In an effort to enhance the efficiency and effectiveness of higher education, some governments have redefined their funding relationships with universities as contracts rather than grants in order to develop stronger incentives for university performance. This paper explores this approach, as adopted in Australia, within a principal agent framework. It examines the conditions necessary for success in achieving government objectives as well as preserving university autonomy.Recent trends in public finances, international economic markets, and political currents, have led governments to focus on ways to enhance the efficiency and effectiveness of universities, in particular those which they control and finance.t It is the responsibility of the government to make sure that public funds are wisely used and, thus, it is the govemment's obligation to hold its institutions accountable for their performance. Redefining the funding relationship between the government and its institutions as a contract rather than a grant is one strategy to enhance performance which has not only generated a lot of interest, but which has been a popular policy choice. This paper explores the contractual approach to higher education funding from a principal agent framework. The objectives are threefold: (1) to reveal the theoretical underpinnings of the approach; (2) to highlight the various factors relevant to the design of contractual relations; and (3) to examine the likely effects of such a policy strategy. This paper seeks to accomplish this by providing an exposition of principal agent theory and its application to higher education and an examination of the recent Australian higher education reforms within the principal agent framework.
I. Agency theory and its application to higher educationThe principal agent model A cornerstone of the recent economics literature on contracts focuses on the principal agent problem 2. The principal contracts with an agent to act on the principal's behalf. A potential agency problem arises when the principal and the agent have different objective functions, and the principal lacks the information necessary to assess the performance of the agent. If the objective functions are