The Region

Coping with volatile labor markets

There are high costs to not optimally adjusting government policy to deal with earnings volatility

Douglas Clement | Editor, The Region

Published March 4, 2019

Ciranno Marcon Soares/Minneapolis Fed

To anyone with a job or looking for one, “labor market risk” is a fact of life, and sometimes a painful one. To an economist, the term means volatility in individual earnings, whether it’s due to job loss, a pay cut or—on the upside—a salary bump or more lucrative career. Labor market risk has increased considerably in the United States and elsewhere, and economists are focusing on causes and consequences.

They’ve turned in particular to indications that another salient trend, rising inequality, is driven partially by this rise in earnings volatility. And they’ve studied how governments can help workers cope. Most governments set up social insurance programs to help workers who lose their jobs—unemployment insurance, job training, wage subsidies, for example.

In a new study from the Opportunity & Inclusive Growth Institute, visiting scholar Tom Krebs of University of Mannheim and Martin Scheffel of Karlsruhe Institute of Technology seek answers to two interrelated questions: What are the effects of rising labor market risk on people’s welfare (quality of life) when governments optimally adjust social insurance systems? And what are the costs of not adjusting?

“An increase in labor market risk increases social welfare if the government adjusts the tax-and-transfer system optimally.”

Their answers are clear. “An increase in labor market risk increases social welfare if the government adjusts the tax-and-transfer system optimally,” they write, in “Optimal Social Insurance and Rising Labor Market Risk.” And “the welfare cost of not adjusting the social insurance system optimally can be substantial.”

More specifically, the optimal policy response to higher labor market risk is to increase social insurance markedly, so that displaced workers have to absorb (through lower consumption) just a quarter of the 30 percent earnings loss they suffer. That’s a significant safety net, and it means work effort and economic well-being barely change—a true smoothing effect.

Secondly, they estimate that the cost of doing nothing—a passive response that sustains social insurance as is—is large. The jump in job displacement risk leads to a large drop in consumption by displaced workers. This is true even though workers can adjust their behavior by increasing their work hours, savings rate, or job training effort. A further estimate shows that if the rise in labor market risk doubles earnings losses of displaced workers, the costs of not adjusting “are very large indeed—about 1.5 percent of lifetime consumption.”

Many moving parts

These straightforward results belie the complex model behind them. To shape the optimal response, and to measure the price of passive policy, Krebs and Scheffel build a mathematical economy with households that can invest in risky human capital—specifically, through on-the-job training. The investment is risky because the worker may lose his or her job; training will lower the chance of being laid-off, but not eliminate it.

“The welfare cost of not adjusting the social insurance system optimally can be substantial.”

And then there’s the issue of moral hazard. It’s the idea that insurance, by lowering the cost of a certain behavior, might actually encourage it; car insurance could lead drivers to be less cautious. So if unemployment insurance hands a check to a laid-off worker, that worker might not look for work because the stipend covers the bills. In the economists’ model, moral hazard means that the government designing its social insurance policy doesn’t know exactly how much effort workers are putting in. Insurance could diminish work effort and thereby lower overall economic output.

The government is the other crucial player in this model. It can tax or subsidize income from capital (interest, dividends) and from labor and, crucially, it provides social insurance, which—through moral hazard—may affect work effort. Government then has to decide how much it will tax each form of income and how much social insurance it will provide to workers in the face of labor market risk. That, of course, is the focus of the analysis.

Results

The most striking result of this research is the painful cost of doing nothing: If government remains passive in the face of increased job displacement, individuals who are laid off would be forced to spend much less. “Almost the complete increase in job displacement risk shows up as an increase in consumption risk,” the economists observe. The welfare loss of not adjusting the social insurance system is 0.20 percent of lifetime consumption.

The most striking result is the painful cost of doing nothing: If government remains passive, laid-off individuals would be forced to spend much less.

This may overestimate the real-world effect, say Krebs and Scheffel. Many countries, including the United States, have progressive income tax systems that provide implicit insurance against earnings declines due to job loss. If you earn less, you pay lower taxes. “This feature of the actual tax system implies that part of any risk in labor market risk is implicitly insured even if transfer payments remain constant,” they write.

So, in a second run-through, the economists calculate the welfare effects of passive policy assuming a progressive tax system wherein 20 percent of earnings losses are implicitly insured by lower taxes. In this scenario, the cost of not adjusting policy is “still substantial, but significantly smaller.” Lifetime consumption drops 0.13 percent instead of the 0.20 percent decline without the implicit insurance of progressive taxes.

Roads ahead

So, in a second run-through, the economists calculate the welfare effects of passive policy assuming a progressive tax system wherein 20 percent of earnings losses are implicitly insured by lower taxes. In this scenario, the cost of not adjusting policy is “still substantial, but significantly smaller.” Lifetime consumption drops 0.13 percent instead of the 0.20 percent decline without the implicit insurance of progressive taxes.

Research could also explore the political economy of social insurance: Why aren’t social insurance systems more generous given the large increase in labor market risk since the 1980s? Perhaps, Krebs and Scheffel observe, it’s because college-educated workers are less affected by that risk than those with less education and, therefore, “less likely to rely on the welfare-state. … These differences might explain why college-educated workers prefer a rollback of the welfare state even in times when labor market risk has been increasing.”

Top