Congress Spends Billions on Ineffective Job-Training Programs

Report Jobs and Labor

Congress Spends Billions on Ineffective Job-Training Programs

October 1, 2002 11 min read Download Report
David Muhlhausen
David Muhlhausen
Research Fellow in Empirical Policy Analysis
David B. Muhlhausen is a veteran analyst in The Heritage Foundation’s Center for Data Analysis.

President George W. Bush's proposed U.S. Department of Labor (DOL) budget for fiscal year (FY) 2003 is intended to promote effective programs while reducing or eliminating programs that are unproven or duplicative.[1] The Administration proposes $4.975 billion for job-training programs authorized under the Workforce Investment Act of 1998 (WIA)--a $505 million reduction when compared to FY 2002.[2] The budget also proposes decreasing funding for Youth Formula Funds, Youth Opportunity Grants, and Adult Employment and Training Activities.[3]

Given the absence of empirical evidence supporting the effectiveness of these programs, the Administration's request for decreases in funding for WIA programs is a move in the right direction. However, the Administration's budget still requests nearly $5 billion in funding for questionable DOL job-training programs,[4] including $1.54 billion for the Job Corps (which it has deemed a "highly successful" program[5] despite evidence to the contrary.[6] Job Corps program has failed to increase participants' wages to any substantial degree or to move participants into full-time employment.[7]

The President's goal of linking funding to program effectiveness has been further compromised by the U.S. Senate's Labor, Health and Human Services, Education Appropriations Bill (S. 2766). This bill continues the federal government's long history of funding job-training programs without regard to effectiveness.

The Senate intends to spend over $5.6 billion on Labor Department job-training programs--a 13.2 percent increase over the Administration's request.[8] Rather than continuing to spend tax dollars on job-training programs that are unproven or have negligible positive effects on participants, the funding could be more effectively allocated to reduce the federal government's budget deficit (which is projected to be $145 billion for FY 2003[9] ) or to support vital homeland security efforts.

In crafting their version of the Labor, HHS, Education Appropriations Bill for FY 2003, Members of the House now have an opportunity to rectify the failure of both the Administration and the Senate to curb wasteful spending. The call for reduced funding for federal job-training programs is justified by the inability of these programs to bring about substantial increases in the incomes of participants and move participants into full-time employment. While WIA programs have yet to be rigorously evaluated, similar programs that were funded under the Job Training Partnership Act (JTPA) of 1982 were found to be largely ineffective.[10]

With the return of budget deficits and the high cost of combating terrorism, Congress should reaffirm its commitment to de-fund ineffective programs by:

  • Further reducing funding for WIA programs beyond the Administration's request, and
  • Eliminating funding for the Job Corps.

Evolution of Federal Job-Training Programs

Federal programs that were intended to improve the lives of the economically disadvantaged[11] through work initiatives, including the Works Progress Administration (WPA) and the Civilian Conservation Corps (CCC), were originally created during the 1930s.[12] Despite the fact that civilian unemployment just prior to World War II was about the same as it was when the WPA and CCC were created, support for job programs continued to build after the war.[13]

During the 1960s, the federal government established job-training programs for the unemployed and economically disadvantaged. These programs provided a combination of remedial education, vocational training, on-the-job training, subsidized work experience, basic life-skills training, and job search assistance. Programs funded under the Manpower Development and Training Act (MDTA) of 1962 were originally intended to re-train workers dislocated by technological advances, but MDTA was converted into a job-training program for economically disadvantaged persons.[14]

In 1973, MDTA programs were superseded by the Comprehensive Employment and Training Act (CETA), which was designed to decentralize control of federally sponsored job-training programs.[15] Nearly a decade later, charges of corruption and mismanagement contributed to Congress's decision to replace CETA with the JTPA.[16]

In 1998, the Workforce Investment Act (WIA), which superseded the JTPA, was signed into law. WIA emphasized a "one-stop" approach in which an array of job-training, education, and employment services were provided to communities.[17] In contrast to the JTPA, state and local governments were given authority to tailor the content of job-training programs to meet their specific needs.

