by Peter Gottschalk"
This essay reviews the evidence on the extent to which low-skilled workers gain from labor market experience. Entry-level jobs for workers with few skills are viewed by some as stepping stones to better jobs. Others see them as the first in a string of dead-end jobs. Is the empirical evidence consistent with either of these views?
The answer to this question is particularly important given the recent shift to a work-based welfare strategy. Will the work requirements lead to early labor market experience that will in turn raise future wages Will the increase in wages be sufficiently large to lead to self-sufficiency This hope for the success of work programs designed under the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) is clearly articulated in the following program descriptions:
- Work First programs share a common philosophy regarding work: any job is viewed as a good job and program efforts should be geared toward helping recipients enter the paid labor force as quickly as possible (Holcomb et al. 1998, p. 4).
- The best way to succeed in the labor market is to join it. It is believed that job advancement and higher wages will come from the experience of working rather than from first building skills through education and/or training (Holcomb et al. 1998, p. 13).
If early labor market experience does have long-term consequences, then these and other work-oriented programs will provide the initial push that will lead to self-sufficiency.
This essay is divided into three parts. The first describes the difficult methodological issues that must be addressed in any study trying to measure the impact of early labor market experience for low-skilled workers. The two sections that follow provide evidence based on experimental and nonexperimental evidence. The final section contrasts the state of knowledge on this question to the state of knowledge about the effects of human capital programs.
The Evaluation Question
Before turning to the evidence on whether work-based programs are likely to have long-term impact on wages, it is important to understand exactly the question being asked and the conditions under which different types of evidence answer this question. The question is much more difficult to answer than is commonly assumed.
The claim that taking a low-wage job will improve future wages is a claim about an actual event — the person's actual outcome — and a hypothetical outcome — what would have happened if the person had taken the job. Clearly, the counterfactual is not observable. The inability to see people in both states (having taken the job and not having taken the job) is the fundamental issue in evaluation studies. If we could only rewind the tape and have the person follow the alternative path, we would have the answer. This is, however, not possible, so the question can only be answered by making assumptions that allow the researcher to compare other people's outcomes to infer what would have happened to the person had he/she followed the alternative path. These "identifying assumptions" are the key to answering the evaluation question.
Randomized experiments require the weakest assumptions. Under this approach, people are randomly assigned to a "treatment" group and a "control" group. A treatment (e.g., job placement) is randomly assigned to a subset of the participants. In a classical experiment, the effect of the treatment is measured by the difference in outcomes between the treated and the control groups. The only assumption necessary is that the treated group would have had the same outcomes as the control group had they not received the treatment. This assumption is justified by random assignment. If people are truly assigned randomly to the treatment and control groups, then the only systematic difference between the two groups is the treatment.(1) Therefore, differences in outcomes between the two groups reflect the pure effect of the treatment, not unobserved differences in outcomes that would have taken place even without the treatment.(2)
The second method of measuring the impact of early labor market experience on future outcomes is to use nonexperimental data. For example, entry-level jobs may be more plentiful in some areas than others. The resulting variation in early work experience, which may affect future wages, potentially takes the place of random assignment. It is, however, now necessary to assume that this nonexperimental assignment is random. If people with skills live in areas with more plentiful jobs or higher wage growth, then it is not possible to separate the effects of having obtained a job from the effects of having more skills.(3) Therefore, if one were to tabulate the adult wages of persons classified by their initial employment status, or to estimate regressions with adult wages as the dependent variable and initial employment status as one of the independent variables, one would implicitly be assuming that initial employment status was random. However, more motivated people are more likely to find jobs initially and are more likely to receive high wages later in life. Comparing their wages with the wages of less motivated people who did not have the early labor market experience tells us little about the wages the latter group would have received had they landed a similar job.(4)
While nonexperimental studies have these drawbacks, they may still be useful if they can be used to obtain bounds on the possible future effects of work on the target population. For example, it might be reasonable to assume that nonworkers would have gained less from early labor market experience than those who did work. In this case, nonexperimental evidence would give us an upper bound on what the nonworkers would have gained from work had they had early market experience. For example, knowing the impact of work experience on future wages for the welfare mothers who did work may give us an upper bound on the potential gains from work for those who did not work.
During the 1980s, states were given wide latitude to experiment with programs aimed at increasing future employment and earnings of welfare recipients. Many states set up services, such as job search workshops, that made it easier for participants to find work. As part of these welfare-to-work projects, states were required to monitor the experiences of participants and a control group.
