Unemployment Insurance is Not a Safety Net for
Unemployed Former Welfare Recipients[1]

Heather Boushey and Jeffrey B. Wenger[2]

 December 4, 2003

Executive Summary

            Unemployment insurance (UI) is an important source of security for the nation’s workers, providing them with an income if they are laid off from their job through no fault of their own. It is an insurance system, not a government transfer program – workers pay for UI themselves through a payroll tax.

            The stated goal of welfare reform was to get people off welfare and into work. While employment rates did rise substantially among those leaving welfare in the late 1990s, many former welfare recipients who have lost their jobs in the current labor market downturn remain ineligible for UI. In the late 1990s, nearly one-half (49.1 percent) of women who left welfare failed to meet the monetary requirements for unemployment insurance, compared to only one-of-five (20.7 percent) women who were employed but had not been on welfare. In spite of increased employment rates, and rising wages in the late nineties boom, former welfare recipients were actually less likely to qualify for UI in the late 1990s, than in the period prior welfare reform.

            This paper suggests two simple policy changes that would substantially increase eligibility rates for former welfare recipients:

1) The use of an alternate base period.  This alternative base period (which is already used by 17 states) would include the most recent quarter of work in assessing workers’ earnings history. Since most former beneficiaries have short work histories, this change can substantially increase eligibility rates for UI.

2) The use of an hours requirement instead of an earnings requirement. Since most former welfare recipients work in low paying jobs, they often have low earnings even if they work regularly. Substituting an hours requirement, such as the 400 hours per quarter level used by Oregon, will allow many more former beneficiaries to qualify for UI.

Simulations in this study show that the general adoption of the proposed alternative base period would increase eligibility for UI by 6.6 percentage points for former beneficiaries in the third quarter that they are off welfare. Switching to an hours requirement would increases eligibility by 6.1 percentage points in the third quarter off welfare.

These proposals do not change the fact that the UI benefit is earnings related – those who earn more get higher benefits. They also don’t change the fact that workers who lose their job for cause (including difficulties finding adequate child care) are not eligible for benefits. However, they will increase the probability that former beneficiaries will be eligible to collect a benefit that they have already paid for through their taxes. 

Introduction

The unemployment insurance (UI) system is supposed to provide relief for unemployed workers, in order to help them make ends meet until they find another job. However, for former welfare recipients, hard-hit by the foundering U.S. job market, the UI system does not provide them with a much-needed safety net. In the late 1990s, nearly one-half (49.1 percent) of women who left welfare failed to meet the earnings requirement for unemployment insurance, compared to only one-of-five (20.7 percent) women who were employed but had not been on welfare. Welfare reform led to increased employment, and the buoyant economy raised low-wage worker’s earnings over the 1990s; nevertheless, former welfare recipients were actually less likely to qualify for UI in the late 1990s, as compared to the early 1990s.

As unemployment remains high, states must consider what will happen as low-wage and former welfare recipients lose their jobs. Policymakers should look to expand UI coverage so that all workers – including those who work intermittently and often for low wages – are eligible for assistance if they lose their job. This paper provides two sets of simulations to demonstrate the alternatives available to state policymakers. States could change the time period over which the person’s earnings are examined from the standard base period (the four completed calendar quarters one quarter prior to the quarter of unemployment) to the alternate base period (the four completed calendar quarters prior to the quarter of unemployment). So far, 17 states have adopted an alternate base period policy. If all states did so, in the late 1990s, 55.2 percent of former welfare recipients would have been monetarily eligible for UI in their first quarter off welfare; this is nearly four percentage points higher than was actually the case. Second, states could follow Oregon’s lead and move to an hours-only requirement in lieu of an earnings requirement. If every state required individuals to have worked at least 400 hours over the past year (the “alternate base period”), then, in the late 1990s, one year after leaving welfare, UI eligibility would have fallen by half a percent in the first quarter after leaving welfare, but would have increased by 3.5 percentage points by the second quarter off welfare and 6.1 percentage points in the third quarter off welfare.

