Jobs do not appear from thin air. They are mostly created by entrepreneurs whose objective is to make money. Entrepreneurs will create jobs when they expect the job to generate profit and will not create a job that is expected to generate a loss. The expected profit margin — the difference between the value of worker’s output and the wage the worker is paid — determines the amount of job creation. When profit margins rise, more jobs are created. When they fall, job creation falls, too.
What does this have to do with the impact of unemployment benefit extensions on the labor market? A lot.
Those arguing for benefit extensions in recessions have the best intentions: They want to improve the well-being of the unemployed. An improvement in the well-being of the unemployed puts an upward pressure on wages that squeezes profit margins, hindering the creation of new jobs.
The effect is particularly strong in recessions, when the value of workers’ output is relatively low. As a result, job creation collapses and unemployed workers cannot escape unemployment. Unemployment rises and persists. Workers, who would have found a job fast in the absence of benefit extensions, remain unemployed for a long time. Long-term unemployment rises; discouraged workers drop out of the labor force and stop searching altogether. The best-intentioned policy kills job creation incentives of entrepreneurs and condemns the unemployed to long periods of life with no chance of finding work.
Do we somehow imply that U.S. workers became lazy after the last recession? Absolutely not!
Most unemployed are desperate to get a job. But they are given no chance. The mere expectation of higher wages potential employers would have to pay to newly hired workers prevents employers from creating jobs. With no vacant jobs, the unemployed do not have a chance of getting one.
Is this a radical new idea? No. The mechanism we describe is at the core of modern labor market theory. Economists all along the ideological spectrum have for years ridiculed sclerotic European labor markets because generous subsidies to unemployed workers priced them out of the labor market. Today, it is Germans who are ridiculing the sclerotic U.S. labor market.
While the mechanism we describe is simple and clear, is there evidence that it is large in magnitude? Yes. In our research, we compared the labor market performance in counties that border each other but belong to different states. Border counties are similar in terms of economic fundamentals but not in terms of policy. We observe that counties that belong to states that extend unemployment benefits experience sharp rises in unemployment, declines in job vacancies and employment relative to the counties next door that belong to states where benefits duration stayed the same or did not increase as much. And yes, wages go up in the counties with longer benefits. The change in wages is small. Yet it is sufficient to induce a large response of job creation, just as predicted by theory. In fact, our estimates suggest that most of the persistently high unemployment after the Great Recession can be traced back to the impact of the unprecedented unemployment benefit extensions.
Finally, how about the stimulus? The popular argument for unemployment benefit extensions is that it not only helps unemployed workers but provides the much needed stimulus to the economy in the time of a recession. The argument is simple. The unemployed spend all unemployment benefits, which stimulates aggregate demand. But the cost of this policy is that these workers do not get a job where they would have produced and spent considerably more.
Our estimates imply that the cost of unemployment benefit extensions exceeds the most optimistic estimates of their stimulative effects by a factor of ten. Getting people jobs is the best stimulus policy we know.
The diehard skeptics may perhaps find the experience of North Carolina suggestive in this respect. North Carolina lost access to all federally financed benefits as of July 1, 2013. The loss of hundreds of millions of dollars in pure transfers to unemployed workers in the state from the federal government was predicted to have a sharp negative impact on aggregate demand and dire consequences for employment. Nothing like that happened. The available data on North Carolina’s labor market (available on our websites) illustrates that its labor market performs at least as well as that of comparable states, and likely much, much better, although the limited duration of currently available data is too small to draw scientifically definitive conclusions.
Marcus Hagedorn is a professor at Oslo University. Iourii Manovskii is an associate professor of economics at the University of Pennsylvania, where Kurt Mitman is a doctoral candidate in economics. Their research team just released a report last month on this topic as a National Bureau of Economic Research Working Paper.