM-2007-4,” August 1, 2007, available at www.acf.hhs.gov/programs/ofa/particip/2005/IM05RATE.htm. O’Neill, June, and Sanders Korenman, “Child Poverty and Welfare Reform: Stay the Course,” Civic Report No. 44, May 2004, Manhattan Institute for Policy Research, available at www.manhattan-institute.org/html/cr_44.htm. Parrott, Sharon, Liz Schott, Eileen Sweeney, Allegra Baider, Evelyn Ganzglass, Mark Greenberg, Elizabeth Lower-Basch, Elisa Minoff, and Vicki Turetsky, Implementing the TANF Changes in the Deficit Reduction Act: ‘Win-Win’ Solutions for Families and States, Second Edition, Center on Budget and Policy Priorities (CBPP) and Center for Law and Social Policy (CLASP), February 2007, available at www.clasp.org/publications/tanfguide.pdf. Sawhill, Isabel, and Ron Haskins, “Ending Poverty in America: Using Carrots and Sticks,” The American Prospect, May 2007, available at http://www.brookings.edu/articles/2007/05poverty_haskins.aspx. Urban Institute, Welfare Rules Database, Washington, DC., n.d., available at www.urban.org/Content/Research/NewFederalism/Data/ANFData.htm. Zellman, Gail L., Jacob Alex Klerman, Elaine Reardon, and Paul Steinberg,. Welfare Reform in California: State and County Implementation of CalWORKs in the First Year, Santa Monica, California: RAND Corporation, 1999. 17 PUBLIC POLICY INSTITUTE OF CALIFORNIA Board of Directors Walter B. Hewlett, Chair Director Center for Computer Assisted Research in the Humanities Mark Baldassare President and Chief Executive Officer Public Policy Institute of California Ruben Barrales President and Chief Executive Officer San Diego Regional Chamber of Commerce John E. Bryson Retired Chairman and CEO Edison International Gary K. Hart Former State Senator and Secretary of Education State of California Donna Lucas Chief Executive Officer Lucas Public Affairs Ki Suh Park Design and Managing Partner Gruen Associates Constance L. Rice Co-Director The Advancement Project Thomas C. Sutton Retired Chairman and Chief Executive Officer Pacific Life Insurance Company Raymond L. Watson Vice Chairman of the Board Emeritus The Irvine Company Carol Whiteside President Emeritus Great Valley Center PUBLIC POLICY INSTITUTE OF CALIFORNIA 500 Washington Street, Suite 600 San Francisco, California 94111 phone: 415.291.4400 fax: 415.291.4401 PPIC SACRAMENTO CENTER Senator Office Building 1121 L Street, Suite 801 Sacramento, California 95814 phone: 916.440.1120 fax: 916.440.1121 www.ppic.org" } ["___content":protected]=> string(102) "

R 409CDR

" ["_permalink":protected]=> string(99) "https://www.ppic.org/publication/sanctions-and-time-limits-in-californias-welfare-program/r_409cdr/" ["_next":protected]=> array(0) { } ["_prev":protected]=> array(0) { } ["_css_class":protected]=> NULL ["id"]=> int(8567) ["ID"]=> int(8567) ["post_author"]=> string(1) "1" ["post_content"]=> string(0) "" ["post_date"]=> string(19) "2017-05-20 02:38:43" ["post_excerpt"]=> string(0) "" ["post_parent"]=> int(3792) ["post_status"]=> string(7) "inherit" ["post_title"]=> string(8) "R 409CDR" ["post_type"]=> string(10) "attachment" ["slug"]=> string(8) "r_409cdr" ["__type":protected]=> NULL ["_wp_attached_file"]=> string(12) "R_409CDR.pdf" ["wpmf_size"]=> string(6) "271489" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(64414) "Sanctions and Time Limits in California’s Welfare Program Caroline Danielson Deborah Reed, Mathematic Policy Research, Inc. with research support from Qian Li and Jay Liao April 2009 The Public Policy Institute of California is dedicated to informing and improving public policy in California through independent, objective, nonpartisan research on major economic, social, and political issues. The institute’s goal is to raise public awareness and to give elected representatives and other decisionmakers a more informed basis for developing policies and programs. The institute’s research focuses on the underlying forces shaping California's future, cutting across a wide range of public policy concerns, including economic development, education, environment and resources, governance, population, public finance, and social and health policy. PPIC is a private, nonprofit organization. It does not take or support positions on any ballot measures or on any local, state, or federal legislation, nor does it endorse, support, or oppose any political parties or candidates for public office. PPIC was established in 1994 with an endowment from William R. Hewlett. Mark Baldassare is President and Chief Executive Officer of PPIC. Thomas C. Sutton is Chair of the Board of Directors. Copyright © 2009 by Public Policy Institute of California All rights reserved San Francisco, CA Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source and the above copyright notice is included. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. Contents Summary Tables Acknowledgments Acronyms 1.  INTRODUCTION 2.  THE WELFARE CASELOAD AND FEDERAL WORK REGULATIONS 3.  FAMILY ECONOMIC CIRCUMSTANCES 4.  SANCTION AND TIME-LIMIT POLICIES 5.  EFFECTS OF SANCTIONS 6.  EFFECTS OF TIME LIMITS 7.  CONCLUSIONS Appendix. Data and Methods References v  iv  v  viii  1  3  6  7  8  10  12  14  15  Tables Table 5.1 – Effects of Work-Related Sanctions......................................................................... 8  Table 6.1 – Effects of Welfare Time Limits ............................................................................. 10  Summary California’s welfare program ‒ the California Work Opportunity and Responsibility to Kids (CalWORKs) program ‒ provides cash assistance to needy families while helping them gain self-sufficiency. Toward this end, most adults receiving CalWORKs are required to work; they may also (with some restrictions) combine work with education or training. If they do not work or do not seek employment and lack a valid exemption, CalWORKs adults risk losing a portion of their welfare grants. Federal rules require the state to have close to half of all adults on welfare working at least part-time, or engaged in a limited set of activities intended to lead to employment. Failure to meet this standard (the so-called “work participation rate”) can result in substantial fiscal penalties for the state. The most recent official statistics indicate that only about one-fifth (22.2%) of CalWORKs families required to comply with the federal standard actually did in 2006. In his 2007, 2008, and 2009 budget proposals, Governor Schwarzenegger suggested major changes to the sanction and time-limit policies in the CalWORKs program, seeking to boost the share of welfare adults who are working. Current state law allows cash assistance to continue to children whose parents have been removed from aid (“sanctioned”) for failing to meet work requirements. Similarly, current law limits adults to a maximum of 60 months of cash assistance, but their children’s eligibility is not time limited. The governor‘s proposals entailed eventually eliminating benefits to the entire family if parents are not working sufficient hours. To-date, the governor’s sanction and time-limit proposals have not been included in an enacted budget. This report examines the likely effects that increasing the severity of sanction and timelimit policies would have on the welfare caseload, the state’s work participation rate, and the economic circumstances of vulnerable families. Comparing California’s caseload and work participation rate to those in other states, we find that the state’s caseload would be substantially lower, and its work participation rate significantly higher, if the state adopted stricter sanction policies for adults who fail to meet work requirements. The effect on child poverty depends on whether a grant-elimination sanction simply removes welfare benefits from children or whether recipients respond to the policy in ways that ultimately increase family resources. Controlling for differing state characteristics, we find that states that implemented grant-elimination sanctions in the 1990s reduced child poverty somewhat, compared to states that imposed the less stringent sanction of reducing grant amounts. Our estimates imply that poverty among children in single-mother families in California would be slightly lower if the state adopted a gradual or immediate grantelimination sanction policy. Our findings for welfare time limits are somewhat different. We find that the state would be