An Experimental Comparison of Transfer Pricing Methods Under High and Low Private Information

Abstract

Transfer pricing is an important managerial decision in a decentralized firm. When there is no competitive market for an internally traded good or service, a fair transfer price must be established for the buying and selling divisions in the firm. Analytical research has proposed different transfer pricing methods for different information circumstances in a firm. For example, when information asymmetry is high, the information advantage of the negotiated method generates higher firm profits that the cost-based method. But, when information asymmetry is low, the cost-based method provides more firm profits because there is no information advantage through negotiation. Yet so far little empirical evidence exists that identifies when a particular scheme performs better than others in practice. The main challenge remains that it is difficult to apply analytical prescriptions without proper considerations of managerial behavior issues. Behavioral studies on transfer pricing indicate that perceptions of fairness influences manager's decisions, willingness to trade and thus the firm's performance. This study conducts a laboratory experiment based on the results of analytical and behavioral research to compare three transfer-pricing methods (cost-based equal profit distribution, cost-based equitable cost distribution, negotiated). We experimentally investigate which method provides higher perceived fairness, more truthful cost reporting, willingness to trade and thus higher firm profits under conditions of high/low private information asymmetry. We first test the analytical prescriptions regarding whether high/low information asymmetry is a key factor in determining the preferred transfer pricing methods. We then further test our hypotheses with considerations of behavioral factors such as perceived fairness, truthful reporting, and willingness to trade in transfer pricing decisions. Overall, we find that when information asymmetry is high, firm profit is similar across transfer pricing methods, inconsistent with prior analytical studies. Conversely, when information asymmetry was low, both cost-based methods provided more firm profit than the negotiated transfer pricing method, consistent with prior analytical studies. In addition, firm profits were higher overall when information asymmetry was high than when it was low. Further, we found that perceptions of fairness significantly influenced participant's truthful reporting, willingness to trade, and firm profit. Finally, perceptions of fairness was a significant variable when perceptions of fairness, truthful reporting, willingness to trade and transfer pricing method was regressed on firm profit, indicating that perceptions of fairness had the most significant influence on firm profit in this study.