Anti-Poverty Design: The Cash-Out Option
The poverty debate largely ignores the success of the Earned Income Tax Credit and, by extension, the insights it might hold for future policy design.
Published June 1, 2003 | June 2003 issue
Few policy topics generate as much passion, disagreement and even some cognitive dissonance as povertyor more accurately, anti-poverty.
For as long as U.S. policymakers have sought to eradicate poverty, debate has raged over the relative effectiveness of various anti-poverty programs. Lyndon Johnson's War on Poverty was considered a mixed success at best. Social Security, on the other hand, has clearly prevented a great deal of poverty among the elderly.
Whatever their faults, antipoverty programscombined with a growing economyhave had a dramatic effect. Poverty rates have dropped significantly in recent decades, declining by half from 1959 to 2000 for all families, according to the Bureau of the Census, and dropping from 43 percent to 29 percent for female-headed households.
During the Clinton administration, policymakers again revisited national welfare programs and in 1996 significantly reformed their provisions and, by extension, the nation's approach toward poor families and children. The debate over the 1996 reforms continues in full force today, thanks to the fact that the 1996 measures are up for reauthorization, and President Bush, Congress and various interest groups are debating changes to current programs.
But amid the argumentative din, one particular programthe Earned Income Tax Credit (EITC)has been conspicuously absent, despite having managed to accomplish many of the goals of both welfare advocates and critics. It rewards work, is well targeted to low-income households (and therefore resource-efficient) and provides string-free resources to low-income families.
Most important, as an anti-poverty program, the EITC is demonstrably effective: In 1999, for example, EITC payments were directly responsible for lifting almost 5 million peoplehalf of them childrenabove the poverty threshold, according to an analysis of census data by the Center on Budget and Policy Priorities. That's a greater impact than any other government transfer program.
The EITC is a tax code provision, administered not by the Department of Health and Human Services but by the Internal Revenue Service. And while IRS code is not generally known for its straightforward nature, it is precisely the simple and singular design of the EITC that, say analysts, results in its success as an anti-poverty program: If the biggest problem facing the poor is their lack of money, then give them more money and allow them to spend it as needed.
In all, federal EITC spending tallied about $32 billion in 2002, and supplemental state programs added another $1 billion or so onto the paychecks of low-income workers. (See a description of the EITC's payment design.) That compares with total state and federal welfare spending of $30 billion in 2001, according to a report last year by the National Council of State Legislatures.
Like its anti-poverty predecessors, the EITC's just-give-them-money approach comes with trade-offs and unintended consequences. But a substantial body of research suggests that no other program can claim the anti-poverty success and administrative efficiency of the EITC when it comes to poor families. In a paper on the political history of the EITC, Dennis Ventry of the Brookings Institution wrote, "The overwhelming majority of economic evidence suggests that the EITC constitutes a uniquely effective and viable anti-poverty program."
A winning design?
In the last several years, an armful of studies has shown that the EITC has had a large, positive effect on reducing welfare use and increasing employment among single parents. In a 2001 paper for the National Bureau of Economic Research (NBER), Jeffrey Grogger of the University of California, Los Angeles, found that the EITC was a "particularly important contributor to both the recent decrease in welfare use and the recent increase in employment, labor supply and earnings."
Specifically, he found that the EITC explained as much as one-third of the decline in welfare use from 1993 to 1999 and close to half of the increase in employment. EITC expansions in the 1990s "have had substantial effects on almost all dimensions of behavior. The EITC may be the single most important policy measure for explaining the rise in work and earnings among female-headed families in recent years."
Follow-up research published this year by Grogger is particularly notable because he investigated factors behind both entries into and exits out of welfare. While this might sound obvious, the vast majority of research looks only at welfare exits, treating the entire welfare pool as "lifers" as it were, rather than as a mixed, mostly transient pool where some recipients are long term but most come and goall for different reasons.
By disaggregating the welfare population, Grogger found that much of the decline in welfare caseloads was driven by a reduction in welfare entries and reentries, which were heavily influenced by the EITC, the economy and the decline in welfare benefits. Exits, on the other hand, were affected most by the economy and welfare reform (like time limits and lower benefit levels), though the EITC appears to have had a small influence here as well.
