POOLING AND SEPARATING EQUILIBRIA. Maha likes the large numbers argument for government-mandated (and provided)? health insurance.

Insurance works by risk pooling — everybody throws money into a pot so that there’s money for people who are hit with unexpected expenses. In order for this to work, in any given year most of the people in the pool throw more money into the pot than take it out. Generally, the bigger the pool, the better it works. Insurance companies invest the premium money, and they make most of their profits from investments.

Thus, a single carrier is by definition able to pool risks more effectively than competing carriers (leaving aside for the moment the problem of valuing risk without a market test) and a single carrier enjoying the force of law has higher exit costs for people who know they are subsidizing others. Pooling equilibria are therefore fragile, as participation is not incentive-compatible for everybody, and that's why automobile insurance (required by most states) comes with experience rating and with choices.

I left that last sentence in as it is also instructive. Many advocates of the government insurance company suggest that, as there is no residual claimant, the operating costs of that insurer will be lower. Is that accurate? Perhaps such an insurer, if it did crowd out all private insurers, could act as a monopsonist with respect to risk managers. (Higher education acts as a monopsonist with respect to humanities Ph.Ds. Discuss.) On the other hand, an insurance company cannot do better with its investments, adjusted for risk, than an investment bank. If that is the case, is it the insurance premiums that provide the profits, or is it the investments of premiums?

In a column dealing primarily with the inefficiency effects of favorable tax treatment for health insurance as a fringe benefit, Martin Feldstein identifies the incentive for participants in a pool to self-select out.

The best solution to this problem of private overconsumption of health services would be to eliminate the tax rule that is causing the excessive insurance and the resulting rise in health spending. Alternatively, Congress could strengthen the incentives in the existing law for health savings accounts with high insurance copayments. Either way, the result would be more cost-conscious behavior that would lower health-care spending.

But unlike reductions in care achieved by government rationing, individuals with different preferences about health and about risk could buy the care that best suits their preferences. While we all want better health, the different choices that people make about such things as smoking, weight and exercise show that there are substantial differences in the priority that different people attach to health.

And perhaps, worst case scenario, in the priority that an insurer attaches to a policyholder receiving treatment?