Estimating the Benefits of Agri-environmental Policy: Econometric Issues in Open-ended Contingent Valuation Studies

This paper reports on an open-ended Contingent Valuation Method study of the conservation benefits of Environmentally Sensitive Areas (ESAs) in Scotland. The ESA scheme is a central component of agri-environmental policy in the UK, and an interesting policy question concerns the extent of non-market benefits generated by such ESAs. The econometric issues we raise in this paper revolve around bid curves. Bid curves are estimated in open-ended Contingent Valuation Method (CVM) studies for three reasons. These are: (1) as a test of theoretical validity; (2) as a test of discriminant validity; and (3) as a means of benefits transfer. Within the first and last of these aims, the partial relationship between willingness to pay (WTP) and independent variables such as income is of interest. There are several econometric issues involved in estimating such relationships, First, the selection process implicit in obtaining positive WTP bids should be explicitly modelled. Second, many CVM surveys suffer from item non-response with respect to 'sensitive' questions such as the respondent's income; these non-responses may be non-random in nature. Finally, it is possible to dis-aggregate the effect of marginal changes in, say, income on WTP into two elements, namely: an effect on the probability that the individual will be willing to pay something; and secondly, an effect on how much they are willing to pay.