|
Article Excerpt
In this paper, we examine the utilization of a state earned income credit program among current and former welfare recipients, where earned income credit utilization is characterized by two measures: (1) the receipt percentage, which is the percentage of all current and former welfare recipients that receive the state earned income credit, and (2) the participation percentage, which is the percentage of those eligible for the credit that receive the credit. Both measures are useful, depending upon the policy issue. If the issue is use of the earned income credit among all current and former welfare recipients, then the receipt percentage is relevant. On the other hand, if the issue is making sure those eligible for the credit receive it, then the participation percentage is the relevant measure.
Researchers have correlated the earned income credit to higher labor force participation among single mothers and to lower welfare caseloads. We add to this literature by examining how many welfare recipients receive the earned income credit. Also, previous researchers have established that current public assistance recipients tend to participate in the earned income credit at low rates. We find similar results for our sample, and examine who is less likely to receive the state earned income credit. We further add to the literature by examining (1) receipt percentages in conjunction with participation percentages, (2) the number of years parents spend before taking up the credit and the number of years parents receive the credit, and (3) whether increases in the generosity of the state earned income credit in the form of changes in its parameters increase receipt and participation.
We find that in any given year, roughly 38 percent of our sample of current and former welfare households receives the state earned income credit, and that 65 percent of current and former welfare households who are eligible for the credit participate in the state earned income credit. The percentage of parents that receive the state earned income credit increases with earnings, rising to rates comparable to national statistics when looking at all eligible taxpayers. Although parents' transitions from welfare to higher earnings levels are somewhat chaotic, we find that, as time passes, roughly 75 percent of households who are at least temporarily observed on welfare may eventually receive the state earned income credit and that most will receive the credit for four years or less. Current and former welfare households who are less likely to receive the credit are non-whites and household heads without a high school diploma, which may roughly correspond to persons who face work barriers. In regressions estimating the probability of receipt and participation, we find some evidence that increasing the generosity of the credit by changing the individual earned income credit parameters correlates with increased receipt and participation probabilities.
Our plan for future research is to extend our analysis of program utilization to other tax credits and low-income benefits. Our hope is to characterize parents' paths to higher income levels. The research will fulfill two purposes. One is to identify which programs or combinations of programs are utilized by parents to increase their standard of living. The second is to examine parents who face work barriers, tracing their program utilization and outcomes over time. Our efforts will help identify the benefits and shortcomings of current policies and hopefully provide guidance for future policy innovations.
INTRODUCTION
During the 1980s and 1990s, Congress and the states realigned programs for low-income families, which placed greater emphasis on work as a means of raising their standard of living. Central to federal and state efforts was the transformation of the welfare program Aid to Families with Dependent Children (AFDC) into Temporary Assistance for Needy Families (TANF) and expansions to state and federal earned income credits. In turn, researchers evaluated these program innovations and correlated expansions of federal and state earned income credits to lower welfare caseloads and higher employment rates among single mothers.
The rewards and consequences of an emphasis on work are still under scrutiny as researchers more completely characterize the sometimes successful, sometimes troublesome transitions of individuals off of welfare. An evaluation of programs affecting such transitions depends partly upon who is intended to benefit from the program, which may be unclear or may include several intended pools of beneficiaries. Nonetheless, appropriate measures of program utilization hinge on choosing the pool of beneficiaries that comport with the perceived goals of the policy. In this paper, we examine two pools--all current and former welfare recipients regardless of their earned income credit eligibility, and the subset of only those eligible for the credit. If policymakers want to know the extent to which the earned income credit earnings incentive reaches all current and former welfare recipients in any one year, then the relevant pool is the first one. On the other hand, if policymakers want to know the extent to which the earned income credit reaches those eligible for the credit, then the relevant pool is the subset of the first pool that includes only those households with earnings within the earned income credit eligibility range. Ultimately, the measures of utilization associated with the two pools may have substantively different policy implications. For example, to raise utilization among all current and former welfare recipients, policymakers may consider alleviating barriers to work. To raise utilization among the subset of current and former welfare recipients with earnings eligible for the earned income credit, policymakers may consider outreach efforts.
To facilitate our discussion on utilization, we define two measures: (1) the receipt percentage, which is the percentage of all current and former welfare recipients that receive the state earned income credit in a given year regardless of eligibility for the credit, and (2) the participation percentage, which is the percentage of current and former welfare recipients who receive the credit among only those eligible for the credit. In this paper, we report results on the utilization of the credit in three parts: (1) an investigation of the household characteristics that determine receipt and participation in the state earned income credit program among current and former welfare recipients, (2) an analysis of the number of years current and former welfare recipients spend before taking up the credit and the number of years current and former welfare recipients receive the credit, and (3) an examination of whether current and former welfare recipients respond to changes in earned income credit programs. The first issue addresses the fact that policymakers may not only care whether those eligible to participate in the earned income credit do so, they may also want to know the extent to which all current and former welfare recipients use earned income credits as they transition away from welfare, especially given the current policy emphasis on work incentives. The second issue recognizes that receipt and participation percentages are snapshots of a dynamic process as parents pass through a window of eligibility for the credit. The third issue addresses the fact that not only will having earnings within the eligibility window affect receipt and participation, the amount of the benefit received, determined by the parameters of the earned income credit, may matter.
