Does Forecast-Accuracy-Based Allocation Induce Customers to Share Truthful Order Forecasts? Pelin Pekg € un* Moore School of Business, University of South Carolina, Columbia, South Carolina 29208, USA, pelin.pekgun@moore.sc.edu Minseok Park Perdue School of Business, Salisbury University, Salisbury, Maryland 21801, USA, mspark@salisbury.edu Pınar Keskinocak H. Milton Stewart School of Industrial and Systems Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, USA, pinar@isye.gatech.edu Mani Janakiram Intel Corporation, Chandler, Arizona 85226, USA, mani.janakiram@intel.com T hrough a behavioral study, we investigate buyers’ strategic order forecasting behavior in a multi-period setting under a forecast accuracy-based allocation policy, where the supplier allocates (proportionally) more inventory to the buyer with the better order forecast accuracy in case of scarce supply. We developed an interactive game that simulates a supply chain in which one supplier sells a key component to two buyers, who in turn sell to consumers. In each period, buyers share forecasts of future orders with their supplier. The participants in the game play the role of a buyer, while the sup- plier is automated. Our experimental findings suggest that rewarding forecast accuracy in allocating inventory can signifi- cantly improve the order forecast accuracy of the buyers and reduce their forecast inflation behavior. Interestingly, even without communication of the policy, buyers learn over time that more accurate forecasts lead to better service from their supplier and improve their order forecast accuracy. Key words: soft orders; forecast sharing; forecast inflation; capacity allocation; behavioral experiments History: Received: February 2017; Accepted: May 2019 by Elena Katok, after 2 revisions. 1. Introduction One of the key elements of supply chain collaboration is the sharing of forecast information between supply chain partners. In many industries, buyers (who may sell to end consumers or other buyers) often submit forecasts for future orders to their supplier to help with capacity planning. Sharing order forecasts (also referred to as “soft orders”) with the supplier can also help reduce the order variability and the bullwhip effect (Chen and Lee 2009). However, buyers may update the soft order multiple times until the firm order is placed, and as a result of this “forecast volatility” (Terwiesch et al. 2005), the supplier takes the risk of cancellation or holding costs, while the buyers take the risk of the delay costs (Cohen et al. 2003). Since soft orders represent the intent of pur- chasing and are often not legally binding, buyers tend to submit inflated order forecasts to secure capacity from their supplier. The supplier, knowing this fore- cast inflation behavior of the buyers, may discount or even ignore the forecasts received, leading to lower service levels and longer delivery times, which may further drive the buyers to inflate their forecasts (Ter- wiesch et al. 2005). Capital-intensive industries, such as semiconductor and aerospace manufacturing, par- ticularly suffer from this ( € Ozer et al. 2011). While buyers of semiconductor equipment expect a high degree of responsiveness, the high value and cus- tomized nature of such equipment make it risky for the equipment supplier to start production early, or even keep finished goods inventory. Victims of inflated forecasts include the networking giant Cisco Systems who had to write off $2.2 billion worth of inventory in 2001 due to its failure to recognize dupli- cate orders from its customers (Bloomberg 2002). An important issue is how the supplier should allo- cate his capacity when the total order quantity from buyers exceeds the available capacity (Krishnan et al. 2007). Certain allocation policies can motivate a buyer to submit orders higher than their optimal or realistic levels so as to compete with other buyers for scarce Please Cite this article in press as: Pekg€ un, Pelin, et al. Does Forecast-Accuracy Based Allocation Induce Customers to Share Truthful Order Forecasts? Production and Operations Management (2019), https://doi.org/10.1111/poms.13066 Vol. 0, No. 0, xxxx–xxxx 2019, pp. 1–14 DOI 10.1111/poms.13066 ISSN 1059-1478|EISSN 1937-5956|19|00|0001 © 2019 Production and Operations Management Society