Sales and Perishable Products

Abstract

This paper examines the role of sales (temporary price reductions) in the pricing of perishable products. When products can be stored, periodic sales are commonly explained using inventories: the ability to store lets consumers wait for better prices, and so they have a more elastic demand. When consumers differ in their ability to wait, firms periodically reduce prices to keep the price regularly high, thereby targeting the low prices to just the most elastic (patient) consumers. However, this explanation is not reasonable for perishable goods, since they cannot be stored. Using a large retail dataset, I show that a cyclic pattern of sales is a major feature of how perishable products are priced, which cannot be explained by product expiration. To explain this, I develop a dynamic model of loss leadership. In my model, consumers purchase baskets of goods which contain different mixtures of perishable and storable products. Shopping costs induce consumers to buy their entire basket at a single store. Because these baskets differ, firms can use them to discriminate between different groups of consumers. Specifically, firms want to attract perishable-buying consumers when they are also purchasing relatively many storable goods. In order to do this, they offer these consumers a better total price for their whole basket. It is optimal to do this by lowering just the price of the perishable good, since this targets the price reduction to the group the firm wishes to attract, keeping the total basket price high for other consumers. Firms time these price reductions to allow the target group of consumers to run down their inventory, which entails trading off present and future profit in an inter-temporal optimization problem. The link between storable good inventories and perishable products creates, in equilibrium, periodic sales on the perishable good. I test my model by linking the retail data to consumer choice data, a process which requires the development of a data-driven method of classifying prices into sales. The results validate the central prediction of the loss leadership model: when consumers buy perishables on sale they also buy more of other products, particularly storable ones, relative to their purchasing in non-sale periods. These findings highlight the role multi-product competition has on pricing dynamics, and rationalize an empirical finding which is difficult to explain with most models of sales.