Map of life expectancy at birth from Global Education Project.

Tuesday, November 23, 2004

The Strange Ways of Insurance

Okay, so we've figured out that there are important reasons why we don't pay as we go for medical services, but rather buy something called Health Insurance. Most insurance is called casualty insurance -- we pay a premium, and we hope to get nothing for it. But if our house burns down or our car crashes or our spouse dies, the insurance company pays off. The insurance company does not, generally, in the meantime pay for a sprinkler system, driving lessons, or cancer surgery for our spouses. Health insurance as most of us know it today is different: it does pay for some routine expenses including regular physician visits when we don't even have a complaint, and medications such as statins and antihypertensives that are intended to prevent illness in people who aren't yet evidnetly sick.

So most health insurance is not casualty insurance. There is an increasingly rare version that is, major medical insurance that only pays if we have very large bills. And GW Bush wants to encourage more of us to buy that kind of insurance, and pay our regular, smaller bills on our own with the benefit of tax advantaged savings. What's wrong with that?

In order to understand what's wrong with it, we need to think about how insurance companies make money and what strategies they can use to make the most money on health insurance plans. One way you can try to make more money is to charge more for your product, or provide fewer services, but employers -- who do most of the buying -- will obviously try to get the most for their money, so it's tough to compete that way. What you can do, however, is try to sell your product selectively to cover people who are less likely to have expensive medical problems. You could do this in various ways.

You could charge more to employers (or other groups such as unions or voluntary associations, or individuals if they're buying) whose plan members are more likely to cost you money. For example, employers with many older workers, or just employers who cost you more last year. This is called "risk rating."

You could try to market your plan selectively to buyers who will cost you less. You could just refuse to sell to some buyers, or not return their phone calls, or keep the lowest possible profile in certain neighborhoods.

However you do it, if you succeed you can charge a bit less than the competition and attract more buyers who will enroll relatively healthy plan members. That leaves your competitors with customers who are going to be even more expensive for them, so they'll have to raise their prices. They're going to compete for the least costly segment of those remaining, more expensive potential customers, and as they cream them off, the remaining buyers will be the most expensive to cover. Nobody will want their business, and they won't be able to buy insurance for their employees, or it will cost exorbitantly. This phenomenon has a cute name: The Death Spiral.

So, since the purpose of insurance, after all, is to spread risk, this must be prevented by regulation. The cleanest way to prevent the death spiral is by requiring what's called community rating: premiums must be based on the average cost of a resident of the community, and all comers must be able to buy into the plan.

This post is too long as it is. But soon, very soon, I promise, we'll get to current events and actual, pressing political issues.

Comments, questions, and thought experiments based on all this are welcomed.


Hatter: 10/6 said...

Ok, I'll jump ahead and make some links to current events.

From the perspective of the insurance companies, Medicare managed care is all about marketing to the healthy. Such "cherry picking" increases the cost per person for the rest of Medicare, of course, but even with that the companies can't make enough profit. The Medicare Prescription Drug, Improvement, and Modernization Act of 2003 provided $14 billion in subsidies for "Medicare Advantage" (HMOs) to do this.

What will this mean? Optimists say this will increase the financial efficiency of care, limiting the rapid increase of Medicare costs. This ignores the fact that the overhead for traditional Medicare is about 4%, whereas insurance overhead usually runs in the 15-20% range.

Unfortunately, it's more likely that (a) more relatively healthy seniors will go into HMOs which will (b) drive up costs per person for traditional Medicare and (c) make it more politically palatable to move toward completely privatized medical coverage.

That wouldn't be so bad, but (1) a private system will never adequately cover the sick or the poor, and (2) for profit health care empirically is of lower quality overall than not-for-profit care.

Cervantes said...

Thanks. Yes, this cherry picking problem has always plagued the concept of Medicare HMOs. There are stories of plans putting their offices in third floor walk-ups so that only vigorous people could sign up, that sort of thing. The Medicare HMOs tried to recruit customers by offering things like prescription drug benefits, figuring they'd pay for it by squeezing out costs elsewhere, but by and large they didn't make money and most got out of the business. I'm not up-to-date with what's been happening with that, so thanks for your comment.

Since we're all paying for Medicare anyway, it makes sense for society to roll it into a single universal system. One political strategy that people have been pushing is to expand Medicare as a step in the right direction. But there are major problems with the way Medicare is structured. I don't think we'd want a universal system that looked like Medicare.