This article is from Dollars & Sense: Real World Economics, available at

issue 298 cover

This article is from the January/February 2012 issue of Dollars & Sense magazine.

Subscribe Now

at a 30% discount.

Different Anti-Poverty Programs, Same Single-Mother Poverty

Fifteen Years of Welfare Reform

By Randy Albelda

Four years into a period of deep recession and persistent economic crisis, only now has the p-word—poverty—finally surfaced. The Census Bureau’s September 13 announcement that the U.S. poverty rate had increased to 15.1% in 2010, up from 14.3% in 2009, put the issue of poverty onto page one, albeit briefly. In fact, poverty and how to address it have not been prominent items on the national agenda since the “welfare reform” debates of the 1980s and early 1990s.

“Welfare queens” may have disappeared from politicians’ rhetoric, but poor people, disproportionately single mothers and their children, are still around. Single-mother families have been and continue to be particularly vulnerable to being poor. The September report showed the poverty rate for single mothers and their children rose as well: from 32.5% in 2009 to 34.2% in 2010.

It is remarkably hard to be the primary caregiver and garner enough income to support a family. This reality was built into the design of the first generation of federal anti-poverty programs in the United States. Developed beginning in the New Deal era, these programs were aimed at families with no able-bodied male breadwinner and hence no jobs or wages—single mothers, people with disabilities, and elders. Putting single mothers to work was thought to be undesirable. Or, white single mothers—there was much less reluctance in the case of black single mothers, who were largely excluded from the various anti-poverty programs until the 1960s.

The most important of the anti-poverty programs for single mothers was the cash assistance program, Aid to Dependent Children (later renamed Aid to Families with Dependent Children, or AFDC), established in 1935—also commonly referred to as “welfare.” Other programs developed in the succeeding decades included Food Stamps, Medicaid, and housing assistance.

Then, in 1996, with the support of President Clinton, Congress abolished AFDC, replacing it with a block grant called TANF (Temporary Assistance to Needy Families), and passed a spate of other changes to related programs. The new anti-poverty regime implied a new social compact with the non-disabled, non-elder poor, supported by both conservatives and liberals: to require employment in exchange for—and ultimately be weaned off of—government support. In other words, the new mandate for poor adults, especially single mothers, was to get a job—any job.

And, in fact, in the ensuing years the number of poor families with wages from work increased. Moreover, welfare rolls dropped. And, in the first four years following welfare “reform,” the official poverty rate for single-mother families fell too. (It has been increasing since 2000, although not quite back to its 1996 level.) But despite their higher wage income, many single-mother families are no better able to provide for their basic needs today than before the mid-1990s. Even the lower poverty rate may not reflect the real material well-being of many single moms and their children, given that their mix of resources has shifted to include more of the kinds of income counted by poverty measures and fewer of the uncounted kinds.

While TANF and the other legislative changes promote employment in theory, they did not reshape anti-poverty programs to genuinely support employment. Key programs are insufficiently funded, leaving many without access to child care and other vital work supports; income eligibility requirements and benefit levels designed for those with no earnings work poorly for low-wage earners; and the sheer amount of time it takes to apply for and keep benefits is at odds with holding down a job.

Ironically, there has been little or no talk of revisiting these policies despite the massive job losses of the Great Recession. With job creation at a standstill, in 2010 the unemployment rate for single mothers was 14.6% (more than one out of every seven). For this and other reasons it is time to “modernize” anti-poverty programs by assuring they do what policy makers and others want them to do—encourage employment while reducing poverty. And they must also serve as an important safety net when work is not available or possible. But changes to government policies are not enough. If employment is to be the route out of poverty, then wages and employer benefits must support workers at basic minimum levels.

Ending “Welfare,” Promoting Employment

Among the changes to U.S. anti-poverty programs in the 1990s, the most sweeping and highly politicized involved AFDC, the cash assistance program for poor parents. The 1996 legislative overhaul gave states tremendous leeway over eligibility rules in the new TANF program. For the first time there was a time limit: states are not allowed to allocate federal TANF money to any adult who has received TANF for 60 months—regardless of how long it took to accrue 60 months of aid. And the new law required recipients whose youngest child is over one year old to do some form of paid or unpaid work—most forms of education and job training don’t count—after 24 months of receiving benefits.

To accommodate the push for employment, Congress expanded the Earned Income Tax Credit, which provides refundable tax credits for low-income wage earners; expanded the Child Care Development Block Grant, which gives states money to help provide child care to working parents with low incomes, including parents leaving TANF; and established the State Children’s Health Insurance Program (S-CHIP), in part out of a recognition that single mothers entering the workforce were losing Medicaid coverage yet often working for employers who provided unaffordable health insurance coverage or none at all. Even housing assistance programs started promoting employment: the Department of Housing and Urban Development encouraged local housing authorities to redesign housing assistance so as to induce residents to increase their earnings.

