News > Skills Blog

The CEA Training Report: Very Wide of the Mark

  ·   By Senior Fellow in Economic Studies at the Brookings Institution, LaFarge SJ Professor at the McCourt School of Public Policy at Georgetown and former NSC board member Harry J Holzer
The CEA Training Report: Very Wide of the Mark

By: Senior Fellow in Economic Studies at the Brookings Institution, LaFarge SJ Professor at the McCourt School of Public Policy at Georgetown and former NSC board member Harry J Holzer

Georgetown University, August, 13 2019-- The White House Council of Economic Advisers (CEA) has issued a report that claims to assess the available evidence on government employment and training programs, and to offer policy implications based on this assessment.[1]

But the document is highly flawed. It clearly misrepresents basic facts about federal job training programs in the US, and it misinterprets research evidence; it appears more driven by ideological and political agendas rather than what is best for US workers. In short, it is very wide of the mark as an evaluation of federal training in the US.

For instance, on the fundamental question of how much the US spends on workforce development: Figure 2 of the CEA report implies that federal spending on workforce development has been rising over time. But it does so without adjusting for inflation – an astounding feature in a report written by economists. In the text, it acknowledges “real (i.e., inflation-adjusted) spending in 2018 is nearly unchanged from the 2014 levels;” but it fails to note dramatic declines in such funding over the past four decades (by almost two-thirds), while the US labor force has roughly grown in size by half.[2] It quietly acknowledges that the nearly $19B of federal funding for such programs, constituting less than one-tenth of one percent of US GDP, is a paltry sum in comparison to spending in most European Union countries on “active labor market policy” (which often falls in the range of .5 to 1 percent of GDP, above the numbers it cites), while not acknowledging how low such spending is for an American economy with 160 million workers.[3]

When reviewing evaluation evidence, the report cites a range of studies using widely respected methodologies that show more or less positive results for programs funded by the Workforce Innovation and Opportunity Act (WIOA) and its earlier incarnations, with many (including mine) showing positive impacts.[4] Yet the CEA concludes that “Government job training programs (with the exception of apprenticeships) appear to be largely ineffective” (p. 23), in a leap of logic that clearly runs counter to the much more mixed evidence the report provides.

When discussing the most important recent study with negative findings on training – by Fortson et al. in 2017 – the CEA report fails to highlight the evidence that intensive workforce services have positive impacts on worker earnings (of 7-20 percent, depending on the source). These results strongly imply that such services are cost-effective – while federal funding for them remains extremely modest.[5]

And, when discussing the lack of positive training impacts in the Fortson study, the CEA report omits important caveats highlighted in the study itself – like the fact that relatively few workers in the “treatment” group actually received training while many in the “control” group received it with funding from other sources – that render the lack of estimated training impacts very hard to interpret and “inconclusive,” as indicated by the authors. The CEA also ignores other well-known and rigorous studies showing impressive training impacts for adult or dislocated workers.[6]

But the most egregious aspect of the CEA report is that it completely fails to acknowledge a growing evaluation literature on highly effective “sector-based” or “career pathway” programs that show large and lasting impacts on disadvantaged worker earnings. These mostly local (though now spreading) programs – like Per Scholas, Project QUEST, the Wisconsin Regional Training Partnership, the Jewish Vocational Services-Boston, and Year Up – have generated large, statistically significant earnings impacts in several randomized controlled evaluation studies.[7] It’s worth noting that these programs all make substantial investments in the skills of their participants, and work closely with employers to ensure those skills are relevant in the labor market. These results offer a strong counterpoint to the somewhat disappointing results for training in the WIOA study. Though they are not explicitly “government” programs, they have received financial support from a range of state and federal (as well as private) sources.[8]    

Given the very clear successes of these programs, a sensible policy discussion would focus on how to replicate and scale the best sector-based efforts at community colleges or other training providers with available or new federal and state funding. Instead, the CEA completely ignores this strong body of evidence on programs that work, while presenting misleading facts on federal job training funding over time and a skewed portrait of evidence on its impacts. Furthermore, the CEA report makes no evidence-informed recommendations for future policy directions in workforce development.

This report should not be taken seriously as the basis for any discussion of federal funding for workforce policy in the future.



[1] Government Employment and Training Programs: Assessing the Evidence on their Performance. The Council of Economic Advisers, Executive Office of the President, June 2019.

[2]CETA Training Programs – Do They Work for Adults? Congressional Budget Office, 1982.

[3] Such policies include training, job placement assistance, and subsidized work experience. See Chad Brown and Caroline Freund. Active Labor Market Policies: Lessons for the US. Peterson Institute for International Economics, 2019.

[4] Frederik Andersson et al. “Does Federally-Funded Job Training Work? Nonexperimental Estimates of WIA Training Impacts Using Longitudinal Data on Workers and Firms.” NBER Working Paper, 2013; and Carolyn Heinrich et al. “New Estimates of Public Employment and Training Program Net Impacts: A Nonexperimental Evaluation of the Workforce Investment Act Program.” IZA Discussion Paper, 2009. Across studies, the estimates of training impacts on earnings per quarter are in the range of $320–$887 per quarter for participants, which indicates fairly strong agreement given the varying study samples and methodologies Estimated effects of training on the probability of employment are also positive and statistically significant across a majority of studies. These estimates of employment increases range from about 5 to 29 percentage points (measured monthly or quarterly), with some differences observed between women and men, and by specific training type and time following program entry. 

[5] See Kenneth Fortson et al. Providing Public Workforce Services to Job Seekers: 30-Month Impacts Findings on the WIA Adults and Dislocated Worker Programs. Mathematica Policy Research, 2017. As the CEA notes, “Wagner-Peyser” funding for such services at over 3000 job centers across the US is under $.7B now and has changed little in recent years despite their clear cost-effectiveness. Providing intensive services increased earnings over the follow-up period by $3,300 to $7,100 (7 to 20 percent) per customer depending on the data source. The benefit-cost analyses demonstrate that providing intensive services is cost-effective from the perspectives of customers, taxpayers, and society as a whole (Fortson et al., 2017).

[6] For instance, see Louis Jacobson et al. “The Impact of Community College Retraining on Older Workers: Can We Teach Old Dogs New Tricks? Industrial and Labor Relations Review, 2005.

[7] See Anne Roder and Mark Elliott, Nine-Year Gains: Quest’s Ongoing Impact, Economic Mobility Corporation, 2018; David Fein and Jill Hamadyck, Bridging the Opportunity Divide for Low-Income Youth: Implementation and Early Impacts of the Year-Up Program, US Department of HHS, 2018; and Sheila Maguire et al. Tuning Into Local Labor Markets, PPV, 2010. To take one example, The Year Up experimental evaluation found that young adults in the treatment group saw a 53% increase in initial earnings, which remained strong over time, with 40% earnings gains two years out.

[8] Public funding sources for these programs have included WIOA (and its predecessor), federal Social Innovation Funds, and state funding for community colleges.

Posted In: Work Based Learning, Temporary Assistance for Needy Families, SNAP Employment and Training