Executive Summary
The importance of pricing decisions for firms has fueled an extensive stream of research on price elasticities. In an influential meta-analytical study, Tellis (1988) summarized price elasticity research findings until 1986. However, empirical generalizations on price elasticity require modifications because of important changes since that time in market characteristics (i.e., features of brands, product categories, and economic conditions) and research methodology that assesses price elasticity. Therefore, the authors present new empirical generalizations of price elasticity and its determinants based on a significantly expanded data set, new predictors, and updated methodology. Across a set of 1851 price elasticities based on 81 studies published between 1961 and 2004, the average price elasticity is -2.62.
This meta-analysis reveals several key suggestions for managers who set prices. A particularly salient finding is that during the past four decades, sales elasticities have significantly increased in magnitude, whereas share and choice elasticities have remained fairly constant. This indicates that primary demand elasticities are increasing (i.e., consumers seem to base their timing and quantity decisions increasingly on price promotions). Managers may not appreciate the implied deal-to-deal buying behavior, because this suggests that products are decreasingly sold at full margin. An equally relevant finding for managers is that price elasticities are the strongest in the growth stage of product categories, both for durables and for groceries. This suggests that for introductions to a new category, a penetration pricing (i.e., low to high) is more effective than a skimming price strategy (i.e., high to low). An important null result is the nonsignificant effect of brand ownership on price elasticities. This implies that manufacturer brands are not necessarily more differentiated than private labels if brand differentiation is operationalized as the own-brand price elasticity.
The economic environment in which the manager operates has ramifications for pricing as well. If a country’s inflation rate is high, demand becomes more price elastic in the short run. Thus, a brand may gain additional sales or share if the item is promoted or if it keeps its regular price increases at a pace that is low relative to the inflation rate. The effect of economic growth rate on elasticities is not significant. This implies that during economic booms, managers should not believe that price increases are less harmful than they are during recessions. Finally, the absence of significant effects of the region dummies (i.e., North America, Europe, and Japan/Australia/New Zealand) on price elasticities may provide an argument for price strategies that are similar in developed countries across the globe.
To some extent, the results of this meta-analysis are reassuring for researchers because several factors that pertain to the data set and model options (i.e., data source, temporal aggregation, functional form, estimation method, and heterogeneity) do not have a significant effect on price elasticity estimates. Yet this is not the case for all factors, and the analysis identifies important potential biasing factors. Across all determinants in the study, the authors find that accommodating price endogeneity has the strongest (magnitude-increasing) impact on price elasticities. Therefore, the authors recommend that researchers at least test for price endogeneity before estimating price elasticities. Furthermore, in addition to a price variable, researchers should aim to include advertising and sales promotions as predictor variables in response models because omitting the variables leads to a strong omitted variable bias in the price elasticity estimate. It is also essential to clarify the exact definition of price that is used. Preferably, price is decomposed into regular price and price index (actual/regular price); this enable the disentanglement of the short- and long-term effects of both components.
Biography
Tammo H.A. Bijmolt is Professor of Marketing at the University of Groningen. He received his Master of Science degree and his doctoral degree in Economics (cum laude) from the University of Groningen and subsequently became a professor at Tilburg University. His research interests include multidimensional scaling and meta-analysis. His work has appeared in Journal of Marketing Research, Journal of Consumer Research, and International Journal of Research in Marketing, among others. He is currently working on the effects of loyalty programs on consumer behavior.
Harald J. van Heerde is Professor of Marketing at Tilburg University. He holds an Master of Science in Econometrics and a doctoral degree (cum laude) from the University of Groningen. His work has appeared in Journal of Marketing Research, Marketing Science, Quantitative Marketing and Economics, and International Journal of Forecasting. His research has addressed building econometric models in various substantive domains, such as sales promotions, pricing, and new product introductions, and he is branching out into new areas, such as building brand equity, loyalty programs, product crises, and price wars.
Rik G.M. Pieters is Professor of Marketing at Tilburg University. He holds a Master of Science in Psychology from TillburgUniversity and a doctoral degree (sin laude) from the University of Leiden. He is interested in the coordination of consumer behavior by attention and emotion. His research deals with advertising effectiveness, eye tracking, social influence, and specific emotions. His work has appeared in Journal of Marketing Research, Journal of Consumer Research, Journal of Marketing, Journal of Personality and Social Psychology, and PINT Nieuws.
J Marketing Research, Volume 42, Number 2, May 2005
View Table of Contents.