Findings—Retailer Promotion Pass-Through: A Measure, Its Magnitude, and Its Determinants

Abstract

We use data on all manufacturer funding and promotion activity by a major U.S. retailer during a two-year period to compute promotion pass-through and assess its magnitude. Then, we estimate a two-tiered probit and lognormal regression model to study drivers of the large variation we observe in pass-through rates. Although our analysis is based on data from a single retailer, it provides a much more complete picture of the magnitude and variability of pass-through than has been available to date. Some key insights from our work are as follows. First, the retailer passes through more than 100% of the total manufacturer funding it receives in aggregate, but the median pass-through rate for individual manufacturers is much lower than 100%. Second, some manufacturers are promoted even without funding. This is more likely for private label and high-share manufacturers in high-lift and high-margin categories. Third, a small number of manufacturer and category characteristics explain a significant amount of the variation in pass-through. In particular, pass-through is higher for private label. It increases with the share of the manufacturer in the focal category but also with the sales of that manufacturer in other categories. Categories with high sales, high promotion lift, low concentration, and low margin get more pass-through. We corroborate some recent conclusions in the literature, e.g., that some pass-through rates are much higher than 100% and that high-share manufacturers get more pass-through. We add several new insights, e.g., on the magnitude of aggregate pass-through, on cross-pass-through across and within categories, on pass-through for private label, and on the category drivers of pass-through.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 66 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.