The strategy of promoting employment was remarkably successful at getting single mothers into the labor force. In 1995, 63.5% of all single mothers were employed; by 2009, 67.8% were. This rate exceeds that of married mothers, at 66.3%. So with all that employment, why are poverty rates still so high for single-mother families? The answer lies in the nature of low-wage work and the mismatch between poverty reduction policies and employment.

Single Mothers and Low-Wage Jobs Don’t Mix

There are two fundamental mismatches single mothers face in this new welfare regime. The first has to do with the awkward pairing of poor mothers and low-wage jobs. In 2009 over one-third of single mothers were in jobs that are low paying (defined as below two-thirds of the median hourly wage, which was $9.06). In addition to the low pay, these jobs typically lack benefits such as paid sick or vacation days and health insurance. Many low-wage jobs that mothers find in retail and hospitality have very irregular work hours, providing the employers with lots of flexibility but workers with almost none. These features of low-wage work wreak havoc for single moms. An irregular work schedule makes child care nearly impossible to arrange. A late school bus, a sick child, or a sick child-care provider can throw a wrench in the best-laid plans for getting to and staying at work. Without paid time off, a missed day of work is a missed day of pay. And too many missed days can easily cost you your job.

Medicaid, the government health insurance program for the poor, does not make up for the lack of employer-sponsored health insurance common in low-wage jobs. Medicaid income eligibility thresholds vary state by state, but are typically so low that many low-wage workers don’t qualify. Only 63% of low-wage single mothers have any health insurance coverage at all, compared to 82% of all workers. The new Patient Protection and Affordable Care Act (a/k/a Obamacare) may help, depending on the cost of purchasing insurance, but for now many low-wage mothers go without health care coverage.

Finally, there is the ultimate reality that there are only 24 hours in a day. Low wages mean working many hours to earn enough to cover basic needs. Yet working more hours means less time with kids. This can be costly in several ways. Hiring babysitters can be expensive. Relying heavily on good-natured relatives who provide care but may not engage and motivate young children also has costs, as does leaving younger children in the care of older brothers and sisters, who in turn can miss out on important after-school learning. Long work hours coupled with a tight budget might mean little time to help do homework, meet with teachers, or participate in in- and out-of-school activities that enrich children’s lives.

Poverty Remeasured

According to the Census Bureau, 46.2 million Americans were poor in 2010. But what exactly does “poor” mean? The academic and policy debates over how to measure poverty fill volumes. Some questions relate to the establishment of the poverty threshold. On what basis should the poverty line be drawn? Is poverty relative or absolute—in other words, if the average standard of living in a society rises, should its poverty threshold rise as well? Other questions concern measuring income. What kinds of income should be counted? Before or after taxes and government benefits? Who is included in the poverty assessment? (For example, those in institutional settings such as prisons are excluded from the official U.S. poverty measure—not a minor point when you consider that nearly 2.3 million adults were incarcerated in the United States as of the end of 2009.)

Established in 1963 by multiplying an emergency food budget by three, and adjusted solely for inflation in the years since, the official U.S. poverty thresholds are notoriously low. A family of four bringing in over $22,314—including any TANF cash assistance, unemployment or workers’ comp, Social Security or veterans’ benefits, and child support—is not officially poor. In many parts of the United States, $22K would not be enough to keep one person, let alone four people, off the street and minimally clothed and fed.

An interagency federal effort to develop a more realistic poverty level has just released its new measure, known as the Supplemental Poverty Measure. The SPM makes many adjustments to the traditional calculation:

  • It counts the Earned Income Tax Credit and non-cash benefits such as food stamps and housing assistance as income.
  • It subtracts from income out-of-pocket medical costs, certain work-related expenses (e.g., child care), and taxes paid.
  • Its thresholds are adjusted for cost-of-living differences by region and are relative rather than absolute—basic expenses that are the building blocks of the threshold are pegged at the 33rd percentile of U.S. households.

The SPM poverty rate for 2009 was 15.7%, somewhat higher than the 14.5% official rate. More dramatic differences between the two poverty rates appeared in some subgroups, especially the elderly: 9.9% by the traditional measure versus 16.1% by the SPM, largely due to their high out-of-pocket medical expenses.
—Amy Gluckman

Sources: “Measure by Measure,” The Economist, January 20, 2011; Ellen Frank, “Measures of Poverty,” Dollars & Sense, January 2006; Center for Women’s Welfare, Univ. of Wash. School of Social Work, “How Does the Self-Sufficiency Standard Compare to the New Supplemental Poverty Measure?”; U.S. Census Bureau, “How the Census Bureau Measures Poverty” and “Poverty Thresholds by Size of Family and Number of Children: 2010.”

