Using experimental auctions for marketing applications: A discussion

Journal of Agricultural and Applied Economics, Aug 2003 by Lusk, Jayson L

So how should researchers handle problems with multiple attribute valuation? Clearly, thought should be given (and perhaps pre-studies should be conducted) to identify whether there are attributes that are significant substitutes for the attribute of interest. If significant interactions between product attributes exist, experimental designs can easily be constructed to fully identify utility functions such as that in equation (1). And what of the diminishing marginal attribute hypothesis? First, some research should be conducted to determine whether and to what extent such a phenomenon may exist. If this conjecture holds, then the value of an attribute such as GM-free or guaranteed tender is only accurate and extendable to the "real world" when experimental procedures inform subjects of other attributes a good already possesses, such as brand names, nutritional labels, etc.

Conclusions

Given the increased emphasis in product differentiation at the retail level, there has been an increased interest in marketing research techniques such as experimental auctions. Given the increased interest in this area, the present article discussed general issues associated with EA and their relative advantages and disadvantages over other marketing research techniques. The relative advantage of EA over other methods is that they create an active market environment with feedback in which subjects exchange real goods and real money. The article also discussed four experimental design issues associated with EA: auction mechanism, market feedback and bidder affiliation, demand reduction and wealth effects, and multiple attribute valuation. The hope is that the quality and contribution of future EA studies will be enhanced by carefully considering each the experimental design issues.

1 Many of these studies actually focused on policy rather than marketing issues. Nevertheless, results from these studies have food marketing implications in addition to any policy implications that were presented in the papers.

2 Clearly, there a wide variety of other tools used in marketing research such as structural equations modeling, LISREL, multidimensional scaling, cluster analysis, etc. This discussion focuses on a few marketing tools used to determine product pricing, which is likely of greater interest to economists.

3 Attention is given to "choice-based-conjoint-analysis" because it is consistent with economic theory and more closely mirrors actual shopping behavior. See Louviere, Hensher, and Swait for an excellent book on this method.

4 This is by no means an exhaustive list of incentive compatible elicitation mechanisms. The discussion of elicitation mechanisms is limited to those frequently appearing in the literature.

5 Obviously, Figure 1 is a simplistic illustration. The demand for GM food is not likely to remain constant if a participant bought a non-GM food in round 1 but not a GM food. Because the goods are substitutes, the demand for the GM food is likely to shift downward by some amount in round 2.