Are single mothers finding jobs without displacing other workers? – Statistical Data Included
Robert I. Lerman
Despite a large influx of single mothers into the labor force following the passage of welfare reform in 1996, metropolitan areas generated more than enough jobs to employ these new entrants without deleterious effects on competing groups of workers
Moving welfare recipients from welfare to work was the primary goal of the Personal Responsibility and Work Opportunity Reconciliation Act. Four years after the passage of this Act, the Nation had achieved considerable success in reaching that goal. Together with a thriving economy, the Act has generated unprecedented increases in employment among mothers heading families (single mothers), the group most likely to receive welfare.
Despite worries that the economy could not absorb the more than 1 million welfare recipients that were expected to enter the job market,(1) enough jobs materialized to employ not only those welfare mothers who began looking for work, but also other single mothers who had been unemployed as well. Between early 1996 and the middle of 1998, when about 741,000 additional never-married mothers entered the labor force,(2) the economy generated enough jobs for a 40-percent rise in employment for this group. The 40percent job growth figure dwarfed the 9-percent increase in employment for the economy as a whole.(3)
Notwithstanding these impressive national gains, three serious concerns have emerged. First, single mothers in large metropolitan areas may not be faring as well as those in the rest of the Nation. A recent study published by the Brookings Institution found that reductions in welfare cases were lower in counties with large central cities than in other counties in the same State.(4) As a result, the 89 urban counties containing the largest 100 cities increased their share of the Nation’s welfare caseload from 47.5 percent to 58.1 percent between 1994 and 1999. Second, the increase in jobs for welfare recipients may be coming at the expense of jobs for other, less skilled workers. Third, even if low-skilled jobseekers actually find employment, the enormous inflow of low-skilled single mothers into the job market may be depressing the wages of all low-skilled workers? After developing a detailed analysis and projections, Timothy J. Bartik concludes that welfare reform’s stimulus to the low-skilled labor force will exert substantial effects that will lower the wages or employment opportunities of female heads of households and female high school dropouts. Still, he acknowledges that there is little evidence yet of such negative effects.(6)
The current article builds on an earlier study that examined the potential of 20 large metropolitan areas to absorb the expected inflow of welfare recipients over the next 5 years.(7) Projections from that analysis indicated that, while average growth in low-skilled jobs would be sufficient to employ the inflow of recipients and to reduce the unemployment rate of low-skilled workers, four areas–Baltimore, the District of Columbia, New York City, and St. Louis–could very well experience rising unemployment.
The text that follows describes what actually took place in the labor markets of the same 20 metropolitan areas. The article examines labor market outcomes of single mothers and of low-skilled workers with whom they likely compete. The rationale for focusing on single mothers is twofold: they are the group most likely to have been affected by changes in the welfare program, and the monthly CPS data do not specify who among the single mothers are welfare recipients. The following questions are addressed:
* Did large metropolitan areas experience substantial increases in the labor force participation of single mothers?
* Were large metropolitan areas able to generate sufficient jobs to employ the rapidly rising number of single mothers in the labor force?
* Was the increased labor force participation of single mothers associated with increased joblessness among competing low-skilled workers, including less educated men?
* Did the rapid increase in low-skilled single mothers in the labor market lead to a reduction or growth in their wages or in the wages of subgroups of low-skilled workers?
* Which metropolitan areas experienced the most serious problems absorbing the inflow of single mothers and other low-skilled workers?
These questions are answered by comparing labor force measures prior to the passage of the Personal Responsibility and Work Opportunity Reconciliation Act with the same measures 3 years later. While, certainly, State waiver programs stimulated substantial increases in work effort among individuals eligible for welfare, it was the passage of national welfare provisions mandating work, rigorous requirements for remaining eligible for welfare, and time limits that drew, and has continued to draw, the most attention and most concern about the implications for the low-skilled labor market.(8)
The economic context
In the 3 years after the enactment of welfare reform, the remarkably fast growth of the economy stimulated a rapid increase in the demand for labor. Real gross domestic product jumped by 12.5 percent between the first three quarters of 1996 and the first three quarters of 1999.(9) A decline in the unemployment rate from 5.6 percent to 4.3 percent, together with an increase in the labor force of 5.6 million people, resulted in 7 million more jobholders.
