Confidence Game
Could gambling on your health be the best way to safeguard it?


May|June 2010 contents
Gazette Home

First Person

Notes from the Undergrad The new face of En Portada
Alumni Voices Oprah, OJ, and me
Elsewhere In a mineshaft in Bolivia
Expert Opinion Gambling on your health

A Brief History of Tontines:
The Most Popular Insurance Product You’ve Never Heard Of


The last year of debate on health-insurance reform has suffered no shortage of proposals, but Tom Baker’s is probably the only one inspired by a 17th-century lottery scheme. It may also be the only one whose name is likelier to pop up in detective fiction than in policy papers. Baker, the William Maul Measly Professor of Law and Health Sciences in the Law School, wants to bring tontines—named after the banker who popularized them in the mid 1600s—into 21st-century healthcare.

“Historically, a tontine was an arrangement in which everybody in a pool would contribute an amount of money, and the last one alive got it,” Baker explains. A cash incentive for murder would seem a poor guarantor of good health, but Baker, along with Peter Siegelman of the University of Connecticut Law School, thinks that a variation on the scheme might actually be a great way to lure 18- to 29-year-olds into the health-insurance marketplace. These so-called “young invincibles” have the lowest insurance rate of any age group in America, accounting for one-third of uninsured adults below retirement age, in part because many don’t believe sickness or calamity will strike them. Essentially, they gamble that their health is good enough that insurance premiums are a waste of money. But what if they could choose a health-insurance policy bundled with a tontine that would pay an end-of-term bonus to subscribers whose optimism proved well founded, and indeed didn’t run up hospital bills? Not long after the White House healthcare summit, Baker chatted with Gazette associate editor Trey Popp about the potential benefits of that sort of gamble, for personal health and public policy alike.

What makes tontines attractive in the current health-insurance context?

They would be appealing to people who think of themselves as very low-risk and not needing health insurance. The idea of a tontine health-insurance policy would be that people who don’t use it would get a bonus at the end, and the size of the bonus would be based on how many people [in the insurance pool] used it and how many didn’t. And we think that people who believe they’re low-risk will think they’re likely to get the bonus. So they’re going to be overweighting the possibility of getting the bonus, while underweighting the likelihood that they’re actually going to need the insurance. And the gambles will kind of cancel out.

In some states, “young invincibles” have access to barebones insurance policies with premiums of around $70 a month. That’s pretty affordable. What’s wrong with these?

People aren’t buying them! Think about it. If you don’t think you’ll need health insurance to begin with, the fact that it’s a barebones policy makes it seem even less likely you’ll use it. And that’s even though the barebones ones, from an economic perspective, are really good—because what they are is catastrophic coverage, which is what a young invincible needs.

How much would you have to increase the premium on a policy in order to support a sufficiently attractive tontine element?

We think that for 18- to 29-year-old men, if they paid an extra $25 a month [on a three-year policy], and if we set the three-year spending threshold at $750—which could perhaps cover some preventive or well-care visits—then at the end of the period you would get $1,600 back [if you didn’t exceed the threshold]. That, by the way, is assuming that only 18- to 29-year-olds signed up. You could open it up. It might be appealing to some older people, too.

It’s also potentially a way to make managed care more palatable. Instead of a deductible, or mandatory managed-care providers, you could have a program where as long as you followed a certain set of procedures, you’d be eligible for a bonus after the fact. The idea would be that it’s not that they’re making me stay in-network; they’re giving me that option, and if I choose it I get a bonus.

Do you think that the very prospect of a prize, or deferred dividend, would change the likelihood that subscribers would seek medical care?

I think it would change their behavior. And that’s both a plus and a minus. Long-term, one of the big problems with healthcare is how do we control costs? Our usual mechanism for controlling costs is that people pay for things and they ration for themselves. But we don’t do that with healthcare. So looking at the positive side, there’s lots of evidence that a lot of healthcare is unnecessary—at least among the haves, and this is a program for the haves, not the have-nots, because it targets people who have enough money that they can spend the extra $25 a month.

On the other hand, there would be concerns.  Suppose I’m 26, and I’ve got these nagging headaches, and if I just don’t go to the doctor for the next year—they’re not that bad, they’re just bugging me—I’ll get a $3,200 bonus. And then God forbid it’s a brain tumor. That’s the nightmare idea. This is the concern that I’m mindful of. My wife raises it. She basically thinks this is evil!

But there’s never going to be a perfect situation. And the other thing is, there would be nothing preventing that person who had the headache from going to the doctor and paying out-of-pocket. That would be permissible.

Evil aside, are these things legal?

In the individual health insurance market, there’s no problem with it. Blue Cross or Aetna could decide tomorrow to include this option. The laws that got rid of life-insurance tontines in the early 20th century [click here for more on that intriguing history] don’t apply to health insurance. In the [group health insurance market], its legality is an open question. You’re not allowed, in group insurance policies, to price differently according to people’s risk. So if you understand the bonus as being a kind of after-the-fact risk-based pricing, it wouldn’t be permissible. But it’s an open question. Nobody’s tried it.

Say Congress passes something close to the health-insurance reform package that they’ve got in front of them now. Would there still be room for health-insurance tontines?

As I understand the current reform idea, from a universal-healthcare perspective the best we can expect is a mandate [for people to buy insurance] … and I think this could be a really good option. Because if people are going to be made to do something, it’s much better to have them be made to do something they’re more happy about.

A mandate might actually increase the appeal of this. When all of a sudden everyone is going to get a fine—like in Massachusetts—if they don’t buy it, some people might think, well if I’m going to buy it, I’d rather buy the one where I get the big check at the end if I don’t use it.

And perhaps it could be a way for an insurer to win the loyalty of consumers at a young age and keep them as customers for the long haul.

That’s another benefit—for both insurers and a public-policy perspective: binding people to a company for the long term. Companies like it. But the reason it’s good from a public-policy perspective is that right now, it’s not worth it for a company to invest in preventive care for their policyholders. Because there’s no guarantee that they’ll be policyholders in the future. Preventive care pays off in the future, not this year or next year.

So if this idea ever really took off, you could imagine people having 10- or 15- or 20-year tontine periods. That would probably only work in this Obama individual-mandate world. You can’t control what your employer does, but if we shift to a market where people are choosing their health insurance individually, even if it might be paid for through their employer, you could imagine people having a 15- or 20-year relationship with a company—which people do have with life-insurance companies, by the way. And the company could reward you more effectively. They could say: here’s the 20-year dividend for people who don’t smoke for the 20-year period. Or who keep their weight down. Because let’s face it: keeping your weight down for one year isn’t going to affect your health very much—but keeping your weight down for 20 years, now you’re talking about a serious payoff.

Tom Baker is the William Maul Measly Professor of Law and Health Sciences in the Law School. He wrote about health-insurance tontines with co-author Peter Seigelman in the Winter 2009/2010 issue of Regulation, a journal published by the libertarian Cato Institute.

©2010 The Pennsylvania Gazette
Last modified 4/29/10