Supplier Performance Management: A Behavioral Study

Abstract

This dissertation studies supplier performance management and explores practical approaches to improve supplier performance. We use the theoretical model prediction as a benchmark and we test our analytical findings using behavioral experiment. The result typically is a strategy or policy recommendation to OEM in order to induce desirable performance of its suppliers. This dissertation analyzes three important problems in supplier performance management in three essays, Chapter 2 to 4. Below, we briefly summarize each problem: In the first essay, Chapter 2, we study the coordination problem in a setting with multiple contractors working on a project. Specifically, we analyze the risk-sharing contract in which the payment to the contractors depending on the minimum effort among the contractors. Following the conventional wisdom of minimum effort games, we show that the contractors may fail to coordinate their efforts. For a project with parallel tasks, there exist multiple Pareto-ranked equilibria, where all contractors exert the same efforts, and the lowest equilibrium efforts are observed when the contractors play the secure equilibrium. To mitigate this coordination failure, viii we propose an information feedback policy, and show that the contractors’ efforts in the secure equilibrium increase in the information feedback frequency. Therefore, the OEM may induce the contractors to increase their effort by providing feedback with frequency that increases linearly in the number of contractors. To test our analytical findings, we conducted a behavioral experiment that varies the feedback frequency and the number of contractors. Our experimental result confirms that a higher information feedback frequency leads to a better project outcome and a higher contractors’ payoff. In addition, coordination among contractors gets more difficult with more contractors working on the project and in turn results in a lower contractors’ payoff. In the second essay, Chapter 3, we considered supplier’s scorecard system which is a tool for manufacturers to track supplier performance. We investigate the effectiveness of two approaches for a manufacturer to incentivize desirable performance of suppliers based on the evaluation of their scorecard, the Absolute and Relative approaches. Under the Absolute approach, the manufacturer incentivizes the supplier if the supplier achieves a prespecified targeted score. Under the Relative approach, the manufacturer incentivizes suppliers based on the suppliers’ scorecard ranking in the supplier base. We consider a two-period supplier-manufacturer contractual agreement where the manufacturer evaluates and reviews supplier performance (acceptable or unacceptable) at each period. Supplier performance outcome at each period is determined by its binary effort decision (high or low) where acceptable performance is more likely to be resulted from exerting the costly high effort than from exerting low effort. We derive the optimal targeted score and the optimal reward scheme that results in the highest supplier performance under the Absolute and Relative approach respectively. We derive that under the Relative approach allocating the whole reward pie to the supplier with the highest overall score maximizes supplier performance. Comparing the suppliers’ resultant performance under the two approaches, we characterize conditions on which each approach is preferable. To test our theoretical findings, we conducted a human-subject experiment that varies the reward as well as the incentive approach. We observe that subjects over-provide effort compared to the theoretical model benchmark under both approaches. We further observe that subjects exert more effort under the Relative approach compared to the Absolute approach indicating that in the Relative approach, competition motivates suppliers to provide more effort. The result shows that the manufacturer incurs less cost for each unit of effort it receives from the supplier under the Relative approach compared to the Absolute approach. Finally, in the third essay, Chapter 4, we study a long-term contractual agreement following the Absolute approach. We derive that supplier’s optimal effort decision at each period follows a threshold strategy where exerting high effort is optimal when its score is between a lower and upper threshold, and exerting low effort is optimal otherwise. We experimentally test the theoretical findings using human-subject in the role of supplier. The result shows that subjects overexert effort compared to the theoretical model prediction. The data are explained well by a behavioral model that incorporates three behavioral concepts; joy of wining, perceived probability and bounded rationality. The result suggests that subjects gain psychological utility as they reach the target and win the reward. Second, subjects’ perceived probability follows a function that is shaped like an inverted "S". We observe that subjects overweight small probability and underweight large probability. Finally, the behavioral model estimation shows that subjects make noisy decisions. The result suggests that likelihood of a decision increases in the expected utility of the decision.