Cross-Brand Pass-Through in Supermarket Pricing

We investigate the sensitivity of cross-brand pass-through estimates to two types of pooling: across stores, and across regular price and promotional price weeks. Using the category data from Besanko, Dubé, and Gupta (2005), hereafter BDG, we find consistent support across all 11 categories for the predictive power of the wholesale prices of substitute products for retail shelf prices. A Bayesian procedure is used to address the small sample issues that arise in the absence of pooling. Even though the unpooled results render our inferences for specific cross-brand pass-through magnitudes reported in BDG as imprecise, consistent with McAlister (2007), we find significant empirical support for cross-brand pass-through. We next assess the sensitivity of cross-brand pass-through estimates to pooling. This requires us to construct a much longer time series of 224 weeks for the refrigerated orange juice category, in contrast with the 52-week samples used in BDG and McAlister (2007). We find strong empirical support for the predictive power of wholesale prices of substitute products for retail shelf prices. In addition, we find evidence of nonzero own- and cross-brand pass-through elasticities for which our inferences are much more precise. These findings are robust to the separation of regular and promotional price weeks. However, the magnitudes of own-brand and cross-brand pass-through are quite different during promotional and regular price weeks. Our results clearly show that with longer data series and more robust models that can handle small sample sizes, there is evidence of cross-brand pass-through, substantiating the findings in BDG. Finally, we comment on why our results are entirely consistent with both the theoretical and empirical literatures on category pricing and retailer behavior.