At one level, these experiments provide support for the proposition that early labor market experience yields subsequent benefits. Three years after receiving job placement services, the earnings of experimentals were higher than the earnings of the control groups.(5) The annual gains were, however, small. Most differences were in the $200 to $300 range, with few of the programs showing gains over $600.(6) While gains of this size are not trivial for people with limited incomes, they are not large enough to lead to self-sufficiency. Even after reaping the benefits of the program, the earnings of the experimentals were still in the $2,000 to $3,000 range.
Two more pieces of evidence from the welfare-to-work experiments are relevant. First, the gains in earnings of the experimentals were primarily the result of increased hours, not increased wages. Women were working more but at the same low rate of pay, indicating that initial jobs were not stepping stones to higher-paying jobs. Second, the benefits of almost all these programs did not continue through the fifth year.(7) The five-year follow-up evaluation indicates that the earnings gains of experimentals over controls during the first three years largely reflected a shorter time to obtain the initial job, not better future outcomes for people who obtained these jobs. Almost all of the employment gains were from a one shot faster job placement for experimentals, not from a lasting increase in wages or hours. By the fifth year, the difference between experimentals and controls had largely vanished for all but one site. And for this site (Baltimore's Options program) it is difficult to disentangle the effects of the training component from the work component, since it included a substantial human capital component and fewer work mandates.(8) Furthermore, the gains were still not large, $475 in the fifth year, but at least they came from higher wages and were lasting.(9)
The experimental evidence does not point to large and lasting benefits from employment services for less-skilled workers. Putting a person in a job does have the immediate impact of raising their earnings, which may be sufficient grounds for advocating a work-based-strategy.(10) But it does not support the hope that initial labor market experience will lead to self-sufficiency.
In this section, two types of nonexperimental evidence are reviewed. The first comes from the broad-based studies of the impact of experience on wages. The second set of studies are more tightly focused on persons with limited skills.
Claims about the long-term benefits of work partially reflect the well-documented fact that the average wages of persons with more labor market experience are higher than the wages of persons with less work experience. The growth in average wages are particularly large in the first few years on a job. Loprest (1992) estimates that the average gain in earnings during the first four years of full-year work is 7.1 percent per year for males and 5.7 for females.(11) Part of this reflects growth in wages on the job, but a large part reflects wage increases that accompany job changes. While the gains from experience decline with age, they still remain substantial. For example, the four-year growth for persons with 10 years of experience is about 10 percent per year for males and 5 percent for females.(12) Thus, if getting the target population started into work world yields these average benefits, then there would be substantial long-run benefits from work.
The problem with using this evidence on the average return is that the rewards to experience vary widely across the population. The estimated average return includes persons who gain little or nothing from greater work experience. Their below average return to experience is offset by others who gain substantially from working an extra year, thus pulling up the average. Likewise, the average gains from job changes may not be relevant for the population of policy concern. What may be stepping stones to better jobs for some may be dead-end jobs for others. Thus, averages may tell us very little about what would happen if the target population were gaining experience.
More tightly focused studies are, therefore, required to measure future benefits to a low-skilled worker from taking a job. The key question is how to tighten the focus. One approach is to limit the analysis to persons who started with low wages. But this is likely to be misleading for two reasons. First, low-wage workers include a large number of young people who face very different prospects than the population of policy concern.(13) Again, we are faced with a counterfactual question. If a welfare recipient had taken the same lowwage job that went to a college-bound 18-year-old, would she have experienced the same wage growth This seems unlikely. There are too many differences between an AFDC recipient and a college-bound teenager to use one's experience to infer the experience the other would have had.(14) Second, some of these people received low wages temporarily, which would have rebounded in any case, even if the person had not landed the job. This "reversion to the mean" will tend to overstate the gains to those who were not on the rebound.
An alternative is to limit the analysis to persons who have characteristics similar to the target population, such as persons in low-income households or welfare recipients.(15) But even this will not be satisfactory if the workers in these groups are systematically different from the target population. For example, if the welfare mothers who work are more motivated or living in areas with greater access to jobs, then their future outcomes could reflect these differences rather than the causal effects of their lack of early work experience. If these individual characteristics (e.g., greater ability or motivation) are the source of their success, then we cannot use their experiences to infer what would have happened to less-skilled or less-motivated recipients had they gained early labor market experience.
