The objective of this paper is to propose and empirically test a potential mechanism for how consumers form reference prices. The proposed peak‐end rule of reference price formation says that reference prices are formed as a weighted average of the highest observed price and the most recent price.
The authors argue why the peak‐end rule observed in satisfaction contexts may also apply to the process by which consumers form reference prices. They then test the proposed peak‐end rule using IRI scanner panel data for decaffeinated coffee.
Fit and predictive validity of a choice model improves when a reference price term based on the peak‐end rule is added. While the most recent price has a greater impact on reference price, the effect of the highest observed peak price is also significant, managerially and statistically.
The study provides evidence for a novel and behaviorally plausible reference price formation process.
Temporarily charging a high price has a longer‐lasting effect on reference price than would be suggested by other reference price models, which typically involve a quickly decaying lag effect. Temporarily charging a very high price to restore the reference price may therefore be a useful pricing tactic.
While the peak‐end rule is amply supported as a mechanism by which consumers form global satisfaction judgements, its application to reference price formation is novel, and has some potentially useful implications.
