These studies give fairly scattered results. Nevertheless, a tentative conclusion can be drawn here
(one which would obviously need to be verified by more far-reaching research) that passengers and goods
traffic by road and rail generate specific noise costs (in terms of passenger- or tonne-km) that are equivalent
in order of magnitude.
Local pollution
Local pollution is expressed in terms of several components: sulphides, nitric oxides, and
particulate matter (neglecting pollution by CO2, which is global in nature, and which is discussed below).
The methods of evaluation used here are primarily of the indirect type, involving first of all a technical
estimate of the damage done, and then an evaluation of the "cost of repairs or protection". "Substitution
market" methods are less common here than for noise. As for "contingent valuation" methods, these are
occasionally used, but they tend to produce very different results.
Damage refers to human health, material damage and effects on plant life. These have been
separated out in a number of studies. They are summarised in Kågeson (1992a), from which most of the
information in Table 2.7 is drawn. It should be noted that, as for noise, the estimates based on willingness to
pay are generally much higher than those based on damage. For example, the Planco study indicates
pollution costs higher by a factor of 1.5 to 2.
Apart from the UPI estimate, which is very high, the percentages obtained are relatively little
dispersed around a mean value of the order of 0.4 per cent of GNP. The breakdowns by mode and by traffic
unit can be examined either in terms of physical or monetary quantities. In terms of physical quantities, the
following studies have been done: