How Biased Household Inventory Estimates Distort Shopping and Storage Decisions

Abstract

The authors develop a model of how consumers estimate the level of product inventory in their households. Two laboratory experiments and two field studies involving 29 product categories show that (1) consumers anchor their estimates on their average inventory and fail to adjust sufficiently; (2) adjustments follow an inelastic psychophysical power function, leading to overestimations of low levels of inventory and underestimations of high levels; and (3) adjustments are more elastic and, thus, more accurate when inventory is salient. Contrary to the assumptions of practitioners and academic modelers, these inventory estimates, not actual inventory levels, drive subsequent purchase incidence. Simulation results further show that biased estimates increase overstocking and spoilage among stockout-averse consumers but increase stockouts and unmet demand among overstocking-averse consumers. By predicting the magnitude, not just the direction, of estimation biases, the model and the results offer new insights into accelerating the consumption of healthy foods and improving the targeting of stockpiling- inducing sales promotions.