05/14/2002

FAIRFAX, Va.

THE BUSH administration and Congress are ironing out the details of mental-health-parity legislation. This legislation would require health-insurance plans to cover treatments for mental illness on the same terms as treatments for physical illness.

While this sounds like a good idea, policy-makers are ignoring what effect such legislation will have on behavior.

For example, 34 states with mental-health-parity laws on the books by 2000 included treatment for drug and alcohol abuse in their mandates. Examining the per-capita alcohol consumption in the states suggests that introduction of the parity laws led to a significant increase in the amount of alcohol consumed in the state. This result shows up even when other factors are controlled for, such as national trends, changing demographics, and state-specific drinking rates before introduction of the parity laws.

That is, individuals in those states where mental-health parity laws were passed consumed more alcohol after the laws went into effect than they had before the laws' enactment. In effect, individuals recognized that they would no longer face the full future costs of their drinking, leading them to drink more, much as they would buy more candy bars or compact discs if their prices declined.

This response is perfectly rational for the individual, but it leads to higher costs for those left picking up the tab. Supporters of this legislation, or even the detractors, have probably not considered this indirect effect of mandating insurance coverage for behaviorally influenced mental illnesses.

Because of this incentive effect, any estimates being considered in the current debate substantially underestimate the true costs of national mental-health-parity legislation from the perspective of firms who will pay the insurance bill, as well as from a public-health perspective.

Insurance analysts have known for quite some time that insurance coverage induces what they call "moral hazard." From an economic view, moral hazard simply captures the notion that people respond to incentives. If the costs that an individual faces for a particular activity decline, the individual will be more likely to engage in that activity. If someone is insured against the costs of doing something, he will be more likely to do it.

In the context of mental-health parity, with guaranteed coverage of treatments for those illnesses over which individuals have some control, people will be less likely to avoid behaviors that increase their propensity to suffer from the illness. Not only will this moral hazard increase the insurance costs faced by employers but it will generate negative public-health effects as well.

Certainly not all mental illness is substantially affected by behavior, but a greater effort needs to be made to separate those that are primarily beyond an individual's control from those that are not. Unless this is done, there is no way to accurately balance the costs and the benefits of mental-health-parity legislation, and we run the risk of mandating something that might very well lead to more harm than good.

Jonathan Klick is the Dorothy Donnelley Moller research fellow at the Mercatus Center and Robert A. Levy fellow in law and liberty at George Mason University School of Law.