• Working Paper

How Green Are These Valleys? Combining Revealed and Stated Preference Methods to Account for Ecosystem Costs of Deforestation


Pattanayak, S. K. (2001). How Green Are These Valleys? Combining Revealed and Stated Preference Methods to Account for Ecosystem Costs of Deforestation.


The rising tide of worldwide deforestation is irreversibly and substantively impairing ecosystem functions, thereby endangering the flow of several socially valuable goods and services from standing forests. Such seemingly irrational behavior arises, in part, from not considering the full costs of deforestation, particularly in terms of latent and complex ecosystem costs that are not well understood and rarely quantified. In this paper, we account for one forest ecosystem service—drought conditions induced by deforestation by using author-compiled survey data in the quasi-natural experimental setting of Flores, Indonesia. We apply two theoretically consistent economic valuation methods to estimate downstream drought costs of deforestation. Spatial variation in current indices of drought are used to estimate the marginal value of forest-induced droughts in terms of marginal profits for farming households in the affected watersheds. Contingent valuation (CV) surveys complement such measures by measuring support for forest protection that mitigates droughts and by providing information on household perceptions of the size and value of drought mitigation by forest ecosystems. Because the profit and CV models utilize different types of information to explain essentially the same economic optimizing behavior, based on data from the same households, by using common variables, we exploit these commonalities by estimating a combined model with constraints. Although we reject the restrictions across the combined profit and CV models, we achieve consistency across the two types of data by estimating a model that allows households to perceive different sizes of the ecological service. In addition to methodological and policy conclusions, our results provide guidance for understanding the role of forest hydrology in the local economy and a benchmark for further research on estimating the magnitude of local ecosystem costs of deforestation.