A few studies have tried to estimate the returns to experience of welfare recipients and some have tried to adjust their estimates for the fact that the welfare recipients who did work were likely to be the ones who would gain the most from work.(16) Moffitt and Rangarajan (1989) estimate wage trajectories for female heads with children less than 18 years old.(17) Their estimates indicate that the average wage growth of an 18-year-old former recipient is only 2 percent per year.(18) These very flat profiles are not directly comparable to the much steeper experience profiles discussed above (5.7 percent gain), which are for full-time work. Since many former recipients work part-year, their returns are proportionally lower. For example, if full-year work increases wages by 5.7 percent per year, then half-year work would increase wages by roughly 2.9 percent The relevant growth rate depends crucially on the question being asked. Do we want to know the wage gains former welfare recipients would have experienced if they had landed full-time/full-year jobs or the wage gain from the jobs they would typically get?(19)
Again, it is important to recognize that the estimated wage growth of 2 percent per year for welfare recipients who landed a job is an upper bound estimate of the wage growth for non-working recipients. These non-working recipients either had less to gain from working or were less likely to be hired if they did apply for jobs. As a result, their expected wage growth is likely to be lower than the 2 percent per year wage growth for those who did land jobs.
Bartik (1997) also attempts to estimate the returns to work for welfare recipients by using a sample of female heads who received some form of public assistance in the previous year. He estimates that the female heads who worked the average number of hours per week in the previous year (31.5 hours) at the average wage rate ($5.14) experienced a 4.5 percent increase in wages.(20) Part of the difference between these estimates and those from Moffitt and Rangarajan (1989) seem to lie in the high proportion who worked last year (30.4 percent) and the high average number of hours per week worked last year.(21) It is again unlikely that those welfare recipients who did not work would have had as high a wage growth or worked as many hours. Therefore, this again forms an upper bound on what non-working recipients would have gained had they worked.
Two other studies give indirect evidence on the future value of work experience. Burtless (1995b) traces the wage histories of a cohort of young women who received AFDC in the late 1970s. These young recipients had an average wage that fluctuated in the $6 to $7 range throughout the next 10 years. Two things are striking about Burtless's numbers. First, these wages are low. Even if current recipients could earn as much as these women who voluntarily entered the labor market, they would still earn less than $15,000 a year working at a full-time, full-year job. Second, and more importantly for our question, is the low change in wages as these women gain experience. Over a 12-year period the average wages of these former recipients grew by less than $1. This amounts to a growth rate of roughly 1 percent per year, which is even lower than from Moffitt and Rangarajan's (1989) estimate and consistent with the small long-term effects on wages of work found in the experimental literature. Corcoran and Loeb (1999) also find small average wage gains for former welfare recipients (wages rise from $6.32 to $7.86 over the course of the panel). This seems to be largely the result of former welfare recipients working few weeks and being disproportionately likely to work part-time, which has little or no effect on future wages. However, when former welfare recipients do work full-time, their future wages increase by roughly 6 percent for every year of work.
Finally, Newman (1999) has followed a sample of 350 low-wage workers households who applied for entry-level jobs in fast-food establishments.(22) Newman has recently reinterviewed 103 of these low-skilled women to find out whether their initial success at entering the work world had paid off. For roughly a third, the news is good. They have moved up in the fast-food business or obtained jobs in unionized shops and are making more than $10 per hour. Another third is earning between $5.50 and $10 while the remainder of those who initially landed jobs in the fast-food industry now find themselves unemployed or still at the minimum wage. The fact that a third are earning $10 or more and that another third experienced some growth in wages is impressive, but this must be tempered again by the caveat that these growth rates almost certainly overstate the returns to experience of people who do not seek or cannot find entry-level jobs. These job applicants had considerably higher education and experience than the typical welfare recipients. In fact, even the applicants who were turned down for the fast-food jobs had higher education than the typical welfare recipient.(23)
Impact on Earnings
Given this experimental and nonexperimental evidence, what can one conclude about the ability of less-skilled workers to attain self-sufficiency through work While the estimates of returns to experience for less-skilled people vary widely, there is a strong consensus that they all start with low wages and that they work only part-year. Thus, even fairly high rates of growth in wages will leave them with low earnings.