Along with gains in employment came an increase in earnings. Usual weekly earnings (in current dollars) among full-time workers rose from $599 to $665 per week among adult men and from $439 to $494 among adult women. These figures represent a rise of about 1.5 percent to 2 percent per year after adjustments for overall price increases.(10)
Job and wage data
The analysis that follows uses data from the monthly Current Population Survey (cPs) for the 12 months prior to the passage of the welfare reform act (September 1995 through August 1996) and 3 years later (September 1998 through August 1999).(11) The monthly CPS provides a wide range of information, including employment and weekly wage statistics for the U.S. population. Although respondents report their welfare status only in the annual March CPS the analysis uses monthly samples for three reasons. First, the March sample covers only a single month in each year and includes a smaller and less representative sample than the 12 months of data per year used in this study. Second, judging the labor market implications of welfare changes requires taking account of the added employment of past and potential welfare recipients and not simply those reported to have received welfare during the previous year. Therefore, it is best to examine trends among all single mothers and not simply those who report their welfare income during the previous year. Third, because of high and increasing levels of underreporting of respondents’ welfare status, it is difficult to use March CPS data alone to learn about employment and wages of welfare recipients.(12)
The analysis presented in this study covers single mothers and other workers between the ages of 20 and 45.
Single mothers entering the labor force
Between the year before the passage of the Personal Responsibility and Work Opportunity Reconciliation Act and the subsequent 3 years, the percentage of single mothers in the labor force (their labor force participation rate) rose rapidly in the 20 metropolitan areas examined. For the 20 areas as a whole, the share of single mothers working or looking for work jumped from 67 percent to 79 percent, or by about 230,000. During the same 3-year period, the labor force participation rate for all 20- to 45-year-olds increased by only 1 percentage point, from 82 percent to 83 percent.
The increase in labor force activity among single mothers varied widely, with the labor force participation rate rising between 19 percentage points and 20 percentage points in Boston and Jacksonville, compared with 0 percentage points to 2 percentage points in San Diego and San Francisco. (See table 1.) In general, a catching-up process took place: those metropolitan areas with high initial labor force activity had slower rates of growth than areas with low initial activity. The increased labor force activity extended to less educated, as well as more educated, single mothers. Single mothers with a high school degree or less raised their rate of labor force participation from 59 percent to 72 percent.
Table 1. Labor force participation of single mothers,
1995-96 and 1998-99
Labor force Labor force
Metropolitan area participation rate, participation rate,
1995-96 1998-99
Total 0.67 0.79
Atlanta .76 .88
Baltimore .77 .84
Boston .63 .82
Chicago .69 .80
Dallas .86 .93
Detroit .70 .77
Houston .80 .86
Indianapolis .71 .86
Jacksonville .69 .89
Los Angeles .61 .73
Minneapolis .73 .80
New York .49 .64
Philadelphia .65 .80
Phoenix .73 .84
San Antonio .73 .84
San Diego .69 .69
San Francisco .88 .90
San Jose .67 .75
St. Louis .75 .83
Washington, DC .77 .89
Change In labor Percent change In
Metropolitan area force, 1995-96 to labor force, 1995-96
1998-99 (thousands) to 1998-99
Total 231.9 13.7
Atlanta 32.4 32.2
Baltimore -15.3 -22.8
Boston 19.2 36.8
Chicago 23.0 12.8
Dallas 25.5 24.9
Detroit 1.7 1.5
Houston 18.7 18.0
Indianapolis 5.3 13.9
Jacksonville 7.3 25.2
Los Angeles 14.2 7.8
Minneapolis 1.1 2.2
New York 20.5 10.4
Philadelphia 14.8 13.4
Phoenix 12.6 19.4
San Antonio 7.7 22.6
San Diego 9.6 19.5
San Francisco — —
San Jose 2.5 8.9
St. Louis 16.6 24.7
Washington, DC 10.2 10.5
NOTE: Dash indicates too few cases to calculate reliable estimates.
These cells had fewer than 100 unweighted observations in at least
1 year. The labor force participation rate is the percentage of
the population that is either employed or unemployed. The percent
change in the labor force reflects changes in the tendency of
single mothers to work or look for work and
changes in the numbers of single mothers.
