Please use this identifier to cite or link to this item:
Bell, David R.
Boztuæg, Yasemin
Year of Publication: 
Series/Report no.: 
SFB 373 Discussion Paper No. 2001,75
Humboldt University of Berlin, Interdisciplinary Research Project 373: Quantification and Simulation of Economic Processes, Berlin
The authors develop and test a model to study the in uence of inventory-on-hand and price-based reference points on the consumption rate of consumers. The model is motivated by recent theoretical and empirical research which suggests inventory pressure can cause consumers to increase consumption. A second stream of research shows that purchase behavior is affected by consumer expectations about product prices. To date, no study has developed a unified assessment of both the direct (via prices) and indirect (via inventory) effect of marketing activity on consumption. We introduce the concept of \inventory elasticity of consumption to represent the effect of inventory on consumption rates and propose a function that allows consump- tion to vary with time and level of inventory on hand. The consumption function also accounts for the effect of positive and negative deviations from category price expecta- tions. The model is estimated on eleven product categories. The inventory elasticity of consumption is highly significant in all categories and the elasticities range from 0.28 to 2.46. Some categories (e.g., butter, crackers, margarine, paper towels, soft drinks and sugar) are consumption inelastic while others (e.g., bathroom tissue, detergents, hot dogs, and ice cream) are consumption elastic with respect to inventory. While consumption rates in all categories are sensitive to negative deviations from the reference point (i.e., losses) the consumption rates of relatively \discretionary products (e.g., bacon and soft drinks) show the greatest slowdown. Overall, reference effects matter less than the inventory effect in driving exible consumption, but categories with greatest inventory effects also show the greatest reference effects. Implications for managers and researchers are discussed.
Reference Points
Choice Models
Persistent Identifier of the first edition: 
Document Type: 
Working Paper

Files in This Item:

Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.