Measuring the Impact of Job Training

There is still no consensus regarding the ability of MDTA programs to increase the incomes of the economically disadvantaged. Some evaluations of MDTA have found that the program increased incomes,[18] while other studies reported a combination of outcomes--including positive, negative, and/or no effect on income.[19]

The lack of consensus over the effectiveness of job-training programs continued with regard to CETA, where evaluations again indicated mixed results. Some studies found that CETA programs raised incomes in some cases, while, in other cases, the programs had no effect on participants' income.[20] Other studies found that the incomes of some CETA participants actually declined.[21]

While the evaluations of major job-training legislation such as MTDA and CETA have reported mixed results, all of these studies suffered from methodical flaws that reduced their ability to measure the impact of the programs they researched. All of the studies cited above were quasi-experimental in design[22] and, as such, were likely to exhibit selection bias. Selection bias occurs when there are pre-existing differences between the intervention and control groups that affect outcome measures.[23] The evaluations of MTDA and CETA were not rigorous enough to inform policymakers adequately about the effectiveness of job training for the disadvantaged.

Determining the impact of social programs requires comparing the conditions of those who had received assistance with the conditions of an equivalent group that did not experience the intervention. Experimental studies in which eligible participants are randomly assigned either to intervention or to control groups represent the "gold standard" of evaluation designs. Random assignment allows the evaluator to test for differences between the experimental and control groups that are due to the intervention and not to pre-intervention discrepancies between the groups.

The federal government has sponsored experimental evaluations of at least two major federal programs--JTPA and the Job Corps.[24]

The National JTPA Study

The JTPA evaluation tracked program effects for adult men and women, and male and female out-of-school youths, over the course of 30 months.[25] Three types of job-training activities were examined. The first, classroom training, consisted mainly of occupational skills instruction and basic education.[26] The second intervention was a combination of on-the-job training and job-search assistance.[27] The third type of training, "other services," offered different services to participants on the basis of their age.

Adults mainly received job-search assistance, customized occupational-skills training, and on-the-job training.[28] Youths were enrolled in basic education courses along with "tryout employment," where participants were hired on a probationary status to learn the job, and "job shadowing," where they observed regular employees during the workday.[29]

Effect on Income. Members of the intervention and control groups were tracked over 30 months, and results were reported for three periods: months 1-6, months 7-18, and months 19-30.[30] Classroom training for adult men and women failed to raise the incomes of participants over the course of 30 months,[31] and the other JTPA interventions--on-the-job training and "other services"--failed to raise the incomes of adult male participants.[32]

Although, initially, the incomes of adult women participating in on-the-job training/job search assistance increased by an average of $484 (about $81 per month), this impact was fleeting.[33] During the subsequent measurement periods of months 7-18 and 19-30, the incomes of the women who received training did not differ from the incomes of the women in the control group.[34] Only the effect on the income of women in the "other services" category was positive and lasting. Women participants earned, on average, $478 more than non-participants during the first six months, while the impact increased to over $1,700 during the two subsequent time periods.[35]

In general, JTPA training had no significant positive impact on the incomes of female and male youths. Of the three intervention types, only female youths participating in on-the-job training/job search assistance experienced an income increase during the first six months.[36] The average income increases of $762 during the first time period faded away during the remaining 24-month follow-up period.[37] For male youths, the three types of intervention had no impact on income.[38]

In sum, JTPA programs were ineffective in raising the incomes of adult males or male and female youths, while only the "other services" elements appear to have had a sustainable impact on adult women.

JTPA's Effect on Hourly Wages. A program's effect on participants' hourly wages is an important indicator in any evaluation of job-training programs. Human capital theory suggests that if job-training programs do, in fact, increase the skills of participants, then employers would be willing to pay program graduates higher hourly wages in return for increased productivity. The JTPA programs did not have a statistically measurable effect on the wages of adult men and women (hourly wages of youths were not measured).[39] This indicates that, in the opinion of employers, JTPA did not increase the skills of participants.

A question may arise regarding how the incomes of some adult women in certain programs increased in comparison to the control group if their wages did not increase. This contradiction may be explained by the fact that adult women participants worked more hours than adult women in the control group.[40] For adult women and men, JTPA appears to be ineffective in boosting the skills of participants to a level that would allow them to secure higher-paying jobs.

