Markets and the 90% Case

At Marginal Revolution, Alex Tabarrok argues that where laws permit it to do so, the medical insurance industry has been largely successful in overcoming adverse selection.

A few commenters object, pointing out that those with poor health are either denied coverage or required to pay above-average rates. They may be confused about what "adverse selection" is---it's when low-risk customers avoid buying insurance because they get charged the same rates as high-risk customers but are less likely to get their money's worth---but they do have a valid point. That people with congenital health problems face higher insurance rates isn't a market failure, because it's exactly how markets are supposed to work. But it is a problem. And it's a problem that will only get worse as the ability to screen for potentially costly genetic defects improves.

But the left-wing solution to the problem---nationalizing the health care and/or health insurance industries---is wrong. It's wrong not just because runs contrary to libertarian principles, but because, for the sake of a small minority for whom the market doesn't provide a satisfactory solution, it deprives the vast majority of the very real benefits of market-driven medicine.

I'm a software engineer by trade. When writing software, a simple, elegant solution will often work for the vast majority of users and scenarios. Much of the complexity in software comes from trying to handle obscure cases that many users will rarely or never need. At the company where I work, we refer to the simple solution as the 90% case, because it works (very) roughly 90% of the time. It's a variation on the Pareto Principle: A small minority of the problems cause most of the trouble.

Markets solve the 90% case in medicine. For the vast majority of us, they work, and they work well. I agree that it's unfair that some people should have to pay hundreds of thousands of dollars in medical bills just because they inherited the wrong genes. And I can certainly sympathize with the desire to lighten their load somehow. But destroying markets and nationalizing industries will do far more harm than good in the long run.

It pains me to say anything good about government handouts, but I do believe that direct assistance targeted to those who need it most is far more effective and less destructive than the market-distorting and -destroying measures generally proposed and implemented by the left. If some people can't get insurance because of genetic defects, then let the government pay their medical bills if you must, but don't (further) nationalize the health care system. Let the market work for the rest of us.

And the practice of using a flamethrower where a laser would do isn't confined to the medical industry. Take Social Security (please!). Most of us are perfectly capable of saving for our own retirements; aside from charity, the least disruptive way of helping those who can't would be means-tested transfer payments or subsidization of savings. But what we actually have is a $500 billion middle-class welfare program, inexplicably referred to as a "safety net," in which the government takes away money people could be saving, skims a bit off the top for expenses, and then returns it decades later with a small fraction of the interest a productive investment would have returned.

The minimum wage is similarly misguided. The vast majority of us have the skills necessary to earn enough money to support ourselves, and know better than to have children before we can afford them. Only a tiny minority of adult, full-time workers have families which they are not capable of earning enough to support. But the minimum wage, aside from being a poorly-targeted policy which affects primarily teenagers, childless adults, and secondary earners, distorts the labor market and creates unemployment. Here as well, it would be far better to target assistance, perhaps in the form of wage supplements, to the truly needy, while letting the market work for the 90% case.

When designing policies to help the needy, legislators should keep in mind the Hippocratic Oath taken by physicians, popularly summarized as "First do no harm." It is no less relevant to them. The first criterion for any such policy should be that it disrupts market processes as little as possible. As a corollary, the benefits should be targeted as narrowly as possible, allowing the desired goals to be achieved at the lowest possible cost.

None of this should be taken as an endorsement of government handouts. I would much prefer that all assistance to the needy take the form of private charity. However, that's unlikely to happen in our current political climate. But a more benign, more efficient form of government assistance is within our reach, and I think that its advantages over the status quo should be apparent to those on all points of the political spectrum.

Share this

I don't think we're talking

I don't think we're talking about a "small minority" here where a valid solution doesn't exist. Part of the problem, as I see it is that there isn't enough preventative health care provided by health insurance companies. That being said, my empirical calculations indicate that most health problems are caused by eating too much, drinking too much, and/or smoking too much. The trick lies in the insurance companies finding a way for preventing a heart attack to be more efficient that curing one. The genetic defect clause is worrisome to me since I think anything could be considered a genetic defect in a court of law.

I agree that it’s unfair

I agree that it’s unfair that some people should have to pay hundreds of thousands of dollars in medical bills just because they inherited the wrong genes. And I can certainly sympathize with the desire to lighten their load somehow. But destroying markets and nationalizing industries will do far more harm than good in the long run.