Former welfare recipients not as likely as other working women to qualify for UI

The weak recovery has led to a sustained low level of job creation. Unemployment is currently at 6.0 percent (October 2003) and 2.4 million jobs have been lost since February 2001. Individuals who lost their jobs are remaining unemployed for longer than is usually the case in a recession: currently, nearly one-in-four unemployed workers (23.0 percent – October 2003) has been unemployed for six months, indicating that for many, finding employment is not an easy task. Among workers with a high-school degree or less, unemployment averaged 8.1 percent in 2002, as compared to 6.0 percent in 2000. For young workers, those between 16 and 24, unemployment averaged 12.1 percent in 2002, up from 9.4 percent in 2000. Jobs for workers previously on welfare have become increasingly scarce, as evidenced by these increases in unemployment.

When a workers become unemployed, they typically look to the UI system to help them make ends meet until they find a new job. However, nearly half of former welfare recipients are ineligible for UI because they have low wages or intermittent work histories. Table 1 shows former welfare recipients are nearly one-third less likely to qualify for UI, compared to other women workers. In the late 1990s during their first quarter off welfare, 50.9 percent of women met the earnings thresholds for UI eligibility, while 79.3 percent of other women workers were monetarily eligible.[3] Although increased employment among welfare recipients followed the implementation of welfare reform and the buoyant economy raised low-wage worker’s earnings during the 1990s, former welfare recipients were actually less likely to qualify for UI in the late 1990s, as compared to those leaving welfare in the early 1990s. For women formerly on welfare, eligibility fell by 1.8 percentage points compared to leavers in the early 1990s, while for women who worked and were never on welfare, eligibility fell by only 0.9 percentage points. Among those not on welfare, women are less likely than men to qualify for UI, and this is more so in the late 1990s than in the early 1990s. The remainder of the paper focuses exclusively on women and women welfare leavers.
 

            That former welfare recipients are less likely to be monetarily eligible for UI relative to other women is disconcerting, especially when we consider the likelihood that many women will be ineligible for unemployment insurance benefits due to non-monetary reasons such as voluntary leaving, quits, and justified dismissals. Historically, welfare played the role of a “poor woman’s unemployment insurance” by providing income support to poor women who were not working. Now, with many poor women off welfare and working, the expectation is that these women will earn enough to receive UI benefits, rather than going back on welfare. Welfare is not an option for many low-income women because welfare reform made it more difficult for poor women to use welfare as an unemployment safety net. The 1996 Personal Responsibility and Work Opportunity Reconciliation Act – commonly known as welfare reform – made it more difficult for families to get on and stay on welfare. Congress tightened the rules for applications, added substantial work requirements, and implemented a five-year lifetime limit on receiving welfare payments. Thus, many women who left welfare will not be able to go back to welfare if they lose their job. For these families, UI should help them make ends meet until they find another job.

            It is also important to note that this is the best-case scenario in terms of eligibility. Many, if not most, former welfare recipients leave their jobs for reasons that make them ineligible for receiving UI, even if they meet the monetary requirements.[4] UI eligibility can be broken down into three components: monetary eligibility, initial eligibility and continuing eligibility. This paper focuses exclusively on monetary eligibility – that is, determining if a worker earned enough income to qualify for UI benefits. Initial eligibility is determined by the reasons for a worker’s job separation. Workers initiating a job separation (i.e. quitting) will typically be ineligible for UI. There are circumstances under which a worker may quit and still be eligible for benefits. For example, in a few states, if a worker quits as the result of sexual harassment or domestic violence she will be eligible. However, in most states, quitting because of childcare problems makes one ineligible for benefits. Continuing eligibility is determined by the effort to find work during unemployment and the acceptance of job offers. UI benefit recipients must undertake a documented job search and must accept reasonable offers of employment. Continuing eligibility is predicated on being “available” for work; the unemployed who restrict their job searches exclusively to part-time employment are not considered available in most states.