We examine these measures with administrative data on Minnesota's earned income credit program, the Working Family Credit (WFC), which includes information on household heads receiving welfare, receipt of the earned income credit, and earnings covered under the unemployment insurance (UI) system. The sample includes all households first observed on welfare from 1992-1999, with an additional observation for each year afterward. As discussed later, some children will become too old to be eligible for welfare or the EIC, which implies that the "parents" in our sample sometimes include households that claim zero qualifying dependents for the state earned income credit. Our household-level data include over one million observations, capturing the wide range of circumstances that face current and former Minnesota welfare recipients. Although we examine a state earned income credit, readers may find our analysis relevant to the federal earned income credit, although effects on utilization of state credits will differ from those at the federal level since migration across states affects only state credits.
To summarize our results, we find that participation percentages for current and former welfare recipients are lower than the participation percentages for all taxpayers under the federal EIC, which may be partly due to a higher concentration of lower earnings among current welfare recipients. In turn, receipt percentages are substantially lower than participation percentages for all current and former welfare recipients, which is in part due to the number of ineligible parents who have moved out of the state, have no earnings or have too much in earnings. In any given year, 40 percent or less of current and former welfare households in our dataset receive the state earned income credit. However, a large proportion of parents that successfully transition from welfare to work eventually participate in the earned income credit, which implies state funding is transferred from welfare to tax credits. Examining such welfare transitions reveals that some former recipients eventually find themselves with earnings too high to be eligible for the earned income credit. Those less likely to transition quickly from welfare to earned income credits and to income levels above the earned income credit eligibility limits are non-whites, younger parents, and those without a high school diploma, factors that may be related to barriers to work. Further examination finds that expansions to state earned income credits may have raised credit utilization rates, effectively reaching more welfare recipients and providing further incentives to work. However, even with a large micro-level dataset, the results are less than perfectly robust.
LITERATURE REVIEW
Researchers find that state and federal earned income credits may contribute to a greater likelihood that parents will transition from welfare to work. Some find that increases in federal and state earned income credits increase the employment rate of single mothers (Eissa and Liebman, 1996; Meyer and Rosenbaum, 2000 and 2001; Ellwood, 2000; Blank and Schmidt, 2001). (1) Grogger (2003) finds that more generous earned income credits contributed to lower welfare caseloads in the 1990s. Using data from 1986-1995, Neumark and Wascher (2001) report a positive correlation between higher earned income credits and the proportion of parents with incomes above the poverty guidelines.
Research on the impact of earned income credits on employment rates and welfare caseloads raises the question of how many welfare recipients get earned income credits. With respect to the receipt percentage, Hirasuna and Stinson (2007a, 2007b) and Hotz and Scholz (2003) use administrative data from Minnesota and California and find that between 20-55 percent of welfare recipients received state or federal earned income credits between 1992-1999, with higher receipt percentages in later years. In this analysis, we examine all current and former welfare recipients because some may never use the earned income credit but rely on their welfare grant, child support, other income, and in-kind assistance. Because earned income credits extend benefits to those who transition from welfare to work, we contribute to the literature by examining receipt rates after parents transition away from welfare.
Several researchers have examined the participation percentage for the entire eligible population; Scholz (1994) matches Survey of Income and Program Participation (SIPP) data with federal income tax information and estimates that 80-86 percent of eligible taxpayers participated in the earned income credit in 1990. Liebman (1998) combines 1990 federal income tax returns and 1991 Current Population Survey (CPS) data and finds participation percentages equal to 70 percent over the phase-in segment of the credit and 83-86 percent in the maximum and phase-out segments of the credit. (2) In a study for the U.S. Internal Revenue Service, SB/ SE Research (2002) uses 1996 SIPP data and estimates that 83 percent of eligible taxpayers filed a federal tax return and consequently may have received the earned income credit. Finally, the U.S. General Accounting Office (GAO) (2001), using CPS data, estimates that 75 percent of eligible households claimed the federal earned income credit in 1999.
Researchers generally find that public assistance recipients participate at lower rates than those for the larger eligible population reported in the previous paragraph. Hill et al. (1999) match a weighted sample of California welfare recipients with federal income tax records for the years 1993 and 1994 and find that between 42-54 percent of single parents in California on AFDC and eligible for the federal earned income credit actually receive it. For married parents, they estimate between 61-84 percent of AFDC parents participate in the credit. SB/SE Research (2002) also report lower participation percentages of approximately 65 percent for public assistance recipients and 71 percent for food stamp recipients, where they defined participation as eligible recipients who file an income tax return. Scholz (1994) includes a number of demographic and institutional characteristics in a regression analysis of participation rates and finds that taxpayers on public assistance are less likely to receive the federal earned income credit. Also, taxpayers who receive Social Security, who have larger families, who are unmarried, who are of Hispanic origin, who have a smaller earned income credit, or who live in states without a state income tax are less likely to receive the credit.
Our research adds to the literature by more completely characterizing: (1) the difference between receipt and participation rates, (2) parents' transitions from welfare to earned income credits or other outcome paths, and (3) whether expansions in earned income credit parameters raise the probability of receipt and...
|