A New Mismatch

The first generation of anti-poverty programs were designed on the assumption that recipients would not be working outside the home. Unfortunately, their successor programs such as TANF and SNAP, despite their explicit aim of encouraging employment, still do not work well for working people.

What does it mean that these programs are not designed for those with employment? There are two important features. First, income thresholds for eligibility tend to be very low—that is, only those with extremely low earnings qualify. For example, only two states have income thresholds above the poverty line for TANF eligibility. To get any SNAP benefits, a single mother needs to have income below 130% of the poverty line. Working full-time at $10 an hour (that’s about $1,600 a month in take-home pay) would make a single mother with one child ineligible for both programs in all states. Moreover, even if you are eligible, these benefits phase out sharply. With TANF (in most states), SNAP, and housing assistance, for every additional dollar you earn, you lose about 33 cents in each form of support. This means work just does not pay.

Second, applying for and maintaining benefits under these programs often takes a great deal of time. Each program has particular eligibility requirements; each requires different sets of documents to verify eligibility. While some states have tried to move to a “one-stop” system, most require separate applications for each program and, often, one or more office visits. Recertification (i.e., maintaining eligibility) can require assembling further documentation and meeting again with caseworkers. If you have ever applied for one of these programs, you have probably experienced how time-consuming—and frustrating—the process can be.

In short, the programs were designed for applicants and recipients with plenty of time on their hands. But with employment requirements, this is not the right assumption. Missing time at work to provide more paperwork for the welfare office is just not worth it; there is considerable evidence that many eligible people do not use TANF or SNAP for that reason. Even the benefit levels assume an unlimited amount of time. Until recently, the maximum dollar amount of monthly SNAP benefits was based on a very low-cost food budget that assumed hours of home cooking.

Unlike cash assistance or food assistance, child care subsidies are obviously aimed at “working” mothers. But this program, too, often has onerous reporting requirements. Moreover, in most states the subsidy phases out very quickly especially after recipients’ earnings reach the federal poverty line. This means that a worker who gets a small raise at work can suddenly face a steep increase in her child-care bill. (Of course, this is only a problem for the lucky parents who actually receive a child-care subsidy; as mentioned earlier, the lack of funding means that most eligible parents do not.)

The Earned Income Tax Credit is a notable exception. The refundable tax credit was established explicitly to help working parents with low incomes. It is relatively easy to claim (fill out a two page tax form along with the standard income tax forms), and of all the anti-poverty programs it reaches the highest up on the income ladder. It even phases out differently: the credit increases as earnings rise, flattens out, and then decreases at higher levels of earnings. Most recipients get the credit in an annual lump sum and so use it very differently from other anti-poverty supports. Families often use the “windfall” to pay off a large bill or to pay for things long put off, like a visit to the dentist or a major car repair. While helpful and relatively easy to get, then, the Earned Income Tax Credit does not typically help with day-to-day expenses as the other anti-poverty programs do.

Has Employment-Promotion “Worked”?

The most striking change in the anti-poverty picture since welfare reform was enacted is that the welfare rolls have plummeted. In 1996, the last full year under the old system, there were 4.43 million families on AFDC nationwide; in 2010, amid the worst labor market in decades, the TANF caseload was only 1.86 million. In fact, when unemployment soared in 2008, only 15 states saw their TANF caseloads increase. The rest continued to experience reductions. Plus, when the TANF rolls fell sharply in the late 1990s, so did Medicaid and Food Stamps enrollments. These programs have since seen increases in usage, especially since the recession, but it’s clear that when families lose cash assistance they frequently lose access to other supports as well.

Welfare reform has worked very well, then, if receiving welfare is a bad thing. Indeed, advocates of the new regime tout the rapid and steep decline in welfare use as their main indicator of its success. In and of itself, however, fewer families using anti-poverty programs does not mean less poverty, more personal responsibility, or greater self-sufficiency. During the economic expansion of the late 1990s, the official poverty rate for single mothers and their children fell from 35.8% in 1996 to 28.5% in 2000. It has risen nearly every year since, reaching 34.2% in 2010. But if a successful anti-poverty effort is measured at all by the economic well-being of the targeted families, then that slight drop in the poverty rate is swamped by the 60% decrease in the number of families using welfare over the same period. Far fewer poor families are being served. In 1996, 45.7% of all poor children received some form of income-based cash assistance; in 2009, only 18.7% did. The Great Recession pushed 800,000 additional U.S. families into poverty between 2007 and 2009, yet the TANF rolls rose by only 110,000 over this period.