SOURCE: Authors’ tabulations of data from the Current Population
Survey, September 1995-August 1996 and September 1998-August 1999
This sharp growth in the labor force activity of single mothers accounted for a substantial share of the growth in the total labor force: whereas single mothers in the 20 metropolitan areas made up only about 6 percent of the total labor force in the 1995-96 period, they accounted for 20 percent of all labor force growth in the 3 years after the passage of the welfare reform act. Still, as of the period between September 1998 and August 1999, only 7 percent of the labor force in these metropolitan areas consisted of single mothers. In the absence of any increase in the labor force participation of single mothers, the growth rate of the labor force would have been about 1.31 percent per year, as opposed to the actual growth rate of 1.53 percent per year. The impact on the labor supply would have been even larger but for the reduction in the population of single mothers in these areas. Between the 1995-96 and 1998-99 periods, single mothers as a proportion of all 20- to 45-year-olds declined from 7.7 percent to 7.2 percent.
The contribution of single mothers to the change in the labor force varied widely across metropolitan areas. In 5 of the 20 areas, the total labor force declined slightly in absolute terms, largely as a result of the declining total population, although only Baltimore experienced an absolute drop in the number of single mothers in the labor force. (See table 1.) The growth in the participation of single mothers in the labor force accounted for more than 20 percent of labor force growth in Atlanta, Boston, Phoenix, St. Louis, and San Antonio. In New York City, an area with a large number of welfare recipients and with a relatively high initial unemployment rate, the inflow of single mothers into the workforce represented only 14 percent of the growth of the total labor force.
The effects of welfare reform on single mothers might be expected to influence the job market for less educated adult workers more dramatically than the job market for all workers. The reason is that the single mothers most likely to enter the labor market because of welfare reform have lower educational attainment than the average worker. Their most plausible adult competitors are women and unmarried men with a high school degree or less. While single mothers made up 13 percent of this less educated segment of the workforce, they accounted for 24 percent of its labor force growth.
Absorption of added workers into jobs
The growth in the number of people looking for work posed a challenge to metropolitan area labor markets. Would jobs suitable for the 230,000 additional single mothers in the workforce materialize? Or would most of the new jobseekers become unemployed?
As it happened, labor markets responded well: they supplied the 230,000 jobs necessary to absorb the single mothers entering the labor force, plus an additional 36,000 jobs to move unemployed single mothers into employment. While the participation of single mothers in the labor force grew by 14 percent, the number of jobs going to single mothers increased by an even larger 18 percent. As a result, the unemployment rate of single mothers in these metropolitan areas fell dramatically, from 12 percent in the 1995-96 period to 8 percent in 1998-99. An overwhelming number of single mothers worked full time. The average number of hours worked by single mothers was virtually unchanged, at about 38.6 per week.
Three years after the passage of the Personal Responsibility and Work Opportunity Reconciliation Act, the percentage of employed single mothers increased in each of the 20 metropolitan areas and in all 20 metropolitan areas combined (from 59 percent to 73 percent; see table 2). Double-digit rates of job growth for single mothers took place in 16 of 19 metropolitan areas.(13) The exceptions were Baltimore, Detroit, and Minneapolis, areas in which the absolute number of single mothers declined. Thus, where job growth was modest, it resulted from a reduced labor supply and not inadequate demand.
Employment of single mothers, 1995-96 and 1998-99
Employment-to- Employment-to-
Metropolitan area population ratio, population ratio,
1995-96 1998-99
Total 0.59 0.73
Atlanta .70 .82
Baltimore .68 .80
Boston .58 .79
Chicago .60 .70
Dallas .78 .88
Detroit .62 .72
Houston .75 .80
Indianapolis .65 .82
Jacksonville .54 .84
Los Angeles .52 .66
Minneapolis .69 .74
New York .41 .55
Philadelphia .56 .72
Phoenix .68 .81
San Antonio .64 .76
San Diego .60 .63
San Francisco — .89
San Jose .55 .69
St. Louis .65 .77
Washington, DC .67 .83
Change In
employment, Percent change
Metropolitan area 1995-96 to 1998-99 In employment,
(thousands) 1995-96 to 1998-99
Total 268.2 17.6
Atlanta 31.5 33.5
Baltimore -9.0 -14.7
Boston 20.8 42.5
Chicago 21.6 13.7
Dallas 28.8 30.4
Detroit 6.2 6.1
Houston 17.6 18.3
Indianapolis 6.5 18.3
Jacksonville 10.9 43.9
Los Angeles 19.4 12.0
Minneapolis .4 .8
New York 19.3 11.5
Philadelphia 18.1 18.7
Phoenix 14.1 22.9
San Antonio 8.0 26.3
San Diego 11.1 25.3
San Francisco 4.5 16.2
San Jose — —
St. Louis 18.4 30.4
Washington, DC 15.5 17.8
NOTE: Dash indicates too few cases to calculate reliable estimates.