This is illustrated in table 1, which shows estimated wages, earnings and earnings plus the Earned Income Tax Credit minus child care costs for a woman with two children.(24) This example uses an entry wage level of $5.50, which is consistent with evidence on the wages received by welfare recipients mandated to work.(25) Future wages are calculated under alternative assumptions about growth rates. Columns 1 and 2 use a wage growth rate of zero, as suggested by the experimental literature. Columns 3 and 4 assume that low-skilled workers who did not work would have experienced growth rates half as much as those estimated by Moffitt and Rangarajan (1989) for former welfare recipients who did work. Columns 5 and 6 are labeled "high growth" since these columns assume that low-skilled persons who did not work would have experienced the same growth as those who did work. In order to translate wage rates into annual incomes, earnings for persons working quarter-time and half-time are used. This covers the range of hours worked in the experiments reviewed in Friedlander and Burtless (1995).(26)
Growth in Wages and Earnings Net of EITC and Child Care Cost for a Mother with Two Children
(Poverty Line of $12,641)
|No Growth||Low Growth||High Growth|
|Earnings plus EITC|
|Earnings plus EITC minus Child Care|
|Full time adj||0.5||1||0.5||1|
Since columns 1 and 2 assume no growth from experience, wages are constant at $5.50 per hour at all ages. This translates into annual earnings of $2,503 for quarter-time work and $5,005 for half-time work. The EITC raises this to $3,504 for quarter-time and $7,007 for half-time work, but the cost of child care brings net earnings back down to $2,254 and $4,507 for these two groups. With a poverty level of $12,641 for a single woman with two children, this leaves the family in poverty, even if the mother works half-time year-round. In fact, even full-time work would leave a three-person family in poverty. Thus, if the experimental evidence is correct that wages do not grow with experience, then even a full-time job will leave the family poor.
As discussed above, the nonexperimental evidence is more optimistic. If low-skilled workers experienced half the growth in wages of working recipients, then a mother working quarter-time year-round in all years between ages 18 to 40 would find herself with a wage of $6.14 by the time she reached 40, which yields annual earnings of $2,795. Her income net of child care costs plus the EITC would be $2,663. A woman working half-time would have higher earnings both because she worked more hours and because she gained more experience. As a result, the $5.50 wage would grow to $6.84 by the time this half-time worker reached 40. After the EITC and child care cost, this woman would have an income of $6,209, or 49 percent of the poverty level.
If one were to assume that the gains from work would be as high for the target population as it was for the sample of working former recipients, then the $5.50 wage at age 18 would have grown to $8.50 by the time this half-time worker reached 40. While this rapid rate of growth would have raised wages by 54 percent over this 22-year period, the person started with such a low wage that net earnings would still only be 70 percent of the poverty level for a family of three.
This essay has reviewed the evidence on the impact of work experience on future wages for individuals with limited skills. The fundamental difficulty in obtaining any empirical evidence is that one must infer what the person's future earnings history would have been had the person had more early labor market experience. Since it is impossible to observe persons both with and without early labor market experience, it is necessary either to use experimental evidence or to assume that the nonexperimental evidence gives an upper bound on the benefits of experience.
The experimental literature indicates that gains are small at best and not long lasting. Earnings may go up by $300 in the first few years, but even these gains are the result of finding jobs more quickly while on the program or working more hours after the program, not receiving higher pay per hour. The nonexperimental evidence offers some evidence of future effects of work but wages start from such a low base that even rather optimistic assumptions about growth in wage rates still leads to earnings below the poverty level for families working less than full time.
Does this evidence mean that work programs have few benefits The answer depends on what is counted as a benefit of work. Society (and the former welfare recipients) may value work in its own right. Work does not need to raise future wages to be valued. It is, however, important not to overstate the future benefits of work.
An important lesson can be learned from the U.S. experience with training programs.(27) During the 1960s and 1970s training was viewed as the key to self-sufficiency. Training would provide skills that would lead people off the welfare rolls. The often-cited claim that teaching people how to fish is better than giving them fish became the conventional wisdom. But in the final analysis, teaching people how to fish well enough to support a family turned out to be considerably harder than was initially thought. The early evaluations of training programs showed little or no gains, while the more recent evaluations show modest gains for some, but not all, segments of the population. Fairly ambitious training programs raised earnings of welfare recipients on the order of $500 to $1,000, which is in the same order of magnitude as some of the higher estimates of the future returns to work programs.(28) While the benefits were small, they were larger than the costs in many cases. Thus, they met basic cost-benefit criteria. What they did not meet was the exaggerated claims that human capital programs would lead to self-sufficiency. Having discarded these programs in favor of a jobs-based strategy, we now stand the very real chance that in 10 years, after careful evaluations, we will conclude that work is also not the magic bullet.
