Falling through the safety net

Low-income single mothers in the jobless recovery

by Jeff Chapman and  Jared Bernstein

When President Clinton signed the welfare reform bill that created Temporary Assistance for Needy Families (TANF) in 1996, the country was poised to benefit from one of the strongest labor markets in decades. Now Congress is preparing to reauthorize TANF, but rather than being on the brink of robust labor market growth, the country is mired in a jobless recovery.

Because of this weak job market, low-income single mothers are having a much harder time finding work than they did in the first four years of welfare reform.¹ The unemployment rate for low-income single mothers has risen more than the overall rate, and their real incomes—low to begin with—have fallen. Historically, when the economy hit a recession, the social safety net caught those families with few resources who were exposed to the vicissitudes of a downturn. Yet the national TANF caseload has not increased, and low-income single mothers actually received less from public assistance, on average, in 2001 than in 2000.2 While the U.S. unemployment insurance system picked up many job losers (Primus 2002), it is unlikely that former welfare recipients made it onto those rolls (Boushey and Wenger 2003).

Yet the debate over TANF reauthorization continues to pay insufficient attention to the facts about the labor market and the safety net’s inadequate response to the needs of low-income, mother-only families. First, it appears clear that TANF is not functioning effectively in the current recession, as public assistance income has fallen even while jobless rates among the target population have risen sharply. Second, rather than recognizing that the current labor market is less able to absorb TANF recipients, the bill that recently passed the House of Representatives imposes stricter work requirements on participants. Instead of fixing the holes in the safety net that have become apparent in the recession, the bill raises the barriers to participation in TANF and provides inadequate funding to meet the growing need for income support. And when the economy eventually revives and unemployment begins to fall, this bill will still exacerbate the difficult situation faced by low-income single mothers.

The data reveal that low-income single mothers are struggling in the current labor market and the safety net is not responding:

• After falling faster than the national unemployment rate over the boom, the jobless rate of low-income single mothers rose faster than the overall rate between 2000 and 2002, averaging 12.3% last year.
• As the labor market weakened and the real annual earnings of low-income single mothers declined by $343, the safety net failed to adequately respond. Nationally, TANF caseloads have not gone up, and income from public assistance, primarily TANF, fell, on average, by $137 for this group in 2000-01 (most recent available data).
• Due to an increase in paid work for low-income single mothers, the Earned Income Tax Credit has become a more important source of income than welfare for this group. But as earnings fell, so did this group’s tax credit (i.e., the EITC functioned in a pro-cyclical manner in the downturn).
• The jobless recovery has led to job losses in key sectors where low-income single mothers work. Vacancies in retail trade, for example, fell by over 20% last year, and continue to decline.
• The key to the earnings gains of low-income single mothers in the latter 1990s was the full employment labor market that prevailed, as the unemployment rate dropped to 4% in 2000. But forecasts reveal that, even when the economy recovers, the unemployment rate is unlikely to fall below 5%. If so, TANF recipients will be hard-pressed to meet the proposed new work requirements, despite their best efforts.

The ups and downs of employment among single mothers
Driven by the push of welfare reform and the pull of the strong labor market, the share of low-income single mothers that were employed grew quickly in the latter half of the 1990s, from 59.1% to 68.5% between 1995 and 2000. As employment rates have declined over the last two years, some policymakers have argued for the need to raise the work requirements on TANF participants, as done in the House bill. The implication is that these recent declines are caused by inaction by state welfare administrators and by low-income women deciding to work less. In fact, there is no evidence that work effort has declined. On the contrary, the recent decline in employment rates is driven by weak demand for labor, and not by personal decisions to work less or less push from state programs. In this economic climate, ratcheting up work requirements makes no sense.

In fact, low-income single mothers have proven to be highly responsive to the increase in demand for low-wage labor, and they were among the many beneficiaries of the booming economy of the latter 1990s. Of the 12.2 million jobs added to the economy between 1996 and 2000, over three-fifths (7.7 million) were in the two industries most likely to hire former TANF participants—retail trade and services.

These conditions led to a marked decline in the jobless rate of low-income single mothers (Figure A). Nationwide, the overall unemployment rate fell 1.4 percentage points from 5.4% in 1996 to 4.0% in 2000. While the unemployment rate of low-income single mothers remained well above the national average, it fell by 3.8 percentage points—more than twice that of the overall rate—from 13.6% in 1996 to 9.8% in 2000.

When the boom ended, the nation’s unemployment rate rose from 4.0% to 5.8% from 2000 to 2002. Despite weakening economic conditions, low-income single mothers have continued endeavoring to be employed; the share of this group in the labor force fell only slightly from 2000 to 2002.3 However, they found a much less accommodating labor market, and their unemployment rate rose to 12.3% (see Figure A), undoing much of the progress made in previous years. 4

Many economists believe that the recession ended in 2001, yet the first year of recovery has done little to improve job prospects for low-wage workers. Between 2001 and 2002, the number of unemployed low-income single mothers grew by 19.2%, as shown in Figure B. At the same time, the number of job openings in retail trade and services declined by 22.5% and 16.1%, respectively (Figure B).

The labor market downturn won’t last forever, but few expect the labor market to become as tight as in the late 1990s. In the Bush Administration budget for fiscal year 2004, the Office of Management and Budget predicts that the nation’s unemployment rate will remain above 5% through 2008.

A future of slower economic growth poses a problem for policymakers who focus on growth in employment as a measure of success for TANF. During the strongest labor market in decades, the unemployment rate of low-income single mothers fell barely below 10%. It is highly unlikely that even the most stringent work requirements will lead to lower unemployment, especially in a weaker economy. In fact, if the Bush Administration’s projections are correct and the relationship with the national unemployment rate remains the same, it is unlikely that the unemployment
rate of single mothers will fall below 12% in the next five years.⁵

Family income and components
Rising unemployment has had a direct impact on the incomes of low-income single mothers. Between 2000 and 2001, the most recent year for which data are available, the average family income of low-income single mothers fell by $484, after adjusting for inflation. While data for 2002 are not yet available, we do know jobs became scarcer and wage growth slowed, suggesting that incomes continued to fall.

Welfare reform increased the emphasis placed on three sources of income that appear to rise when the economy is growing and potentially fall when the economy is in a slump—earnings from work, the Earned Income Tax Credit (EITC), and child support payments. From 1996 to 2000, all three sources of income rose, on average, for low-income single mothers, contributing to an annual overall increase in average income of about $817 (see Figure C and Table 1).

But from 2000 to 2001, income from all three sources fell. Earnings from work can be expected to fall because of rising unemployment and slower wage growth. Because rising unemployment also affects non-custodial fathers, child support may also fall. And since the amount one receives from the EITC is tied to earnings, it can also fall as earnings decline.⁶ In 2001, low-income single mothers received $343 less, on average, from earnings, $75 less from the EITC, and $61 less from child support.

Figure C also shows changes in income from the major safety net programs one would rely on to make up for some of the loss in earnings during a recession—unemployment compensation, public assistance (primarily TANF), and Food Stamps.

Since earnings fell primarily due to rising unemployment, one may expect that the unemployment insurance program (UI) would pick up some of the slack. The average unemployment compensation for low-income single mothers did grow from 2000 to 2001, but only by about $66, not nearly enough to offset the losses from the pro-cyclical income supports. Wenger and Boushey (2003) argue that UI is unlikely to meet the needs of this group because their low earnings and lack of job tenure make many of them ineligible for benefits.

Since it unlikely that UI is filling in for much of the income lost from wages and the EITC, an important question is whether or not TANF is acting as a safety net. In fact, even though the need increased, average public assistance dollars to low-income single mothers actually fell by $137 from 2000 to 2001 (Table 1). While this is a lesser decline than the steep loss of TANF income that occurred over the four previous years (over $300 annually), it nevertheless reveals a flaw in the current safety net, as TANF income should have increased over the downturn.

The average value of Food Stamps also fell, though only slightly, and housing subsidies remained the same. On net, the value of safety net programs fell by $74 in 2001, as the increase in unemployment compensation failed to offset the losses in the other safety net components, not to mention the loss of earnings, EITC, and child support (Figure C). Clearly, our weakened support system failed to adequately lift the incomes of low-income single mothers hurt by the recession.