What would be the problem with parents buying "genetic insurance" before they even have children? I'm no biologist, but isn't heritability of many of these defects probabilistic? Or would the presence of even one "bad" gene in one of the parents be cause for thousands in premiums?

"It pains me to say anything

"It pains me to say anything good about government handouts, but I do believe that direct assistance targeted to those who need it most is far more effective and less destructive than the market-distorting and -destroying measures generally proposed and implemented by the left."

While I don't disagree with the general thrust of the statement, the problem with its application is lies in the distribution of health care costs through the system. They are so heavily concentrated as to leave virtually no distinction between having the government "pay for those who can't afford it," and having the government "pay for everything." In the U.S., 1% of patients account for 1/3rd of all heath care costs. 5% of patients account for half. About 50% of patients account for 95% of all health care costs, while the other half of patients account for just 5%.

Moreover, for most people, more than half of the health care costs they will face will come in the last six months before they die.

To address the cost of health insurance, you have to address the cost of health care itself. And to address the cost of health care itself, there must be structures in place that allow the determination that some life-extending treatments cost more than they are worth. On a fundamental level, any change to the health care system -- either within the public or private sector -- that does NOT do that cannot possibly address the real cost drivers of the system.

This does not necessarily have to mean simply allowing people to die, although at the margins, it does have to be an option. Before ever getting to that stage, however, a major first step would be the creation of structures (again, either within the private or public sector) to actually collect information on exactly how effective various procedures are, how they compare to alternatives, how much they cost, and how often they are employed. Lay people who are less familiar with health care tend to assume that such data must, of course, be available. They are shocked to discover how little actually exists, and how various patient "privacy" laws make collecting such information nigh on impossible, even by the current third party payors -- insurance companies and the government.

You cannot have a functioning market without some basic level of information about what a product or service is, what it costs, and how it compares to alternatives. In the health care market, it is not just that consumers don't have access to this information -- it's that NOBODY does.

"I agree that it’s unfair

"I agree that it’s unfair that some people should have to pay hundreds of thousands of dollars in medical bills just because they inherited the wrong genes."

I don't. What do you mean by "unfair" here.

Aren't the genes just one aspect of the the self? You can have bad parents, bad friends, a bad personality, bad cultural background, bad education, and so forth. All aspects of the self.

If one gets preferential treatment (not having to pay the costs) for bad genes then why not for bad personality? Sure bad personality can be compensated for with charm school, but who is going to pay for that?

If it is unfair that some people have bad genes then isn't the reverse true? Wouldn't these lead you to consider other means to make things more fair. For instance, why should all the pretty girls get the rich guys? Why should all the good looking guys get all the one night stands? (I'm using 'all' here in a colloquial manner).

Not only do I not believe it is "unfair" to have bad genes but I don't think people should automatically get compensated for the "unfair".

Insurance is suppose to spread risk, not generate compensation for "unfairness". If 100 people have a 2% risk of incurring losses of $10,000 each year then can limit their risk to $200 per year if they co-insure. Over the long run they are still paying for their risk however. It isn't' a free lunch. Over 50 years each person would be paying $10,000 and on average one such loss. Each person is paying his own way. The same is true if the size of the loss is doubled and the odds of it occuring half. Each person is still paying for their own risks.

If someone were to join the group in the prior example who knew they were going to incur losses on $1000 on average every year then they would be cheating the other participants if they payed only $200. They wouldn't be paying their own way. In order to pay their own way they would need to up their payment to $1000. They could still participate but with a different insurance cost for the same coverage. A concrete example being that people in flood plains should pay more than people living on mountaintops for flood insurance. There is nothing unfair about this, and even if it is 'unfair' that some people live on flood plains, there is no reason someone not living on one should pay for it.

If you think a having a genetic disease is unfair I have a question for you. Who was being unfair? Unless it was due to some involuntary exposure to radiation or pollution, the only people I can think of being responsible for this are the parents. They are the ones who decided to bring you into this world. Another reason why they should be responsible is that they carry (or may carry) the gene in question. It is part of their 'self'. I don't see why the consequences of the gene shouldn't bear upon the carriers. Shouldn't the parents be the first people we go to when extracting support for incompentent children, after the children themselves? Isn't it their responsibility to know whether they run the risk of having defective children before they create them? Isn't it their responsiblity to ameliorate their risks by taking out insurance in cases where they can't know? If they don't they are freeloading off the rest of us.