unlikely to see a change in the economic circumstances of families headed by single v mothers if grant-elimination time limits were adopted, but we are unable to determine whether the state’s work participation rate and caseload would be higher or lower under more severe time limits. We note three important limitations in our findings. First, our results suggest that moving to grant-elimination sanctions and time limits would not increase child poverty or worsen the economic circumstance of single-mother families on average, but it is possible that some highly vulnerable families would experience worse economic conditions under grantelimination policies. Second, our child poverty measures do not take into account the additional costs families face when moving from welfare to work (e.g., child care costs) although some families obtain subsidies or other assistance to offset those costs. Third, because the major welfare policy changes we examine occurred in the 1990s, our study covers a period of robust economic growth followed by a short recession and a period of slower growth. Using these data, it is difficult to predict what might happen to child poverty under grant-elimination policies during a prolonged recession. vi Acknowledgments In conceptualizing and preparing this report, we consulted numerous welfare policy experts and obtained data and invaluable input from fellow researchers. We would like to acknowledge the contributions of Phil Ansell, Mary Bellamy, Marianne Bitler, Todd Bland, John Cokley, Shantell Denny, Anastasia Dodson, Gene Falk, Mary Farrell, Scott Graves, Mark Greenberg, Jack Hailey, Cynthia Hinckley, Karen Kennedy, Jacob Klerman, Sara McCarthy, Casey McKeever, Charr Lee Metsker, Dean Misczynski, Max Neiman, David Neumark, Don Oellerich, Cathy Senderling, Paul Smilanick, Peverill Squire, Diane Tanaka, William Wascher, and the participants at sessions on TANF work participation at the annual Administration for Children & Families research and evaluation conferences in 2007 and 2008. However, we bear entire responsibility for the ultimate analysis and interpretation of the data used in this report. At PPIC, Gary Bjork and Ellen Hanak gave useful feedback that improved the final product. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. vii Acronyms ACF AFDC CalWORKs CPS DSS DHHS DRA FNS MOE OFA PRWORA SDR SSP TANF TDR USDA Administration for Children & Families, DHHS Aid to Families with Dependent Children California Work Opportunity and Responsibility to Kids Current Population Survey California Department of Social Services U.S. Department of Health and Human Services Deficit Reduction Act of 2005 Food and Nutrition Service Maintenance of Effort Office of Family Assistance, ACF Personal Responsibility and Work Opportunity Reconciliation Act SSP-MOE Data Report (form ACF-204) Separate State Program Temporary Assistance for Needy Families TANF Data Report (form ACF-199) U.S. Department of Agriculture viii 1. Introduction The creation in 1996 of the Temporary Assistance for Needy Families (TANF) program represents a watershed in federal welfare policy. The legislation that created TANF blockgranted a program (Aid to Families with Dependent Children, or AFDC) that had long been uncapped, and it gave states extensive control over program eligibility and on-going requirements. Along with the capped block grant that states could use flexibly, the federal government stipulated penalties if states fail to show that enough welfare adults are meeting federal work requirements. The reauthorization of TANF in 2006 made work requirements considerably more stringent, increasing the likelihood of penalties against the states. Hence, California and other states continue to seek ways of increasing work among adults receiving welfare. In this report, we examine the effects of changing two policies over which states now have control: sanctions (the penalties families face if parents log too few hours in approved welfare-to-work activities) and time limits (restrictions on the length of time adults can receive aid).1 We focus on these policies because Governor Schwarzenegger has suggested modifications to the policies in his 2007, 2008, and 2009 budget proposals (California Department of Finance, 2007; California Department of Finance, 2008; California Department of Finance, 2009). In his most recent proposal, the governor recommended eliminating grants to families if parents who have exceeded their allowed 60 months of assistance are not working for an adequate number of hours.2 Earlier proposals recommended similar changes to the time limit policy and proposed eliminating family grants if sanctioned parents fail to reverse their sanction within a limited number of months. Currently, the state continues assistance to children regardless of whether their parents forfeited assistance due to a sanction or time limit. To date, the governor’s proposals have not been adopted as part of an enacted budget. Some of the impetus for increasing the severity of sanction and time-limit policies in California has been the expectation that doing so would increase the share of welfare adults meeting federal work requirements (the so-called work participation rate). We evaluate whether this would be the case and how policy changes would affect the number of families on welfare.3 1 Because we use other states’ experiences with similar but not identical sanction and time-limit policies to estimate the likely effect of changing these policies in California (and because the governor’s proposals have changed somewhat from year to year), we present the estimates as illustrative of the direction of the changes the state would experience if policymakers established more severe financial penalties for noncompliant or timed-out CalWORKs families. 2 In 2009, the governor also proposed eliminating a family’s welfare grant if adults not working for sufficient hours fail to attend an in-person, biannual self-sufficiency review. This proposal would have applied to sanctioned adults, but also to many other cases in which adults are not aided, but children are (Legislative Analyst’s Office, 2009). 3 Specifically, we examine changes in the number of families containing at least one “work eligible” adult—that is, an adult required to be included in the federal work participation rate calculation. The primary reason we draw attention to effects on the caseload is that a change in the work participation rate provides no information about caseload change: An increase, for instance, could occur in conjunction with a caseload increase, decrease, or no caseload change at all. Caseload declines resulting from modifications a state makes to its eligibility rules (e.g., modifying the time limit) cannot be claimed as a 1 However, we do not focus simply on expected effects on the state’s work participation rate. California’s TANF program—California Work Opportunity and Responsibility to Kids (CalWORKs)—is intended to be a safety net for families facing difficult economic circumstances. A major concern with increasing the severity of sanctions and time limits is that the economic circumstances of vulnerable families would suffer. Thus, we also examine the implications that changing these policies would have on child poverty, maternal employment, and household income in single-mother families.4 We construct statistical models relating this set of outcomes to state TANF policies in all fifty states, holding constant a broad group of economic, demographic, and political factors that may also differ across states. We focus on the period prior to the federal reauthorization of TANF in 2006. In drawing our conclusions, we compare the caseload, work participation rate and family economic circumstances in states that have California-like policies with the same outcomes in states that have more severe policies.5 We derive these outcomes from two sources: the TANF and Separate State Program (SSP) data reports that states collect and submit to the federal government in evidence of their work participation rates and other compliance requirements, and the Annual Social and Economic Supplement to the Current Population Survey (CPS), collected by the U.S. Bureau of the Census.6 We used the TANF and SSP data to determine the size of states’ welfare caseloads and work participation rates; we used the CPS data to measure poverty, income, and employment among vulnerable families. We obtained details on sanction and time-limit policies from the Urban Institute’s Welfare Rules Database. In the following sections, we discuss the shortfall in California’s work participation rate under new federal regulations, characteristics of single-mother families in California, and the likely effects that increasing the severity of sanction and time-limit policies would have on the welfare caseload, the state’s work participation rate, and the economic circumstances of single mothers and their children. We then conclude with a discussion of our findings. The appendix briefly summarizes the data and methods. Detailed descriptions of the data and methods are provided in the technical appendix. credit against the required work participation rate (such “caseload reduction credits” are described in further detail in the following section). 