How Knowing More Can Make Us Worse Off
Some people are born with bad hearts, some good. As long as nobody knows which is which it is possible to insure against the risk of having a bad heart. What happens if a genetic test is invented that distinguishes people who are likely to have a heart attack from people who are not?
Consider some possible legal rules that might be proposed to deal with such an invention:
1. The test is banned; nobody is allowed to use it.
2. Testing is legal. Insurance companies are permitted to make testing a condition of insurance and take account of the result in setting rates.
3. Individuals are permitted to get tested; the results are confidential. Insurance companies are forbidden to make testing a condition of insurance.
4. Individuals are permitted to get tested. Insurance companies are not permitted to require testing as a condition of insurance but are permitted to know whether or not a potential customer has been tested and to take account of that fact in setting the rate they charge him.
What are the consequences of each rule? Is it possible that, under some rules, the invention of the test makes us worse off? Under all rules?
To see why the answer is “yes,” compare rules 1 and 2. Under rule 1, which corresponds to the situation before the test is invented, neither the insurance company nor the customer knows the condition of the customer’s heart, so the risk of having a bad heart is insurable. Under rule 2, if you try to buy insurance and refuse to be tested the company will take that as evidence that you have already been tested and discovered that you have a bad heart. The company will set the price of the insurance accordingly. You can get tested, show the results to the insurance company and insure against whatever uncertainty is left, but the risk of having a bad heart is now uninsurable.
The result of rule 3 is worse still. Since the price is the same whether you have a good or bad heart, insurance against a heart attack is a much better deal for people who know they have bad hearts. The insurance company, realizing that, prices its policies on the assumption that someone who chooses to insure against a heart attack is probably a bad risk. People with good hearts cannot get insurance unless they are willing to pay far more than the actuarial value of their risk, so many of them don’t buy. That leaves an insurance market mainly populated by bad risks. Now nobody can insure against the risk of having a bad heart; you cannot bet on the dice after you have rolled them. People with bad hearts can insure against the residual uncertainty of when their hearts will fail. People with good hearts cannot insure even against that unless they are willing to do it at a bad heart price.
The classic article on the problem is “The Market for Lemons” by George Akerlof. It described the same logic in the context of the used car market. The seller knows if his car is a lemon or a creampuff, the buyer doesn’t. Buyers offer the same price for lemons and creampuffs since they cannot tell which is which. At that price, owners of creampuffs are much less willing to sell than owners of lemons, so if the seller accepts your offer the car is probably a lemon. Knowing that, you offer a lemon price. In the limiting case, the result is a market where only lemons sell.
The problem is called adverse selection. Its origin is asymmetric information: One party to an exchange knows something about the value of the good being sold that the other does not.
Rule 4 provides the best outcome. People who want to insure against the risk that they have a bad heart can buy insurance before being tested. Since the insurer knows that they have not been tested, the price will be based on the cost of insuring a random customer. After they are insured they can decide whether the advantage of better information about what health precautions they should take and how long they can expect to live outweighs the risk of learning something they do not want to know.
Rule 4 may not be an option, especially in a world of many countries. Even if the United States insists that all tests be recorded and successfully suppresses the black market in secret tests, American citizens can still get their genes tested somewhere with less restrictive rules. The same problems apply to rule 1. So it is possible that the invention of the test, by moving us from the world of rule 1 to the world of rule 2 or 3, may make the risk of being born with a bad heart uninsurable, just as the risk of being born poor is now. If that effect is large enough to outweigh the benefits that individuals get from knowing more about their own health risks the invention of the test will have made us worse off.
I was introduced to this problem by a commencement speech proposing rule 3 as a way of protecting people from the use of genetic information by their insurance companies. I concluded that the speaker had never heard of adverse selection.
Suppose we expand our example by assuming that genetic testing provides good information not only on the risk of a heart attack but on the risks of most other causes of mortality. Neither the insurance company nor I can predict the date of my death precisely but we can predict it accurately enough to eliminate much of the benefit of life insurance in case 2, accurately enough so that adverse selection destroys the market for life insurance in case 3.1
I have been assuming a world where mortality is largely determined by genetics and we know enough to use genetics to predict it. The first assumption is at least doubtful, the second, at present, false. But the same issue played an important role in the real world controversy over the Affordable Care Act, aka Obamacare.
There are at least two things about a potential customer for medical insurance that tell us a good deal about how much he can be expected to cost the insurance company. One is how old he is; young people are much less likely to require expensive care than old people. The other is his current medical condition. If you have survived one heart attack, the odds that you will have another are high. In an unregulated marketplace for health insurance, insurance companies will take account of those effects. They will be happy to insure everyone — but not at the same price. Young people will be charged much less than old, healthy people less than sick.
