This article is based the research of Kelley Assistant Professor of Business Economics and Public Policy, Boyoung Seo. This work was co-authored by R. Andrew Butters, Daniel W. Sacks (University of Wisconsin).
Introduction
It makes intuitive sense that product consumption fluctuates seasonally. Items like ice cream, hot dogs, beer, and canned soup are obviously in much higher demand at specific times during the year.
The law of supply and demand dictates that prices will change when either the supply of products or the demand for them changes. So when the demand increases for a product seasonally, we would expect prices for those items to increase too.
But what’s surprising—and actually counterintuitive—is that prices fall slightly, or remain stagnant, for products during seasonal demand peaks. Are retailers not maximizing their profits?
This puzzling phenomenon is called countercyclical pricing. Economists have long speculated about the reason for this countercyclical pricing pattern. Many answers lean on exclusively supply-side, such as a decrease in wholesale price, or cross-category explanations like “loss leading,” when retailers sell a product at an unprofitable price to attract new customers or sell additional products. Some suggest that firms are making mistakes by not setting their prices at the optimal level.
But are they really behaving suboptimally? Could any demand-side dynamics play a role? Could seasonal fluctuations in demand incentivize retailers to price on a more elastic part of the demand curve? Could brand preferences shift seasonally? Does the composition of shoppers change throughout the year? Could seasonal shoppers exhibit more price sensitivity?
To answer these questions, researchers from Indiana University and the University of Wisconsin designed a study to explain the pattern of decreased pricing during peak demand periods using data about widely sold products spanning dozens of categories.
Statement of Problem
The authors wanted to understand the prevalence of seasonal fluctuations in demand across product categories for typical retailers. And most importantly, they wanted to use data to explain the pattern of decreased pricing during periods of peak demand.
Data Sources
The authors used three main sources of data from NielsenIQ Datasets for their research. First, they gathered retail scanner data for nearly 1,500 widely available products that are sold in roughly 25,000 food, drug, and mass merchandise retailers in the contiguous U.S. This data details prices and revenue for each product with positive revenue for the study period (2006-2014), including store-level information on promotional activities.
Next, the authors looked at data recording household-trip purchasing patterns for over 150,000 households across the U.S for the study period. This consumer panel data includes each shopping trip, the store visited, the prices and quantities for each product purchased, and whether the consumer perceived the purchase as a deal.
Last, they gathered information on wholesale prices and promotional activities, as well as county-level unemployment rates from the Bureau of Labor Statistics, which were used as a benchmark to assess observed seasonal fluctuations.
Analytic Techniques
To measure seasonality in demand, the authors identified month-to-month variations in quantities of products sold and measured the difference between the maximum and minimum quantity each month. They also broke down household-level seasonal fluctuations into whether each household purchased in the category, and if so, how much.
Using a seasonal pricing model that could be applied to a wide variety of product categories, the authors compared price changes from the month of trough demand to the month of peak demand. A negative price change would mean that prices fell when demand peaked, indicating countercyclical pricing.
For their analysis, the authors excluded all weeks that include a major holiday, including New Years, the Superbowl, Memorial Day, July 4th, Labor Day, Thanksgiving, and Christmas. These holidays are the time periods where demand for all grocery products increase. Therefore, they are excluded when measuring the seasonality of a specific product category.
Results
Most products—in obvious categories like soup and frozen novelties, but also in less obvious categories like cookies—experience large seasonal fluctuations in demand. Seasonality in demand is largely driven by extensive margin changes in the number of households more likely to purchase a product in the category. In fact, seasonal changes increase the likelihood that a household will make a purchase in a product category, which accounts for nearly 90 percent of the seasonal fluctuations in quantity on average.
The authors also found that for the majority of seasonal product categories, demand becomes more price-elastic during seasonal demand peaks. In other words, the average market elasticity also fluctuates over the seasons. Shoppers that exclusively shop during the peak month of demand tend to purchase products on promotional price discounts. To cater to these seasonal shoppers, firms have incentives to lower the prices by offering promotional discounts, generating countercyclical pricing.
When it comes to seasonality in pricing, the authors found that countercyclical pricing is common, but the changes in prices are substantially smaller in magnitude than the seasonal fluctuations in demand they coincide with. Roughly two thirds of seasonal product categories exhibit slight price decreases during demand peaks, which are driven by a combination of more frequent sales and lower base prices.
Product categories such as hot dogs, canned soup, and ice cream not only have some of the largest seasonal fluctuations in demand, but they also experience the largest declines in prices from trough to peak periods of demand.
The statistical fact that demand fluctuates a lot more than price is perfectly consistent with standard economic theory of firms’ profit maximization. In theory, profit maximizing firms set prices according to elasticities. The authors’ findings show that the small magnitude of price fluctuation is likely to be a result of a small fluctuation in elasticities over the seasons.
These patterns suggest that countercyclical pricing can partly, if not all, be accounted for by a demand-side explanation: Seasonal changes in the composition of consumers shopping in a category result in demand being more elastic during demand peaks.
Supply-side explanations are rejected because wholesale costs do not move countercyclically for most products. Brand preferences do shift seasonally but it does not explain the countercyclicality in prices. Loss-leading explanation is not required to explain the countercyclical pricing but cannot be entirely ruled out when it comes to pricing patterns observed around major holidays.
Business Implications
There has long been debate in industrial organization about the sources of countercyclical pricing around seasonal demand peaks. By studying a broad and comprehensive set of categories and retailers, the authors established that seasonal fluctuations in demand and countercyclical pricing are pervasive, economically important, and not exclusive to a select set of product categories, holidays, or coordinated advertising activities. These fluctuations typically arise from a changing composition of households shopping in a category and often coincide with demand becoming more elastic as it peaks.
Although countercyclical pricing may appear at first, it is consistent with profit maximizing behavior. For business decision makers, the results suggest that countercyclical pricing is not necessarily suboptimal if they face more elastic consumer pool during peak demand.
The results also establish a benchmark for the degree of seasonal demand fluctuations and the extent of countercyclical pricing—consistent with the macroeconomic theory that suggests retailers use temporary discounts to price discriminate between low- and high-price elasticity consumers.
Importantly, seasonal fluctuations in demand are a central part of the retail sector and likely play an economically significant role in the effectiveness of retail pricing decisions.
Leave a Reply