The National Job Corps Study

Created in 1964, the Job Corps is a residential job-training program that serves disadvantaged youths aged 16-24 in more than 100 sites across the nation. In 2001, the National Job Corps Study was published. Its primary hypothesis relating to employment and earnings is that "youth who obtain Job Corps education and training will become more productive and, hence, will have greater employment opportunities and higher earnings than those who do not."[41]

While the Job Corps was founded to raise the incomes of participants, its limited impact has not justified the Administration's declaration that the program is effective. The estimated average increase in the weekly incomes of participants was never more than $25.20.[42]

Participants responded to the Job Corps differently by age, race, and ethnicity. The Job Corps did not increase the earnings of 18 and 19-year-olds, who represent 32 percent of the population served by the program.[43] During the fourth year of the follow-up period, whites and blacks experienced small increases in average earnings per week, while the Job Corps had no impact on income for Hispanics and other minorities.[44] These findings are alarming, given that Hispanics comprise 18 percent of all youths served by the program.[45]

Over the course of the 48-month study period, Job Corps participants actually worked less than the control group. Measured in the number of weeks employed over the course of the study, Job Corps participants were employed 45.2 percent of the time, while non-participants were employed 46.9 percent of the time--a difference of 1.7 percent.[46] Despite being enrolled in the Job Corps, participants failed even to put in a full year's worth of work.

Impact of Job Corps on Hourly Wages. Again, the best way to determine whether the Job Corps raises the skills of its participants is to look at the hourly wage of its graduates. If the Job Corps is effective in improving the skills of its participants, then it should have substantially raised the hourly wages they received. The National Job Corps Study measured hourly wages during two time periods. During the 10th quarter, Job Corps participants earned $0.24 more per hour than non-participants.[47] By the 16th quarter, this difference had decreased to $0.22 per hour.[48]

The study revealed that the Job Corps also had little impact on increasing the number of hours worked per week. During the course of the study, the average time participants spent working each week never rose above 28.1 hours.[49] Participants never averaged working more than two hours per week more than the control group.[50]

Those who did benefit from the Job Corps experienced only slight gains in income. The Job Corps appears to be incapable of providing the skills and training needed to bring about substantial increases in the wages of participants. Despite costing the taxpayer $16,500 per participant over an average of eight months, the program has failed to move a significant number of participants into full-time employment.[51]


To enhance the impact of every taxpayer dollar spent on job-training programs, Congress and the Administration should continually review programs and reduce funding for those that are ineffective. Especially at a time when the budget is strained and the need for security heightened, Congress should redirect the funding of any program that has been unproductive (i.e., has not achieved its stated or implied goals). Research indicates that this will mean reducing funding for a number of job-training programs.

To date, no evaluations of the effectiveness of WIA programs in improving the lives of participants have been conducted. Given the history of federally funded job-training programs, it would be unwise to increase funding for these programs as the Senate has recommended until credible evaluations of WIA have been conducted. Congress should reduce funding for WIA programs beyond the Administration's request.

Given the poor performance of the Job Corps, the spending recommendations of both the Administration and the Senate seem unreasonable. The Job Corps is not performing as the "highly successful" program it is touted to be. Congress should move to eliminate this wasteful and unproductive program.

David B. Muhlhausen is a Senior Policy Analyst in the Center for Data Analysis at The Heritage Foundation.

1.U.S. Department of Labor, Employment, and Training Administration, " Summary of ETA Fiscal Year 2003 Request," at (August 27, 2002).

2.Departments of Labor, Health, and Human Services, and Education, and Related Agencies Appropriation Bill, 2003, Report 107-216, U.S. Senate, July 22, 2002, p. 4.

3.U.S. Department of Labor, Employment and Training Administration, " Summary of ETA Fiscal Year 2003 Request."

4.Departments of Labor, Health, and Human Services, and Education, and Related Agencies Appropriation Bill, 2003, p. 4.

5.U.S. Department of Labor, Employment and Training Administration, " Summary of ETA Fiscal Year 2003 Request."

6.Peter Z. Schochet, John Burghardt, and Steven Glazerman, National Job Corps Study: The Impacts of Job Corps on Participants' Employment and Related Outcomes (Princeton, N.J.: Mathematica Policy Research, Inc., June 2001).


8.Departments of Labor, Health, and Human Services, and Education, and Related Agencies Appropriation Bill, 2003, p. 4.

10.Larry L. Orr, Howard S. Bloom, Stephen H. Bell, Fred Doolittle, Winston Lin, and George Cave, Does Training for the Disadvantaged Work? Evidence from the National JTPA Study (Washington, D.C.: The Urban Institute Press, 1995).

11.The term "economically disadvantaged" has class warfare connotations that there are structural impediments in society that prevent lower-class individuals from moving up the socioeconomic ladder. While the term is used throughout the paper, the theory that individuals lack the ability to overcome poverty and joblessness through a strong work ethic is rejected.