Other recent research reinforces the particular influence of the EITC-among a crowd of potential factors-on moving welfare recipients into the workforce. Bruce Meyer of Northwestern University and Dan Rosenbaum of the University of North Carolina at Greensboro, writing in The Quarterly Journal of Economics in 2001, stated that "financial incentives have powerful effects on single mothers' employment decisions." They attributed about 60 percent of the increase in work by single mothers from 1984 to 1996 "to the EITC and other tax changes, with smaller shares for welfare benefits cuts, welfare waivers, training programs and child care programs."
Looking at an anti-poverty demonstration project in four California counties for the Joint Center for Poverty Research, V. Joseph Hotz, Charles Mullin and John Karl Scholz (2001) found "striking, positive effects of the EITC on employment." Evidence suggested that the EITC "has played an important role in increasing the employment rates among low-skilled workers," particularly those that are current or past welfare recipients. From 1993 to 1998, the researchers calculated, the EITC was responsible for 21 percent of the increase in labor force participation among one-child, single-parent welfare recipients, and 45 percent of the increase among single-parent households with two or more children.
That doesn't mean that the EITC has a proportional effect on the amount of work. In fact, for those already in the labor force, economic theory suggests that the EITC should have a slightly negative effect on the number of hours worked, because the phase-out portion of the program reduces the financial premium for working longer hours.
In a 1996 paper, Nada Eissa of the University of California, Berkeley, and Jeffrey Liebman at Harvard University stated that while theory predicts an expansion of EITC benefits should increase labor force participation and reduce hours worked by those already working, their empirical analysis found that only the first prediction held true when the EITC was expanded in 1987: Labor force participation did increase, but there was no reduction in hours worked. The authors noted, "compared with other elements of the welfare system, the EITC appears to produce little distortion of work incentives."
A March 2003 study by Eissa and Hilary Williamson Hoynes of the University of California, Davis, put a finer point on it: The EITC had a modest disincentive for married women to work, "effectively subsidizing married mothers to stay at home," while single women with children have been shown "to substantially increase labor supply in response to the EITC."
Indeed, it appears that the EITC's most important lever is inducing nonworkers into the labor force, which has important lasting implications. For one, it limits long-term unemployment, which has significant consequences for the job skills and other human capital of workers that determine ultimate earnings. More immediately, poor families today retain access to numerous support programs if they decide to worka major shift from the old welfare system that reduced assistance as work earnings increased.
Rebecca Blank of the University of Michigan and the NBER pointed out in a December 2002 paper, "The evidence on income and poverty suggests that most single mothers had higher incomes by the end of the 1990s, despite a loss in [direct] government assistance." Research over the last decade "unambiguously" showed that "steady work, even at a relatively low wage, when combined with EITC, with child care assistance, with access to Medicaid, with Food Stamp and Child Support assistance, would leave a woman substantially better off."
Keep it simple
Much of the success of the EITC can be traced to its simple, tax-based design. It achieves its unique objectivegetting more resources directly into the pockets of poor families, no strings attachedwith great efficiency.
High participation. As many as 85 percent of qualified individuals are believed to claim the EITC; in contrast, the Urban Institute estimated that participation among eligible families for Temporary Aid to Needy Families (the main federal welfare assistance program created in the 1996 welfare reform law) ran about 55 percent in 1998, possibly because of the bureaucratic headache of applying and receiving benefits, and partially because of the stigma associated with welfareboth of which are avoided by the EITC.
Well targeted to low-income populations. Because it subsidizes only welfare recipients who are working, the EITC could be viewed as poorly targeted to welfare populations. But in the larger view of an anti-poverty program, the EITC is well targeted because the vast majority of resources go to those with low incomes. In 1999, almost 20 million people received more than $30 billion in EITC payments. According to Hotz, Mullin and Scholz, 40 percent of payments go to workers making $6.25 or less; 80 percent go to workers with wages below $9.78.
Low administrative costs. The General Accounting Office estimates that EITC administrative costs are about 1 percent of payments. In contrast, administrative costs for traditional cash assistance programs and food stamps run as high as 15 percent, according to several sources.
