Table 8 gives a summary of the marginal social costs of Belgian emissions of air pollu-
tants. When using these figures, one should keep in mind the assumptions made in the
previous paragraphs and the uncertainties associated with them. This table gives the
marginal social cost, per gram, for each pollutant. It can be seen that the dominant cost
category is the health costs, especially the health costs caused by PM,, concentrations.
In the first part of Table 9, we give the emission factors of a few car types and of the
urban public transport modes. These emission factors are computed with the functions
of VIA (1995) in which the travel speeds of 2005 in Brussels are used. The emission rates
are computed for the urban driving cycle defined by the EC in the framework of their
car emission standards. The urban driving cycle corresponds to one average speed.
Emission rates for other average speeds were obtained by homothetically increasing or
decreasing the maximum speeds defined in the urban driving cycle. This might be a poor
approximation as decreased average speeds originating from increased congestion might
result in a more irregular driving pattern. For gasoline cars, the emissions of PM,, and
SO, are negligible. The second part of Table 9 gives the external costs of air pollution
per vehicle km. Even among private cars, the variation in air pollution costs between
different car types and uses is quite important (e.g. the external cost of a large gasoline
car in the peak period is twice as high as the external cost of a small diesel car driving in
the off peak period). For gasoline vehicles, NO,Y is the dominant cause of air pollution
costs. For diesel vehicles, SO,,., PMlo and NO., are the most important costs.