| Since the 20th century 70s, a revolutionary change has been taken place in the Western principal-agent theory. The theory of information, contract and incentive have been drawn into the research of the principal-agent theory. Researches focused on the simple description of the phenomenon of separation between ownership and business in the past. Now it turns to researching how the owners to constrain and incentive the operators. Basing on the previous studies, it talks about the principal-agent theory coming from the separation between ownership and management in the modern enterprise system. It puts forward the concept of performance measure congruity, and then discusses the optimal incentive contract.Firstly, the paper talks about the background and Significance, and proposes the performance measure congruity as the starting point. It expounds the related theories of Information Economics and principal-agent, in order to make clear the connotation and significance of incentive contracts, and clarify the status and role of the performance evaluation, laying a theoretical basis for this paper.Secondly, it studies the optimal design of incentive contract basing on a balanced scorecard. This firstly provides a formal definition of scorecard balance, and compare it to the property of congruity as defined in the multitask agency setting. It is shown that balance is a sufficient condition for performance measure congruity, and first best can be achieved if the agent is risk neutral. It is shown that under complete verifiability, a properly designed balanced scorecard is capable of perfectly aligning the interests of owners and employees by means of an explicit contract. This result, however, builds on the assumption that all scorecard measures can be used in an explicit contract. Since contractibility could be in doubt for nonfinancial performance measures, I also investigate whether incongruity of the verifiable measures may be compensated for by using a subjective performance evaluation. To this purpose, I adapt the model of relational contracts proposed by Baker, Gibbons, and Murphy (1994) to the multitask agency setting. I extend their work by considering multidimensional performance measures for which the design of an aggregate performance score becomes part of the contracting issue. It emerges that congruity of the contractible scorecard measures constrains a purely implicit incentive contract, but the first-best solution may still be obtained through a combination of formal and relational contracts. Furthermore, a purely explicit contract in most cases can be improved by incorporating subjective rewards.Thirdly, it sets a virtual enterprise and analysis this example by the use of Balanced Scorecard.Reflecting the difference of optimization among formal contracts, relational contracts and hybrid contracts. Make the Theory clarity.Finally, the paper summarizes the conclusions and research deficiencies, and look into the distance of future research. |