4 See also the syntheses presented in Blank (2002) and Grogger and Karoly (2005). These studies review the literature examining the effects of TANF policies on welfare use, poverty, employment and earnings, and other measures of family and child well-being. 5 In the case of the work participation rate and the caseload, we compare outcomes across states with more and less severe policies. In the case of family economic circumstances, we compare changes before and after implementation of more severe policies with changes in states that implemented California-like policies. 6 The TANF and SSP data reports cover federal fiscal years 1998-2005 (October 1997-September 2005), and we use calendar years 1990-2005 from the CPS. As allowed by federal rules before the 2006 reauthorization of the TANF program, states operated SSPs to move whole segments of the caseload out of the work participation calculation. By 2005 California and thirty-one other states had created such SSPs. These were often created for two-parent families (Office of Family Assistance, 2007a). Such families were typically required to meet state program requirements, which included work requirements. The SSP data reports include the same information about work participation as the TANF data reports. 2 2. The Welfare Caseload and Federal Work Regulations The federal legislation creating TANF in 1996 required that states engage 50 percent of most welfare adults for 20 or 30 hours a week in work or in related activities (job search or education related to employment, work experience, and so on)—or face fiscal penalties. States were also required to engage 90 percent of those families with both parents on welfare in work activities. These are known, respectively, as the “All Families” and the “Two Parent” work participation rate requirements in the federal legislation.1 Prior to the reauthorization of TANF in early 2006, most states easily met these requirements because they were extended credits for caseload reductions that reduced the rate they were required to meet.2 Between 2004 and 2006, California’s All Families rate credit averaged 45.5 points (Office of Family Assistance, 2006; Office of Family Assistance, 2007a; Office of Family Assistance, 2007b).3 The size of the state’s caseload reduction credit implies that the CalWORKs caseload fell dramatically in the late 1990s and early 2000s, and indeed it did: The caseload dropped an average of 10 percentage points each year between 1998 and 2000, and then continued to decline at a slower pace for most of the rest of the period between 2001 and 2005, although it rose by 3 percent in 2004 (California Department of Social Services, n.d.).4 Changes in the legislation governing TANF, effective fiscal year 2007, have made it more difficult for states to meet federal work participation requirements because of changes in the ways states must calculate their credits. In California, the Department of Social Services estimated in 2007 that the caseload reduction credit under the new rules would drop to about 4 percentage points in coming years, implying that California’s effective All Families work 1 Single parents of children under age 6 fulfill the federal requirement by working at least 20 hours each week, while most single parents of older children, and one adult in two-parent families with children of any age, must work at least 30 hours a week. States face limitations on the length of time adults can remain in some work-related activities, as well as on the fraction of adults that can be engaged in some of these activities, and still have the adults’ effort count. Rules governing the calculation of the two-parent rate differ somewhat. For details on current program rules and restrictions on countable work activities, see Department of Health and Human Services (2008). California has its own work rules for CalWORKs adults, which differ somewhat from the federal rules. They are summarized at www.dss.cahwnet.gov/CDSSWEB/PG141.htm. 2 The so-called caseload reduction credit is a one percentage point reduction in the required work participation rate standard for every percentage point drop in the welfare caseload as compared to the baseline year. The original baseline year was 1995 and the current baseline year is 2005. Before the 2006 reauthorization, states could also exclude from the rate calculation those families receiving state-funded SSP assistance as well as families with children, but not parents, receiving welfare because the latter had been sanctioned or had reached a time limit. 3 Because two-parent CalWORKs families were in the state’s SSP, California had no Two Parent rate to meet until October 2006. While we focus in this report on the All Families rate, we include two-parent families in the rate calculation because their work effort must now be counted in both rate calculations. 4 Danielson (2008) describes changes in the education, citizenship, and family characteristics of welfare adults who were required to work in California and in the rest of the nation between 1999 and 2005. 3 participation rate requirement would be about 46 percent.5 It appears that the weak economy may reduce this credit to zero in the immediate future: As of October 2008, the CalWORKs caseload had exceeded its 2005 average (California Department of Social Services, n.d.). The penalty for failing to meet the All Families rate is a reduction of 5 percent in the state’s federal TANF block grant. This penalty can be increased by up to 2 percentage points each year, to a maximum of 21 percent of the state’s federal block grant. The state must make up this penalty out of its own funds and, in addition, must increase its required Maintenance of Effort (MOE) welfare expenditures.6 To put these penalties in perspective, California’s TANF block grant is $3.7 billion for the current and upcoming fiscal years, and the state’s minimum CalWORKs and related program spending is $2.7 billion. The penalties imply that California may be required to make up a $149 million shortfall in its TANF block grant, which could grow over time, and must also increase its minimum required state expenditures by $180 million each year it is out of compliance (Legislative Analyst’s Office, 2007; Parrott et al., 2007).7 A state can avoid penalties (but it must still increase its MOE expenditures) by entering into a compliance plan with the U.S. Department of Health and Human Services and achieving an agreed-upon increase in its work participation rate. In 2005, the final year before the reauthorization legislation passed, California achieved its adjusted All Families rate, but it would not have done so absent its caseload reduction credit. With their caseload reduction credits, 49 states and the District of Columbia met their adjusted standards. Absent these credits, only nine states would have been in compliance; three more were within 5 percentage points of meeting the rate absent the credit (Office of Family Assistance, 2006).8 Thus, most states, including California, face a substantial shortfall under the new federal requirements. 5 This caseload reduction credit is the so-called “natural reduction” credit and is distinct from the “excess MOE” credit that California has considered. The final federal rule published in February 2008 limits flexibility to claim such a credit. See www.acf.hhs.gov/programs/ofa. Both are discussed further in Legislative Analyst’s Office (2008). 