An empirical investigation on causes and effects of the Bullwhip-effect: Evidence from the personal care sector

This paper analyses the empirical demand data for fast moving consumer goods to measure the Bullwhip-effect. The data consist of the sell-in from a large manufacturer to the retailers and the sell-out from a retailer to the consumers. Our findings show that the Bullwhip-effect can be substantial. Indeed, in more than 50% of the cases, the demand upstream (sell-in) is twice as variable as the demand downstream (sell-out). However, in other cases it can be negligible (and in one case, the demand upstream is slightly less variable than the demand downstream). So, while the Bullwhip-effect can be very large, it need not be so. Finally, we attempted to delve into the dynamics that generate a significant Bullwhip-effect and discovered that among the various causes that create the Bullwhip-effect discussed by Lee et al. (1997a), price fluctuations and forward buys driven by sales targets play a decisive role. We show that the flatter the final consumer demand (sell-out), the more room there is for the retailers to forward buy in order to take advantage of the deals offered by the manufacturer's sales force toward the end of the sales period. On the contrary, the more variable the final consumer demand, the less willing the retailer is to make a risky inventory investment and, thus, the more the retailer's orders closely follow consumer demand. This result at the product family level is very consistent with the results that Cachon et al. (2007) obtained at the company/industry level.