Three essays on the theory of environmental taxation in a second -best setting

This dissertation consists of three essays on the theory of environmental taxation in a second best setting.;In the first essay, I extend previous models of the “double dividend hypothesis” by incorporating tax-favored consumption goods (e.g., housing, medical care). The efficiency gains from recycling environmental tax revenues are therefore larger because pre-existing income taxes distort both consumption decisions and factor markets. In this setting a revenue-neutral emissions tax (or auctioned permits) produces a double dividend. Moreover, the overall costs of environmental tax swaps are negative, for modest emissions reductions. The efficiency gains from emissions taxes over grandfathered permits are also much larger than previously recognized.;In the second essay I explore the interactions between taxes on work-related traffic congestion and pre-existing distortionary taxes in the labor market. A congestion tax raises the overall costs of commuting to work and discourages labor force participation at the margin, when revenues are returned lump-sum transfers. We find that the resulting efficiency loss in the labor market can be larger than the Pigouvian efficiency gains from internalizing the congestion externality. In contract, if congestion tax revenues are used to reduce labor taxes the net impact on labor supply is positive, and the efficiency gain in the labor market can raise the overall welfare gains of the congestion tax by as much as 100 percent. Recycling congestion tax revenues in public transit subsidies produces a positive, but smaller, impact on labor supply.;In short our results indicate that the presence of pre-existing tax distortions, and the form of revenue recycling, can crucially affect the magnitude, and possibly even the sign, of the welfare effect of road-pricing schemes. The efficiency gains from recycling congestion tax revenues in other tax reductions can amount to several times the Pigouvian welfare gains from congestion reduction.;The third essay uses analytical and numerical models to illustrate how the presence of other distortions within the transport system change the overall welfare effect of a congestion tax. These other distortions include a transit fare subsidy, congestion on competing (unpriced) routes, accident externalities, gasoline taxes and pollution externalities.;Each of these pre-existing distortions can substantially alter the welfare effect of a congestion tax that would be predicted by a first-best analysis. To the extent that congestion taxes encourage additional travel on other congested routes, or on subsidized transit systems, they can produce sizeable indirect welfare losses. In addition, induced reductions in the demand for gasoline can lead to substantial welfare losses when, as appears to be the case for European countries, gasoline taxes greatly exceed marginal pollution damages. On the other hand, congestion taxes can produce significant welfare gains by offsetting accident externalities—though these are partially offset by increased accidents on competing roadways.;The importance of other distortions varies considerably across different transport systems and across different countries. Our generic analysis illustrates the proportionate change in the welfare effect of a congestion tax due to each of these distortions, under a wide range of alternative parameter scenarios.