6 MOE is the minimum a state must spend of its own funds in order to qualify for its entire TANF block grant. The level is pegged at 75 percent of its fiscal year 1994 AFDC expenditure, or 80 percent if it fails to meet federal requirements. 7 Before penalties are calculated, the state’s block grant is first adjusted downward to reflect amounts transferred to the child care or social services block grants and amounts spent through tribal TANF programs. If the state is in compliance with the All Families rate but not the Two Parent rate, the penalty is reduced to reflect the share of the caseload in two-parent families. In California, this is approximately 8 percent. 8 Although federal fiscal year 2007 was the first year rates were calculated under the new rules, we use the 2005, not the 2006, statistics to represent the status quo ante because the new legislation was enacted in February 2006, less than half way into fiscal year 2006. Indiana, the one state that failed to meet the federal requirement, was within 3 percentage points of meeting its adjusted rate. Two of the nine states that achieved a 50 percent or higher rate had a waiver to work participation rules in effect. In the absence of the waiver, both states’ rates would have been well below 50 percent, and one of the states (Tennessee) would not have met its adjusted rate. 4 However, California’s work participation rate was particularly low, ranking 36th among the fifty states in 2005 (Office of Family Assistance, 2007b).9 The difference does not appear to be attributable to California’s distinctive demographic characteristics: Adjusting the state’s share of immigrants and single mothers, educational attainment, and race/ethnic mix to match that in the rest of the nation pushes the rate even lower.10 In sections 5 and 6, we assess the potential for proposed policy changes to increase the work participation rate by shifting welfare adults in and out of compliance as well as on and off the caseload. After adjusting for demographic, economic, and political differences across states, we examine whether states with stricter sanction and time-limit policies have higher work participation rates and lower caseloads. 9 These rankings were computed under the old work participation rules, implying that they exclude all cases in SSPs (most of which were two-parent cases). It does not take into account waivers to federal rules in effect in two states in 2005. Using those waiver-adjusted work participation rates, California ranked 38th in the nation. California’s official rate for 2005 was 25.9 percent, and its rate for 2006 (the most recent available) was 22.2 percent (Office of Family Assistance, 2007a, 2007b). 10 Our simulation uses the same model as used for the sanction and time limit estimates reported in sections 5 and 6. The number of adults in the denominator of the rate is lower in the simulation, but the simulated number meeting work requirements is even lower. The result is a lower simulated work participation rate. Further details of the variables used in the simulation are provided in the technical appendix, available at http://www.ppic.org/content/pubs/other/409CDR_appendix.pdf. 5 3. Family Economic Circumstances Although raising California’s work participation rate has been of concern to state policymakers, getting adults on welfare to work is not the sole objective of the CalWORKs program. CalWORKs was crafted to assist low-income families in becoming self-sufficient while providing a safety net for children (Zellman et al., 1999). In this section, we review the circumstances of families liable to be affected by welfare policies. We define these families as single women between the ages of 16 and 46 who have less than a four-year college degree and who have children living with them. Most of these families do not receive a welfare check in any particular year, but far more will obtain welfare assistance over their lifetimes. Thus, their actions are likely to be affected by changes in welfare policies, regardless of whether they are current recipients of CalWORKs benefits. For instance, they may work more or less if sanction or time-limit policies are more or less strict, or they may take other action to change the resources available to their families (such as moving in with friends or family or obtaining more education). In contrast, mothers with more resources (those with more education and married mothers) are much less likely to substantially alter their behavior in the wake of a welfare policy change.1 Between 2003 and 2005, the endpoint in our period of analysis, an average of 8.4 percent of families in California were headed by a single mother who lacked a four-year college degree. These families had less than two children (1.7 on average), and close to half (47.6%) had a child under age 6. Slightly over two-fifths of mothers (43.3%) had some college experience, while about three out of ten had a high school diploma or equivalent (29.3%) and about the same share had not completed high school (27.4%). These families had higher incomes in the mid-2000s than they did in the early 1990s before the advent of CalWORKs: In 2003-2005, an average of 34.9 percent had incomes below the federal poverty line, down from 44.1 percent in 1990-1992.2 Single mothers were also more likely to be working: In 2003-2005, 72.9 percent had worked at some point during the reporting year, up from 61.2 percent in 1990-1992.3 Earnings among the group of single mothers who had worked at all during the year averaged $23,132 in 2003-2005, compared to $15,436 in 1990-1992.4 Clearly, at least by some measures these vulnerable families were better off in the mid-2000s than they were in the early 1990s. Although welfare reform policies may have contributed to these improvements, it is not clear whether the specific sanction and time-limit policies that California adopted are important for family well-being. We explore this issue in the sections that follow. 1 Although some have argued welfare policies can affect the incidence of single motherhood by providing an alternative to marriage, the existing research finds little or no effect of welfare policies on marriage or fertility among single women. However, research is still ongoing in this area (Grogger and Karoly, 2005). While married couples are eligible for welfare, the vast majority of welfare families (approximately 95 percent nationwide) include only one parent in the assistance unit. 2 Examining poverty from the child’s point of view, 40.9 percent of children living with single mothers who did not have a college degree lived in poverty in 2003-2005. 3 CPS respondents were also asked whether they worked in the previous week. Using this metric, an average of 62.3 percent were working in 2003-2005 while 48.4 percent were working in 1990-1992. 4 Dollar values are adjusted for inflation to 2005 levels. 6 4. Sanction and Time-Limit Policies Sanctions are the reduction or elimination of a family’s aid payment if parents fail to meet work-related requirements, while time limits restrict the length of time adults can receive welfare assistance. When state lawmakers fashioned the CalWORKs program in 1996 and 1997, they established policies intended to further the state’s “work-first” approach for adults on welfare while preserving a safety net for children (Zellman et al., 1999).1 Consequently, CalWORKs sanction and time-limit policies focus penalties on adults. California continued the sanction policy established under AFDC, the federal cash assistance program preceding TANF. That policy calls for a reduction in the family’s aid payment equal to the sanctioned adult’s share for as long as that adult remains out of compliance with requirements.2 Most states established more severe penalties.3 Likewise, California is one of four states that have not applied the federal 60-month time limit on aid to families on welfare. In California, adults have a 60-month time limit on welfare benefits, but children can continue to receive welfare benefits after parents have reached time limits, as long as they remain otherwise eligible.4 California uses state funds to continue assistance to these children. The governor’s 2008 budget proposal included a recommendation to eliminate family grants over a period of 12 months if sanctioned parents fail to reverse their sanction (the 2007 proposal suggested a grace period of 3 months). Likewise, the 2007, 2008, and 2009 budget proposals recommended eliminating family grants once a parent has reached the time limit unless that parent meets federal work requirements (California Department of Finance, 2007; California Department of Finance, 2008; California Department of Finance, 2009).5 1 We emphasize that the CalWORKs program is far more than its sanction and time limit policies. CalWORKs incorporates a range of policies designed to promote work among adult welfare recipients, ranging from assistance with job search to subsidized child care for those who become employed. 