Personal Selling Elasticities: A Meta-Analysis

This article presents a meta-analysis of prior econometric estimates of personal selling elasticity—that is, the ratio of the percentage change in an objective, ratio-scaled measure of sales output (e.g., dollar or unit purchases) to the corresponding percentage change in an objective, ratio-scaled measure of personal selling input (e.g., dollar expenditures). The authors conduct a meta-analysis of 506 personal selling elasticity estimates drawn from analyses of 88 empirical data sets across 75 previous articles. They find a mean estimate of current-period personal selling elasticity of .34. They also find that elasticity estimates are higher for early life-cycle-stage offerings, higher from studies set in Europe than from those set in the United States, and smaller in more recent years. In addition, elasticity estimates are affected significantly by analysts’ use of relative rather than absolute sales output measures, by cross-sectional rather than panel data, by omission of promotions, by lagged effects, by marketing interaction effects, and by the neglect of endogeneity in model estimation. The method bias–corrected mean personal selling elasticity is approximately .31. The authors discuss the implications of their results for sales managers and researchers.