Contingent valuation experiments for strategic behavior

Abstract

The existence and extent of strategic disclosures in field contingent valuation surveys have been debated but few empirical studies are available. This paper summarizes the results of an experiment to identify strategic behavior using alternative disclosure mechanisms with closed-ended and open-ended formats. In addition, the experiment examined the effect of mechanism design on respondent's ability and willingness to disclose their personal valuation. Statistical results using censored discrete regression analysis and nonparameteric means tests provide some evidence of systematic differences due to strategic behavior. But the comparisons are inconclusive due to the significant influence of sample selection on valuation results using closed-ended formats and the inherent limitations of comparative means test procedures. Evidence from this analysis cannot refute the hypothesis that respondents try, to the best of their ability, to provide truthful information. But the procedures used to interpret this information can have a significant effect on aggregate valuation results.