Isn’t it their

Isn’t it their responsiblity to ameliorate their risks by taking out insurance in cases where they can’t know? If they don’t they are freeloading off the rest of us.

That was sort of the point I raised earlier; couldn't you fix the "unfairness" element by having parents pay insurance premiums for carrying bad genes, or would they still have to pay the exhorbitant costs anyway?

Stefan, I didn't see your

Stefan,

I didn't see your comment before I posted my reply to the article. I agree with you. No they wouldn't have to pay exhorbitant costs anyway. By paying for the genetic insurance up front they are spreading those probablistic risks across the pool of those who have insured. They have paid for their coverage and should get the benefit. Of course, it all depends on how the policy you proposed is written.

Perhaps the example I used made you think I would expect people to pay anyway. I wasn't really expecting everyone to get back exactly what they put in. The example was simplified for illustrative purposes. After all even in my example, with sufficient policy holders, someone is going to collect the first year. The purpose of the example was to illustrate the fact that the policy is not shifting responsiblity but only reducing risk, variablity in bad consequences.

A simple example of this is someone who buys million dollar life insurance policy in good faith and dies from a covered accident a short time later. They not freeloading because they bought the insurance. Had they not bought the insurance and still expected some payout for their beneficiaries then that would be unjustified.

I assume you noticed that making people responsible for their own genetic heritage is not the same as eugenics. You're not preventing them from breeding, or even worse exterminating them, you are just making them bear the burden of there own heritage. Yes it has a selective effect but it is not an arbitrary one.

I don't think there is a set rule on this of force them to pay either. People are bad evaluators of risk. There are going to be cases where individuals are going to make mistakes for which they are incapable of taking responsiblity. I think charity needs to take up the slack in these cases.

I am still sitting on the fence over whether and to what extent charity should be voluntary or that it is a duty. I think it depends on many factors, including the state of the society you live in. I certainly think that if you are walking along and see a baby drowning in a 5 gallon bucket you have a duty to help the baby, no matter what the society. I think I can justify that with sufficient verbiage. I have thought about it and have at least convinced myself. I think the duty is so strong that to not to save the baby is actually a criminal act. In fact, it was my pondering of this very issue that first caused me to pull back from the positions of at least some libertarian thinkers.

I sorry if I seem to waste a lot of time going off on tangents but I constantly working on my ethical and philosophical positions. I think this issue does impinge upon charity and I have unconventional ideas in these areas. So unconventional that I haven't run across anyone making the arguments I have come up with. I am not sure if it is because of my originality or just plain ignorance.

assume you noticed that

assume you noticed that making people responsible for their own genetic heritage is not the same as eugenics. You’re not preventing them from breeding, or even worse exterminating them, you are just making them bear the burden of there own heritage. Yes it has a selective effect but it is not an arbitrary one.

That's true, and I should probably state that I don't see anything wrong with "eugenics" per se, if what is meant by that is any effort to improve human beings through changes to their genetics. Everyone engages in eugenics everyday when they choose a mate, as evidenced by such comments as "Oh, that person would make good father", etc. But if there's another sense in which the word is being used here I'm not aware of it.

I am still sitting on the fence over whether and to what extent charity should be voluntary or that it is a duty... I think the duty is so strong that to not to save the baby is actually a criminal act. In fact, it was my pondering of this very issue that first caused me to pull back from the positions of at least some libertarian thinkers.

The question is fundamentally whether acts of omission ought to punishable by the law, since presumably we both agree that actually shoving the baby into the water constitutes criminality. The libertarian answer to the question "What is justice?" is a negative one, saying only that certain actions constitute injustice - it says nothing about inequality in socioeconomic status, or different conditions of risk, or freak natural disasters, or failures to act.

