Moral hazard

Moral hazard is in Economics some non observable actions or behaviours that can lead to assymetric information between market participants. It is usually associated with situations where one side of the market can not see the action of the other (see Varian (1990)).

It has been first been analyzied in the insurance industry. Kenneth Arrow (1963) had first introduced Moral Hazard in the litterature in a seminal paper. Since then, many researchers have tried to defined it. Buchanan (1964) defines it as "every deviation from correct human behavior that may pose a problem for an insurer".

In the insurance context, it can be defined as the effect of having an insurance and insured's incentives to reduce expected losses.It arise when one market participant do not bear all the full consequence of its actions. Therefore, he will be more inclined to act with more risk or less care than otherwise.

For example, if a car accident cost a person $10.000 but the insurer pays $9.500, the insured person would have less incentive to avoid the accident.

Moral hazard in Insurance

Basically, when completly insured, any benefit of additional precautions will accrue to the insurer rather than to the policyholder. Since insurers understand well that insurance reduces the policyholder's incentives to prevent losses, they have responded to this problem in multiple ways. Moral hazard arise in insurance as one side of the transaction in not able to correctly monitor the other side (or that latter is trying to hide some of his actions).

In most cases, insurance will not offer full coverage. Instead, only some part of the losses will be covered by insurers. To reduce the moral hazard, insured will have to bear some risk.

Another principle used in the insurance industry to mitigate the moral hazard is to rate future premium according to:

• the number and gravity of losses already paid by the insurance

• and to the policyholder's experience (seniority, habits, etc,...).

Holmström (1979) and Shavell (1979) have first studied the impact of moral hazard in insurance. They demontrate that setting deductibles, or a limitation of coverage for small losses is optimal if policyholder's effort can mitigate loss probability.

Similarly, insurers can setup co-payment clauses un which the policyholder will have to pick up some percentage of the loss when there is a claim.

It had been distinguished by Didone and St-Michel (1991) that two different types of moral hazard exist in the insurance context:

• Incentives to not have optimal precaution level to avoid losses (ex-ante)
• Incentives to exagerate the loss level when incurred (ex-post).

In that framework, one could consider insurance fraud as result of extrem ex-post moral hazard.

For insurers, it is very difficult price premium correctly as too high premiums to cover client with risky behaviour would hurt less-risky client. Hence, nobody would buy that insurance.

Moral hazard in Finance

=> centrals banks with deposit insurance, IMF and bailout, LTCM,.. to be developped.

References

Arrow, K.J. (1963), «Uncertainty and the Welfare Economics of Medical Care». American Economic Review, 53, December, p. 941-973.

Buchanan, J. (1964), "The Inconsistencies of the National Health Service", Institute of Economic Affairs (Occasional papers n°7), London.

Dionne, G. et P. St-Michel (1991), «Workers compensation and moral hazard». Review of Economics and statistics 83(2), p. 236-244

Holmstrom, B. (1979), "Moral Hazard and Observability", Bell Journal of Economics, 10, p. 74-91

Shavell, S. (1979), "Risk-sharing and Incentives in the Principal-Agent Relationship" , Bell Journal of Economics 10, p. 55-73.

Varian, H. R.(1990), "Intermediate Microeconomics: A Modern Approach" , 2nd ed, W W Norton and Co., New York.