Under Obamacare, insurance companies were forbidden to make the price they charged depend on the patient’s state of health; premiums had to be the same for everyone of a given age, independent of pre-existing conditions. They were permitted to charge old people more than young but the ratio of prices could not be more than three to one. That made insurance a bad deal if you were young, a good deal if you were old, a bad deal if you were healthy, a good deal if you already had expensive medical problems. Both customer and insurance companies had the information that predicted how much the insurance company would have to pay for the customer’s medical expenses but the insurance company was forbidden to use it in deciding at what price to sell the customer insurance.
If customers had remained free to decide whether or not to buy insurance, many for whom it was a bad deal at the price, the young and healthy, would have chosen not to. That would have raised the average cost to the insurance companies. Since they had to cover their costs, the price of insurance would rise. The more it rises, the more of the young and healthy choose to go uninsured.
The solution to that problem, obvious to the economists involved in designing the program, was to forbid the customers as well as the companies from using the information both had, to require everyone to buy insurance whether or not he thought it was a good deal for him. So the program included a mandate, a requirement that everyone buy insurance and a penalty for not doing so. That ended up as one of its most controversial elements.
I have simplified the details of ACA in my description but not, I think, in ways that affect its essential logic.
My web page, with the full text of multiple books and articles and much else
Past posts, sorted by topic
A search bar for past posts and much of my other writing
A draft of my next book, Consequences of Climate Change, webbed for comments.
I feel that this analysis conflates two impacts of insurance: as a hedge against risk, and as a social safety net.
In world 1, good-hearted folks are subsidizing the medical costs of those with bad hearts. They do this because they have a strong desire to hedge against their own heart-attack risk, and the only hedging mechanism available to them (purchasing insurance) bundles an implicit subsidy.
In world 2, genetic tests make it possible to purchase a hedge on one's own risk without subsidizing the bad-hearters. World 2's good-hearters still have the option to subsidize bad-hearters, e.g. by donating money to a nonprofit which defrays the inevitable medical costs of those with bad hearts. If each good-hearter donates the difference between their world-1 premium and world-2 premium to this nonprofit, nobody in world 2 is any worse off than world 1.
Of course, we expect almost no good-hearter would choose to make this donation. We could force them to do so via a tax. This would be a sort of social safety net, where wealth is reallocated from the good-hearted to the bad-hearted. (Analogous to how society solves the issue of the born-poor, another non-insurable condition you mentioned.)
So it seems to me a world 5, with tests as a prerequisite for insurance and a redistributive tax on heart quality, could be at worst equivalent to world 1. But as we can now adjust the amount of redistribution to choose the optimal value for society (there is no reason to believe a priori that the amount of implied redistribution in world 1 is the optimal amount), I expect that world 5 could be made better than 1.
It is not obvious that a world where there is perfect knowledge about who has a bad heart, but it is uninsurable, is worse than the world where there is asymmetric knowledge of bad hearts and it is insurable.
Consider a world where nobody knew whether they were hungry or not, their hunger was random, and their body weight remained constant regardless of their calorie intake, but they died or were struck ill the next day if they didn't eat when they were hungry. Now in such a world, hunger would be insurable. But would that world be better than the current world? I would venture no.
In the current world, insurance exists because of uncertainty. It produces value only because of that uncertainty. I am willing to pay some money because of the chance I could get a heart attack. In a world where that uncertainty disappears, the specific value add that an insurance program gives is gone. If we legislate information asymmetry to make a bad heart insurable, it transforms insurance into a redistribution program, and creates externalities between the those with good hearts and those with bad hearts. Insurance would make only those with bad hearts better off, while those with good hearts will be worse off.
Consider the long-run effects of this. When the costs of something are insulated against, that thing is more desirable, other things equal. On the other hand, when getting a bad heart or having children with bad hearts costs more, people will do more to avoid that outcome. So redistribution to favor the less healthy ultimate incentivizes being unhealthy.
The cost of having a bad heart, when there can be certain knowledge of who has it, is information too, and information that is effectively destroyed or attenuated when we legislate insurable bad hearts into existence. With that, those who have or could have bad or worse hearts are made worse off because they can't take those costs into account themselves. So this scheme harms both those with good and with bad hearts. It makes everyone strictly worse off in the long run even if it appears to benefit one group at the expense of another in the short run.