most doctors favor a public option; why non-profit co-ops are not enough

Posted on: Tue, 09/15/2009 - 13:44 By: Tom Swiss

A recent New England Journal of Medicine poll found that 62.9 percent of physicians favor a public option as part of health care reform. An additional 9.6 percent favored a completely government-run single payer option.

The alternative to the public option being put forth by industry shills is some sort of co-op. Now normally, I'm all in favor of co-ops and other non-profits: they're a great illustration of how a non-capitalist organization can function in a free market. But the idea of for-profit competing with non-profit organizations doesn't work for insurance, because of the nature of pooled risk.

Let's say you've got a bunch of people, half of whom -- call them group A -- are going to get significantly sick this year, and half -- group B -- that aren't. Pulling numbers out of the air for illustrative purposes, let's say that a group A person consumes $1,500 worth of health care over some period of time (including the necessary administrative costs), and a B consumes $500.

In an efficient non-profit insurance system, one that does the best job of spreading the risk, we set everyone's cost at the average, at $1,000.

But now, let's introduce a for-profit provider into the model. If you can tell ahead of time who's an A and who's a B -- based on things like medical history and age -- you can offer Bs a plan priced at $900. That's cheaper than the $1000 they're paying now, and since Bs only consume $500, you make $400 on the deal. Wow! What's not to love?

The problem is that this cherry picking takes Bs out of the risk pool. Say that, with that $100 incentive, half of the Bs leave the non-profit and buy into the for-profit plan. Then the non-profit's risk pool now has twice as many As as Bs. The average cost to provide care to the non-profit group jumps from $1,000 to $1,167, so that's the new cost of the non-profit plan. (Either that, or the non-profit plan has to start kicking As off its rolls.)

And so more Bs leave the non-profit pool to join the for-profit plan. And -- here's the fun part -- with that increased demand, and the costs of the general pool rising, the for-profit plan can raise its price! At, say, $1,050, it might be more expensive than what a B was paying before the for-profit plan came along, but it's cheaper than the non-profit's new prices.

Go through a few cycles of this, and eventually, if there were no other factors involved, the non-profit would have only As paying $1,500, and the Bs would all be in the for-profit plan, paying just under that $1,500 mark. Everyone's price has gone up, and the only happy people are the shareholders of the for-profit plan.

Now, in a more realistic situation where you have more than one for-profit provider, competition among private plans would keep things from running to this extreme. And of course age and medical history are not perfect predictors of how much care a person is going to need. But this simplistic model illustrates the cherry-picking problem that makes a non-profit player in an insurance system unable to drive down costs.