Stemming the Big Crash: A Triple Hazard Analysis of Price and Sales Crashes of New Products

This paper proposes a novel triple hazard model that can statistically untangle three important sources of variation in the timing of events that may arise from i) observed covariates, ii) unobserved and correlated heterogeneity, and iii) causality between events. We apply the model to analyze the relationship between the sales crash, price crash, and sales recovery of new products. A sales crash is a significant and permanent drop in the sales of a new product whereas the sales recovery is a modest sales peak that follows the crash. Similarly, the price crash is a deep and permanent reduction in the price of a new product. We find that the depth of the price crash is close to 23% while the depth of the sales crash is close to 60%. In addition, we find that the occurrence of the price crash causally and significantly increases the hazard of a sales crash whereas the occurrence of a sales crash causally and significantly increases the hazard of a price crash. These causal effects are asymmetric and the latter is significantly stronger. The findings and model have important implications for managers of new products.

This event is organised by the Econometric Institute.
Twitter: @MetricsSeminars