Because these people have higher actual travel times, the value of the coefficients is lower in
absolute size. However, the quotient of travel time and travel cost, i.e. the VOT, can give
more insight into the relative importance of travel time and travel cost for respondents having
a higher actual travel time. Looking at the VOT, respondents with a longer travel time in
reality (longer than 50 minutes) have a lower value of time. Thus, respondents who in the
current situation live relatively far from their work want to pay less to save a certain amount
of travel time than people who live closer to their work. This might be explained by self-
selection: people who prefer low travel times already live closer to their work. This finding
seems to be in contrast to the finding of Gunn (2001), who expects a higher VOT with
increasing travel time. However both results do not have to be in conflict. The travel times
shown within the choice experiment used were tailored on basis of the actual travel distance.
Thus, respondents with lower travel times observed relatively lower travel times within the
experiment than those with higher commute distances. In the end the self-selection effect
might well overshadow a possibly increasing VOT with travel time for individual persons.
Closer inspection of the travel time and travel cost coefficients finally seems to lead to the
conclusion that the lower VOT for higher actual travel time classes is particularly caused by
the relatively higher travel cost (discounted for travel time) disutility for these higher travel
time classes.
6. Conclusions
Road pricing may play a role in both stages of the relocation process: the decision whether or
not to relocate and the choice of the new residential location. This paper focused on getting
more insight into the probability to relocate on the one hand and on studying the relative
influence of trip and location related variables in the actual residential location choice on the
other hand. With respect to this last point special emphasis has been put on the comparison of
the importance of travel cost (especially due to road pricing) versus housing cost and travel
time in location decisions.
The probabilities of moving to a residential location closer to work and/or searching for
another job closer to home under influence of different types of road pricing measures have
been studied. Roughly half of the sample consisted of car commuters confronted with traffic
congestion on a regular basis. The other part of the sample was selected on basis of the
criterion that people had to possess a car. Therefore, results may not be directly transferable to
commuters in general. On average roughly 5 percent of the respondents indicated a quite high,
24