Moral Hazard
Moral Hazard
The term moral hazard describes a situation in which two or more parties voluntarily interact, but the value of this interaction to one or more of the parties can be adversely affected by actions the other party may take for personal benefit. This phenomenon is particularly troublesome in instances in which the affected party can neither (contractually) control nor perfectly monitor these actions (directly or indirectly).
While many examples exist, perhaps the most common and compelling ones come from the insurance industry. Consider, for example, theft insurance, in which the owner of some valuable items contracts with an insurance carrier that in the event of the theft of the property, the insurance company will pay the owner the entire value of the property. The insured party, in most instances, can carry out actions that will lower the probability that the property will be stolen—for example, installation of a security system, hiring guards to protect the property, installing better locks on doors and windows, and so on. Each of these actions is costly, but some of them are likely to nonetheless be worthwhile investments. Specifically, if the decrease in the expected loss from theft is greater than the cost of the action, then the investment is socially desirable (not taking into account the utility of the thief). However, if the lost property is fully covered by the insurance policy, the owner has little reason to expend these resources. What triggers the problem is the shifting of the risk from the party that can most efficiently protect the property to a party that cannot. If this problem cannot be overcome, the cost of the insurance policy will be such that the insurance company can cover its expected cost given that the desired precautions will not be taken, and there may even be market failure, whereby the insurance premium is so high that the parties cannot reach agreement.
The theoretical underpinnings of this problem are attributed largely to a series of articles written by J. A. Mirrlees in the early to mid-1970s, with one of the most influential not being actually published until 1999 (although completed in 1975). Recognition of the problem predates these writings, going back to Mark Pauly (1968), Kenneth Arrow (1971), and Richard Zeckhauser (1970), who was the first to model the phenomenon. The idea of moral hazard has been shown to have implications for a large variety of fields in economics, including banking, environmental economics, health economics, international trade and lending, regulation, and growth. Most central, however, is the development of optimal-contracts literature, which has its roots in the moral hazard literature.
To this end, consider, once again, the insurance problem. In some instances, steps can be taken, either contractual or other, to attempt to overcome the moral hazard problem. For instance, the insurance company may try to control the problem by refusing to insure the property unless some of the steps mentioned are taken, or the insurer can condition payment of a claim on proof that the steps were carried out. Unfortunately, such contracts are unlikely to guarantee optimal behavior on the insured party’s part. This is because: (1) theft can occur even if the desired steps are taken, so there is no (indirect) way to be sure that the desired precautions were not taken; and (2) it is unlikely that the company will be able to (directly) perfectly check the level of security given by the system, the number and quality of guards hired, or the quality of the locks installed, and even if it can, it is likely to face difficulties in proving these things in a court of law.
Alternatively, the insurance company may try to shift some of the risk back to the property owner by, for example, selling a policy with a large deductible or only paying a percentage of the loss, thus better aligning the interests of the insured party with those of the insurance company. This, however, lowers the value of the insurance policy, and does not allow for the pooling of risks that are at the base of the insurance industry. As another possibility, the insurance company could provide its own security (and charge for it), but the insurance company is almost certainly not as well situated as the insured party to provide such services or to ensure that they are carried out as desired. Thus, it seems almost inevitable that the moral hazard problem will persist in such markets.
The issue of optimal contracts has been well developed in principal-agent settings, of particular importance in the accounting literature. Consider a situation with two parties—a principal (the owner of a firm or manager) and an agent (worker). The agent is hired to carry out some task, and the level of success depends upon the level of effort invested by the agent. The agent prefers to exert himself as little as possible, but the principal desires that the agent put in a large amount of effort in order to maximize the principal’s expected return. The agent, left to his own, is likely to exert as little effort as he feels he can get away with, and the principal, therefore, is interested in finding a way to align the worker’s interests with his own. If the effort can be observed directly, either through observation of the agent’s working habits or by the fact that there is a one-to-one relationship between effort and output, then the principal can write an employment contract that will have the effect of causing the agent to choose to exert effort (assuming there are no other problems with this solution, such as a requirement to prove the effort level in court). However, if effort cannot be directly observed, and the outcome of the worker’s effort is uncertain, so that the output level can be low even in the face of great effort, then the principal is faced with a more difficult situation because the worker can always say that he exerted effort, but was unlucky.
In some circumstances, writing a contract that shifts the risk from the principal to the agent can eliminate this problem. If successful, such a contract has the effect of making the agent take direct consideration of the effects of his effort, and causes him to exert the optimal amount of effort. However, this is not always possible; if, for instance, the agent is risk-averse and the principal is risk-neutral, it may be the case that the worker will be unwilling to bear risk that is actuarially worthwhile for the principal.
There are those who feel that the importance of the moral hazard concern is overstated. Specifically, they feel that in instances in which the interaction between the principal and the agent is of a repeated game nature (long-term employment), this problem may disappear. Even if not, if reputation is important to the agent, he is unlikely to slack even when the principal cannot prove bad faith.
SEE ALSO Information, Asymmetric; Insurance; Insurance Industry; Principal-Agent Models
BIBLIOGRAPHY
Arrow, Kenneth J. 1971. The Economics of Moral Hazard: Further Comment. In Essays in the Theory of Risk Bearing. Amsterdam: North Holland.
Kreps, David M. 1990. A Course in Microeconomic Theory. Princeton, NJ: Princeton University Press.
Mirrlees, J. A. 1999. The Theory of Moral Hazard and Unobservable Behavior: Part I. Review of Economic Studies 6 (1): 3-21.
Pauly, Mark V. 1968. The Economics of Moral Hazard. American Economic Review 58: 531-537.
Zeckhauser, Richard. 1970. Medical Insurance: A Case Study of the Tradeoff Between Risk Spreading and Appropriate Incentives. Journal of Economic Theory 2 (1): 10-26.
Avi Weiss