Why did UI eligibility fall over the 1990s for former welfare recipients?

            With strong labor markets in the latter 1990s, a key question of this research is why more workers, particularly those at the bottom of the earnings distribution, did not become eligible for UI. We test two hypotheses about why former welfare recipients were less eligible for UI after welfare reform than before. First, it is possible that workers who left welfare prior to welfare reform had better economic opportunities, were more motivated, or had fewer barriers to leaving. Alternatively, some of the decline in monetary eligibility could be because many states increased their monetary eligibility requirements during the late 1990s. We conclude that the reason for the decline in the proportion of former welfare recipients eligible for UI post-welfare reform is more about labor market outcomes than UI rule changes. Those who left welfare after the implementation of welfare reform had lower employment rates and lower earnings for those in the bottom third in their first few quarters off welfare compared to those who left welfare in the early 1990s, before welfare reform. Had states not raised their requirements for eligibility, more former welfare recipients (0.2 percentage points) would have been eligible for UI in the late 1990s compared to the early 1990s.

Labor market outcomes pre- and post-welfare reform

The combination of welfare reform and the strong economy led to a rapid decline in the number of families on welfare in the late 1990s. However, women who left welfare in the late 1990s had slightly lower employment rates and greater dispersion of earnings, relative to those who left in the early 1990s. Figure 1 shows employment rates for women by quarters off welfare in the early and late 1990s. Women who were never on welfare had higher employment rates in the late 1990s than in the early 1990s. However, women in their first quarter off welfare had much higher employment rates in the early 1990s than in the late 1990s.

Employment rates only tell half the story. Table 2 shows average and median earnings for women who left welfare by quarter off welfare. The decreased eligibility for former welfare recipients in the late 1990s is surprising given that the average wages of former welfare recipients increased over the 1990s. In their first year off welfare, inflation adjusted average quarterly earnings of women ad left welfare rose from $2,544 to $2,800 from the early to late 1990s – a gain of 10 percent. This increase in earnings should, all else equal, have increased monetary eligibility. However, when we look to median earnings, the story becomes more complicated: as with the employment rates, the trend for welfare leavers differs from other women. In the first quarter off welfare, inflation-adjusted median earnings were higher in the early 1990s than they were in the late 1990s and, further, earnings dispersion is larger in the late 1990s.

            Figure 3 shows how greater dispersion helps to explain lower UI eligibility in the late 1990s: for women who left welfare, a year and a half after they left, those at the 30^th percentile (including those with no earnings) who left welfare in the early 1990s earned $1,075, while those who left in the late 1990s earned only $524. We find that there are more welfare leavers with very low earnings in the late 1990s, leading to the lower qualifying rates for UI. Higher earnings dispersion is partially because there are more women with no earnings in the late 1990s

In some states, changes to UI rules made qualifying more difficult

Just after passage of welfare reform, between 1996 and 1997, nine states increased their minimum earnings requirements necessary to qualify for UI, making it more difficult for low-wage workers to meet the monetary eligibility requirements.[5] On average, these states raised their minimum earnings requirements by nearly 14 percent. Some states, such as Alabama, more than doubled the minimum earnings requirements necessary to qualify for UI benefits between 1996 and 1997. These increases in earnings requirements created significant barriers to eligibility for previous welfare recipients. Overall, from 1996 to 2001, 15 states raised their earnings requirements faster than median wages. Higher earnings thresholds mean that low-wage workers and former welfare recipients are less likely to qualify for UI. In eight states,[6] a worker employed at 20 hours per week for a full year and earning their state’s minimum wage would fail to qualify for UI. In each case, these workers would fail to meet their state’s high quarter earnings requirements (Emsellem et al. 2002).

We simulated the proportion of individuals meeting the monetary eligibility requirements in the late 1990s if the UI rules had not changed since the early 1990s. As it turns out, changes in eligibility requirements did contribute to the lower proportion of former welfare recipients qualifying for UI in the late 1990s, although the effect is very small. Table 3 shows that if the individuals in the late 1990s had faced the same UI rules as those in the early 1990s, then UI eligibility would have increased by 0.2 percentage points among women who left welfare after their first year off welfare and 0.2 percentage points by women who were not on welfare in the late 1990s.