Administrative costs for the EITC do go up, however, if one includes the cost of tax preparation to claim the credit. It's estimated that about $2 billion in EITC payments were diverted for tax preparation, electronic filing and high-cost loans (which put the money in filers' hands about eight to 10 days earlier than a standard tax return, but for an elevated fee). That's about 6 percent of the EITC's final payout to eligible families, according to a 2003 Brookings Institution report. But even when such costs are includedbringing total administrative costs to about 7 percentthe EITC is still more administratively efficient than traditional welfare programs.
Still not perfect
The design of the EITC involves some important tradeoffs Most importantly, as an anti-poverty program it misses those most in neednamely, those families without a working adult. But then again, other welfare programs remain in place for such families, and the inclusion of nonworking families in the EITC would likely undermine the work-oriented utility of the program. But the EITC has other drawbacks as well.
Higher cost. As noted earlier, total spending for the EITC is more than for traditional welfare programs because so many meet its eligibility requirements. Whereas about 2.5 million people are on welfare, about 20 million claim the EITC. As such, the EITC's design creates "windfall benefits" for those workers who would have been in the labor force anyway, absent the programprobably about 90 percent of all EITC recipients. But from a policy standpoint, the design is intentional: to provide incentives to help welfare recipients into the workforce; to help low-income workers avoid going on or returning to welfare; and to increase the living standards of the working poor regardless of whether they were ever on welfare.
Noncompliance. Another Achilles' heel of the EITC is compliance, making sure that only those eligible receive payment. As a tax credit, the program receives significantly less government oversight than a traditional anti-poverty program, which leaves the door wide open to fraud and other noncompliance issues. A 2002 report by the Treasury Department found that about 27 percent to 32 percent of the $31 billion paid out in 2000 "should not have been paid." (Interestingly, two-thirds of all tax returns claiming the EITC were prepared by professional tax preparers, but the IRS has found their error rates to be similar to those returns that are self-prepared.)
This isn't a new problem. The Balanced Budget Act of 1997 gave the IRS more than $700 million over five years solely to improve EITC compliance through expanded customer service and public outreach programs, better research and stronger enforcement. The IRS claims that efforts over the five-year budget period generated $5 billion in credits that were either protected (did not have to pay) or collected (from incorrect or fraudulent returns).
In its latest effort, the IRS announced in late April that it would require more conclusive proof of parental or guardian relationships (which determine EITC eligibility) from certain subgroups known to have high error rateslike fathers, grandparents and foster parents who do not meet child custody requirements but still claim the credit. The move is expected to affect as many as one in five recipients and was criticized by sources in a New York Times article as an unfair burden of proof and "wasteful allocation" of scarce IRS compliance funding, given the scale of the problem in context with other tax fraud.
What about its anti-poverty policy competitors? Does the EITC have any advantage over other programs or strategies? The two most prevalent strategies are the minimum wage and narrowly focused, end-use subsidy programs, like those for child care, transportation and job training.
The research suggests that while each of these other programs has positive attributes, none appears to offer the net results or benefitsthe reduction of poverty, the targeted and flexible use of resources and the efficiency of deliveryof the EITC. Child-care subsidy programs, for example, are widely (and accurately) lauded for helping welfare recipients find and hold jobs (though the full magnitude of their contribution is still a matter of debate). But success from this narrowly defined program comes at a considerable cost.
About $9 billion per year is spent on child care through federally funded, means-tested programs with a work requirement, according to a variety of research sources. (Close to half of that total is spent by states, but the funding comes mostly through federal block grants.) Recent estimates by several government and academic studies show that about 1.7 million children receive subsidized care. That translates to better than $5,000 per child and includes administrative costs and care services rendered. Actual reimbursement rates (as of 1999) vary widely by state, ranging from $13 a day in Louisiana (equivalent to about $3,000 annually) to $711 monthly in Maryland ($8,500 annually), according to a 2001 study by the Joint Center for Poverty Research.
Because of limited resources, only one in seven eligible children receives the childcare subsidy. By contrast, the EITC casts a broad net, and the credit is available to any eligible worker for the small price of filling out the necessary tax form. Somewhat ironically, as many as 20 percent of those believed to be eligible for the EITC fail to claim the credit.
But the EITC's real advantage comes in its flexibility and the absence of any requirements about how benefits can be used. With limited resources, poor working people face a multitude of financial challenges, including (but hardly limited to) day care, transportation, nutritional needs and health care. If child care is the most urgent need, then the EITC can help pay for such needs; if not, the money can be spent on other needs, like fixing a car.
