Incentive Incompatibility and Starting-Point Bias in Iterative Valuation Questions

We test for incentive incompatibility and starting-point bias to describe the effects of iterative valuation questions on willingness to pay. We compare double-, triple-, and multiple-bounded models with data from two surveys with similar designs of the valuation questions. We find that incentive incompatibility is present in both sets of data and starting-point bias is present in one. The efficiency of the willingness-to-pay estimate is improved in only one set of data. The potential loss from using iterative questions without controlling for both incentive incompatibility and starting-point bias is biased willingness-to-pay estimates. (JEL Q26)