From Welfare
to the Low-Wage Labor Market
Sheldon
Danziger and Rucker C. Johnson 1
Russell Sage Foundation Forum on the Future of Work
May 27, 2004
The 1996 Personal Responsibility and Work Opportunity Reconciliation
Act (PRWORA)
is widely praised for "ending welfare as we knew it." The
1996 Act does not allow a single mother to reject an available low-wage
job and remain on welfare, as those who refuse to actively search
for work or co-operate with the welfare agency can lose both cash
assistance and food stamps. Welfare no longer provides a steady source
of income, in part because of the 60-month lifetime limit on benefit
receipt, and in part because of the requirement that most recipients
quickly search for work or engage in other qualifying activities specified
by the state.
Together, welfare reform and the economic boom of the1990s contributed
to a dramatic decline in welfare caseloads, an increase in the employment
rate of single mothers, and a modest decline in family income poverty.
Nonetheless, the poverty rate of single mothers remains very high
and many of them lost health insurance as they made the transition
from welfare to work.
In this note, we present some new evidence on the employment and
wage rates in 2003 of a sample of women, all of whom received cash
welfare in 1997. Many fewer of them received welfare and many more
of them worked in 2003 than in 1997. However, only about two-thirds
had jobs in August 2003 and only about half of them had "good
jobs," ones that paid enough so that a mother and two children
could escape poverty and have health insurance. Women who leave welfare
to work face frequent spells of nonemployment and continuing economic
hardship over all phases of the business cycle, from the low unemployment
rates of the late-1990s to the higher rates during the recession.
The Transition from Welfare to Work
Greater numbers of single mothers entered the labor force after welfare
reform. From the late 1960s to the late 1980s, about 70 percent of
single mothers with children under age 18 worked at some time during
the year. After welfare reform, this increased to about 85 percent
during the late 1990s. Annual employment fell during the recent recession,
but has remained close to 80 percent. Employment increases were greatest
for the racial/ethnic groups that had been most likely to receive
cash welfare. For example, between 1989 and 1999, the annual employment
rate of white single mothers increased by about 9 percentage points,
whereas that of African-American and Hispanic single mothers rose
by about 18 points.
Many evaluations of current and former welfare recipients in the
aftermath of the 1996 reform show that about two-thirds were at work
in any month.
2 In these comments, we focus on data from the Women's
Employment Study (WES) conducted by the Michigan Program on Poverty
and Social Welfare Policy at the Gerald R. Ford School of Public Policy
at the University of Michigan. Although WES interviewed only women
from one urban Michigan county, it has more extensive information
on work and welfare receipt than any other welfare reform study. A
random sample of single mothers who received cash assistance in February
1997 were interviewed in their homes five times between Fall 1997
and Fall 2003. The interviews provide information on whether or not
the respondent worked in each of the 79 months between February 1997
and August 2003. 3
Figure 1 shows the monthly employment rate for Wave 5 respondents
over the entire study period. In February 1997, all respondents received
cash assistance and 42 percent were employed; in August 2003, only
about one-fifth of respondents received cash assistance and the monthly
employment rate had increased to 64 percent. 4
The employment rate climbed rapidly from February 1997 to a peak of
76 percent in November 1999 during the height of the economic boom.
The rate then fluctuated around 72 percent until late 2001 before
falling to 64 percent in August 2003.
The transition from welfare to work was characterized by a monthly
employment rate increase of 34 percentage points (from 42 to 76 percent)
in the period just after reform, one-third of which (12 points, from
76 back to 64 percent) was eroded in the aftermath of the recent recession.
The typical respondent worked in about two-thirds of the 79 months.
The most successful, about one-third of respondents, worked in 90
percent or more of the months; in contrast about one-sixth of the
respondents worked in less than one-third of the months.
5
The typical respondent who took a job at some point between 1997
and 2003 could expect to work for 10 months before she experienced
a month of unemployment.
6 This median employment spell was 12 months for
a woman with a high school diploma, but only 7 months for a woman
without a diploma.
In the early years of the study, about one-fifth of women who exited
from a job to nonemployment had been laid off or fired; by the end
of the study period, about one-third of exits were due to firings
or lay-offs. Voluntary quits fell from about one-fifth of job exits
in the late-1990s to about one-tenth between 2001 and 2003.
Welfare reform contributed to increased employment among current
and former welfare recipients. However, during economic booms, but
especially during periods of recession and slow aggregate employment
growth, a significant minority of these women do not have either wages
or cash assistance.
From Welfare to Low-Wage Jobs
In Fall 1997, working respondents earned a median wage of $6.62 (all
wage rates are expressed in $2003). As they accumulated work experience
over the 6 ½ year period, their median wage increased by 16
percent to $7.66 in Fall 1999 and by 8.5 percent to $8.31 in Fall
2001. Over the next two years, the median wage was virtually constant
in real terms--$8.33 in Fall 2003. Over the entire period, the median
wage grew by 26 percent.
There is no official definition of what it means to have a "good
job." We define a good job as one that provides a net annual
income that exceeds the poverty line for a single mother with two
children ($14,824 in 2003) and allows her to maintain health insurance.
If a woman works full-time full-year (1800 hours), she will reach
this threshold after paying social security and Michigan state income
taxes on her wages and after receiving the earned income tax credit
and food stamps. Thus, a respondent has a good job if she works full-time
(at least 35 hours per week), has an hourly wage of at least $7.75
(2003 $), and is offered health insurance benefits either immediately
or after a trial period.