2 Adult welfare recipients were subject to so-called “durational sanctions” until 2007, which meant that those who had been sanctioned more than once had to spend a minimum number of months in sanction status before their full grant would be restored. Pursuant to AB 1808 (2006), this is no longer the case. 3 For a more detailed typology of the sanction and time-limit policies across states, see Table A.2 of the technical appendix, available at http://www.ppic.org/content/pubs/other/409CDR_appendix.pdf. 4 Children continue to be eligible if their total family income remains below an income threshold, which varies by family size. The threshold is lower after adult(s) reach their time limit. 5 The intent of these changes, at least in part, is to increase California’s work participation rate (California Department of Finance, 2007, 2008). In the wake of the federal reauthorization of TANF in 2006, California must include in the work participation rate calculation sanctioned adults whose children continue receiving assistance. Sanctioned adults who have been sanctioned for three months or less in the previous twelve months can be excluded. This means that roughly half of California’s sanctioned caseload can be excluded from the calculation (Danielson, 2006). Federal rules in the wake of TANF reauthorization require California to count time-limited parents whose children continue to receive aid when calculating the work participation rate. That is, these parents will be added to the denominator of the calculation and may also be included in the numerator (if they are working an adequate number of hours). 7 5. Effects of Sanctions How would the welfare caseload, the work participation rate, child poverty, and single mothers’ employment differ under a more severe sanction policy? In this section, we examine the implications of moving from California’s current grant-reduction sanction to a policy of gradual or immediate grant elimination.26 We use statistical methods to account for a range of other conditions in states—strength of the economy, political climate, and demographic characteristics—that, apart from policies, may also cause states’ work participation rates to differ. In sum, we find that increasing the financial penalties associated with noncompliance with work rules would reduce California’s welfare caseload, substantially increase its work participation rate, and slightly reduce poverty among children living with single mothers (Table 5.1). Table 5.1. Effects of work-related sanctions Gradual grant elimination (%) Immediate grant elimination (%) Caseload Work participation rate Child poverty Mother’s employment Mother’s earnings Household income -27+ 50* -4* 1 6 0.03 -52* 94* -5* 4 3 4* SOURCES: Authors’ calculations from TANF and SSP-MOE data reports (1998-2005), the Annual Social and Economic Supplement to the CPS (1990-2005), and the variables described in technical appendix Table A.3. NOTES: Significance is in comparison to grant-reduction sanction. Table entries derived from model estimates in technical appendix Table A.4, column 1, and Table A.8, columns 1 and 2. Caseload refers to the portion of the caseload required to be counted in the federal rate calculation. Child poverty, mothers’ employment, and mothers’ earnings are among families headed by single mothers ages 16-46 who have less than a four-year college degree. For non-linear models (poverty and employment), we simulate the effects at the means of explanatory variables for California in 2003-2005. Household income is cash income from all sources for all residents of housing units that contain such a family. + Significantly different at the 10% level or lower. * Significantly different at the 5% level or lower. 26 The gradual grant elimination category we use includes several variants of this type of sanction. Some states eliminated the grant at the instance of a second or higher sanction. Some eliminated the grant after a sanction lasting more than a certain number of months. This number of months ranged from 1 to 12, with a median of 4 months. Several states imposed a grant reduction sanction if the family had received aid for fewer than 24 months, and a grant elimination sanction if it had received aid for more months. 8 Adjusting for other differences, states with gradual or immediate grant-elimination sanctions have adult-headed caseloads that are lower by about a quarter to a half as compared to states with grant-reduction sanctions. 27 The work participation rate is sharply higher, by 50 percent in the case of a gradual sanction and by over 90 percent in the case of an immediate sanction. In other words, if the state had had a gradual sanction in place, California’s work participation rate would have been 34 percent in 2005 rather than 23 percent. Under an immediate sanction, it would have been 44 percent. These differences are large; however, a portion of the change (about half in the case of a gradual sanction) is due to a federal rule change in 2006. This rule change requires more of California’s sanctioned adults to be counted in the state’s work participation rate, thus lowering the rate even though the state has not altered its sanction policy. Our estimates suggest that moving to some form of a grant-elimination sanction will result in a smaller caseload and a higher work participation rate. A serious concern with such a change is that it would be harmful to low-income families. If the policy simply removes all benefits from families who fail to meet work requirements, or discourages such families from ever applying for benefits, it could increase poverty among at-risk families. On the other hand, it is possible that a grant elimination sanction would encourage work, either because sanctioned adults would be able to replace lost assistance with work or because the more stringent sanction would motivate more parents on assistance to comply with work requirements while on welfare, or both.28 We explore these issues by investigating the effect of more severe sanction policies on the economic circumstances of single-mother families.29 We find that, relative to California-like policies, grant-elimination sanctions reduce child poverty by a small amount. About 41 percent of California children in single-mother families were poor in 2003-2005. With grant elimination sanctions, the percentage would have been about 39 or 40 percent (a reduction of approximately 4 to 5 percent, as shown in Table 5.1). Single-mother employment and earnings are higher, but the effects are not statistically significant.30 Implementation of an immediate (but not gradual) grant-elimination sanction increases income in the households in which these families live by a small amount (4 percent). Overall, the results suggest that grant-elimination sanctions do not negatively affect the cash resources of single-mother families.31 27 Previous studies examining the effects of welfare sanctions on caseload find mixed evidence of a negative effect of gradual full-family sanctions and a negative effect of immediate full-family sanctions (Danielson and Klerman, 2008; Grogger and Karoly, 2005). These studies were not able to exclude childonly cases, which are little affected by work-related TANF policies. 28 To be consistent with our finding that the smaller caseload is smaller when more severe sanctions are in place, in the latter scenario it must be the case that some on welfare who work more leave assistance. 29 If we restrict the sample to mothers with less than a high school degree, a larger fraction of whom received income from public assistance, estimates remain substantially the same, but sample sizes are smaller and coefficients are estimated less precisely. 