One of the oddest flaws in American public policy is our persistent belief that unemployment and poverty can be dealt with in the main by improving the “employ-ability” of potential workers – in other words, by improving the quality of labor supply.
In the 1960s and 1970s, “manpower development” and “work training” were seen as the solution to moving the poor into a labor market that would surely sweep them up. When Reagan dismantled the War on Poverty and CETA, he left in place training programs as an acceptably bootstrapping policy sufficient to deal with the 1981-1983 recession. Clinton leaned especially hard on training and education as his solution to all problems – the early 90s recession, displacement from NAFTA, welfare reform, and rising inequality. While Obama has gone a step beyond mere job training with the public works of the stimulus bill, his more recent comments about “winning the future” show the persistent strength of supply-oriented labor market policy.
What a shame that it doesn’t work.
The Flaw in Labor Supply Policy:
Gordon Lafer’s The Job Training Charade is one of the most comprehensive critiques of the labor supply paradigm. Lafer points out that the core arguments of labor supply advocates – that sufficient demand exists for all workers (and that high demand exists for skilled workers), that increasing education and skills will lower unemployment, and that job training is an effective method of improving human capital – are empirically unfounded. As longitudinal studies between 1980 and 2000 show, even in periods of low unemployment, there are 3-4 applicants for each vacancy on average, and that the situation for above-poverty wage jobs is even worse (with 22 applicants per opening). Similar studies conducted between 1967-1993 show that 66-80% of earnings are determined by factors other than education, with differences within educational groups being much larger than differences between educational groups. And given that wages for college graduates have stagnated over the last 13 years, there seems to be little evidence of labor shortages among the well-educated and skilled.
Given these factors, it’s somewhat unsurprising that the evidence on America’s longest-running jobs training program, the Job Training Partnership Act, is weak at best. The vast majority of wage increases for participants, which are generally modest, are due to increased employment rather than increased wages – but no group of participants actually succeeded in raising their income above the poverty line. At most, job training programs seem to move people from the ranks of the non-working to the working poor at great cost. Namely, for every ten dollars spent in training, workers gain one dollar in wages – meaning at least $108,000 per worker to assure an above-poverty income.
As I’ve argued before, the basic problem we have right now is a persistent weakness in labor demand in the face of growing supply. Rather than expand their permanent workforces, employers increasingly are turning to higher productivity demands, mechanization, offshoring, and casualized labor. This has in turn lead to gradually higher “floors” of unemployment, wage stagnation, and a general decline in the power of labor.
A New Strategy:
In this environment, what is needed is not more labor supply policy, but more labor demand policy, to counter-act this defect in both the short term and long term. The surprising thing is that in the last thirty years, given the negative effects that persistent weakness in labor demand has had on the Democratic Party’s base, and the effects that the inability to provide greater economic security has had on the Party’s ability to win elections, the Democratic Party has not as a whole embraced labor demand policy in the same way that it has embraced health care reform as a cause d’être.
Nor has the Democratic Party a reason to fear investing in labor demand policy. In the short-term, it provides extra bang for any jobs program without having to spend too many bucks, because financing can be re-purposed from Unemployment Insurance, Trade Adjustment Assistance, and other programs, and because much of it is regulatory as opposed to fiscal policy. Hence why Obama’s American Jobs Act includes a provides to allow states to use UI funds to finance work-sharing programs. In the long-term, labor demand policy gives the government an invaluable tool in dealing with the persistent weakness in the labor market I discussed above. It softens the blow of any future recession, which in turn reduces the cost of any future stimulus, and it allows the government to tailor anti-unemployment policies to specific firms and industries before a general recession emerges.
We can think of labor demand policy as consisting of two kinds of mechanisms – policies that create incentives for employers to hire more workers and fire fewer, and policies that create incentives for workers to offer less labor.
The first suite of labor demand polices operate on employers to provide incentives to refrain from laying off workers and to hire more workers in good times.
The foundation of these incentives begins with an expanded WARN Act (which currently requires employers to give 60 days’ notice of layoffs, but only covers firms of 100 workers or more and only certain sizes of layoffs), covering all employers with more than 20 employees and any layoffs of 10 workers or more. This would give time for the Labor Department and state agencies and labor unions (hopefully) to coordinate with employers to find alternatives to firing large numbers of workers.