Income from other sources grew between 2000 and 2001, offsetting some of the losses from the sources mentioned above, though not offsetting the decline in overall family income. This increase is primarily driven by increases in Social Security and Supplemental Security Income, programs that are not typically considered part of the social safety net that protects workers from recession (Table 1).

Conclusion
As the Senate moves forward with TANF reauthorization, it is vital that policymakers understand that the low-wage labor market in 2003 and beyond is fundamentally different than the labor market of the latter 1990s. While low-income single mothers continue to make a valiant effort to get and keep jobs, at the end of 2002 there were 3.2 unemployed workers for every job opening, compared to 1.3 at the end of 2000 (these ratios refer to all persons, not just single mothers).⁷ The downturn and sputtering recovery significantly amplify the demands on state agencies with the task of helping people move from welfare to work. In addition, the fiscal crises faced in many states have led them to cut back on programs designed to support those leaving welfare (see Johnson et al. 2003). Rather than focusing reauthorization efforts on ratcheting up work requirements, the debate should focus on fixing the holes in the safety net and ensuring that those who need help the most have access to public assistance.

EPI’s living standards program gratefully acknowledges the support of the Charles Stewart Mott Foundation, the John D. and Catherine T. MacArthur Foundation, the Foundation for Child Development, the Public Welfare Foundation, the Rockefeller Foundation, and the Joyce Foundation. We received helpful comments from Wendell Primus, Mark Greenberg, Elise Richer, and Arloc Sherman.

Bibliography
Boushey, Heather and Jeffrey Wenger. 2003. In the Policy Nexus:Unemployment Insurance Eligibility Before and After Welfare Reform. Forthcoming working paper. Washington, D.C.: Economic Policy Institute.

Johnson, Nicholas, Iris J. Lav, and Rose Ribeiro. 2003. States Are Making Deep Budget Cuts in Response to the Fiscal Crisis. Washington, D.C.: Center on Budget and Policy Priorities
(http://www.cbpp.org/3-19-03sfp.htm).

Primus, Wendell. 2002. How Will the Safety Net Perform in this Recession? Unpublished presentation at the 2002 APPAM Annual Conference.

Endnotes
1. For income analysis, as in Table 1, we define low-income single mothers as those age 20-55, with incomes in the bottom half of the income-to-needs distribution (the source for these data is the March Current Population Survey (CPS)). This distribution is derived by dividing family income by the poverty threshold. Thus, it is a measure of income controlling for family size. Employment analysis comes from the monthly CPS, which does not include detailed information on income. Respondents are asked to estimate their annual income, and we select single mothers (same age restriction as above) who report annual incomes of less than $25,000.

2. Data on income from public assistance beyond 2001 are not yet available. However, based on flat TANF caseload trends through the first three-quarters of 2002, it is unlikely that average public assistance income increased in 2002. Note that new data from CLASP do show national caseloads about 1% higher in December 2002 relative to September 2002 (not seasonally adjusted). But year over year, the national TANF caseload was 2% lower in December 2002 than a year earlier.
(see http://www.clasp.org/DMS/Documents/1049386524.1/doc_caseload_2002_Q4.pdf)

3. The labor force participation rate of low-income single mothers fell from 76% in 2000 to 74.6% in 2002.

4. The unemployment rates for all single mothers, age 20-55, fell 3.4 points, from 10.2% to 6.8% between 1996 and 2000, and rose 2 points, from 6.8% in 2000 to 8.8% in 2002.

5. The ratio of low-income single mothers’ unemployment to overall unemployment from 1996 to 2002 was 2.4. This suggests that, if the nation’s unemployment rate falls to a low of 5.1% (“full employment,&
rdquo; according to the Office of Management and Budget), the unemployment rate of low-income single mothers will be around 12%.

6. Obviously, those who lose their job will lose their tax credit, but for those whose earnings decline (due to lower hourly wages or fewer hours), the amount of their credit can rise or fall, depending on where they are located on the EITC schedule.

7. These ratios come from Bureau of Labor Statistics JOLTS (job vacancies) and unemployment data.