            Thus, holding work experience constant, fewer workers were monetarily eligible under the UI rules in the late 1990s. However, these differences are quite small. This is likely the case for three reasons. First, monetary requirements are generally very low resulting in most people who work being eligible for benefits. Second, changes in policy are marginal; they only affect a small portion of those who have incomes in the relevant earnings categories. The fact that the changes are statistically significant overall and for those without a history of welfare use, but not for welfare recipients, indicates that these policy changes affected former welfare recipients more so than other workers.

States should rethink UI eligibility rules to assist low-wage workers and former welfare recipients

            There are a number of policies to address the lower eligibility for UI among former welfare recipients. The basic premise is that earnings thresholds must be consistent with the earnings patterns of low-wage, part-time, and intermittent workers and not only for the earnings patterns of full-time workers. First, states could change their eligibility rules so that those who search exclusively for part-time employment are eligible for UI. In 2002, twenty-eight states had rules prohibiting workers who search exclusively for part-time employment from receiving UI. Other options would be to include the most recent calendar quarters in the calculation of monetary eligibility or to move away from an earnings requirement, to an hours only requirement. We simulate the effects of using more recent calendar quarters in determining eligibility and moving to an hours only requirement and find that both would increase eligibility for UI among former welfare recipients.

Simulation 1: Alternate base period[7]

            Many states have sought to increase eligibility among low-wage and part-time workers by using an “alternate base period” that includes the most recent calendar quarter of earnings for those who fail to qualify for UI benefits under the standard base period. This alternate base period would increase the likelihood that former welfare recipients are eligible for UI by taking into consideration the most recent earnings for those who would otherwise fail to qualify. Figure 3 shows the proportion of individuals meeting the monetary eligibility requirements by quarters off welfare as the rules were in the late 1990s[8], as well as how many would qualify if every state adopted the alternate base period. For former welfare recipients in their first quarter off welfare, adopting an alternate base period would increase UI eligibility by 4.3 percentage points. In the next two quarters off welfare (quarters three and four) the alternate base period would increase eligibility by 5.8 and 6.6 percentage points, respectively.

Simulation 2: Hours requirement

            In addition to using the alternate base period, states could remove the earnings requirement and move to an hours-only requirement. Oregon has adopted an hours-only requirement and the alternate base period, allowing workers with 500 hours of work experience over the previous four calendar quarters to achieve “monetary” eligibility. Benefit payments would depend on previous earnings, but eligibility would not. For part-time workers to qualify, we estimate the proportion of former welfare recipients meeting an hours-only requirement[9] of 260 hours, or an average of 20 hours per week for a full quarter. Figure 4 shows that reducing the hours requirement to 260 hours per quarter substantially increases eligibility among former welfare recipients, however reducing the hours requirement to 500 hours per quarter does not. The 260 hours policy would increase eligibility more than the alternate base period. The largest increases are in the first three quarters of employment, as many former welfare recipients start their employment in short-term or part-time work and then accumulate enough hours to qualify. Changing the initial eligibility requirement from an earnings requirement to an hours-only requirement increases the equity of the program. All workers, regardless of income, must demonstrate similar attachment to the labor force. Changing the program to focus on hours will improve program equity; it will also simplify the program, making it easier for workers to understand their eligibility status.

            Figure 5 shows eligibility levels if all states adopted both the alternate base period and an hours-only requirement of 400 hours in the base period. This means that over the past year, an individual would have to work an average of eight hours per week, or at least part-time (20 hours) for 20 weeks. Four hundred hours strikes a compromise; this requirement is sufficient to demonstrate labor market attachment, while recognizing that many workers – about one-in-eight – work only part-time.

