18 October 2016

Contract theory - The Nobel Prize in economics, 2016

Bhaskar Dutta 

The Nobel Prize in economics for 2016 has gone to Oliver Hart, a British economist at Harvard, and Bengt Holmström, originally from Finland but since 1994 a professor at the Massachusetts Institute of Technology, for their work establishing contract theory as a "fertile field of basic research". Neither economist was a front runner this year, perhaps because their most influential work was completed roughly two decades ago. However, very few will question the Nobel committee's choice.

Contracts are ubiquitous in our daily lives because contractual arrangements have to be approved by the parties involved in transactions in a wide array of contexts, including employment, borrowing and lending, insurance and private provision of public services. In several of these cases, many subtle issues are involved in distinguishing "good" or "optimal" contracts from ones that are not so good. Like most theoretical research in economics, contract theory has been useful in understanding why or how existing institutions evolved as well as in designing better institutions. Formal contract theory has been shaped to a great extent by the contributions of Hart and Holmström. Although their most important work was done in the two decades following the late Seventies, their body of work remains just as important today. Indeed, it has become part of the DNA of contract theory.

Perhaps the most fundamental principle in designing contracts is the need to strike a proper balance between two competing objectives - the need to provide proper incentives versus the need to obtain the right allocation of risk. The trade-offs involved between these two objectives are apparent in any share-cropping contract between a tenant and the typically absentee landowner. The landlord cannot monitor the number of hours put in by the tenant. So, the possibility that the tenant may shirk his duties rules out the use of pure wage contracts. In other words, the need to provide proper incentives requires the contract to give the tenant a share of the output.

The strongest incentive to work would be provided by a pure rental contract in which the tenant pays the landlord a fixed sum of money and keeps the entire output. However, rental contracts are not observed in practice because they imply that the tenant bears all the risk associated with production. For instance, a crop failure can wipe out the poor tenant, whereas the richer landlord is better placed to bear the lion's share of risk. So, the optimal allocation of risk alone would imply the use of wage contracts in which the tenant bears no risk at all. The need to balance the two objectives results in output-sharing contracts. Interestingly, an equal-sharing arrangement, which is typically observed in West Bengal share-cropping contracts, is not necessarily the optimal arrangement.

Oliver Hart

Starting from 1979, Holmström wrote a series of papers analysing the nature of such incentive contracts in a variety of settings. He pointed out that payments should be based on all the outcomes that can provide information about a worker's level of effort. For instance, a senior manager's efforts do influence a firm's profit and hence share price. But, this does not mean that remuneration should depend only on the share price of the firm employing the manager - a firm may be doing well simply because of a booming economy or very favourable overall expectations, which have nothing to do with the specific firm in question. Holmström's analysis suggested that managerial remuneration should depend instead on the share price of the firm relative to that of others in the same industry. Unfortunately, this relatively simple lesson seems to have been ignored in actual practice, since it is common to provide bonuses that depend on firm-specific performance criteria. Holmström, in collaboration with Paul Milgrom (another theorist who is tipped to win the prize every year), later extended his 1979 analysis to incorporate the framework of an employee performing multiple tasks, which are again hard to monitor. In many such cases, optimal remuneration schemes should not depend too narrowly on performance in specific tasks that are easy to measure. For instance, if student performance measured by, say, their marks in annual examinations has a large weight in teachers' remuneration, then teachers will spend a lot of time teaching students to do well in examinations, and less time in teaching them to think independently or in promoting creativity.

Another area in which Holmström has made a seminal contribution has been incentive contracts in which workers form a team and produce output jointly. In such cases, if remuneration is based on average output or performance of the group as a whole, as in a workers' cooperative, then individual workers may "free ride" in the terminology of economists. In such cases, outside ownership and control may allow for more flexible and appropriate remuneration schemes.

Oliver Hart is best known for initiating work in the area of incomplete contracts. This term describes situations in which it is impossible to describe all the contingencies that may arise in the future. The key insight was that the best contract must in such cases specify who has the right to decide what to do when circumstances arise that have not been explicitly contracted upon. Of course, the right to decide then provides greater bargaining power to the party who has the right to take decisions. This then serves as a powerful incentive device.

The incomplete contracting paradigm has been applied to a variety of contexts, many of these being developed by Hart and his collaborators. For instance, if not all uses of an asset can be specified in advance - a typical case of incomplete contracts - then any contract negotiated in advance must leave some discretion over the use of the asset. The "owner" of the "firm" or asset is then the party to whom the residual rights of control have been allocated when the contract was written. The optimal allocation of property rights or governance structure is one that minimizes efficiency losses. This produces a theory of ownership and vertical integration as well as a theory of the firm. These principles have been used in discussions of the internal organization of firms and financial contracts, to mention just a few examples.

One application of the incomplete contracting framework that has some topical relevance in India is the issue of the private provision of public services. For instance, publicly operated enterprises such as hospitals are frequently criticized for wasting resources and running up huge operating costs. One suggestion sometimes made is that such enterprises should be handed over to private entrepreneurs who should not charge customers, but be adequately compensated by the State. Hart pointed out that there are two, often conflicting, forces at work in such situations - improvements to quality and cost reduction. If the government's main priority is to improve quality it should retain control, since private providers usually have too strong an incentive to cut costs, resulting in an unacceptable drop in quality. One example cited by Hart was the management of American prisons, some of which are run privately. Interestingly, the government of the United States of America now accepts that conditions in privately run prisons are significantly worse than those that are publicly owned and has decided to end their use.

The author is professor of economics, Ashoka University 

No comments: