So far in the Job Insurance series, I’ve been largely focusing on how a Job Insurance program would function in the abstract, rather than dealing with the harder questions of how things could go wrong. On one level, it’s important to start by outlining in full the parts of how the program should operate, so that people understand the ideal and be able to recognize departures from it.
At the same time, it’s important to be realistic and acknowledge that any Job Insurance program would face problems both of policy design and political opposition, and that it’s important to grapple with them if we’re going to make sure that a future program would succeed.
At the risk of annoying my former teacher, Eric Foner, who used to say that there are no such things as the lessons of history, I do think there are insights we can clean from historical examples of jobs programs.
While I have talked about some of these programs before in other contexts, it’s important to consider them side-by-side to get a handle on what strengths and weaknesses jobs programs can have, and especially what recurring issues they have to deal with.
Works Progress Administration (1935-1943):
The WPA was the longest-running and most successful jobs program in American history, and it was also the most politically popular. It ran for a bit over eight years, employed an average of three million people at any given time (the WPA peaked at 3.5 million, and hit its nadir at about 1 million; at its height, this meant that 7% of the American labor force worked for the WPA, and that the WPA had directly absorbed 30% of the total unemployment caused by the Great Depression), and was a major constructive element of the New Deal coalition (especially in regards to African-American voters in the North).
The key strength of the WPA was that it operated through “force account,” which is a technical term referring to work carried out with one’s own workforce and equipment. Most often, force account is used to describe work done by the government directly as opposed to the normal practice of letting out contracts for bidding by private contractors. Under the WPA’s administrative setup, the entire process of running a WPA project was done by public sector workers – the U.S Employment Service handled the application and assignment of unemployed workers, the Army Corps of Engineers was responsible for the technical design and engineering work on construction projects, the U.S Treasury Department was responsible for printing out checks, and the Federal, State, and Local WPA Administrations were responsible for the management of the workforce and the completion of projects. This gave the WPA several enormous advantages – first, the WPA could move incredibly quickly to put people to work, because it didn’t have to go through the normal process of circulating contracts, assessing bids, and so on. Second, the WPA had an enormous amount of control over the number of jobs it created, with the ability to increase or decrease the total level of employment within a matter of weeks.
In addition to force account, the WPA’s other major advantage was its focus on “light construction.” As opposed to heavy construction projects like the Hoover Dam, which required huge amounts of machinery, materials, and skilled labor to make use of the most advanced techniques in constructing such complicated and large-scale works, light construction refers to smaller-scale projects like building roads, bridges, sewers, sidewalks, and buildings – in other words, work that could be done by large numbers of unskilled workers using basic hand tools. While it seems counter-intuitive that a less efficient form of construction predominated, the reality is that heavy construction projects by their very nature focus their resources towards machinery, materials, and other non-labor costs, and tend to rely on a relatively smaller and more skilled workforce, which means that they generate less employment per dollar spent. Precisely because the WPA designed its construction projects to maximize the use of human manpower, the WPA was able to replace machinery with thousands of workers, creating far more jobs per dollar and keeping non-labor costs low; similarly, the choice to favor basic tools and techniques allowed the WPA to make use of unskilled workers who wouldn’t have been hired on traditional public works or by professional construction firms.
And yet the WPA had major limitations that should be avoided. First, when the WPA was formed, a decision was made to limit 90% of the jobs created to people who were already on the relief rolls, unlike the situation with the Civil Works Administration (CWA) which the WPA’s administrators had operated in the winter of 1933 and spring of 1934, where half the jobs were open to anyone who was unemployed. On one level, this decision was designed to make sure that the WPA was helping those who had been unemployed the longest, who had exhausted whatever economic resources they might have had, and who were completely destitute. However, it had the effect of labeling WPA workers as welfare clients, however much the WPA insisted that they weren’t, and meant that workers held their jobs as a matter of public charity, not by right. This opened the program up to attack from conservatives,who wanted to stigmatize the program as much as possible (for example, Congressional conservatives proposed that WPA workers should be barred form voting under the old poor law theory that people on relief were wards of the state and not independent citizens). Second, the WPA’s wages reflected this ambivalence between WPA workers as workers or as welfare clients – at various times, the WPA provided either a “security wage” based on poor relief models of minimum necessary budgets for families in poverty, or a “prevailing wage” based off of union wage scales in the building trades. The average WPA wage of $50 a month ($776 in 2008 dollars, or $9312 a year), while a huge improvement over poor relief, was still barely adequate to keep people alive, which also kept the program from having as much of an impact on purchasing power as the New Dealers wanted.
