theory can be implemented in the real world. Very often there are restrictions on the instruments the government can use. We explore the implications of these for the optimal tax rules. Finally, we briefly discuss some dimensions of political acceptability.
External costs of transport
The marginal external costs of transport use correspond to the costs caused by an additional transport user that are not borne by the user himself but by others. Transport use causes four main categories of externalities: congestion, accidents, environmental costs (including air pollution, global warming and noise) and road damage externalities. These costs consist not only of costs in the monetary sense, but also of, for example, time losses, pollution, noise and so on. The marginal external congestion costs are present whenever an additional vehicle in the transport network reduces the speed of the other transport users.A lower speed has several effects.It affects the operating costs of the other transport users, and their time costs. Their time costs increase not only directly because of lower speed, but also indirectly due to schedule adjustment. The marginal environmental costs include the costs imposed by the emission of air pollutants and noise on society in general and on future generations (in the case of air pollution and global warming).
The marginal external accident costs are not straightforward to define. When an additional vehicle joins the traffic flow, it causes three types of costs to society. First, the transport user himself is exposed to an accident risk. The social costs of this consist of his own utility loss due to the accident risk (which is internal), the so-called pure economic costs associated with the accident risk (net output loss, medical costs, police costs etc.), and possibly also the utility loss of relatives and friends. Secondly, the additional transport user may have an impact on the accident risk of the other infrastructure users and therefore on the associated costs for society and these other users. Thirdly, other transport users will adapt their behavior when confronted with a changed traffic situation. These avoidance costs should also be taken into account. How much of these three types of costs is external depends on the liability and compensation rules that are in vigor, on the type of insurance pricing, etc. The external road damage costs are discussed extensively in Newbery (1988) and Small et al. (1989). They arise when the passage of heavy vehicles
causes damage to the road surface. Two types can be distinguished: the increased repair cost of the road, borne by the road infrastructure provider, and the increased vehicle operating costs for the other road users. Under a number of conditions Newbery shows that the second type of road damage costs are negligible in all reasonable cases. The different externalities are not independent. For example, a reduction in speed will have an impact on air pollution and accident risk. But the externalities are not perfectly correlated. Therefore different instruments will be needed to tackle them. Nevertheless, when designing the policy instruments account should be taken of the interactions between the externalities.
A number of studies have tried to determine the marginal external costs of (mainly road) transport use. A selection of recent studies include Small and Kazimi (1995), Delucchi (2000), Friedrich and Bickel (2001), Mayeres and Van Dender (2001) and Beuthe et al. (2002). The relative importance of the different externalities is highly situation dependent. The marginal external costs vary widely in respect of the network considered, the volume of traffic and the vehicle type. As an illustration Figure 1 gives the marginal external costs that are expected to be generated by interregional road transport in Belgium in 2005. It shows that in this particular case congestion is the dominant external cost category in the peak period. In the off-peak period air pollution costs account for the largest share for diesel vehicles. This is due to their relatively high emission of particulate matter which is associated with high health costs. For other vehicle types accident costs are the most important in the off-peak period. It should be noted that the marginal external cost calculations are still subject to many uncertainties. Research in this area is continuing and is expected to improve the quality of the estimates.
Figure 1: Marginal external costs of interregional road transport in Belgium in 2005 (business as usual scenario)
PK = peak, OF = off peak, GAS = gasoline car, DIES = diesel car Source: Mayeres and Van Dender (2001)
Towards better transport pricing
First best
The initial analyses in the transport economics literature of optimal pricing of transport were conducted in the first best framework. The first best pricing solution in the presence of negative externalities can be traced back to Pigou (1920): given the appropriate convexity conditions, an efficient allocation of resources can be obtained by setting the price of a good for consumption or production according to the marginal costs incurred by society.This way the generator of the externality is confronted with the damage his consumption or production imposes on others. In the transport economics literature the first best approach is discussed in partial equilibrium models. The main interest is in urban transport and the determination of optimal tolls in the presence of congestion. The optimal pricing rules are derived by maximizing the net social benefits for trips over a road. The result can be summarized as follows: the price of the trip must equal its short run marginal social costs. To make each transport user face the marginal social costs of his trip, it is necessary to levy a toll equal to the difference between the marginal social cost and the average cost that is already borne by the user.
In practice the necessary conditions for the first best rules to be valid are not satisfied. The first best analysis assumes that the government is concerned with the sole objective of controlling the (congestion)
externality and that it can make use of perfect instruments. Externality taxes are analyzed without taking into account the existence of implementation constraints or the presence of distortionary taxes. In reality these conditions do not hold.
The implications of pre-existing distortionary taxes
Controlling externalities is not the only objective of the policy-maker. He also provides public goods and services–that have to be financed–and has distributional objectives. In the first best models it is assumed that the government can make use of non-distortionary taxes in order to achieve these objectives. A tax is non-distortionary if and only if individuals can do nothing to change the tax they have to pay. Taxes for which this is not the case are distortionary.Distortionary taxes are not efficient: if the government could replace them by non-distortionary taxes it could gather the same revenue while at the same time increasing welfare.The assumption that the policy maker can make use of non-distortionary taxes is not realistic.In practice the government has to use labor taxes, capital taxes, indirect taxes etc. Therefore one has to develop a tax system such that the government’s objectives (revenue raising, distributional objectives and tackling the externalities) can be realized at the lowest cost. Since transport taxes do not only have an impact on external costs but also raise revenue and have an impact on distribution, they should be considered within the broader framework of the general tax system.
Economic theory tells us what the optimal tax structure should look like in this case. Bovenberg and Goulder (2002) give an excellent overview of the theory of environmental taxation. Mayeres and Proost (1997) derive optimal tax rules for passenger and freight transport in the presence of externalities with a feedback effect, in an economy with non-identical individuals. It is assumed that the government can make use of indirect taxes on the consumption of all commodities (including transport), a tax on intermediate inputs in production and of a uniform transfer to households. For simplicity we assume that only transport generates externalities. Transport is used both by households and firms.
The indirect tax on the consumption of transport goods then consists of two components. The first component is aimed at revenue raising. The second component corrects for the external effects. If there are no other restrictions on tax instruments, this second part is present only for goods causing externalities, in our case transport goods. Distributional considerations play a role in both components.
The first component is related to revenue raising. It makes a trade-off
相关推荐：
loading