12.For background on the origins of federal employment programs, see Frank Freidel, Franklin D. Roosevelt: Launching the New Deal (Boston: Little, Brown, 1973); William E. Leuchtenburg, Franklin D. Roosevelt and the New Deal, 1932-1940 (New York: Harper Colophon Books, 1963); Broadus Mitchell, Depression Decade: From New Era Through New Deal, 1929-1941 (New York: Harper Torchbooks, 1969); and John Joseph Wallis and Daniel K. Benjamin, "Public Relief and Private Employment in the Great Depression," Journal of Economic History, Vol. 41 (December 1993), pp. 97-102.

13.U.S. Department of Commerce, Bureau of the Census, Historical Statistics of the United States: Colonial Times to 1970, Part 1, 1975, p. 126.

14.Daniel Friedlander and David H. Greenberg, "Evaluating Government Training Programs for the Economically Disadvantaged," Journal of Economic Literature, Vol. 35 (1997), p. 1812.



17.The Workforce Investment Act, Public Law 105-220.

18.Orley Ashenfelter, "Estimating the Effect of Training Programs on Earnings," Review of Economics and Statistics, Vol. 6, No. 1 (1978), pp. 47-57; Howard S. Bloom, "Estimating the Effects of Job-Training Programs, Using Longitudinal Data: Ashenfelter's Findings Reconsidered," Journal of Human Resources, Vol. 19, No. 4 (1984), pp. 543-556; and Michael E. Borus, "Indicators of CETA Performance," Industrial and Labor Relations Review, Vol. 32, No. 3 (1978), pp. 3-14.

19.Orley Ashenfelter and David Card, "Using the Longitudinal Structure of Earnings to Estimate the Effect of Training Programs," Review of Economics and Statistics, Vol. 67, No. 4 (1985), pp. 648-660, and Robert S. Gay and Michael E. Borus, "Validating Performance Indicators for Employment and Training Programs," Journal of Human Resources, Vol. 15, No. 1 (1980), pp. 29-48.

20.Edward C. Bryant and Kalman Rupp, "Evaluating the Impact of CETA on Participant Earnings," Evaluation Review, Vol. 11, No. 4 (1987), pp. 473-492; Laurie Bassi, "The Effect of CETA on Post Program Earnings of Participants," Journal of Human Resources, Vol. 38, No. 4 (1983), pp. 539-556; and Howard S. Bloom, "What Works for Whom? CETA Impacts for Adult Participants," Evaluation Review, Vol. 11, No. 4 (1987), pp. 510-527.

21.Katherine P. Dickinson, Terry R. Johnson, and Richard W. West, "An Analysis of the Impact of CETA Programs on Participants' Earnings," Journal of Human Resources, Vol. 21, No. 1 (1986), pp. 64-91; Katherine P. Dickinson, Terry R. Johnson, and Richard W. West, "The Impact of CETA Programs on Components of Participants' Earning," Industrial and Labor Relations Review, Vol. 40, No. 3 (1987), pp. 430-441.

22.Quasi-experimental design involves intervention and control groups that are formed by procedures other than random assignment.

23.Peter H. Rossi, Howard E. Freeman, and Mark W. Lipsey, Evaluation: A Systematic Approach (Thousand Oaks, Cal.: Sage Publications, 1999), pp. 241-242.

24.Orr, Bloom, Bell, Doolittle, Lin, and Cave, Does Training for the Disadvantaged Work?, and Schochet, Burghardt, and Glazerman, National Job Corps Study.

25.Orr, Bloom, Bell, Doolittle, Lin, and Cave, Does Training for the Disadvantaged Work?

26.Ibid., p. 30.



29.Ibid., pp. 30-31.

30.Ibid., p. 103.

31.Ibid., p. 153.





36.Ibid., p. 178.



39.Ibid., p. 109.


41.Schochet, Burghardt, and Glazerman, National Job Corps Study, p. 30.

42.Ibid., p. 130.

43.Ibid., p. 155.

44.Ibid., p. 170. Earnings by race and ethnicity were not provided for other time periods.

45.Ibid., p. 171.

46.Ibid., p. 126.

47.Ibid., p. 139.


49.Ibid., p. 127.


51.John Burghardt, Peter Z. Schochet, Sheena McConnell, Terry Johnson, R. Mark Gritz, Steven Glazerman, John Homrighausen, and Russell Jackson, Does Job Corps Work? Summary of the National Job Corps Study (Princeton, N.J.: Mathematica Policy Research, Inc., June 2001), pp. 4 and 25.


David Muhlhausen
David Muhlhausen

Research Fellow in Empirical Policy Analysis

More on This Issue