Bartik, Timothy J. 1997. "Short-Term Employment Persistence for Welfare Recipients: The Effects of Wages, Industry, Occupation, and Firm Size." W.E. Upjohn Institute for Employment Research, Mimeo, June.
Blank, Rebecca M. 1997. It Takes a Nation: A New Agenda for Fighting Poverty. Princeton, N.J.: Princeton University Press.
Burtless, Gary. 1995a. "The Case for Randomized Field Trials in Economic and Policy Research." Journal of Economic Perspectives (Spring): 63-84.
------. 1995b. "Employment Prospects of Welfare Recipients." In The Work Alternative: Welfare Reform and the Realities of the Job Market, edited by Demetra Smith Nightingale and Robert H. Haveman. Washington D.C.: Urban Institute Press.
Corcoran, Mary, and Susanna Loeb. 1999. "Welfare, Work Experience, and Economic Self-Sufficiency." Mimeo, University of Michigan.
Friedlander, Daniel, and Gary Burtless. 1995. "Five Years After: The Long-Term Effects of Welfare-to-Work Programs." New York: Russell Sage Foundation.
Gottschalk, Peter. 1997. "Inequality, Income Growth, and Mobility: The Basic Facts." Journal of Economic Perspective 11 (2, Spring): 21-40.
Gueron, Judith M., and Edward Pauly. 1991. From Welfare to Work. New York: Russell Sage Foundation.
Heckman, James J., and Jeffrey A. Smith. 1995. "Assessing the Case for Social Experiments." Journal of Economic Perspectives (Spring): 85-110.
Holcomb, Pamela, La Donna Pavetti, Caroline Ratcliffe, and Susan Riedinger. 1998. Building an Employment Focused Welfare System: Work First and Other Work-Oriented Strategies in Five States. The Urban Institute, June.
LeLonde, Robert. 1995. "The Promise of Public Sector-Sponsored Training Programs." Journal of Economic Perspectives (Spring): 149-68.
Lennon, Chauncy, and Katherine Newman. 1995. "Finding Work in the Inner City: How Hard Is It Now How Hard Will It Be for AFDC Recipients?" Russell Sage Foundation Working Paper 76, October.
Loprest, Pamela J. 1992. "Gender Differences in Wage Growth and Job Mobility." The American Economic Review, Volume 82, Issue 2. Papers and proceedings of the 104th annual meeting of the American Economic Association, May.
Moffitt, Robert, and Anuradha Rangarajan. 1989. "The Effect of Transfer Programmes on Work Effort and Human Capital Formation: Evidence from the US." In The Economics of Social Security, edited by Andrew Dilnot and Ian Walker. London: Oxford University Press.
Newman, Katherine. 1999. No Shame in My Game: The Working Poor in the Inner City. New York: Knopf/Russell Sage.
Schiller, Bradley R. 1994. "Moving Up: The Training and Wage Gains of Minimum-Wage Entrants." Social Science Quarterly 75 (3, September): 623-36.
U.S. Department of Health and Human Services, U.S. Department of Education. 1998. "National Evaluation of Welfare-to-Work Strategies: Implementation, Participation Patterns, Costs, and Two-Year Impacts of Portland (Oregon) Welfare-to-Work Program." Manpower Demonstration Research Corporation, May.
U.S. Congress. 1993. House Committee on Ways and Means. 1993 Green Book: Overview of Entitlement Programs. July 7.
1. In practice, few experiments achieve these ideals. For a discussion of these issues, see Burtless (1995a), and Heckman and Smith (1995).
2. In practice, the randomized experiments that have been implemented have offered job placement services to the experimental group but not to the control group. Differences in future outcomes, therefore, reflect both the impact of the services on the probability of obtaining a job and the impact of having a job on future outcomes.
3. While it is possible to control for differences in some indicators of skills, such as education, it is not possible to control for unobserved differences, such as ambition.
4. "Natural experiments" in which differences in job availability creates the initial differences in work experiences but not future differences in wages are difficult to imagine. For example, people living in depressed areas will have a lower probability of working initially and they would also have lower wages if they worked.
