As environmental economists, we are often asked about how environmental regulations or policies will affect jobs and the unemployment rate. This should be no surprise as employment effects often play a central role in the political debate over environmental policy, with “green job” arguments on the left and “job killer” claims on the right. But existing economic models are not well-suited to answering the question of how regulation affects jobs.
Two types of studies have attempted to measure employment effects. Empirical analyses of existing regulations have measured the effect on jobs in regulated industries but not the effect on jobs in unregulated industries—and thus can’t address the overall effect. (Even worse, some studies use the unregulated sector as a control group, leading to potentially biased estimates.) General equilibrium models do account for both the direct and indirect effects of environmental regulation, but almost always assume full employment (in other words, that everyone who wants a job has one) and ignore labor market dynamics.
Over the past two years, we have developed a new general equilibrium model specifically designed to study the effects of environmental regulation on employment and unemployment. This model adds a job search requirement (unemployed workers must search for open jobs) to a general equilibrium model of environmental regulation. Like other general equilibrium models, it can measure both direct and indirect effects, and adding a job search avoids the full-employment assumption and lets the model look at labor market dynamics.
In a new RFF discussion paper, we introduce the model and apply it to look at policies aimed at reducing carbon dioxide (CO2 ) emissions in the United States. To simplify the analysis, we begin with a two-sector model with a “clean” and a “dirty” industry (we will be expanding the model to 22 sectors in the near future). The model estimates that a tax of $20 per ton on the CO2 emissions of the dirty sector, with revenues rebated to households in lump-sum fashion, would increase the unemployment rate by 0.26 percentage points. In comparison, the unemployment rate increased 4.5 percentage points during 2007–2009. This effect is much smaller if the revenues are used to finance reductions in the payroll tax. That carbon tax/payroll tax swap raises the unemployment rate by only 0.02 percentage points. These results suggest that the net effect of carbon pricing on unemployment is likely to be very small.
However, looking only at the overall change in unemployment does not tell the whole story. Our model indicates that a carbon tax would cause a notable shift in jobs between the two sectors. We find substantial job losses in the polluting sector—not via layoffs, but rather through natural attrition and a big drop in hiring in that sector. These are offset by a similarly sized gain in jobs in the clean sector (driven both by demand, as consumers shift away from pollution-intensive consumer goods, and by reduced competition for workers from the dirty sector).
We also model an environmental performance standard (a limit on carbon emissions per unit of output). Although this policy would cause a slightly larger increase in unemployment (and have a slightly higher overall cost) than the carbon tax/payroll tax swap, it would lead to a much smaller sectoral shift in employment. To the extent that policymakers want to minimize such sectoral shifts, performance standards may be an attractive option.