1945 Full Employment Bill:
For New Deal liberals looking to the post-war years, the WPA served as a massive social scientific experiment in how the government could provide jobs to counter mass unemployment. The WPA’s “Final Report to the President” touted lessons learned for future programs and even offered model administrative setups. The National Resources Planning Board (NRPB)’s report, “Security, Work, and Relief Policies,” that recommended a cradle-to-grave welfare state and the right to a job, was written partly by academics and bureaucrats who were veterans of the WPA, who naturally looked to their own experiences as a model for how to construct a new mechanism for full employment.
The legislative embodiment of their work was the 1945 (and later, 1946) Full Employment Bill, introduced by such progressive champions as Senators James Murray of Montana and Robert Wagner of New York (the two were frequent collaborators on New Deal legislation, and were two of the three sponsors of the Wagner-Murray-Dingell Bill for universal health insurance in 1943). The Bill had several key elements – first, it stated that it was the policy and the responsibility of the U.S government to ensure full employment at all times; second, it stated that every U.S citizen had the right to a job; and third, it created a legislative mechanism for the realization of this right. By law, the President would be required to submit, as part of the yearly Federal budget, a report forecasting the size of the workforce and the number of jobs that private industry would create. If the projected unemployment rate exceed the full employment rate (at that time, 3% or below was assumed to be the rate of “frictional” unemployment caused by people moving between jobs), the president would be required to submit to Congress a Full Employment Budget that included proposals for the stimulation of private industry and the provision of public employment necessary to bring aggregate investment and employment up to full employment; a Federal institution, like the NRPB, would be tasked with constructing a shelf of public works for the government to use in shaping aggregate employment.
In many ways, the Full Employment Bill was designed to create advantages out of the WPA disadvantages. Unlike the WPA, which as an “emergency” program had to battle for re-authorization every year or face termination, the Full Employment Bill envisioned a permanent process which, because it mandated government action and was inserted into the middle of the normal Federal budgetary process, couldn’t be derailed by a sudden change in political majorities. Similarly, the Full Employment Bill replaced the WPA’s problems with relief status with a forthright declaration of employment by right – just as Social Security has been made politically inviolable by the assumption that senior citizens have a right to their Social Security benefits, so too would the full employment budget be made politically inviolable.
Ironically, the Full Employment Bill sailed through the Senate in 1945 by a vote of 71-10 only to be stalled in the House, where the bill was diluted to unrecognizability into the Maximum Employment Act of 1946, having stripped out all three major elements of the bill (although it did establish the Council of Economic Advisers). This ultimately points to a major weakness of the Bill – any Full Employment program would have required both Presidential and Congressional approval, which would have created problems in periods of conservative dominance or split government. That may be unavoidable, given the nature of American government.
CETA (Comprehensive Employment and Training Administration, 1973-1979):
Unlike the WPA, which is generally well-regarded in historical circles, CETA, which was established to deal with the persistent economic crisis of the 1970s, has a terrible reputation. Indeed, the failure of CETA ushered in and may have ultimately been responsible for the death of direct job creation policy for a generation.
To give CETA its due, the program did have two advantages over previous programs: first, it included more service jobs as opposed to just construction jobs, which created far more jobs for women and prepared people better to find jobs in the expanding service sector. Second, it provided jobs for a guaranteed period of 12-24 months, which gave workers more economic security and stability than before.
However, the disadvantages of the CETA model far outweighed its few advantages. The major problem was that there simply weren’t enough jobs – CETA peaked at 750,000 jobs at a time when the unemployment rate was reaching as high as 9%. Thus, while the program did make some headway, but not enough to really dent the unemployment rate. Its next major flaw was the extreme devolution – CETA programs were operated by state and municipal governments, with Federal funding. The problem was that, during a prolonged economic recession when state and local budgets were deep in the red, far too many governments fell prey to “fiscal substitution” – hiring CETA workers they didn’t have to pay for in favor of regular public employees they did, which vitiated any new employment created by the program. This also gave rise to huge problems with political corruption and patronage, as well as messy political fights with public employee unions who felt that CETA workers were being used as scab labor.
So how does Job Insurance measure up to its historical predecessors?
- It borrows from the WPA the use of force account and the emphasis on light construction (combined with the social services of CETA). This should make the program both fast-acting, labor-intensive, and have low non-labor costs.
- Because it’s nationally operated, it should completely avoid the fiscal substitution effects that crippled CETA.
- Because it has its own funding stream, it should be less vulnerable to Congressional antipathy than the Full Employment Bill.
- Because it’s sized relative to the unemployment level, it should have a much greater impact than the relatively small CETA.
Obviously, Job Insurance as described previously is hardly free of flaws, and I welcome any and all suggestions about how to improve the program. Given the sheer magnitude of the crisis we’re facing, and the implications if we fail, there really is no excuse for not getting this right.