Why an individual mandate to buy health insurance? We've heard a lot of debate on this question lately. But it seems a lot of the talk doesn't reflect much knowledge of why some think an individual mandate might be a good idea. Let's run through the various reasons one hears:

Reason 1 » It's because of uncompensated care. Uncompensated care is when an uninsured person receives care, usually at an emergency room. Uncompensated care, the story goes, has to be paid for by someone, and so it ends up inflating prices for health insurance generally. Force people to buy insurance, and uncompensated care goes away. But it turns out that uncompensated care is really quite a small fraction of our nation's overall health care costs. MIT economist Jon Gruber puts the number at $30 billion, which sounds like a lot until you note that it's about 1.5 percent of our overall health care expenditures.

Reason 2 » Because you have to get auto insurance, right? Maybe, but the reason we require auto insurance is really quite different. If you drive a car you might harm someone else in an accident. And that's the risk we're forced to insure by buying auto insurance. A mandate makes sense here because there's a general tendency to think too little about harm we might impose on others. But this reasoning doesn't apply for health insurance.

Reason 3 » It's a plot by the Democrats to take away our liberties. I guess I can't definitively disprove this one, but if so then it's a Democratic plot they stole from Mitt Romney. The fact is that economists of all stripes (even those of us who try to stay away from politics altogether) agree that insurance markets are unusual (for reason discussed next). Former Gov. Romney's Massachusetts health plan has an individual mandate -- and it's because his economists were telling him pretty much the same thing that President Obama's economists are saying now.

Reason 4 » To prevent unraveling of the insurance market. This one's kind of complicated, but it's also the right answer. And to understand it, one needs to understand the effect of "asymmetric information" on insurance markets. Asymmetric information is when I know something and you don't. If I'm selling you a used car that I've owned for 10 years, I probably know a thing or two about the car that you don't. And that might give you pause. You might worry, in particular, about why I'm selling the car, and whether my decision to sell is related to what I know. Is the transmission about to go? Is the frame bent?

Let's think about how this applies to health insurance. If you're a firm selling an insurance policy, you know you'll lose money on some of the people you cover. People who get cancer, well, that's expensive treatment. But you hope to make it up on people who pay premiums but don't get sick. They pay in more than they get out.

Put this together with asymmetric information, and we can see where the potential problem lies. Suppose potential policy-holders are better informed than you are about their health status. Some people know they're relatively less likely to get sick, and, for them, insurance is a bad deal. They figure they'll pay in more than they get out, so they don't buy. And if healthy people don't buy, it's harder for insurers to make up the costs of those cancer drugs, which means the price for insurance goes up, to a level far higher than it would be in a market where people weren't asymmetrically informed about their health status.

It's this unraveling of the insurance market that makes it so expensive for, say, the self-employed to buy health insurance. Insurers figure (rationally) that if you're willing to pay that much for health insurance then you must be expecting to use it. A mandate may stop this unraveling, and help bring prices down.

Scott Schaefer is associate dean for academic affairs and David Eccles Professor of Finance at the University of Utah's David Eccles School of Business. He blogs about economics at utah-economist.blogspot.com.