These cells had fewer than 100 unweighted observations in at least 1
year. The employment-to-population ratio is the proportion of the
population that is employed. The change in employment reflects
both the change in the tendency for single mothers to have jobs
and the change in the number of single mothers.
SOURCE: Authors’ tabulations of data from the Current Population
Survey, September 1995-August 1996 and September 1998-August 1999.
Even the largest metropolitan job markets–New York, Los Angeles, Chicago, and Philadelphia–were able to respond to the jump in labor force participation that took place after welfare reform. These metropolitan areas alone added nearly 80,000 new jobs for single mothers. The case of the New York metropolitan area is especially interesting. Before the passage of the Act, the share of single mothers who were either in the labor force or employed was low, and New York’s overall unemployment rate stood at 8 percent, more than 25 percent above the level averaged across the 20 metropolitan areas. Further, single mothers constituted a high share of the 20- to 45-year-old population (11 percent), a proportion 36 percent higher than the 20-metropolitan-area average. As a result, moving a high percentage of single mothers into the workforce would generate a relatively large impact on both the low-skilled labor force and the total labor force. The share of single mothers with jobs in the 1995-96 period was only 41 percent, the lowest employment rate by far among the 20 areas and 18 percentage points below the average. Given these indicators, it would be reasonable to expect the New York labor market to have great difficulty stimulating large numbers of single mothers to enter the workforce and absorbing them into jobs if they did begin looking for work. But in fact, over the 3-year period examined, the labor force participation rate of single mothers in New York jumped by 26 percent, from 48 percent to 64 percent, expanding the single-mother workforce by 10 percent.(14) Furthermore, the New York economy generated enough jobs to raise the employment rate of single mothers by 14 percentage points, to 55 percent, and to reduce their unemployment rate from 15 percent to 14 percent.
The entry of recipients from the large welfare caseloads in the Los Angeles metropolitan area, along with an above-average unemployment rate of 14 percent for single mothers in 1995-96, posed a potential problem similar to the one in New York. However, single mothers in Los Angeles did not make up an unusually high share of the adult population, and the absolute number of single mothers declined between the 1995-96 and 1998-99 periods. In this case, a decline in the absolute number of single mothers meant that only about 13,000 additional jobs were required to absorb the new entrants. By raising the employment of single mothers by 19,400, the Los Angeles labor market ended up lowering the unemployment rate of single mothers from 14 percent to 10 percent.
Overall, metropolitan job markets were able to absorb the additional labor force growth induced by changes in the welfare system. In the 20 metropolitan areas examined, 77,000 single mothers entered the labor force per year, a figure within the 48,000-to-162,000 range projected in an earlier study.(15) Growth in the number of jobs held by single mothers amounted to 89,000 per year in the 20 metropolitan areas, resulting in a decrease in the number of unemployed single mothers by 12,000 per year.
Unemployment among other workers
Competition with single mothers could have weakened the market position of other workers, especially less educated ones. To examine this issue, the changes in employment and unemployment of competing groups were compared with the labor force inflows of single mothers. Several groups of adult workers, all with at most a high school degree, are likely to compete with single mothers for jobs: unmarried women who are not mothers, married women, and unmarried men.
The unemployment rate and the employment-to-population ratio do not indicate any decline in job availability. (See table 3.) The unemployment rate of all three groups declined between 1995-96 and 1998-99, and for two of the three groups the percentage of those holding jobs increased. Only the share of less educated married women who work dropped, even as this group saw its unemployment rate fall.
