How Retailers Determine Which Products Should Go on Sale: Evidence From Store-Level Data

Recent theoretical research on retail pricing dynamics provides an explanation of why retailers periodically put items on sale, even when their costs are unchanged. The authors extend this research to show that more popular items (i.e., those that appeal to a wide range of consumers) are more likely to go on sale. One implication of the proposed model is that a good is more likely to be on sale when demand for the good is at its season peak (e.g., eggs at Easter). This implication is tested using store-level retail price data, and the prediction is borne out for the categories of goods that are examined. Additional tests also support the premise that popularity and frequency of sales are positively related.