Analysis of the transmission of carbon taxes using a multi-sector DSGE

Carbon tax has attracted increasing attention as a means of curbing greenhouse gas (GHG) emissions. While the implementation of carbon taxes necessarily involves consideration of the impact across different sectors and different periods, most existing studies use models which do not provide a detailed account of either sectoral interaction or the dynamic nature of the responses of households and firms. To fill this gap, we construct a New Keynesian multi-sector dynamic stochastic general equilibrium (DSGE) model with an input–output structure of intermediate inputs and an investment network calibrated to Japan’s economy. We study the impact over time of carbon tax on different sectors, on aggregate GDP, and on GHG emissions. We then consider the long-term implications through a steady-state analysis, and the short- to medium-term implications by a simulation from 2020 to 2050, under various scenarios with different tax base compositions and announcement timings. We show that the impact on the trade-off between output and GHG emissions is importantly affected by inter-sectoral interactions among firms, and by the intertemporal decisions of households and firms. For example, an orderly implementation of carbon tax may mitigate a decline in value added of the most affected industry and the aggregate economy in 2050 by about 7% and 2% points, respectively, and contain an increase in the stock of GHG in 2050 by 3% points compared with the case of disorderly implementation. Also, keeping GHG emissions constant, GDP in 2050 can vary by 0.5% points depending on tax sources.