Unemployment rates and employment-to-population ratios of less
educated workers, 1995-96 and 1998-99
Single women with no Married women with
children and a high school education
school education or or less
Metropolitan area lees
1995-96 1998-99 1995-96 1998-99
Unemployment rate
Total 9.1 7.7 5.8 5.3
Atlanta 6.6 8.0 3.4 1.2
Baltimore 10.1 13.5 5.3 3.8
Boston 75.0 8.0 4.8 1.1
Chicago 7.6 9.9 5.0 3.0
Dallas 6.3 5.8 3.0 5.5
Detroit 7.7 3.8 4.8 3.0
Houston 12.7 6.0 7.9 4.9
Indianapolis 6.7 — 5.3 5.5
Jacksonville 11.3 — 6.3 4.1
Los Angeles 12.8 10.5 1.9 8.6
Minneapolis 5.4 6.7 .9 2.4
New York 11.1 10.5 7.9 8.8
Philadelphia 12.2 6.9 6.4 5.4
Phoenix 8.9 6.7 1.9 3.4
San Antonio — 5.6 13.1 9.9
San Diego 10.1 5.7 8.8 9.3
San Francisco — — 4.1 2.8
San Jose 4.1 — 4.1 8.7
St. Louis 8.7 6.2 5.1 4.3
Washington, DC 9.0 4.5 4.5 4.1
Employment-to-
population ratio
Total 65.5 68.7 58.9 57.3
Atlanta 71.9 74.1 63.9 65.5
Baltimore 69.4 64.2 76.5 66.1
Boston 65.2 66.6 67.3 70.3
Chicago 65.2 65.3 64.6 61.2
Dallas 77.1 77.2 64.1 58.5
Detroit 62.0 71.2 57.9 60.5
Houston 63.2 72.9 55.4 50.3
Indianapolis — 73.1 65.7 57.3
Jacksonville — 69.1 66.6 71.6
Los Angeles 57.0 63.1 46.3 47.1
Minneapolis 80.3 74.6 80.6 75.5
New York 55.3 60.9 46.2 44.9
Philadelphia 67.2 64.3 64.7 64.1
Phoenix 74.9 70.9 64.9 55.6
San Antonio 75.4 80.2 53.7 59.2
San Diego 67.9 71.9 53.2 51.9
San Francisco 82.6 71.0 62.4 68.9
San Jose 77.2 89.3 62.0 56.3
St. Louis 70.9 73.0 72.3 62.1
Washington, DC 66.7 79.0 70.1 74.9
Unmarried men with
a high school education
Metropolitan area of less
1995-96 1998-99
Unemployment rate
Total 12.3 8.7
Atlanta 8.7 6.5
Baltimore 12.6 16.0
Boston 10.0 5.1
Chicago 15.1 11.6
Dallas 12.3 2.9
Detroit 10.4 7.4
Houston 12.8 7.5
Indianapolis 5.5 4.4
Jacksonville 8.4 4.2
Los Angeles 11.5 9.5
Minneapolis 8.7 4.9
New York 13.9 13.1
Philadelphia 15.0 11.3
Phoenix 11.4 6.5
San Antonio 6.1 6.0
San Diego 17.1 7.0
San Francisco 11.7 10.1
San Jose 12.2 11.2
St. Louis 12.1 7.1
Washington, DC 13.1 7.2
Employment-to-
population ratio
Total 72.6 76.7
Atlanta 70.8 80.8
Baltimore 71.8 68.0
Boston 72.7 84.4
Chicago 71.4 73.9
Dallas 78.4 90.9
Detroit 70.7 78.5
Houston 75.3 80.4
Indianapolis 86.7 82.3
Jacksonville 76.4 85.3
Los Angeles 75.3 74.5
Minneapolis 85.0 84.2
New York 63.9 63.9
Philadelphia 67.2 71.2
Phoenix 81.3 84.4
San Antonio 78.2 77.3
San Diego 69.3 75.9
San Francisco 79.6 79.1
San Jose 71.9 78.0
St. Louis 75.8 80.9
Washington, DC 70.9 81.8
NOTE: Dash indicates too few cases to calculate reliable estimates
for subgroups. These cells had fewer than 100 unweighted
observations in the denominator of the ratio.
SOURCE: Authors’ tabulations of data from the Current Population
Survey, September 1995–August 1996 and September 1998-August 1999.