Data from two federal government reports on TANF, depicted in the figure below, nicely illustrate the dilemmas of the new welfare regime. The figure shows the total average amounts of earnings and the value of major government supports (“means-tested income”) for the bottom 40% of single-mother families (by total income) between 1996 and 2005. It is clear that since welfare reform, these families are relying much more on earnings. But despite the additional work effort, they find themselves essentially no better off. The bottom 20% saw their package of earnings and government benefits fall: their average earnings have not increased much while government supports have dropped off, leaving them with fewer resources on average in 2006 than in 1996. For the second quintile, earnings have increased substantially but benefits have fallen nearly as much, leaving this group only slightly better off over the period. And that is without taking into account the expenses associated with employment (e.g. child care, transportation, work clothes) and with the loss of public supports (such as increased co-payments for child care or health insurance). These women are working a lot more—in the second quintile about double—but are barely better off at all! So much for “making work pay.”

Albelda graph

More hours of work also means fewer hours with children. If the time a mother spends with her children is worth anything, then substituting earnings for benefits represents a loss of resources to her family.

What Might Be Done?

Employment, even with government supports, is unlikely to provide a substantial share of single-mother families with adequate incomes. Three factors—women’s lower pay, the time and money it takes to raise children, and the primary reliance on only one adult to both earn and care for children—combine to make it nearly impossible for a sizeable number of single mothers to move to stable and sufficient levels of resources.

Addressing the time- and money-squeeze that single mothers faced in the old anti-poverty regime and still face in the new one will take thoroughgoing changes in the relations among work, family, and income.

  • Make work pay by shoring up wages and employer benefits. To ensure that the private sector does its part, raise the minimum wage. A full-time, year-round minimum-wage job pays just over the poverty income threshold for a family of two. Conservatives and the small business lobby will trot out the bogeyman of job destruction, but studies on minimum-wage increases show a zero or even positive effect on employment. In addition, mandate paid sick days for all workers and require benefit parity for part-time, temporary, and subcontracted workers. This would close a loophole that a growing number of employers use to dodge fringe benefits.
  • Reform anti-poverty programs to really support employment. To truly support low-wage employment, anti-poverty programs should increase income eligibility limits so that a worker can receive the supports even while earning and then phase out the programs more gradually so low-wage workers keep getting them until they earn enough not to need them. Also, streamline application processes and make them more user-friendly. Many states have done this for unemployment insurance, car registration, and driver’s license renewal. Why not do the same for SNAP, TANF and Medicaid?
  • Support paid and unpaid care work. A society that expects all able-bodied adults to work—regardless of the age of their children—should also be a society that shares the costs of going to work, by offering programs to care for children and others who need care. This means universal child care and afterschool programs. It also means paid parental leave and paid time off to care for an ill relative. The federal Family and Medical Leave Act gives most workers the right to take unpaid leaves, but many can’t afford to. California and New Jersey have extended their temporary disability insurance benefits to cover those facing a wide range of family needs—perhaps a helpful model.

New anti-poverty regime, but same poverty problems. Most single mothers cannot work their way out of poverty—definitely not without the right kinds of supplemental support. There are many possible policy steps that could be taken to help them and other low-wage workers get the most out of an inhospitable labor market. But ultimately, better designed assistance to poor and low-income families, old-fashioned cash assistance, and minimal employment standards must be part of the formula.

RANDY ALBELDA is a professor of economics at the University of Massachusetts, Boston.

SOURCES: Randy Albelda and Chris Tilly, Glass Ceilings and Bottomless Pits: Women’s Work, Women’s Poverty, South End Press, 1997; U.S. Census Bureau, Historical Tables on Poverty; Kaiser Family Foundation, “Income Eligibility Limits for Working Adults at Application as a Percent of the Federal Poverty Level by Scope of Benefit Package,”, January 2011; U.S. Dept. of Health and Human Services, TANF 6th and 8th Annual Report to Congress, November 2004 and July 2009; U.S. Dept. of Health and Human Services, Estimates of Child Care Eligibility and Receipt for Fiscal Year 2006, April 2010; Thomas Gabe, Trends in Welfare, Work, and the Economic Well-being of Female Headed Families with Children: 1987-2005, Congressional Research Service Report RL30797, 2007; Randy Albelda and Heather Boushey, Bridging the Gaps: A Picture of How Work Supports Work in Ten States, Center for Social Policy, Univ. of Mass. Boston and Center for Economic and Policy Research, 2007; Author’s calculations from the U.S. Census Bureau’s Current Population Survey, various years.

Did you find this article useful? Please consider supporting our work by donating or subscribing.

end of article