Branded Promotion and Private Label Share: A Test of Kashyap’s Hypothesis

Abstract

This provides a test of Kashyap’s speculation that branded goods are discounted less often when private label goods have a high market share. This was tested by (1) looking at a cross section across a large set of packaged goods categories and (2) looking at twelve product categories cross-sectionally across markets. The promotion measure used is the proportion of dollar (or volume) sales sold on merchandising (feature, display, price reduction). The evidence is weak if we look at branded promotion relative to private label share without considering private label promotion. However, it is strong for a revised hypothesis that as the level of private label share increases, there is less reason to offer discounts on the branded products and more reason to offer discounts on the private label products. As private label share increases, the proportion of branded dollars on promotion decreases relative to the proportion of private label dollars on promotion. In many categories, this means that the proportion of branded sales on promotion is lower than the proportion of private label sales on promotion.