I have a new working paper available that looks at how job-creating policies should be evaluated in benefit-cost analyses. The paper is technical, and is mainly addressed to economists interested in benefit-cost analysis. However, the paper’s findings have some important implications for how policymakers should think about job creation due to public policies.
Many public policies create jobs. This includes business incentives that boost labor demand. It includes Keynesian macroeconomic policies that attempt to reduce high unemployment. Job-creating policies also include policies that increase the quality of labor supply, including early childhood programs as well as other investments in education and job training.
Job-creating policies provide one obvious measurable benefit: increased earnings. The controversy over the social value of job creation is that such job creation also provides several less tangible benefits and costs, which are hard to measure and value.
On the cost side, one less tangible cost is that job creation increases work time at the expense of non-work time. On a minute-by-minute basis, the empirical evidence, as well as intuition, suggests that most people enjoy their non-work time more than their work time.
On the benefit side, one less tangible benefit for individuals who get jobs is that being employed is important to most individuals’ sense of self-worth and life satisfaction. For example, studies that ask people the question, “how satisfied are you with your life?”, find that being unemployed reduces life satisfaction by much more than the loss of earnings. Being unemployed is also quite damaging to physical and mental health.
Also on the benefit side, another intangible benefit of increased employment is spillover benefits on persons who already have jobs. The life satisfaction of individuals is not only affected by their own employment status, but by the average local unemployment rate. This may reflect in part empathy for the unemployment problems of friends and relatives, or fellow citizens. In addition, this effect of overall local unemployment may reflect that even if an individual is currently employed, they are concerned about the possibility that they may lose their job, in which case the local unemployment rate will affect their probability of finding a suitable replacement job.
If the overall local employment rate increases due to one more person gaining employment, this provides benefits not only to the person getting the job, but to others in that local labor market who place a value on a higher local employment rate. The benefits per person are highest for the person getting a job. But the sum of the “spillover benefits” for all other workers is actually higher than the benefits to the worker getting the job, because of the large number of workers.
Therefore, in valuing policies that involve job creation, policymakers must make some judgment as to how all these intangible costs and benefits add up. Our knowledge from social science of the exact magnitude of all these costs and benefits is less than one might hope.
Given our imperfect knowledge, what advice might be given to policymakers on how to value the costs and benefits of job creation? It seems likely that benefits relative to costs for job creation will be larger for average workers if the unemployment rate is high. If unemployment is low, it is relatively easy for persons with adequate skills to get a new job if they are unemployed. Therefore, if unemployment is low, it seems likely that becoming unemployed will not be as costly to the unemployed, and that the prospect of unemployment will not lead to most current workers placing a large value on additional job creation. In contrast, if unemployment is high, its cost for the unemployed and current workers are high, and additional job creation will have a much higher social value.
However, for disadvantaged workers, the value of job creation is likely to be quite high even when overall unemployment rates are low. Persons with labor market disadvantages may have a hard time getting a job even when unemployment is low, and therefore will highly value programs that make it easier for them to find a good job. In addition, to the extent to which other workers feel solidarity and empathy with disadvantaged persons, they will value programs that improve the disadvantaged’s ability to get jobs even if the overall unemployment rate is low.
What implications does this have for local economic development policy? What this means is that the social value of job creation is likely to vary with (1) whether the local unemployment rate is low, and (2) whether the additional job creation is targeted on the disadvantaged.
For most business tax incentives, the jobs created are not particularly targeted on the disadvantaged. Therefore, in any analysis of the benefits and costs of business tax incentives, these programs make far more sense if pursued by state and local areas with high unemployment rates than by state and local areas with low unemployment rates. If the unemployment rate is already low, then it is unclear why local residents should be willing to increase their taxes in order to give away tax breaks to businesses to create more jobs.
For many early childhood programs, the additional jobs created will tend to be targeted on the disadvantaged. Even for universal preschool programs, it seems likely that the benefits for middle-class participants are largely to improve the quality of jobs obtained as adults, and not the employment rate. The benefits for more disadvantaged participants are likely to come more in the form of an improvement in their employment rates. Therefore, the effects of early childhood programs in creating future jobs are likely to have great social benefits even in local areas where unemployment is low.
Untargeted job creation has its greatest social benefits when overall unemployment is high. In contrast, targeted job creation for the disadvantaged, such as that provided by early childhood programs, has great social benefits in a wider variety of economic circumstances. This difference should cause local areas with more booming economies to target their local economic development strategies more on the disadvantaged. Current policy in most state and local areas does not resemble this ideal: state and local areas vigorously adopt business tax incentives regardless of the local unemployment rate, and early childhood programs are under-invested in almost everywhere.