Altruistic Dynamic Pricing with Customer Regret

A model is considered where firms internalize the regret costs that consumers experience when they see an unexpected price change. Regret costs are assumed to be increasing in the size of price changes and this can explain why the size of price increases is less sensitive to inflation than in models with fixed costs of changing prices. The latter predict unrealistically large responses of price changes to inflation for firms that do not frequently reduce their prices. Adjustment costs that depend on the size of price changes also raise the variability on the size of price increases. Lastly, it is argued that the common practice of announcing price increases in advance is much easier to rationalize with regret concerns by consumers than with more standard approaches to price rigidity. ∗Harvard Business School, Soldiers Field, Boston, MA 02163, [email protected]. The observation of a price change can trigger consumer regret. If the good is storable and the price rises, people regret not having purchased earlier, while they regret not having waited when the price declines. Even when the good is not storable, a price increase can trigger regret if individuals were savoring anticipatory utility before the purchase. These anticipations might make it difficult for consumers not to buy, so they experience regret at having indulged in these anticipations earlier. The purpose of this paper is to study how firms should change their prices if they seek to act as if they empathized with these regret costs of their consumers. While it does not directly absorb scarce physical resources, the regret experienced as a result of a price change is a cost, and the analysis is thus conducted as a comparison of regret costs with the fixed costs of price adjustment postulated in the influential papers of Sheshinski and Weiss (1977) and Golosov and Lucas(2007). There are two ways in which regret costs can be expected to differ from simple fixed costs. The first is that these costs ought to depend on the size of price changes, with regret presumably being larger when prices are changed by larger amounts. The second is that these costs can be reduced if people are told about future price changes in advance. Advance warning can reduce consumer regret both by leading people to change the timing of their purchases and by engaging in anticipatory utility from purchases only when the consumer wants to buy at the actual price. These two simple and intuitive properties of regret have several implications for the pricing by firms that act as if they empathized with their consumers’ sentiments. As I note, these implications appear to realistic relative to the implications of fixed costs of changing prices. The dependence of regret costs on the size of price adjustment makes firms less willing to institute large price changes. This matters in two contexts. The first is the effect of inflation on the size of price changes. In the Sheshinki and Weiss (1977) model, an increase in inflation leads firms to post substantially larger price increases whenever they do decide to raise their price. In practice, however, several papers have shown that the actual size of price increases does not rise substantially when inflation rises. The lack of dependence of price changes on inflation is visible already in the early work on magazine prices by