George W. Bush has repeatedly been described as wanting to privatize pensions and health care, through his plan to bol- ster the use of tax-sheltered health savings accounts and individual retirement sav- ings plans. This is not quite accurate. What Bush is proposing could better be described " with only a slight abuse of terminology " as a ”œdemutualization” of these programs.

Take the plan for individual health savings accounts. What these sorts of proposals assume is that the central reason for state involvement in the health care sector is either that some people can’t afford it, or that they fail to exercise sufficient foresight in sav- ing for their future health care needs. Thus the proposed solution is to offer additional incentives to save, com- bined with some state subsidies.

The benefits that are supposed to flow from such a system come from an expected reduction of moral hazard. Because individuals will be paying for their medical care out of pocket, they will impose greater discipline upon health care providers. They will think twice before undergoing costly procedures that promise only marginal benefits.

Of course, anyone who has ever taken their pet to a veterinarian will immediately see the absurdity of this supposition. But let’s imagine away these problems. Suppose that medical savings accounts magically transform patients into an army of savvy consumers, willing to ask tough questions about whether they really need a central line put in, or willing to specify that they want old- fashioned sutures in their bowel repair, rather than those expensive staples that profligate surgeons like to use.

This still leaves us, the savvy health- care consumers, with one very important decision. ”œHow much should we save?” Well, let’s see. Someday you may need a heart bypass. That will cost about $15,000, assuming you have no complications. You might need kidney dialysis. That costs $2500 a month, for the rest of your life. Or you might need a liver transplant. That costs between $30,000 and $690,000, plus a couple weeks " maybe months " in intensive care, which can easily cost upwards of $5000 a day. (These are all Canadian estimates, for American figures multiply by four or five.)

It all starts to add up pretty quickly. The problem is that we don’t know how many of these expenses we are going to incur. Furthermore, just knowing the background probabilities doesn’t help. Health care spending is characterized by extreme variability between persons, so if you try to save by looking at population averages you are almost guaranteed to save too much or too little. Thus a system in which everyone makes an individual decision regarding how much to save will generate massive inefficiencies.

This is precisely why we have health insurance. While no one individual has any idea whether she will need a coronary bypass or a liver transplant, thanks to the law of large numbers we know almost exactly what percentage of the popula- tion will require bypasses and transplants every year. We therefore know how much we, as a society, should set aside for such procedures. It is precisely because of the utility gains that can be achieved through risk-pooling that we pay for health care through insurance schemes.

(The supposedly ”œredistributive” character of our health care system is actually just a misleading way of describ- ing this risk-pooling function. The tax system is redistributive, the social safety net is not.)

Public pensions have exactly the same structure. There the risk people face is that they will outlive their sav- ings. Individualizing retirement sav- ings essentially guarantees that everyone will save either too much or too little. Yet even though each of us has very little idea when we will die, as a society we are able to state with great precision what percentage of each cohort will die in any given year, and thus how much that cohort will need to save for its retirement. State pen- sions are essentially an insurance scheme that allows us to pool the risk of outliving our savings.

Thus the accusation that propo- nents of health savings accounts and individual retirement accounts are try- ing to privatize health care or public pensions does not get to the heart of the matter. What they are trying to do is eliminate a certain set of risk-pooling arrangements, and thereby to effect a partial demutualization of society.

The official rationale for doing so is the need to control moral hazard, but this is clearly a smokescreen. Every form of insurance generates moral haz- ard. The intelligent way to respond is by tweaking the incentive structure (e.g. by imposing deductibles, or delisting services) not by abolishing the entire insurance system.

The real problem is that, in the case of health care and retirement savings, the most efficient risk-management strategy requires universal risk-pooling. Some people are so offended by the ”œpooling” part of the equation that they are willing to forego all of the ben- efits that come from superior risk-man- agement. This takes a certain meanness of spirit, but there seems to be no shortage of that these days.