30 The table reports results using any maternal work during the past year as the outcome. We found similar results when we examined any work in the last week. Grant elimination sanctions did not have a statistically significant effect on non-work factors associated with poverty: number of children per mother, number of adults in the household, and the probability of a single woman being a mother. 31 The previous research does not specifically examine the effects of more stringent sanctions on singlemother employment and earnings. One study finds that greater financial penalties associated with a sanction reduce deep poverty among single mothers by a small amount but do not have an effect on overall single-mother poverty (McKernan and Ratcliffe, 2006). 9 6. Effects of Time Limits How are the set of outcomes we consider—welfare caseload, work participation rate, child poverty, and single mothers’ employment—different if a grant-elimination time limit is in place rather than a grant-reduction policy? In this section, we examine the implications of moving from California’s current policy, a grant-reduction time limit, to two variants of a grantelimination time limit. The first policy is simply one of halting assistance to families in which an adult has reached the time limit. The second extends additional months of assistance to the family as long as the time-limited adult is working sufficient hours to meet the requirement.32 In sum, due to the data limitations discussed below, we are unable to estimate the effect of changing California’s time-limit policy on the work participation rate and the caseload of those required to be counted in that rate. However, we find that increasing the financial penalty associated with reaching a time limit is not likely to change the economic circumstances of vulnerable families (Table 6.1). Table 6.1. Effects of welfare time limits Grant elimination (%) Grant elimination with extension if meeting work requirement (%) Child poverty Mothers’ employment Mother’s earnings Household income 3 -1 1 -2 1 0 5 2 SOURCES: Authors’ calculations from the Annual Social and Economic Supplement to the CPS (19902005) and the variables described in technical appendix Table A.3. NOTE: Significance is in comparison to grant-reduction time limit (reached). Table entries derived from model estimates in technical appendix Table A.8, columns 1 and 2. Child poverty, mothers’ employment, and mothers’ earnings are among families headed by single mothers ages 16-46 who have less than a four-year college degree. For non-linear models (poverty and employment), we simulate the effects at the means of explanatory variables for California in 2003-2005. Household income is cash income from all sources for all residents of housing units that contain such a family. + Significantly different at the 10% level or lower. * Significantly different at the 5% level or lower. An important rule change in 2006 requires states that continue children on welfare once parents have reached a time limit to include those timed-out parents in their work participation rate calculation. Before 2006, states with grant-reduction policies like California’s excluded 32 Adults may qualify for a time limit exemption or extension for other reasons. Table A.3 of the Technical Appendix lists the reasons that we include in the set of TANF program characteristics that we hold constant in our regression models. 10 time-limited adults from their rate calculations. This rule change implies that we lack a reasonable baseline policy to estimate differences across states with more and less severe time limits using data from before the federal reauthorization of TANF. Thus, we are unable to estimate whether California’s work participation rate and caseload would be higher or lower under a grant-elimination time limit. It is important to note that time limits set into motion potentially complex behavioral changes among welfare recipients and potential recipients, and it is possible that a more severe time limit might either reduce or increase the state’s work participation rate.33 For example, a grant-elimination time limit might increase the incentive of welfare adults to work. However, it is also possible that families considering whether to apply for CalWORKs might defer their applications until they face more extreme hardship in order to avoid exhausting their months of eligibility. Families facing relatively greater hardships may be less able to find and keep employment. Such a change in the types of families on welfare could lower the state’s work participation rate. As with grant-elimination sanctions, a major concern with grant-elimination time limits is that they would harm low-income families. Clearly, the loss of benefits to children without any compensating actions on the part of adults would worsen their economic situation. At the same time, it is conceivable that a full-family time limit might change work incentives or other incentives enough to improve the family’s economic situation. Examining the economic circumstances of single-mother families (in which the mother does not have a college degree), we find that relative to states in which families had reached grant-reduction time limits, states in which families had reached grant-elimination time limits had somewhat higher child poverty rates, but the difference is not statistically significant.34 Grant-elimination time limits, with or without an extension for those complying with work rules, have mixed effects on single mothers’ employment and earnings and on household income, but the differences are never statistically significant.35 In other words, based on the experience of other states (holding constant many other factors), we do not find robust evidence that moving from a grantreduction to a grant-elimination time limit would substantially increase poverty among children in single-mother families.36 33 We note that we estimate the work participation rate to be lower in states with grant-elimination time limits as compared to states with no time limit. We also estimate the caseload to be smaller. However, the differences are not statistically significant. Previous research has found that time limits (as compared to no time limit) reduce single mothers’ use of welfare and the welfare caseload (Danielson and Klerman, 2008; Grogger, 2001; Grogger, 2003; Mazzolari, 2007). 34 Our model controls, among other things, for the length of the time limit and the set of time-limit exemptions and extensions that a state permits. Table A.3 in the technical appendix (available at http://www.ppic.org/content/pubs/other/409CDR_appendix.pdf) lists the full set of control variables. 35 The table shows results for any maternal work during the past year. We found similar results when we examined any work in the past week. 36 Note that California does impose a time limit, so that differences described in the text are in comparison to that time limit. Analyzing data that end in the early 2000s, the previous literature finds mixed evidence that a grant-elimination time limit in comparison to no time limit increases employment among single mothers but no evidence that a grant-elimination time limit alters income, the number of weeks worked, or annual earnings. See Grogger and Karoly (2005) for a summary. Making that same comparison between no time limit and a grant-elimination time limit, our estimates imply no statistically significant change in income, employment, or poverty. 