Why is this? Well, let me explain why I disagree with your approach, as I think you're going in the wrong direction to think that the law should require charity in such a situation. Part of it seems to be a logical problem - would you also advocate income redistribution from the wealthy to the poor on the same principle? What if the poor man lives in the alley behind the rich man's building? For as easily as the man in your example can walk over and save the baby, a millionaire like Bill Gates can just as easily save the lives of many Africans by making a bank deposit from his PC (and surely visiting a website is not much more difficult than walking over to a pale of water?). Or perhaps you're suggesting that the proximity matters somehow, i.e. it's unjust to not help people within 10 meters of your body but just to forget about people living a mile away? I can't think of any reason why justice would depend on geography.

In terms of morals, I agree with you that morality requires the man to save the baby. But morality is not the same thing as justice. The moral requirement to act justly is only one of many that people have, although in interpersonal affairs it seems to be better to act according to justice than not. Are you suggesting the idea libertarians have of justice is flawed, or do you think that other moral concerns can override justice? When, and how do you reconcile them with the example I gave above with Bill Gates?

Stefan, I didn't see your

Stefan,

I didn't see your comment before I posted my reply to the article. I agree with you. No they wouldn't have to pay exhorbitant costs anyway. By paying for the genetic insurance up front they are spreading those probablistic risks across the pool of those who have insured. They have paid for their coverage and should get the benefit. Of course, it all depends on how the policy you proposed is written.

Perhaps the example I used made you think I would expect people to pay anyway. I wasn't really expecting everyone to get back exactly what they put in. The example was simplified for illustrative purposes. After all even in my example, with sufficient policy holders, someone is going to collect the first year. The purpose of the example was to illustrate the fact that the policy is not shifting responsiblity but only reducing risk, variablity in bad consequences.

A simple example of this is someone who buys million dollar life insurance policy in good faith and dies from a covered accident a short time later. They not freeloading because they bought the insurance. Had they not bought the insurance and still expected some payout for their beneficiaries then that would be unjustified.

I assume you noticed that making people responsible for their own genetic heritage is not the same as eugenics. You're not preventing them from breeding, or even worse exterminating them, you are just making them bear the burden of there own heritage. Yes it has a selective effect but it is not an arbitrary one.

I don't think there is a set rule on this of force them to pay either. People are bad evaluators of risk. There are going to be cases where individuals are going to make mistakes for which they are incapable of taking responsiblity. I think charity needs to take up the slack in these cases.

I am still sitting on the fence over whether and to what extent charity should be voluntary or that it is a duty. I think it depends on many factors, including the state of the society you live in. I certainly think that if you are walking along and see a baby drowning in a 5 gallon bucket you have a duty to help the baby, no matter what the society. I think I can justify that with sufficient verbiage. I have thought about it and have at least convinced myself. I think the duty is so strong that to not to save the baby is actually a criminal act. In fact, it was my pondering of this very issue that first caused me to pull back from the positions of at least some libertarian thinkers.

I sorry if I seem to waste a lot of time going off on tangents but I constantly working on my ethical and philosophical positions. I think this issue does impinge upon charity and I have unconventional ideas in these areas. So unconventional that I haven't run across anyone making the arguments I have come up with. I am not sure if it is because of my originality or just plain ignorance.

Hmm, did you click the

Hmm, did you click the 'submit' button accidentally, or is this a subtle way of suggesting I reread your comment? :sleep:

Sorry, I used the same

Sorry, I used the same window to type in my reply, later in the day, had my laptop on my lap and my shirt hit the mouse buttons on the front edge of the laptop. I lost all my text and I was hitting the forward and back buttons to get it back. Somehow in the process I resubmitted the text.

I'm thinking of buying a mac

I'm thinking of buying a mac laptop, but am not sure if I should buy now or later this year after they've switched their products over to intel...

Stefan, This is taking me

Stefan,

This is taking me too long. I will have to pick it up later. I have enough material for an entire article on this so I think that is what I will do over on my blog.

The short of it is that a person who does not act as a Good Samaritan, a 'Levite' in the story, is acting as a freeloader in a reciprocity agreement between reasaonable people. I will show that such implicit agreements should take priority over a conflicting social convention of misanthropic abandonment. I will show a way out of such agreements for unreasonable people who wish to act as Levites, but this "out" has objectively if not subjectively undesireable consequences.

The beginning of my numbered outline for the article is below, I'm already up to item 22). I'm a crappy writer so this may take me awhile to convert this to a easily understandable form. I like credit for my ideas so that's another reason for me to publish fully explicated ideas on my blog instead of in the comments here.

