In Mayeres (1993, 1994), the marginal social costs of NO,, VOC and SO2 emissions by
road transport were determined on the basis of revealed preference of policy-makers, i.e. on
the basis of avoided costs of attaining internationally agreed air pollution objectives. In
this paper, however, we follow Small and Kazimi (1995) in using a direct damage estima-
tion approach. We prefer this method because the robustness of the ‘revealed preference
of the policy-makers’ method is highly dependent upon some very strong assumptions. It
would only be a correct estimate of the social cost if governments were rational, had per-
fect information and represented the preferences of society perfectly.
We begin by describing, in a general way, the methodology underlying the monetary
valuation of the social costs of air pollution. That methodology has been followed as
closely as possible in our analysis. However, as will become clear later on, lack of infor-
mation has often forced us to use a simplified approach.
Valuing marginal road transport emissions generally requires three steps. The first step
consists of establishing a relationship between a change in the emissions and the result-
ing concentration levels of the different primary and secondary air pollutants. This
requires atmospheric dispersion models which predict the spread of pollutants from their
origin, and chemical transformation models which describe how different pollutants react
together to form so-called secondary air pollutants. For some pollutants, these models
are relatively simple, while for others, they are extremely complex. We have encountered
most problems in this first step of determining the link between emissions and concentra-
tions and have been forced to make many assumptions. There is still a great need for
information in the domain of air pollution effects and most particularly for summary
information (such as emission-concentration studies).
The second step consists of relating the change in the concentration level to its effects
on health, vegetation, materials, visibility and ecosystems. This requires the use of
so-called dose-response relationships. ETSU and IER (1994) (hereafter referred to as
ExternE 2) give a broad review of existing studies in the field. However, once again, due
to information problems, only the effects on public health and some vegetation effects
could be included in our study. The final step consists of determining a monetary value
for the different effects of air pollution. For this, we have used Metroeconomica (1994)
(hereafter referred to as ExternE 9).