The Predictive Power of Abnormal Inventory Growth: Application to Earnings Forecasting for Retailers

Abstract

In this paper we test the predictive power of abnormal inventory growth to forecast retailers’ earnings. We demonstrate an inverted-U relationship between abnormal inventory growth and one-year ahead earnings per share for retailers. We find this relationship to be robust to different measures of abnormal inventory growth obtained from operations management literature. Our results also show that equity analysts do not fully incorporate the information contained in abnormal inventory growth in their earnings forecasts resulting in systematic biases in their earnings’ forecasts. We show that incorporating this information in analysts’ forecasts would improve their forecast accuracy. This improvement can be as much as 15.08% for overinventoried retailers that are identified based on previous year’s abnormal inventory growth.

Keywords: Retailing, Econometric Analysis, OM - Accounting Interface

JEL Classification: M43