Modeling starting point bias as unobserved heterogeneity in contingent valuation surveys: An application to air pollution

Up to now, the starting point bias has generally been considered to be a homogeneous phenomenon. In this article, we treat anchoring as an unobserved heterogeneous phenomenon. Our contribution is twofold. First, we show analytically and by way of simulations that assuming homogeneous anchoring can be hazardous and lead to misspecifications. Second, we propose an econometric model that starts with a dichotomous question and then uses an open-ended question. We finally apply our model to a contingent valuation survey on air quality. Our results suggest that how anchoring is modeled in empirical studies deserves more attention.