March 2012

Keith Rolland: Welcome to the Federal Reserve Bank of Atlanta's Economic Development podcast series. I'm Keith Rolland with the Federal Reserve Bank of Philadelphia.

A direct job creation program at the state or local level could help reduce unemployment. In addition, such a government program would support better provision of public goods and services. Philip Harvey, professor of law and economics at Rutgers University, proposed this idea of a direct job creation program as part of the "Big Ideas for Job Creation" project. The project, sponsored by the Institute for Research on Labor and Employment at the University of California at Berkeley, and supported by the Annie E. Casey Foundation, was a call to academics and economic development practitioners to design jobs programs for cities and states that would lead to net new job creation in one to three years. Harvey's is one of five winning ideas we are featuring in this podcast series.

Today I'm speaking with Philip Harvey about his idea. Philip, thank you for speaking with us today.

Photo of Philip HarveyPhilip Harvey: Thanks for the opportunity, Keith.

Rolland: Tell us more about your direct job creation program idea and how it would work, particularly in light of the fact that many state and local governments currently are dealing with budget shortages.

Harvey: To answer that question, Keith, it's important to understand that the goal of my proposal is not just to reduce unemployment to its prerecession level. My goal is to ensure the availability of enough jobs to eliminate involuntary unemployment completely, to secure what President Franklin D. Roosevelt termed "the right to work."

My proposal is designed to enable state and local governments to close the job gaps that exist within their own jurisdictions, and to do so both completely and permanently. The mechanism I propose for achieving this goal is the creation of a state or local trust fund, similar to the trust funds that states currently use to pay for unemployment insurance benefits, except that instead of providing cash benefits to laid-off workers, the trust fund I am proposing would be used to pay for the creation of enough temporary public-sector jobs to completely close the state or local economy's job gap. During recessions, the number of jobs provided in this fashion would be very large; at the top of the business cycle, the number would be much smaller. But however many jobs were needed to make it possible for all job seekers to find work, that's how many would be created. And in the process, the communities in which these jobs were created would benefit from the additional public goods and services the program would make available to them.

Rolland: Obviously, the number and specific types of jobs that this program could create varies by implementing locale. How many and what types of jobs do you project could be created nationwide? Also, what do you expect the cost would be of this type of program?

Harvey: I estimate that, on average, a 14 percent increase in local government revenues would be required to create a trust fund capable of completely closing the job gap in an average local economy over the course of a normal business cycle—from boom to bust to boom and beyond. If the trust fund were established during a period of lower than average unemployment, it would be able to achieve this goal from the outset. It would take longer to achieve this goal if the trust fund was established at a time like the present when unemployment rates are above average. Still, even in times like these, the strategy would reduce unemployment dramatically by creating locally the equivalent of 4.5 million jobs nationally—enough jobs to reduce the unemployment rate almost three full percentage points. And once private-sector employment expanded to its long-term average level, a state or local government that adopted the job creation strategy I advocate now would be able to close its job gap.

How much would local taxes have to increase to achieve the 14 percent increase in local revenue that I estimate would be needed to implement this strategy? To provide an easy comparison, a 1.3 percentage point increase in Social Security and Medicare taxes from their current combined levels of 7.65 percent to 8.95 percent would be sufficient to generate the necessary funds. Although it should be emphasized that it wouldn't matter what kind of taxes were used to pay for the program.

Would tax payers be willing to stomach an increase in their tax burden of that magnitude? When you consider what it would buy, I think they would. The security of knowing that if they, or one of their friends, or a member of their family ever needed a job there would be one available for them.

Rolland: In your paper, you mention real-world applications of this idea in the Works Progress Administration and the Public Service Employment program. Can you describe those programs and the related results? Are there other successful local or state examples of this type of program that you can point to?

Harvey: The direct job creation strategy I advocate has been used by governments in the past, but only occasionally for a large-scale attack on the problem of unemployment. The clearest demonstration of this strategy's effectiveness can be seen in the performance of the direct job creation programs created by the New Deal administration of Franklin D. Roosevelt in the 1930s. The best known of these programs were the CCC, that is, the Civilian Conservation Corps and the WPA, the Works Progress Administration. Unfortunately, both the purpose and effectiveness of these programs is widely misunderstood today.

The CCC and WPA were not designed to end the Great Depression. Instead, their primary purpose was to help people survive the Depression by furnishing them with the jobs the private sector was unable to provide them. Their success in achieving that goal is also widely misunderstood because the unemployment statistics commonly cited by historians in telling the story of the Great Depression mistakenly characterize the persons employed in the New Deal's direct job creation programs as unemployed rather than employed.

If we categorize these workers as they would be in labor market statistics today as employed, we can see more clearly the enormous difference the programs that employed them made in people's lives. During the WPA's first full year of operations, the nation's unemployment rate is supposed to have dropped only 3.3 percentage points from 20.3 percent to 17 percent. But if we count CCC and WPA workers as employed, we see that the unemployment rate actually dropped to 10.8 percent. Achieving the same relative job creation effect today would require the creation of 9.6 million jobs.

The Public Service Employment program established under the Comprehensive Employment and Training Act in the 1970s, commonly known by its acronym as the CETA program, was also based on the direct job creation strategy. But, unfortunately, the CETA program was poorly designed to achieve its goals. The problem was that, unlike the CCC and WPA, the CETA program allowed state and local government officials, in effect, to substitute CETA workers for regular government employees. The result was that CETA created many fewer net jobs than it was designed and funded to create, even though it did prevent the cutbacks in state and local government services that have characterized our own experience the past two years.

In addition to these programs, there are plenty of other instances in which governments have created temporary public-sector jobs to serve other purposes. The College Work Study and Summer Youth Employment Programs are two examples.

Rolland: Thank you for joining us today.

Harvey: It's been my pleasure, Keith.

Rolland: This concludes our podcast. We've been speaking with Philip Harvey, professor of law and economics at Rutgers University. For more podcasts on this topic and others, please visit the Atlanta Fed's website at If you have comments or questions, please e-mail Thanks for listening.

Reference article
Darby, Michael R. "Three-and-a-Half Million U.S. Employees Have Been Mislaid: Or, An Explanation of Unemployment, 1934–1941." Journal of Political Economy 84, No. 1 (February 1976): 1–16.