Many policymakers and others are uncomfortable with cash subsidies, particularly the large, lump-sum payments of the EITC, where the entire credit is paid out at one time rather than tacked on weekly or monthly to a paycheck. Often the perception is that EITC recipients buy stereos or other non-necessities when they "should" be buying better day care for children, or a safer car or more nutritional meals at home.
What recipients actually do with EITC payments has not been studied much, but a 2000 analysis by Timothy Smeeding of Syracuse University, Katherine Ross Phillips of the Urban Institute and Michael O'Connor of Loyola University found that the highest priority for EITC money was bill paying, followed by larger purchases. Recipients can receive EITC payments either as a single refund check or as (limited) advance payments, and almost all choose to receive a lump sum. Both Smeeding et al. and a 2000 study by Jennifer Romich of Northwestern University and Thomas Weisner from UCLA found that recipients use the EITC as something of a forced-savings plan, used later to buy goods like furniture and entertainment but also items like cars and education that can improve economic and social mobility.
Cash payments also encourage recipients to find spending alternatives, which can help poor households retain precious dollars where possible. For example, subsidizing child care penalizes those low-income workers who have arranged informal child care. A 2001 study for the National Center for Children in Poverty found that unpaid, informal care accounts for almost half of all child care used by families with an employed mother and preschool child.
Substantial increases in childcare subsidies might convince low-income mothers using informal caregivers (like grandmothers) to seek program subsidies for formal day care. This would increase a program's use and cost, without necessarily creating any reciprocal benefits: The child is better off only if placed in a better care environment (itself a matter of some interpretation), and the subsidies do not increase the employment of a mother already in the workforce. Day care can also be difficult to find for nontraditional work schedules (like third shift or weekend shifts), which precludes such families from receiving the subsidy because it typically goes to the licensed care provider.
Available evidence on the effect of the minimum wage on poverty suggests that it can't compete with the EITC as an anti-poverty program on several fronts. For starters, at its maximum credit for a family with at least two children, the EITC acts as a wage subsidy of $2 an hour for a full-time worker, and even more for those working part timewell above any minimum wage increases being proposed.
(In fact, one would suspect that employers and their associations would be champions of the EITC, given their disdain for minimum wage increases; but calls to several business associations indicated that the EITC has little visibility among employers.)
Probably more important, from a design standpoint, the minimum wage is fraught with flaws. For example, minimum wage advocates tout its cost savings to government. But those costs ultimately are borne by employers as an implicit tax, which has other negative repercussions. Research has repeatedly shown that increases in the minimum wage lead to overall employment losses. The minimum wage is also poorly targeted, with benefits reaped by teenage workers, second-income earners and other workers in households that might be far from poor.
A 2001 report for the Employment Policies Institute looked at the effect of minimum wage increases on full-time workerswho would see the largest financial gainand found no poverty-reducing effect, probably because many workers are already in the upper part of the affected wage range. Said EPI, "[M]inimum wages, because of inefficient targeting of the poor and unintended consequences on employment and earnings, are ineffective as an anti-poverty device. ... There is virtually no meaningful evidence that higher minimum wages reduce poverty in the United States."
Research in 1999 by David Neumark of Michigan State University and William Wascher of the Federal Reserve Board of Governors indicated that minimum wages increase the net number of poor families because employment loss "tends to exceed the number lifted out of poverty by the higher minimum [wage]." They concluded that the EITC "is much more likely to raise the earnings of poor families than is the minimum wage." The EITC's influence on poverty was stronger mostly because of its ability to induce nonworkers into the workforce and reduce the negative economic consequences of long-term unemployment.
A political mutt, and that's good
The development and expansion of the EITC stem largely from the fact that it has been adapted to conform to changing political breezes. Most important, its underlying make-work-pay philosophy embodies both anti-poverty and anti-welfare characteristics. Even some of the strongest criticisms of the make-work-pay movement are being questioned.