11 7. Conclusions Increasing the severity of CalWORKs sanction and time limit policies has been considered in three successive budget cycles. In this report, we examine the likely effects of modifying these policies. We find that states with stricter sanction policies have substantially higher work participation rates and lower caseloads than states with a California-like sanction policy. Neither of the sanction policies that the governor proposed in 2007 and 2008 would likely make up the expected shortfall in California’s work participation rate; however, the increase in the participation rate in the wake of either one could be dramatic. One important concern is that adopting grant-elimination sanctions will worsen the economic circumstances of children in vulnerable families. We find that implementing a stricter sanction policy actually lowers poverty rates slightly among children living in single-mother families. The evidence we muster indicates states that adopted grant-elimination time limits did not see increased child poverty as a result; we also do not find evidence that grant-elimination time limits increased single mothers’ employment. There are three important limitations of our findings. First, our results suggest that moving to a grant-elimination sanction or time-limit would not increase child poverty or worsen the economic circumstance of single-mother families on average, but it is possible that some highly vulnerable families would experience worse economic conditions under grantelimination policies.37 Such families may be discouraged from applying for welfare, or they may lose benefits under stricter policies and not be able to increase earnings or other sources of income to make up the loss. Second, the child poverty measure does not take into account the additional costs families face when moving from welfare to work (e.g., less parental time with children and work-related expenses). Incorporating these costs could reduce the measured benefits of policies that promote work.38 Third, because the welfare reform period began in the 1990s, our study covers a period of major economic growth followed by a short recession and a period of slower growth.39 Using these data, it is difficult to predict what will happen to child poverty under grant-elimination policies during a prolonged recession in which labor market opportunities are limited.40 37 Bitler, Gelbach, and Hoynes (2006) show that the impact of welfare reform policies varied substantially across families. 38 Families receiving CalWORKs assistance, those that have recently left CalWORKS, and some other lowincome families, are eligible for services that help to offset the additional expenses of work. The most important of these is subsidized child care. 39 During the 1990s, the robust economy and strong labor market were important factors in encouraging single mothers to work, although policies also played an important role (Blank, 2002). 40 Herbst (2008) provides preliminary evidence that some TANF policies have larger effects on work among single mothers when the economy is stronger. The time period examined in this research is similar to the time period we use, implying that it also cannot answer the question of the role of policy during a sustained downturn. In theory, a state’s work participation rate is less sensitive than low-wage employment to economic conditions because welfare adults can count a number of different activities 12 Our findings are at odds with the belief that the economic conditions of children are substantially improved by maintaining their benefits when parents are sanctioned for not working or for reaching the time limit. However, they are similar to the findings of studies examining the introduction of TANF and earlier TANF-like state reform efforts. The 1996 welfare overhaul introduced stronger work requirements and enforced them with financial penalties levied on those who did not meet the requirements. At the time, many argued that the reform would increase poverty among vulnerable children. Drawing on multiple studies, the general consensus is that the 1996 reform did not increase child poverty. Indeed, several studies conclude that the introduction of TANF helped reduce child poverty by encouraging maternal employment.41 Clearly, full-family sanctions and time limits remove welfare benefits from some children. But their economic circumstances will depend on how their parents respond to these incentives to work. toward their required hours. At the same time, employment is the most common activity that welfare adults pursue (Danielson, 2006), and it is also a route that is less expensive for states than supporting subsidized employment, on-the-job training, and related efforts. 41 For summaries of the national studies, see Grogger and Karoly (2005), Blank (2002), and Meyer and Holtz-Eakin (2001). See also Sawhill and Haskins (2007), O’Neill and Korenman (2004), and Haskins and Primus (2002). McKernan and Ratcliffe (2006) find no overall effect of welfare reform on child poverty and mixed effects for specific welfare reform policies. For evidence of welfare program effects on other child outcomes, see Gennetian et al., (2002) and Morris et al. (2005). 13 Appendix: Data and Methods This appendix provides a brief description of the data and methods used in this study. A detailed technical appendix is available on the PPIC website http://www.ppic.org/content/pubs/other/409CDR_appendix.pdf). Here, we first describe our approach to analyzing the work participation rate and welfare caseload, and then our approach to arrive at estimates of changes in employment, poverty, and income among singlemother families. To measure the work participation rate and the number of cases containing an adult required to be counted in the rate calculation, we use official statistics from the Office of Family Assistance reports, 1998-2005(OFA, n.d.). We construct the participation rate from OFA information on the size of the TANF and SSP caseloads meeting federal work participation requirements and the size of the caseloads not meeting these requirements. We use the natural log of the work participation rate and the natural log of the caseload divided by the population of women ages 16-46 in the state and year. We divide by the population statistic in order to adjust for that source of variation in the size of welfare caseload: State populations vary widely. We obtained the denominator for the caseload outcome from U.S. Bureau of the Census population estimates. We estimate the effect of state TANF policies using the variation in policies across states. We control for national changes across years, state-level measures of economic conditions, demographic characteristics, political variables, and a detailed set of welfare policies. For family economic conditions, we use the Annual Social and Economic Supplement of the Current Population Survey. These data, collected by the U.S. Census Bureau, are used for official poverty statistics. We focus our analysis on the main population thought to be affected by welfare policy: single mothers ages 16 to 46 who do not have a bachelor’s degree. We investigate four main outcomes: poverty, maternal employment, annual earnings, and household income. We estimate the effect of state TANF policies using the variation in policies across states and over time. We use a “difference-in-difference-in-differences” approach. We control for state and year fixed effects, individual-level demographic characteristics (such as age, education, age of youngest child, and race/ethnicity) and state-level factors (policies, economic conditions, and political variables). We identify the effect on family economic conditions of the change in state TANF policies using data from 1990-1996 and 1999-2005 (we exclude 1997 and 1998 because during those years many states had only partially implemented their TANF programs). We use single women in the same age range and with the same level of education as a “control group” ‒ we assume that any effect of the TANF policy variables on this group is spurious and that the additional effect on single mothers is causal. In robustness checks, we also use married mothers in the same age range and with the same education as a control group. Using this control group, we find policy effects that are similar in magnitude although they are not statistically significant. 