It is possible that, despite falling unemployment among competing groups, some of the jobs taken by single mothers would have gone to other groups in the absence of welfare reform and led to a larger decline in unemployment among their ranks. If employers did substitute single mothers for less educated adults, though, metropolitan areas with the highest increases in the workforce among single mothers would have experienced the lowest improvement in employment outcomes of competing groups. However, the data reveal no such tendency: there are no significant negative correlations between the increased labor force participation rate of single mothers and the employment of less educated workers.
To gain perspective on the magnitude of the employment changes, one can calculate what would have been the increase in the employed population if half of the new jobs going to single mothers went instead to one of the three competing groups. For the 20 metropolitan areas as a whole, the employed share of the population averaged 65.4 percent in 1995-96 and 67.3 percent in 1998-99 for the three groups combined. If half the gain in jobs held by single mothers went to either or both of the other two groups, the overall ratio of employment to population could have increased to 68.9 percent, raising the employed share by about 1.6 percentage points.
Potential reductions in wages
The accumulating evidence of job growth for single mothers and other less educated workers leaves open concerns about wages. Are wages high enough to allow families to raise their incomes? Are the new entrants to the labor force able to translate their work experience into higher wages over time, or are wages for single parents falling, either because of a shift in the single-parent labor force toward the least skilled or because the added competition for jobs resulting from welfare reform is suppressing the wages of all less educated workers?
According to the laws of supply and demand, an upward shift in the supply of workers should increase employment, but lower wages. Because marginal productivity declines with each additional worker, employers will expand their workforces only if they can match the reduction in productivity with a reduction in wages. In a dynamic economy, the increased supply of workers may only slow the growth in wages. Still, wages end up lower than what they would have been in the absence of the additions to the labor supply.
The macroeconomic context complicates the issue. The supply of labor coming from potential welfare recipients raises the Nation’s capacity to produce. If the economy generates enough demand for goods and services, the additional workers help raise total production, thereby increasing the amounts available for consumption, government spending, and investment. In the current context, additions to the supply of labor are necessary for the U.S. economy to continue to grow at rates of 3 percent to 4 percent per year. For the Nation as a whole, the extra growth in labor force participation of single mothers is raising the total growth of the workforce from about 1.2 percent to about 1.4 percent or 1.5 percent per year. This additional capacity is helping the U.S. economy sustain rapid economic growth without increasing inflation.
While healthy from the perspective of the overall economy, the flow of welfare recipients into the workforce might flood the market for less educated workers, thereby limiting their job opportunities and wage growth. Such negative impacts would be especially troublesome in the context of the declining real wages experienced by less educated workers.
Demographics may help explain the national labor market’s success in absorbing single mothers, as well as other less educated workers, in recent years. Because of the wide gap in educational levels between the relatively less educated retiring workers and the more educated workers entering the labor force, nearly all of the absolute growth in the labor force has been among workers with at least some college education. Between 1989 and 1999, of the 17.1 million persons added to the working-age adult population (25- to 64-year-olds), 15.6 million had a postsecondary education, while the number without a high school degree or the equivalent actually declined by 1.9 million.(16) Thus, far from flooding the market, the additional single mothers in the workforce prevented an even more rapid downward shift in the supply of less educated workers.
In examining wage trends, one must take account of compositional as well as market effects. Wages might fall or rise especially slowly not because of the expanded supply of single mothers, but rather because the new single mothers (and competing groups) entering the labor market have lower skills than single mothers who are already employed at the beginning of the period. This compositional factor could be particularly important for single mothers, because they experienced the most rapid increase in the labor force and, potentially, the largest compositional shift. In that context, it would be a mistake to view wage trends as representing the gains or losses of a fixed group of workers.
Theoretically, while either result–suppressed wages or growing wages–is possible, the conventional prediction is that the wages of single mothers or their competitors will be clearly lower than they would have been with no welfare reform. The results presented herein do not provide a direct test of this hypothesis, because what is measured is only what did happen, not what would have taken place in the absence of the welfare act or similar reforms. That is, what was observed were actual wage gains–they might have been higher or lower had welfare changes not occurred.