For example, many critics have argued that pushing single parents into the workplace can compromise a child's development. But a March 2003 study funded by the National Institutes of Health found that mothers' transition from welfare to work "did not seem to have any negative effects on preschoolers or young adolescents." In general, the research found that preschoolers were unaffected by whether their mothers went on welfare, left welfare or entered or left the workplace. In fact, young adolescents showed slightly reduced levels of anxiety when their moms went to work. In general, the study suggested that each path had positive and negative aspectslike increased income, but less family timethat might ultimately cancel each other out.
Others have predicted that opinions about the work-first model might change when the economy headed south. In fact, Grogger's research indicates that the EITC's effectiveness in reducing initial entry and reentry into welfare, along with other factors, "might explain why caseloads have remained roughly constant despite the increase in unemployment" since the recession that started in March 2001.
What may be most surprising about the current welfare reform debate is that none of the proposals in play makes any mention of the EITC, despite the fact that research has shown it has a significant effect not only on getting welfare recipients into work and keeping them there, but also on helping poor families out of poverty. That's not to say the EITC is the end of poverty, but its track recordparticularly compared with other anti-poverty programsis undeniably strong. It's one of the few programs that puts significant resources directly into the hands of poor familiesand a lot of themand does so at a small administrative cost.
To evaluate alternative strategies for achieving similar policy objectives, one has to compare their benefits and costs. As a tax-based program, the EITC can claim high participation, good targeting to low-income families, low bureaucracy and unique flexibility when it comes to how the resources are actually used by recipients. Traditional welfare programs, on the other hand, can boast pinpoint targeting with few windfall beneficiaries and little fraud, but they carry bureaucratic baggage in their delivery and eligibility, and have significantly higher average costs per recipient.
Judgments about the success of various anti-poverty programs depend on the relative weights policymakers place on the benefits and costs of each strategy, and the debate over the best means of reducing poverty is therefore a difficult and contentious one.
But as Eissa and Liebman stated, "If policymakers want to redistribute income to the working poor and are comfortable with the tradeoffs involved in using the tax system rather than the welfare system to administer transfers, the EITC seems to be a way to do so with minimal efficiency costs."
Berube, Alan. 2003. Rewarding Work Through the Tax Code: The Power and Potential of the Earned Income Tax Credit in 27 Cities and Rural Areas. Center on Urban and Metropolitan Policy, Brookings Institution.
Blank, Rebecca. 2002. Evaluating Welfare Reform in the United States. Journal of Economic Literature 40 (4): 1105-1166.
Eissa, Nada, and Liebman, Jeffrey. 1996. Labor Supply Response to the Earned Income Tax Credit. The Quarterly Journal of Economics 111 (2): 605-637.
Grogger, Jeffrey. 2001. The Effects of Time Limits and Other Policy Changes on Welfare Use, Work and Income Among Female-Headed Families. National Bureau of Economic Research.
_____. 2003. Welfare Transitions in the 1990s: The Economy, Welfare Policy and the EITC. National Bureau of Economic Research.
Hotz, V. Joseph; Mullin, Charles; and Scholz, John Karl. 2001. The Earned Income Tax Credit and Labor Market Participation of Families on Welfare. Joint Center for Poverty Research.
Johnson, Nicholas. 2001. A Hand Up: How State Earned Income Tax Credits Help Working Families Escape Poverty in 2001. Center on Budget and Policy Priorities.
Meyer, Bruce, and Rosenbaum, Dan. 2001. Welfare, the Earned Income Tax Credit, and the Labor Supply of Single Mothers. The Quarterly Journal of Economics 116 (3): 1063-1113.
Neumark, David, and Wascher, William. 2001. Using the EITC to Help Poor Families: New Evidence and a Comparison with the Minimum Wage. National Tax Journal 54 (2): 281-317.
Romich, Jennifer, and Weisner, Thomas. 2000. How Families View and Use the EITC: Advance Payment versus Lump Sum Delivery. National Tax Journal 53 (4): 1245-1266.
Smeeding, Timothy; Phillips, Katherine Ross; and O'Connor, Michael. 2000. The EITC: Expectation, Knowledge, Use and Economic and Social Mobility. National Tax Journal 53 (4): 1187-1209.
Vedder, Richard, and Galloway, Lowell. June 2001. Does the Minimum Wage Reduce Poverty? Employment Policies Institute.
Ventry, Dennis Jr. 1999. The Collision of Tax and Welfare Politics: The Political History of the Earned Income Tax Credit. Brookings Institution.