The travel cost method is based on the estimation of the area beneath an 

 empirical demand curve. In principle and practice, the price of travel or the 

 number of trips to competing outdoor recreation sites may be needed as an 

 independent variable for accurate regression estimates of the travel cost demand 

 curve (the number of trips per thousand residents per annum versus the travel 

 cost). These variables may also be needed to produce unbiased estimates of the 

 net social benefits conferred by the site. 



In practice, TCM estimates of benefits conferred are produced by regression 

 estimates of participation rates on a number of independent variables. The key 

 independent variable is the cost of travel to the site. If the travel cost 

 variable is underestimated by the omission of the cost of foregone wages as a 

 component of total travel costs, benefits estimates based on the empirical demand 

 equation will markedly underestimate the true benefits conferred by the site. 

 Other useful independent variables for the estimation of travel cost demand 

 curves include the real income of the households, and socio-economic variables 

 such as the sex of the head of the household and the level of education of the 

 household. To calculate the benefits conferred by the site, all the other 

 independent variables (except for the travel cost variable) are held at their 

 mean value. 



The estimated travel cost demand curve can be used for other practical 

 purposes, particularly forecasts of future levels of demand that involve shifts 

 in both demographic and economic variables. 



CONCEPTUAL PROBLEMS WITH CONSUMER SURPLUS MEASURES 



There are important shortcomings in the traditional consumer surplus 

 concept. Discussion of certain aspects of these defects is beyond the scope of 

 this paper, but it may be useful to note that the Marshall ian (Marshall 1920) 

 measures of consumer surplus (the area under the demand curve but above the 

 horizontal price line) are nontransitive with respect to multiple price changes 

 in interrelated markets. When prices in two interrelated markets change, the 

 change in the Marshall ian consumer surplus is the sum of the changes in the 

 relevant regions underneath the demand curves of the two goods (Just et al . 

 1982). 



For definiteness, suppose that an exogenous shift in the supply curve of 

 outboard motors leads to a change in the price of outboard motors, in turn 

 causing a shift in the demand (schedule) curve for small boats, and that this 

 latter shift induces a change in the price of small boats. The total change in 

 consumer surplus depends on the sequential order of the analysis; if the consumer 

 surplus change in the market for outboard motors (after the price changes) is 

 estimated first, and then the change in consumer surplus in the boat market is 

 added to this initial value, one value of the total change in consumer surplus 

 emerges from the analysis. But if the sequence of the analysis is changed, and 

 the analysis starts with the motor market first, another value for the consumer 

 surplus results (Just et al . 1982). 