5. See Gueron and Pauly (1991, table 1.1).
6. In almost all cases, the earnings gains were larger than the program cost indicating that these strategies are cost effective.
7. See Friedlander and Burtless (1995) for the five-year follow-up.
8. The Baltimore program offered Aid to Families with Dependent Children (AFDC) recipients a variety of options, including training. The Portland JOBS program, which went even further in this direction by providing training and encouraging people to seek and accept "good" jobs rather than accepting any job, is another of a handful of jobs with multiple year impacts. (See Manpower Demonstration Research Corporation 1998.)
9. See Friedlander and Burtless (1995, table 1-2).
10. These programs also tend to be cost effective. The small benefits are achieved with even smaller costs.
11. The difference between men and women may reflect the combined impact of lower wage growth from part-time work and greater part-time work among women.
12. See Gottschalk (1997). Furthermore, if this age premium continues to increase, the gains will be even larger for current welfare recipients who take jobs.
13. For example, Schiller's (1994) study of wage gains of minimum wage workers largely reflects the expected absolute gains for youth as they move to jobs that are not likely to be available to the target population.
14. Even if one could condition on a large number of observable factors, this would still leave many unobservable differences that are correlated with who lands the job.
15. It is necessary to assume that these variables are exogenous.
16. In ongoing work, Connolly and Gottschalk estimate both wage growth and job duration models for females disaggregated by education. Preliminary results indicate that high school dropouts experience both lower wage growth within jobs and lower growth in starting wages across jobs than females with more education. In this sense, the jobs filled by high school dropouts are "dead-end jobs. " Connolly and Gottschalk also find that high school dropouts are less likely to leave their jobs. Furthermore, those dropouts with the lowest wage growth are the least likely to leave their jobs. In this sense, high school dropouts are "stuck" in "dead-end jobs."
17. Their focus is on whether welfare participation has a causal impact on wages but their estimates yield wage growth rates for recipients, which is the focus of this review. Since welfare participation is based on income, they adjust their estimates to take account of the fact that welfare recipients are likely to have lower wages simply as a result of selection into the program, whether or not being on the program has any causal effect on wages. The methodological problem is how to determine whether low wages cause participation or participation causes low wages. One approach is to find "instruments," which are variables that mimic what random assignment would have done. For example, Moffitt and Rangarajan use the AFDC guarantee and family size as variables that affect participation but not wages. If women with larger families or living in high-benefit states are more likely to participate in AFDC, then this exogenous variation can be used to infer the impact of participation on wages.
18. This is based on the coefficients on age and age squared of .024 and -.0002 in column 1 of table 6.6.
19. Moffitt and Rangarajan answer the latter question.
20. Bartik (1997, table 9). Working more hours or at a higher wage was associated with larger wage gains.
21. Committee on Ways and Means (1993, table 31) indicates that only 6.4 percent of AFDC mothers worked in 1991.
22. About a quarter lived in households receiving public assistance.
23. See Lennon and Newman (1995).
24. Child care costs are taken from Blank (1997, table 7.1) and the EITC is calculated under 1998 rules. The EITC is equal to 40 percent of earnings up to a maximum of $3,656, which is reached when earnings equal $9,140. Persons with income between $9,140 and $11,930 receive the maximum benefit. The EITC benefit is then reduced by 21.06 percent for earnings above $11,930.
25. See Holcomb et al. (1998, table 8.1). While some sites had higher wages, there is a clear inverse relationship between the percent of welfare recipients who found work and their average wage. Culpeper, Virginia, had the lowest average wage ($5.37) but the highest proportion of recipients who found work (66 percent).
26. Friedlander and Burtless (1995, table 4-1) show average total earnings throughout the experiment for four sites. These can be converted into annual earnings. Dividing by $5.50 yields the number of hours a person would have had to work at $5.50 rate in order to achieve the average annual earnings. Experimentals in Arkansas had the lowest average annual earnings ($1,490) and Baltimore had the highest ($4,221). Using 35 hours per week as full-time, this translates to .14 of full-time for Arkansas and .42 for Baltimore.
27. See Gottschalk (1997) for a discussion of this issue.
28. While it was expected that training would raise wages, most of the increase in earnings reflected increases in hours worked rather than higher wages. Programs for adult men showed very little effect. For a review of the literature on evaluation of training programs, see LeLonde (1995).