Predicting Competitive Response to P&G's Value Pricing Move: Combining Normative and Empirical Analyses

This research uses P&G's value pricing initiative as a context for testing the ability of a normative economic model to predict competitor and retailer response. We first estimate the response parameters of a demand function for each brand from the period before value pricing was initiated. We then formulate a dynamic Manufacturer-Stackelberg game theoretic model that includes P&G, a national brand competitor, and a retailer. The model takes P&G's move as given and prescribes the actions that competitors and the retailer should take with respect to price and promotion. We test the predictive power of the normative model by substituting the estimated response parameters into the model to obtain prescriptions for each competitor and the retailer, and then seeing whether these prescriptions are related to the actual moves taken by competitors and the retailer. We find that our economic model, coupled with empirical estimates of its response parameters, has significant predictive power. It is more accurate for competitive response than for retailer response, but the predictive power is statistically significant in both cases. Covariates such as category advertising, category purchase cycle, and multi-market contact also predict competitive response, but our economic model-based prescription is the most important predictor. Overall, the results suggest that competitor and retailer response to a significant policy change by a major "player" is predictable and at least partially rational in its dependence on response parameters processed through a dynamic game theoretic model.