University of Hohenheim, Department of Economics, Germany
This paper analyses possibilities to improve contingent valuation surveys which are used to assess the social value of public goods like e.g. environmental changes. The contingent valuation method (CVM) is a survey based technique which assesses an environmental project by asking a representative sample of citizens their willingness to contribute to that project.
An important requirement for obtaining reliable results in CVM surveys is for respondents to answer honestly to the willingness to pay question. Depending on the payment mechanism respondents often have an incentive to strategically over- or understate their preferences for the public good, thereby systematically biasing CVM results. It is, therefore, essential to design payment mechanisms where it is in the respondent's best interest to answer truthfully according to his preferences. In this study we test the incentive structure of a number of alternative payment mechanisms. In particular we test the Clarke mechanism, which is known to be incentive compatible in theory. Although this mechanism is highly appreciated in theory it has never been implemented in empirical work in the context of the CVM. First, this study analyses theoretical incentive structures of mechanisms conventionally employed in CVM. Second, the Clarke mechanism is adapted to fit the CVM context to create the incentives to make people answer truthfully to the willingness to pay question. The different payment mechanisms are then applied in an empirical survey to compare the theoretical predictions with empirical results.
The public good to be valued was a flood prevention project in the Northern Thai city of Chiang Mai. In this project flood protection is to be achieved by technical measures in the city as well as by reforestation and soil conservation measures in agricultural areas in the upstream watershed. The survey including a total of 1500 face-to-face interviews has been conducted in Chiang Mai in the beginning of 2008. The results of this study will help making CVM surveys a more reliable tool for the valuation of environmental changes.
Keywords: Contingent valuation, environmental valuation, incentives, mechanism design, voting