Monday, October 12, 2020

The Macroeconomic Impact of Europe's Carbon Taxes

Policy makers often express concern about the impact of carbon taxes on employment and GDP. Focusing on European countries that have implemented carbon taxes over the past 30 years, we estimate the macroeconomic impacts of these taxes on GDP and employment growth rates for various specifications and samples. Our point estimates suggest a zero to modest positive impact on GDP and total employment growth rates. More importantly, we find no robust evidence of a negative effect of the tax on employment or GDP growth. We examine evidence on whether the positive effects might stem from countries that used the carbon tax revenues to reduce other taxes; while the evidence is consistent with this view, it is inconclusive. We also consider the impact of the taxes on emission reductions and find a cumulative reduction on the order of 4 to 6 percent for a $40/ton CO2 tax covering 30% of emissions. We argue that reductions would likely be greater for a broad-based U.S. carbon tax since European carbon taxes do not include in the tax base those sectors with the lowest marginal costs of carbon pollution abatement.
Figure 3 shows the IRF (Impulse Reponse Function) for the LP (Local Projection) model for real GDP, estimated using all 31 countries over the full 1985‐2018 sample, where the carbon tax rate is interacted with the share of emissions covered by the tax. Figure 3a shows results from the unrestricted model, that is, the model that allows for a nonzero long‐term effect of the tax on GDP growth. The predicted effect is positive in each year through year 6 except for year 4. In no year, however, is the effect significant at the 5% level (in most years it is within one standard error of zero). The results for the restricted model (figure 3b), in which a zero long‐term effect of the tax rate on GDP growth is imposed, are similar to those for the unrestricted model. Again, the point estimate is generally no more than one standard deviation away from zero.
Figure 4 shows the IRFs for the SVAR (Structural Vector AutoRegression) model. The unrestricted IRF (figure 4a) is always positive and near constant at about 0.4 percentage points.
Figure 6 shows IRFs for total employment. In both the unrestricted and restricted cases, employment initially rises and then subsequently falls. The cumulative impact (Figure 7) is positive over a six year period with a point estimate of 1.15 percentage points in the unrestricted LP model and 0.35 percentage points in the restricted model. In neither model do we reject a zero cumulative impact. Results (in the Appendix) for manufacturing employment are similar to total employment but the estimates are less precise. As with GDP, we can reject significant negative employment impacts from the carbon tax.
by Gilbert E. Metcalf and James H. Stock
National Bureau of Economic Research (NBER)
NBER Working Paper No. 27488l; Issued in July 2020

No comments:

Post a Comment