As you can see some of the assumptions upon which I am basing my principles and the approach I am taking, may not apply the the Bill Gates example. There are additional assumptions I need to make in order to criminalize Levite type behavior. These are not ad-hoc assumptions invented by me to ruin your example. I had these worked out in advance. These other assumptions I have not shown will place further wedges between the baby in the bucket example and the Bill Gates one.

Even after I am finished, my position will be someone who wants to act as a Levite can do so freely under special circumstances and with prior behavior on his part. Just like all killing is not murder, not all abandonment of victims is criminal, not even in the case where help has minimal cost to the rescuer. One is free to be a misanthrop with no fear of criminal prosecution given my 'rules'.

Outline:
1)Proximity is reciprocal. If I tend to be around you then you tend to be around me.
2)Everyone is fallible.
3)It is possible that anyone might get into a desperate situation from which they cannot recover without immediate help by a rescuer.
4)A subset of those situations will be such that it is reasonable to believe the victim could repay the rescuer in some time-shifted manner. Either by actual restitution or by repaying the favor should the rescuer ever be in the same situation, or even if it was likely the victim could have returned the favor in the past. These situations I will call Samaritan situations. If a rescue is attempted it is called a Samaritan Rescue and the rescuer a Samaritan.

...

The short of it is that a

The short of it is that a person who does not act as a Good Samaritan, a ‘Levite’ in the story, is acting as a freeloader in a reciprocity agreement between reasaonable people

I'm sure I don't have to explain to you the numerous problems with implicit contracts, but if I do then I suggest you read the relevant section of Long's anarchism talk.

A subset of those situations will be such that it is reasonable to believe the victim could repay the rescuer in some time-shifted manner. Either by actual restitution or by repaying the favor should the rescuer ever be in the same situation, or even if it was likely the victim could have returned the favor in the past. These situations I will call Samaritan situations.

It sounds to me like you want to justify theft whenever it can be made up for later - which I don't take to be a consistent position. For example, under that logic it would be moral to steal your TV set as long as I pay you back later when I'm rich.

How is rescuing someone

How is rescuing someone equivalent to theft, it's more like a gift.

Suppose someones car is leaking transmission fluid and when he pulls over he notices someone who was hit by a car and dying in the ditch. You seem to think that although morally reprehensible he can just leave the guy in the ditch, and walk for a tow truck and get his car fixed. I on the other hand believe he has a duty to help the guy out and if that involves driving his car without transmission fluid then he should do so. I also believe he should be able to get restitution (a new transmission) from the fellow in the ditch, even if he was unconcious and could not consent to the rescuers behavior.

Oh, and in case you didn't realize it we live in a world of implicit contracts. In fact, respecting property rights is an implicit contract, a reciprocal one.

I already read Long's article a while ago. I wasn't planning to use implicit contracts to justify coercive monopolies over land. In fact, what I was planning to do is perfectly compatible with anarcho-capitalism ... or was it arachno-capitalism ... never could spell.

You seem to think that

You seem to think that although morally reprehensible he can just leave the guy in the ditch, and walk for a tow truck and get his car fixed. I on the other hand believe he has a duty to help the guy out and if that involves driving his car without transmission fluid then he should do so.

What's the difference between a duty to do something and a moral obligation to do something? I've already conceded it's moral to help starving or helpless victims.

I also believe he should be able to get restitution (a new transmission) from the fellow in the ditch, even if he was unconcious and could not consent to the rescuers behavior.

I don't see how a contract is just if there's no consent to it, implicit or otherwise. If they guy gives some indication that he wants help, then you might be entitled to compensation, but if he's unconscious it's hard to see why he owes you anything since unconscious people can't consent to contracts.

Oh, and in case you didn’t realize it we live in a world of implicit contracts. In fact, respecting property rights is an implicit contract, a reciprocal one.

Rights are not the same as contracts on my view. The existence of contracts is a device to ensure the protection of people's rights. Thus "contracts" which involve murder or theft are not just, and hence not contracts at all.

In fact, what I was planning to do is perfectly compatible with anarcho-capitalism … or was it arachno-capitalism … never could spell.

Nah you're thinking of the theory where gigantic spiders own all the means of production (Marx really hated that one).