14 References Assistant Secretary for Planning and Evaluation, “Temporary Assistance for Needy Families (TANF): Data and Documentation,” available at aspe.hhs.gov/ftp/hsp/tanf-data/index.htm. Berry, William D., Evan J. Ringquist, Richard C. Fording, and Russell L. Hanson, “Measuring Citizen and Government Ideology in the American States, 1960-93,” American Journal of Political Science, 1998, 42, pp. 327-48. Bitler, Marianne, Jonah Gelbach, and Hilary Hoynes, “What Mean Impacts Miss: Distributional Effects of Welfare Reform Experiments,” American Economic Review, 2006, 4 (September), pp. 988-1012. Blank, Rebecca M., “Evaluating Welfare Reform in the United States,” Journal of Economic Literature, 2002, Vol. 40, No. 4, pp. 1105-66. California Department of Finance, Governor’s Budget Summary 2007-08, January 2007, available at 200708.archives.ebudget.ca.gov/pdf/BudgetSummary/FullBudgetSummary.pdf. California Department of Finance, Governor’s Budget Summary 2008-09, January 2008, available at 200809.archives.ebudget.ca.gov/pdf/BudgetSummary/FullBudgetSummary.pdf. California Department of Finance, Governor’s Budget Summary 2009-10, January 2009, available at www.ebudget.ca.gov/pdf/BudgetSummary/FullBudgetSummary.pdf. California Department of Social Services, CalWORKs Benefits Model, 2003, available at www.dss.cahwnet.gov/research/res/pdf/BenefitModel.pdf. California Department of Social Services, “CA 237 CW - CalWORKs Cash Grant Caseload Movement Report,” n.d., available at www.dss.cahwnet.gov/research/res/. Citro, Constance F., and Robert T. Michael, eds, Measuring Poverty: A New Approach, Washington, D.C., National Academy Press, 1995. Crouse, Gil, “State Implementation of Major Changes to Welfare Policies, 1992-1998,“ Assistant Secretary for Planning and Evaluation, DHSS, Washington, DC, 1999, available at aspe.hhs.gov/HSP/Waiver-Policies99/policy_CEA.htm. Danielson, Caroline, and Jacob Alex Klerman, “Did Welfare Reform Cause the Caseload to Decline?” Social Service Review, 2008, Vol. 82, No. 4, pp. 703-730. Danielson, Caroline, Patterns of Work Participation in CalWORKs, Occasional Paper, Public Policy Institute of California, San Francisco, California, December 2006. Danielson, Caroline, The CalWORKs Caseload in 1999 and 2005, Working Paper, Public Policy Institute of California, San Francisco, California, September 2008. Department of Health and Human Services, “Reauthorization of the Temporary Assistance for Needy Families Program; Interim Final Rule,” Federal Register, Vol. 71, no. 125, June 29, 2006. 15 Department of Health and Human Services, “Estimated Additional Work Participation Needed Under the Deficit Reduction Act of 2005,” 2007, available at peerta.acf.hhs.gov/pdf/Public_WPR.pdf. Department of Health and Human Services, “Reauthorization of the Temporary Assistance for Needy Families Program; Final Rule,” Federal Register, Vol. 73, No. 24, February 5, 2008, available at www.acf.hhs.gov/programs/ofa/lawreg/finalrule/tanf_final_rule.pdf. Gennetian, Lisa A., Greg J. Duncan, Virginia W. Knox, Wanda G. Vargas, Elizabeth Clark-Kauffman, Andrew S. London, How Welfare and Work Policies for Parents Affect Adolescents: A Synthesis of Research, New York: MDRC, May 2002, available at www.mdrc.org. Grogger, Jeffrey, “The Effects of Time Limits, the EITC, and Other Policy Changes on Welfare Use, Work, and Income Among Female-Headed Families,” Review of Economics and Statistics, Vol. 85, No. 2 May 2003,pp. 394-408. Grogger, Jeffrey, “Time Limits and Welfare Use,” Journal of Human Resources, Vol. 39, No. 2 Spring 2004, pp. 405-424. Grogger, Jeffrey, and Lynn A. Karoly, Welfare Reform: Effects of a Decade of Change, Cambridge, Massachusetts: Harvard University Press, 2005. Haskins, Ron, and Wendell Primus, “Welfare Reform and Poverty,” Brookings Institution, July 2001, available at http://www.brookings.edu/papers/2001/07poverty_haskins.aspx. Herbst, Chris M., “Do Social Policy Reforms Have Different Impacts on Employment and Welfare Use as Economic Conditions Change?” Journal of Policy Analysis and Management, Vol. 27, No. 4, 2008, pp.867-894. Klarner, Carl, “The Measurement of the Partisan Balance of State Government,” State Politics and Policy Quarterly, Vol. 3, No. 3, 2003. Klerman, Jacob Alex, Jeanne S. Ringel, and Beth Roth, “Underreporting of Medicaid and Welfare in the Current Population Survey,” Santa Monica, California, RAND WR-169-3, March 2005, available at www.rand.org/pubs/working_papers/2005/RAND_WR1693.pdf. Legislative Analyst’s Office, “Analysis of the 2007-08 Budget Bill,” Legislative Analyst’s Office, Sacramento, California, February 2007, available at www.lao.ca.gov. Legislative Analyst’s Office, “Analysis of the 2008-09 Budget Bill,” Legislative Analyst’s Office, Sacramento, California, February 2008, available at www.lao.ca.gov. Mazzolari, Francesca, “Welfare Use when Approaching the Time Limit.” Journal of Human Resources, Vol. 42, No. 3, 2007, pp. 596–618. McKernan, Signe-Mary, and Caroline Ratcliffe, The Effect of Specific Welfare Policies on Poverty, The Urban Institute (mimeo.), Washington D.C., April 2006. 16 Meyer, Bruce D., and Douglas Holtz-Eakin, eds., Making Work Pay: The Earned Income Tax Credit and Its Impact on America’s Families, New York: Russell Sage Foundation, 2001. Morris, Pamela A., Lisa A. Gennetian, and Greg J. Duncan, Effects of Welfare and Employment Policies on Young Children: New Findings on Policy Experiments Conducted in the Early 1990s, New York: MDRC, April 2005, available at www.mdrc.org. Neumark, David, and William Wascher. “Minimum Wages, the Earned Income Tax Credit, and Employment: Evidence from the Post-Welfare Reform Era,” Cambridge, Massachusetts: NBER Working Paper No. 12915, February 2007. Office of Family Assistance, “TANF and SSP-MOE Data Reports,” n.d., available at www.acf.hhs.gov/programs/ofa/tanfrpts/index.htm. Office of Family Assistance, 2006, “Temporary Assistance for Needy Families Program Information Memorandum No. TANF-ACF-IM-2006-1,” January 17, 2006, available at www.acf.hhs.gov/programs/ofa/particip/2004/IM04RATE.htm. Office of Family Assistance, 2007a, “Temporary Assistance for Needy Families Program Information Memorandum No. TANF-ACF-IM-2007-5,” August 16, 2007, available at www.acf.hhs.gov/programs/ofa/particip/2006/IM06RATE_REV_CN20070814.htm. Office of Family Assistance, 2007b, “Temporary Assistance for Needy Families Program Information Memorandum No. TANF-ACF-IM-2007-4,” August 1, 2007, available at www.acf.hhs.gov/programs/ofa/particip/2005/IM05RATE.htm. O’Neill, June, and Sanders Korenman, “Child Poverty and Welfare Reform: Stay the Course,” Civic Report No. 44, May 2004, Manhattan Institute for Policy Research, available at www.manhattan-institute.org/html/cr_44.htm. Parrott, Sharon, Liz Schott, Eileen Sweeney, Allegra Baider, Evelyn Ganzglass, Mark Greenberg, Elizabeth Lower-Basch, Elisa Minoff, and Vicki Turetsky, Implementing the TANF Changes in the Deficit Reduction Act: ‘Win-Win’ Solutions for Families and States, Second Edition, Center on Budget and Policy Priorities (CBPP) and Center for Law and Social Policy (CLASP), February 2007, available at www.clasp.org/publications/tanfguide.pdf. Sawhill, Isabel, and Ron Haskins, “Ending Poverty in America: Using Carrots and Sticks,” The American Prospect, May 2007, available at http://www.brookings.edu/articles/2007/05poverty_haskins.aspx. Urban Institute, Welfare Rules Database, Washington, DC., n.d., available at www.urban.org/Content/Research/NewFederalism/Data/ANFData.htm. Zellman, Gail L., Jacob Alex Klerman, Elaine Reardon, and Paul Steinberg,. Welfare Reform in California: State and County Implementation of CalWORKs in the First Year, Santa Monica, California: RAND Corporation, 1999. 17 PUBLIC POLICY INSTITUTE OF CALIFORNIA Board of Directors Walter B. Hewlett, Chair Director Center for Computer Assisted Research in the Humanities Mark Baldassare President and Chief Executive Officer Public Policy Institute of California Ruben Barrales President and Chief Executive Officer San Diego Regional Chamber of Commerce John E. Bryson Retired Chairman and CEO Edison International Gary K. Hart Former State Senator and Secretary of Education State of California Donna Lucas Chief Executive Officer Lucas Public Affairs Ki Suh Park Design and Managing Partner Gruen Associates Constance L. Rice Co-Director The Advancement Project Thomas C. Sutton Retired Chairman and Chief Executive Officer Pacific Life Insurance Company Raymond L. Watson Vice Chairman of the Board Emeritus The Irvine Company Carol Whiteside President Emeritus Great Valley Center PUBLIC POLICY INSTITUTE OF CALIFORNIA 500 Washington Street, Suite 600 San Francisco, California 94111 phone: 415.291.4400 fax: 415.291.4401 PPIC SACRAMENTO CENTER Senator Office Building 1121 L Street, Suite 801 Sacramento, California 95814 phone: 916.440.1120 fax: 916.440.1121 www.ppic.org" ["post_date_gmt"]=> string(19) "2017-05-20 09:38:43" ["comment_status"]=> string(4) "open" ["ping_status"]=> string(6) "closed" ["post_password"]=> string(0) "" ["post_name"]=> string(8) "r_409cdr" ["to_ping"]=> string(0) "" ["pinged"]=> string(0) "" ["post_modified"]=> string(19) "2017-05-20 02:38:43" ["post_modified_gmt"]=> string(19) "2017-05-20 09:38:43" ["post_content_filtered"]=> string(0) "" ["guid"]=> string(50) "http://148.62.4.17/wp-content/uploads/R_409CDR.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) }