The paper will be structured as follows. The next section provides the theoretical framework.
Section 3 focuses on the used data and on the study design. Section 4 discusses the probability
of households changing their residential or work location due to a road pricing measure and
furthermore presents the explanatory variables for the probability to change location. The
importance of trip and location related variables in a location decision is described in section
5. The conclusions finally follow in section 6.
2. Theoretical framework
Figure 1 presents a conceptual model for the relation between road pricing and (re)location
choice. Central to our approach is the observation that relocation decisions consist of several
stages (Devisch et al., 2005). The first stage can be termed awakening. This implies that a
household realises that it can improve its housing conditions by moving to another dwelling at
another location. Awakening can be caused by various triggers. These may relate to changes
in the household, such as changes in household composition, changes in income or changes in
preferences but also to external factors, such as changes in the environment (e.g. socio-
economic status of the neighbourhood). These factors are summarized as ‘other factors’
within figure 1. Besides that, generalized transport costs clearly can also be an external trigger
for awakening (figure 1). Relocation decisions may also be quite dependent on the type of
road pricing measure (not presented in figure 1). More general forms of road pricing, such as
a flat kilometre charge, may especially have an effect on the distribution of people over
locations. The effect of such a pricing measure on the demand where houses or business parks
should be built seems to be lower, as was computed with so-called land-use transport
interaction models (e.g. Eradus et al., 2002). The strongest spatial effects are expected to
occur when spatial dependent forms of road pricing are implemented, such as for example a
spatial differentiated kilometre charge or a cordon charge.
Once the decision to relocate is made a household will evaluate available dwellings on a set of
criteria, including characteristics of the dwelling and the environment. One of the factors in
this respect can be the expected (generalized) travel costs implied by the residential location,
which are affected by road pricing policies (see figure 1). Especially in the spatial economic
theories from the 1960’s (e.g. Alonso, 1964; Muth, 1969) the trade-off between travel costs
for commuting and the housing cost are determining for the residential location in relation to
the work location. Since the 1970’s several authors criticised these classical spatial-economic