Clearly, changing the policy rules for UI eligibility to include part-time work and workers’ most recent employment history would increase the likelihood that former welfare recipients meet the monetary eligibility requirements for UI. However, policymakers must do more than change the monetary requirements. Recent research by Smith et al. (2003) indicates that there are family responsibilities, the largest shares of which are borne by women, which exclude workers from collecting UI. In particular, in most states if workers leave employment for personal reasons such as illness or pregnancy, care of a family member, domestic violence, sexual harassment, or following a spouse who relocates, they are typically ineligible for benefits. While our paper does not address these issues, we note that our estimates of monetary eligibility are a “best case” scenario. In other words, these estimates assume that a worker has a qualifying separation from her employer, and does not leave employment for the reasons stated above.

Conclusions

            Whether or not former welfare recipients meet the eligibility requirements for UI is an increasingly important issue. In August of 1996 when Congress passed welfare reform, unemployment was at 5.1 percent, and during the next four years unemployment declined to 3.9 percent. However, after October 2000, labor market conditions changed for the worse, with unemployment currently at 6.0 percent. For millions of workers, earning enough to qualify for UI became more difficult. Now that the labor market has worsened, analysts anticipate an increase in welfare caseloads. Over the fourth quarter of 2002, welfare caseloads increased in 38 states and by 1.2 percent nationally (Richer, Rahmanou, and Greenberg, 2003). However, year-over-year declines in caseloads are still occurring through March 2003, raising questions about the status of the low-income safety net.

            It is clear that former welfare recipients still face considerable barriers to regular, consistent employment; their inability to become eligible for UI benefits after welfare reform provides additional evidence as to the indeterminate nature of employment. Welfare reform already provides many employment incentives and sanctions for not working. Consequently, many women have left welfare in search of work. Now that national employment has declined and unemployment rates have increased, we should expect UI to fill the gap; unfortunately, this has not happened, because too few former welfare recipients are qualifying for UI.

Appendix

References

Assistant Secretary for Planning and Evaluation. 2000. “Leavers” and Diversion Studies: Summary of Research on Welfare Outcomes Funded by ASPE. Washington, DC. <http://aspe.hhs.gov/hsp/leavers99/ombsum.htm>

Boushey, H, and J. B. Wenger. 2003. “In the Policy Nexus: Unemployment Insurance Eligibility Before and After Welfare Reform.” Center for Economic and Policy Research. October 2003. 

Kaye, K. (2001). Re-Examining the Potential Role of Unemployment Insurance as a Safety Net For Workers at Risk of Public Assistance Receipt. Presented at the National Research Conference on Workforce Security Issues, June 26-27, 2001.

Loprest, P. 1999. How Families that Left Welfare are Doing: A National Picture. Washington, DC: Urban Institute. No. B-1. <http://newfederalism.urban.org/html/series_b/anf_b1.html>

Parrott, S. 1998. Welfare Recipients Who Find Jobs: What Do We Know About Employment and Earnings? Washington, DC: Center on Budget and Policy Priorities

Rangarajan, A, W. Corson & R. Wood. (2001). To What Extent is the Unemployment Insurance System a Safety New for Welfare Recipients Who Exit Welfare for Work? mimeo.

Richer, E., H. Rahmanou, and M. Greenberg, 2003. Welfare Caseloads Increase in Most States in Fourth Quarter Washington, DC: Center for Law and Social Policy, Updated April 29.

Smith, R., R. McHugh, A. Stettner, N. Segal. (2003). Between a Rock and a Hard Place: Confronting the Failure of State Unemployment Insurance Systems to Serve Women and Working Families. New York, NY. National Employment Law Project. 

Wenger, J. B. (2001). Divided We Fall: Deserving Workers Slip Through America's Patchwork Unemployment Insurance System. Washington, DC: Economic Policy Institute. 

Thanks to the Rockefeller Foundation for generous support of CEPR’s project on Labor Markets and Economic Well-Being.

The authors would like to thank Matthew Walters for valuable research assistance on this project.