P. Fix et al., Endogenously chosen travel costs and the travel cost model: an applicationto mountain biking at Moab, Utah, APPL ECON, 32(10), 2000, pp. 1227-1231
The travel cost model is frequently used to estimate net willingness to pay
for recreation at remote sites by using the visitor's travel costs as a pr
oxy for the price of recreation. However, some concern has been raised over
the validity of using the visitor's stated travel costs as a proxy for pri
ce. This paper addresses some of these concerns, by examining the possible
over-estimation of consumer surplus due to endogenously chosen travel costs
. This paper extends past theoretical work for the linear model by developi
ng a correction for endogenously chosen travel costs in more commonly used
nonlinear models such as the Poisson or count data model. Also provided is
the first empirical test of the presence of endogeneity and an estimate of
the magnitude of the error from ignoring endogeneity in travel costs. After
applying this test and the correction to data that was gathered for mounta
in biking at Moab, Utah the estimate of consumer surplus falls from US $153
to US $135, which is a 12% reduction.