Abstract: Blind Boxes, marketed by PopMart in China, are popular in China’s toy markets. PopMart’s publicity states that the term “Blind Box refers to packaging that keeps the toy a mystery until opened.” PopMart markets many series of blind boxes; each series always contains several ‘regular figurines’ and one more valuable ‘hidden/secret figurine’. A blind box is not a complete mystery, as some information is disclosed. One problem is that the stated probabilities of the various figurines cannot be correct because they do not sum to one: this may lead potential buyers to misperceive probabilities, overestimating the probability of getting the ‘hidden/secret/more-valuable figurine’, and therefore purchasing blind boxes. This research uses the Becker-DeGroot-Marschak (BDM) method to elicit valuations of blind boxes, and hence to infer key parameters driving behaviour. The experiment was in two parts: the first to elicit risk preferences, and the second to elicit the probabilities as perceived by subjects. This latter helps to determine whether probability misperception may be a reason for blind box purchases. I fitted four theoretical models: Expected Utility, Rank Dependent Expected Utility, the MaxMin model and the αMaxMin model and ranked them using both the Akaike Information Criterion and the Bayesian Information Criterion. Both imply that EU fits the data the best. With the EU model, over half the participants overvalued the probability of getting the secret figurine. It seems that the marketing may tempt consumers to overestimate the probability of getting the secret figure, and thus being willing to buy blind boxes.
JEL codes: D01, C91, D81, D12