If she works at a full-time job that does not offer health insurance,
we define a good job as one that pays $9.40 per hour. This $1.65 per
hour difference in wages over a full year of work is assumed to be
sufficient for the woman to purchase of a private health insurance
policy and to pay small monthly fees to insure her children under
the State Child Health Insurance Program. Women whose current job
satisfies these wage and health insurance criteria but who voluntarily
work part-time are also considered to have good jobs.
7 All other workers are considered
as having "low-wage" jobs.
Figure 2 shows changes in job quality between 1997 and 2003. In 1997,
only 8.0 percent of respondents had good jobs. The extent of good
jobs increased rapidly to 25.0 percent by 1999, and then increased
slowly to 28.4 percent over the next four years. Movement out of "low-wage"
jobs proceeded steadily, from 51.1 to 32.0 percent between 1997 and
2003, suggesting that, for some women, taking a low-wage job does
provide a stepping stone to a good job. The one-year probability of
moving from a "low-wage" to a good job is about 15 percent.
If we focus only on working respondents (data not shown), the percentage
with good jobs was 47.0 percent in Fall 2003. Thus, 6 ½ years
after welfare reform, about half of all women worked in jobs that
did not pay enough to keep a family of three out of poverty, even
though they accumulated an average of 54 months of work experience.
And almost two-fifths of all respondents had no job.
Unionized jobs and jobs in occupations other than the services were
more likely than nonunionized jobs and service sector jobs to pay
well. For example, in Fall 2003, 17 percent of working respondents
were union members or worked in a job that was covered by a union
contract. Their median wage, $10.15, was 28 percent above the $7.94
median of nonunion workers. Three-quarters of unionized workers compared
to two-fifths of nonunion workers had good jobs. About half of respondents
worked in service sector jobs at this time. Their median wage, $7.61,
was 17 percent lower than the $9.18 median of workers in other occupations.
Service workers were also less likely to have good jobs than workers
in other occupations-37 vs. 57 percent.
Summary
The 1996 welfare reform successfully "ended welfare as we knew
it," reducing the welfare rolls and increasing employment. Another
popular sound bite from the early 1990s was "make work pay."
In a 1992 campaign speech in which he first proposed to "end
welfare as we know it," President Clinton linked these policy
goals, "No one who works full-time and has children at home should
be poor anymore. No one who can work should be able to stay on welfare
forever."
We have reformed our social policies so that no one can stay on welfare forever. However, too many single mothers with children work full-time and live in poverty. Many reasonable policy changes could be adopted that would increase the likelihood that women making the transition from welfare to work and millions of others could move out of low-wage jobs and into good ones.
Figure 1
Figure 2
1 Sheldon Danziger is Henry J. Meyer Collegiate Professor, Gerald R. Ford School of Public Policy, University of Michigan; Rucker C. Johnson is Assistant Professor, Goldman School of Public Policy, University of California-Berkeley. Funds for the Women's Employment Study data collection were provided by the Charles Stewart Mott, Joyce, and John T. and Catherine D. MacArthur Foundations, the National Institute for Child Health and Human Development (HD P50 HD38986) and the Substance Abuse Policy Research Program, Robert Wood Johnson Foundation. See: www.fordschool.umich.edu/research/poverty/publications.htm
2 For reviews,
see, Gregory Acs and Pamela Loprest. 2001. "Final Synthesis Report
of the Findings from ASPE's Leavers Grants." Report to the U.S.
Department of Health and Human Services. The Urban Institute and Robert
Moffitt. 2002. "From Welfare to Work: What the Evidence Shows."
Brookings Institution, Welfare Reform and Beyond, Policy Brief No.
13.
3 Information
on the universe of welfare recipients in the county was provided by
the Michigan Family Independence Agency. A random sample of 874 African
American or White women (due to the demographics of the county) was
selected for the study. Response rates at the five survey waves were,
86, 93, 91, 91 and 93 percent, respectively. Sample size falls from
753 to 536 between waves 1 and 5, but there is no evidence of nonrandom
attrition from the sample.
4 Because welfare receipt declined more rapidly than employment increased, the percentage of WES respondents who had no income from either cash welfare or work increased from zero in February 1977 to almost 20 percent by Fall 2003. See, Lesley Turner, Sheldon Danziger and Kristin Seefeldt. 2004. "Failing the Transition from Welfare to Work: Women Disconnected from Work, Welfare and Other Sources of Economic Support." University of Michigan, unpublished: www.fordschool.umich.edu/research/poverty/publications.htm.
5 The current
and former welfare recipients who had the most difficulty getting
and keeping jobs even during the economic boom are more likely than
others to have low educational levels, few job-specific skills, and
significant physical and mental health problems.
6 Our data will overstate employment spell lengths. For example, we ask a respondent whether or not she worked in every month. A woman who worked during the first week of September was laid off and without a job until the last week of October, would be recorded as working in both months.
7 This definition
of a "good job" was first used in Rucker C. Johnson and
Mary E. Corcoran. 2003. "The Road to Economic Self-Sufficiency:
Job Quality and Job Transition Patterns after Welfare Reform."
Journal of Public Policy Analysis and Management, 22/4: 615-39.