Still, the actual wage trends do provide some indication of plausible effects, especially when wage changes among single mothers and less skilled groups are compared with changes across all adult workers. So far, the outcomes look promising. In the 20 metropolitan areas as a whole, all single mothers (including less educated ones) experienced an increase in wages. Single mothers earned an average of $10.59 per hour in 1995-96 and $11.67 per hour in 1998-99 (unadjusted for inflation), a gain of 9.7 percent. The 1998-99 mean earnings level was 77 percent of the hourly rate paid to all 20- to 45-year-old workers. The median hourly rate for single mothers stood at $8.79 before the welfare reform act and then rose to $10.00 3 years later. Among the less educated single mothers, the median wage reached $8.00 per hour in 1998-99, up from $7.50 in 1995-96.
In all 20 metropolitan areas, some less educated workers saw their wages grow faster than did single mothers. Annual median wage growth was 4.3 percent for all persons, 4.3 percent for single mothers, 3.6 percent for less educated single women without children, and 6.4 percent for less educated unmarried men. (See chart 1.) The lowest wage group of single mothers (the bottom 25 percent) saw nominal wage increases of 4.1 percent per year, a growth rate that was only slightly less than the average growth rate of 4.2 percent for the lowest 25 percent of the total population of adult workers.
[GRAPH OMITTED]
Although a full model of metropolitan area wage determination is beyond the scope of this article, a test can be performed for the presence of a negative relationship between large inflows of single mothers into the labor market and wage growth among less skilled workers. To obtain adequate samples of wages in each of the 20 metropolitan areas, two groupings that act as proxies for less skilled workers were used: workers at the 25th percentile of wages and workers with no more than a high school education. Surprisingly, as charts 2 and 3 indicate, there appears to be no connection between labor force inflows and wage growth among these two groups of workers. The correlations were -.12 for workers at the 25th percentile and -.00 for less educated workers. Thus, wages tended to rise as fast in metropolitan areas with large increases in single mothers joining the labor force as in other areas. In relating the labor force growth of single parents to the wage growth of potential competitors, such as all women with a high school degree or less, again, no wage-depressing effects were found.
[GRAPHS OMITTED]
In the five metropolitan areas with adequate samples to determine the wages of single mothers, the wage trends among working single mothers were similar to trends among all workers. (See chart 4.) Los Angeles, a metropolitan area with above-average unemployment, generated the weakest growth in nominal wages. Surprisingly, wage growth in Detroit was far above average for all workers and for single mothers. As of 1998-99, the median wage of single mothers varied widely across the five cities, from about $10.59 per hour in Chicago and $12.00 in Washington, DC, to between $9.00 and $9.50 in Los Angeles and New York. Again perhaps surprisingly, the gap in median wages between single mothers in low- and high-wage areas was nearly as wide as the gap in median wages between all single mothers and all workers.
[GRAPH OMITTED]
The current flexibility of labor markets
Changes in the Nation’s welfare system apparently did not lead to deleterious consequences for the labor market position of either single mothers or less educated workers as a whole. Despite the substantial flow of single mothers into the job market, metropolitan areas generated more than enough jobs to employ these new entrants, thereby reducing unemployment rates for single mothers and for their potential competitors. Nor has any sign of wage erosion among other less educated adults materialized, although in some areas single mothers themselves have experienced slow wage growth. Still, the wages of single mothers kept pace with the average growth in wages.
Thus, far from weakening the job market, the increased labor force participation by single mothers came at an opportune time. Not all single mothers, though, reaped the benefits of the Nation’s robust economy. Some left welfare without becoming employed.(17) Others stayed on welfare and out of the workforce because of health problems, extremely low skills, or other perceived or actual barriers to employment.(18) A serious recession would certainly weaken the wage and employment picture for single mothers and other less educated workers. Still, the record shows that, at least until now, metropolitan labor markets have been flexible enough to generate sufficient jobs for single mothers and other groups without eroding wages.
Notes
ACKNOWLEDGMENT: This work was financed in part by the Assessing the New Federalism project at the Urban Institute. The views expressed are those of the authors and do not necessarily reflect those of the Urban Institute, its board, or its sponsors. The authors thank Jesse Valente, Patrick Sharkey, and Stephanie Riegg for providing excellent research assistance and Pamela Loprest, Alan Weil, and Corinna Nicolaou for useful comments. The work was presented at the 22nd Annual Research Conference of the Association for Public Policy and Management, Seattle, Washington, November 2-4, 2000.
