How Should Consumers' Willingness to Pay Be Measured? An Empirical Comparison of State-of-the-Art Approaches

Miller, KlausKlausMillerHofstetter, RetoRetoHofstetterKrohmer, HarleyHarleyKrohmerZhang, JohnJohnZhang2023-04-132023-04-132011-02https://www.alexandria.unisg.ch/handle/20.500.14171/9456410.1509/jmkr.48.1.172This study compares the performance of four commonly used approaches to measure consumers' willingness to pay with real purchase data (REAL): the open-ended (OE) question format; choice-based conjoint (CBC) analysis; Becker, DeGroot, and Marschak's (BDM) incentive-compatible mechanism; and incentive-aligned choice-based conjoint (ICBC) analysis. With this five-in-one approach, the authors test the relative strengths of the four measurement methods, using REAL as the benchmark, on the basis of statistical criteria and decision-relevant metrics. The results indicate that the BDM and ICBC approaches can pass statistical and decision-oriented tests. The authors find that respondents are more price sensitive in incentive-aligned settings than in non-incentive-aligned settings and the REAL setting. Furthermore, they find a large number of "none" choices under ICBC than under hypothetical conjoint analysis. This study uncovers an intriguing possibility: Even when the OE format and CBC analysis generate hypothetical bias, they may still lead to the right demand curves and right pricing decisions.enHow Should Consumers' Willingness to Pay Be Measured? An Empirical Comparison of State-of-the-Art Approachesjournal article