Predicting Competitive Response to a Major Policy Change: Combining Game-Theoretic and Empirical Analyses

Abstract

This research uses Procter & Gamble's value pricing initiative as a context for testing whether actual competitor and retailer response to a major policy change can be predicted using a game-theoretic model. We first estimate demand functions for P&G and competitor brands from the period before value pricing was initiated. We then formulate a dynamic manufacturer-retailer Stackelberg model that includes P&G, a national-brand competitor, and a retailer. The model takes P&G's move as given and prescribes the price and promotion response of the competitors and the retailer. We substitute the estimated demand parameters into the model to obtain prescriptions for each competitor and the retailer, and see whether these prescriptions are related to the actual response. We find that the dynamic game-theoretic model calibrated with empirical estimates of demand parameters has significant predictive power. We also test the predictive power of two benchmark models. The first is based on the reaction function approach of Leeflang and Wittink (Leeflang, Peter S. H., Wittink, Dick R. 1992. Diagnosing competitive reactions using (aggregated) scanner data. Internat. J. Res. Marketing 9 39–57.), and the second is a simplification of our dynamic model where the retailer is not strategic. The dynamic game-theoretic model performs better than either benchmark.

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 — 58 citations exist upstream.

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