(1) See, for example, Sheldon Danziger and Jeffrey Lehman, “How Will Welfare Recipients Fare in the Labor Market?” Challenge, March-April 1996, pp. 30-35; and Peter Edelman, “The Worst Thing Bill Clinton Has Done,” Atlantic Monthly, March 1997, pp. 43-58.
(2) The single mothers most likely to participate in welfare programs are those who have never married (as opposed to divorced, separated, or widowed single mothers).
(3) The numbers are from unpublished tabulations provided by the Bureau of Labor Statistics.
(4) Katherine Allen and Maria Kirby, Unfinished Business: Why Cities Matter to Welfare Reform (Washington, DC, Brookings Institution Center on Urban and Metropolitan Policy, 2000).
(5) Robert Solow, Work and Welfare (Princeton, NJ, Princeton University Press, 1998).
(6) Timothy J. Bartik, “Displacement and Wage Effects of Welfare Reform,” in David E. Card and Rebecca M. Blank, Finding Jobs: Work and Welfare Reform (New York, Russell Sage Foundation, 2000), pp. 72-122.
(7) Robert I. Lerman, Pamela Loprest, and Caroline Ratcliffe, How Well Can Urban Labor Markets Absorb Welfare Recipients? Assessing the New Federalism, no. A-33 (Washington, DC, Urban Institute, 1999).
(8) Jared Bernstein, Welfare Reform and the Low-Wage Labor Market: Employment, Wages, and Wage Policies, Technical Paper 226 (Washington, DC, Economic Policy Institute, 1997).
(9) The figure was tabulated with data from National Income and Product Accounts, Second Quarter 2000 GDP Revised Estimates (Bureau of Economic Affairs, 2000); on the Internet at http://www.bea.doc.gov/ bea/newsrel/gdp200a.htm.
(10) Data are from Usual Weekly Earnings of Wage and Salary Workers, second quarter 1996 and second quarter 1999 (Bureau of Labor Statistics, 1996 and 1999); on the Internet at http://www.bls.gov.
(11) Large samples of single mothers and other low-skilled groups are necessary to yield reliable estimates of employment and earnings. Even with 12 months of CPS data in each year and samples of 10,532 and 11,877 single mothers in the two study periods, respectively, the sample size was too small (fewer than 100 cases) to calculate employment levels for some subgroups in several of the 20 metropolitan areas. Because the CPS asks only one quarter of each month’s sample about weekly wages, the sample size for weekly wage data on some subgroups was adequate only in the 5 largest metropolitan areas and all 20 metropolitan areas combined.
(12) Laura Wheaton and Linda Giannarelli, “Underreporting of Means-Tested Transfer Programs in the March CPS,” in 2000 Proceedings of the Section on Goverment Statistics and Section on Social Statistics (Washington, DC, American Statistical Association).
(13) The sample size was too small for the calculation of reliable estimates in the San Jose metropolitan area.
(14) The 10-percent increase in the single-mother workforce was lower than the 26-percent rise in the participation rate of single mothers because of a decline in the population of the group.
(15) Lerman, Loprest, and Ratcliffe, Urban Labor Markets.
(16) Authors’ calculations from CPS, March 1990 and March 1999.
(17) Pamela Loprest, How Families That Left Welfare Are Doing: A National Picture, Assessing the New Federalism, No. B-1 (Washington, DC, Urban Institute, 1999).
(18) Sandra Danziger, Mary Corcoran, Sheldon Danziger, Colleen Heflin, Ariel Kalil, Judith Levine, Daniel Rosen, Kristin Seefeldt, Kristine Siefert, and Richard Tolman, Barriers to the Employment of Welfare Recipients (Ann Arbor, MI, University of Michigan Poverty Research and Training Center, 2000); on the Internet at http://www.ssw.umich.edu/poverty/ wesappam.pdf.
Robert I. Lerman is an economist at the Urban Institute and professor of economics at American University, Washington, DC. Caroline Ratcliffe is an economist at the Urban Institute. E-mail: blerman@ui.urban.org
COPYRIGHT 2001 U.S. Bureau of Labor Statistics
COPYRIGHT 2004 Gale Group
