Glenn C. Blomquist

Eliciting Willingness to Pay Without Bias: Evidence from a Field Experiment

Concern exists that hypothetical willingness to pay questions overestimate real willingness to pay. In a field experiment, we compare two methods of removing hypothetical bias, a cheap talk approach and a certainty approach, with real purchases. We find evidence of hypothetical bias for unadulterated contingent valuation. Contingent valuation with certainty statements removes the hypothetical bias, but the cheap talk approach has no significant impact. Our findings suggest that willingness to pay can be accurately estimated by adding a simple follow-up question about the certainty of responses and that cheap talk is not a generally effective approach.

Experimental Results on Expressed Certainty and Hypothetical Bias in Contingent Valuation

Use of the contingent valuation method is controversial among economists because it is based on hypothetical rather than real choices. Previous experiments have suggested that the commonly used dichotomous choice contingent valuation method leads to hypothetical bias, i.e., overestimates the real willingness to pay. We carried out an experiment to compare the dichotomous choice contingent valuation method with real purchase decisions for a consumer good. We confirm previous findings that hypothetical yes responses overestimate real purchase decisions, but we cannot reject the null hypothesis that definitely sure yes responses correspond to real purchase decisions.

Resource quality information and validity of willingness to pay in contingent valuation

Elicitation of valid statements of contingent value requires survey participants who are familiar with the environmental resource change. A primary purpose of the contingent market must be to assure familiarity by providing information. Information about resource quality is important when incompletely informed respondents, say nonusers, perceive resource quality which diverges from true quality. Differences in perceived quality and true quality can be influenced as respondents learn from information in the contingent market. By presenting survey participants with information about four wetlands of varying qualities we test for information effects in a dichotomous choice contingent market for wetlands allocation. We find that information about quality is a determinant of willingness to pay for wetland preservation. Information about resource quality presented in contingent markets will result in more valid valuations of changes in allocations of environmental resources.

Testing for Non-Response And Sample Selection Bias in Contingent Valuation: Analysis of a Combination Phone/Mail Survey

We use a combination phone/mail survey to test for possible sample biases in contingent valuation. We find no sample selection bias but do find non-response bias. We show how failure to correct for non-response bias distorts aggregate benefit estimates.

Measuring Amenity Benefits from Farmland: Hedonic Pricing vs. Contingent Valuation

The amenity value to Kentucky residents from horse farm land was estimated using both the contingent valuation method and the hedonic pricing method. The hedonic pricing model included both the housing and labor markets. A value function estimated from dichotomous choice contingent valuation responses showed that the value of a change in the level of the horse farm amenity was sensitive to the size of the change, with no evidence of value that is independent of the size of the change. The two methods generated estimates of the external benefits from horse farm land that were within 20 percent of each other.

Calibrating Hypothetical Willingness to Pay Responses

Experimental data comparing hypothetical and real dichotomous choice responses for two different goods were used to estimate a statistical bias function to calibrate the hypothetical yes responses. The probability that a hypothetical yes response would be a real yes response was estimated as a function of the individuals self-assessed certainty of the hypothetical yes response (assessed on a 0–10 scale) and a variable representing the price level. Without calibration the hypothetical yes responses significantly exceeded the proportion of real yes responses, but after calibration the null hypothesis of no difference between hypothetical and real responses could not be rejected in any of the experiments.

Measuring Contingent Values for Wetlands: Effects of Information about Related Environmental Goods

A model of contingent market behavior is developed which emphasizes the role of household information about wetlands and related environmental goods.

Hedonic prices, demands for urban housing amenities, and benefit estimates

Abstract This paper uses a Rosen, two-step, hedonic price-trait demand approach to estimate demand functions for a vector of urban amenities. To ascertain whether this theoretically preferred approach yields benefit estimates which differ from the oft-used Ridker-Henning, one-step, hedonic approach we conduct a sensitivity analysis. We find that the two-step approach does yield different benefit estimates and that the differences are large for some amenities. The estimates are sensitive to the functional form of the hedonic equation when the forms are significantly different according to modified Box-Cox results, but are not particularly sensitive to specification of the amenity demand equation.

Valuing the Arts: A Contingent Valuation Approach

Government funding of the arts has received considerable attention in the United States in recent years. Efforts to cut funding to the National Endowment for the Arts and declining budgets for state arts agencies have raised questions about how much individuals value the arts. This paper applies the contingent valuation method to assess this value, using surveys of random households and of arts patrons. Our analysis estimated a mean willingness to pay (WTP) among all Kentucky households from

Eliciting Willingness to Pay Without Bias Using Follow-Up Certainty Statements: Comparisons Between Probably/Definitely and a 10-Point Certainty Scale

6 to