Do models of vertical strategic interaction for national and store brands meet the market test?

Abstract

This article develops a methodology for empirically examining some of the central assumptions commonly used in the theoretical literature on vertical strategic interaction. This methodology is used to test these assumptions by using data for six individual categories across 59 local markets in 1991 and 1992, focusing on the vertical and horizontal interaction between private label and national brands. There are three central findings. First, the vertical strategic interaction observed for national brands varies considerably across categories (a single form of interaction, "Vertical Nash," tends to be more common for private label brands). Second, we generally reject the use of proportional mark-up behavior by retailers. Third, we reject linear demands in a favor of a more flexible nonlinear form. These results suggest that models specifying proportional mark-up behavior and linear demands do not accurately reflect market reality. Further, because vertical strategic interaction between manufacturers and retailers seems to be idiosyncratic to the category and brand, future research should consider multiple forms of vertical interaction to produce reasonably general results.