Work-Sharing: instead of a voluntary, state-run work-sharing program as envisioned by the American Jobs Act, we should establish a national program within Unemployment Insurance, providing $150/week in UI money for each 20 hour reduction in lieu of termination, for 26 weeks (with the possibility of extension along the lines of traditional UI). This method of labor demand policy has been extremely successful in Germany both at preventing layoffs and cushioning consumer spending against wage loss, thus blunting recessions from the outset.
Continuing Education and Training: to complement the work-sharing program, we should also offer an additional matching credit of $4,500 for each worker who an employer pays to continue their education or complete a vocational training course. Not only does this have the same work-saving effect as the base program, but it also boosts overall skill levels in the workforce.
Termination Disincentives: a possible stick to back up work-sharing or training before firing workers would be a mandate that penalizes employers who do not exhaust all alternatives to layoffs; a mandate that workers automatically receive an extra years salary in severance if they are terminated without attempting either work-sharing or education/training would provide leverage to ensure that employers actually make use of these policies. Just to be clear – this policy would not penalize employers who end up having to lay-off workers after trying to preserve their jobs, but rather singles out employers who treat mass layoffs as a first resort.
Payroll Tax Progressivization: the payroll tax holiday included in the American Jobs Act is a useful addition to our economic tool-box; despite what some have claimed, these things actually do have a positive impact, as Timothy Bartik has shown. At the very least, I think we can agree that the recent statistics on median incomes and poverty suggest that lowering one of the most significant regressive taxes in America would be a step forward. However, progressives aren’t wrong to worry about the long-term effects of a holiday on Social Security’s balance sheet. While this is something of a hobby-horse for me, the solution is quite simple – make the payroll tax progressive. It has the same effect as a holiday, but it does so permanently (thus helping to counter-act long-term deficient labor demand) without harming Social Security’s finances.
Restraining Labor Supply:
One of the ironies of the American labor market is that, while 9% of Americans are unemployed, and a further 7% are underemployed, large numbers of Americans (38% of professional men, 23% of middle income men, and only 8% of low-income men, with lower numbers for women work more than 50 hours a week) are overworked.
These two factors are absolutely linked – in recent years, employers have been able to reduce their labor demand in America chiefly by demanding ever higher productivity from existing workers, with productivity increasing by 2.3% a year between 1995-2005 compared to 1.3% a year between 1973-1995. Reducing the supply of labor from the overworked effectively increases demand, while likely improving the health and productivity of the already employed.
Here, there are a number of options. Establishing a shorter work week, say a maximum of 35 hours before overtime kicks in, has had a modest effect at best in many European countries; work-sharing tends to work better in temporary crises than permanently. Ironically, given the current cry for raising the retirement age and the de-investment in higher education, one of the policies that does work is early retirement for older workers and expanded access to education and training programs for younger workers.
Putting the Pieces Together:
Ultimately, these policies act as a combination of early warning system and shock absorber – allowing us to see the crash coming and cushioning the blow when they arrive. However, the strongest form of labor demand policy is direct job creation; the government providing a new source of labor demand that can soak up workers in a hurry on a scale that work-sharing programs and the like simply can’t compete with.
Long-term of course, some form of industrial policy is necessary. The public sector can’t afford to cover the labor demand gap by itself, hence we cultivate new and existing industries alike to lift labor demand above its current sorry state.
One of the major roadblocks to progressive politics and policies in the U.S and in Europe is that, for the last twenty years at the very least, our major economic policy has essentially been “globalization with a human face.” A fairer distribution of laissez-faire economic growth only works as long as there is economic growth to go around – hence the peculiar phenomenon of center-left political parties being unable to connect with voters following a massive failure of the market. However subconsciously, the electorate can sense that the leaders of the center-left haven’t been offering up an alternative that could actually get the job done.
We need an alternative, and part of that has to be a recognition that, while capitalism is very good at making more widgets at a lower price, it’s not very good at making economic security for the labor force. Labor demand policy should be the foundation of a new economic toolbox for any progressive worthy of the name.