Today is my health econ final exam. I also return their last paper, on faith healing. After an entire semester hearing how we get little health value from a wide margin of medical spending, almost every student (21 undergrads & 9 grads) said that a big argument against legal faith healing is that it can discourage people from going to regular doctors. Most also said it is hard to evaluate faith healer quality, and to know if they are just in it for the money.
1. P.J. O'Rourke on "fairness, idealism and other atrocities": All politics stink. Even democracy stinks. Imagine if our clothes were selected by the majority of shoppers, which would be teenage girls. I'd be standing here with my bellybutton exposed. ...
""Faith healing" should be illegal because it is fraud."
So? People feel like it helps; it boosts their utility. Faith healing is proof that ignorance is bliss. And what's wrong with having blissful idiots? That's certainly better than depressed idiots. And who's to say that they won't blow their money some other way?
Peter, I'd suggest that it is actually more beneficial to learn that seeing the doctor is unlikely to do you much good, and that you can usually get just as much help from an alternative practitioner for much less money, than to learn that you should vote for certain universal health care policies. Learning the former can make a moderately significant difference in my life, in terms of changing my practices with regard to health care. But changing my political views isn't likely to lead to an actual difference in my circumstances, because I am not in charge of national health care policies. So to me, this kind of posting is reasonably helpful.
It's true though that not many people are open to unbiased information about health care, as Robin has pointed out in his other papers. It is a remarkably resistant bias. I find that seeing that graphically illustrated in the comments here has been quite interesting and informative.
I agree with the palm readers bit. Gambling is a bit of a weird case, but the house advantage is well known, so there's really no deception there. (I'd prefer that slot machine owners be required to disclose the odds of every prize-winning outcome, though.) Finally, alcohol also has well-known effects, some of which are desirable, so there's no issue of fraud there either and it probably doesn't count as "irrational" utility in the way that lottery tickets are. (There are other reasons why a ban on ethanol consumption may be desirable, but let's not go there.)
I wish I could ban churches, but the consequences of doing so tend to be worse than the churches. ;)
it serves as a signaling mechanism that they care deeply
Mike (and, to an extent, Paul), that's a bit needlessly negative isn't it? Are you saying there's no chance that people actually want their loved ones to get better, and are prepared to expend whatever's necessary, even to the point of irrationality?
The desire to consume a certain amount of 'treatment' is not difficult to explain, and most likely has little to do with the efficacy of the treatment in question. To put it one way, people look for treatment to feel better, not to get better. Hence all those well-documented placebo studies.
If my nearest and dearest fell seriously ill, I would probably not be at my most rational. I'd spend every penny I had on treatment in order of how likely it was to succeed. Throwing money away as a signal would not be a useful thought.
Fraud addendum: if you're banning people that sell 'irrational utility', after the faith healers, you should start on the palm readers, lottery organisers, churches, breweries....
Paul, I'm fairly certain that Robin would say, and it seems reasonable to agree, that consumers whose family and loved ones fall ill are, precisely as you say, hell-bent on consuming some fixed amount of "treatment" regardless of its efficacy because it serves as a signaling mechanism that they care deeply for the life and well-being of the one who is sick or injured.
Assume someone has a fixed health care budget. They can allocate this budget to medicine or faith healing. Each dollar allocated to either source, assume, has a positive probability of producing some health benefit. That need not be very much -- there's likely to be a lot of waste in medicine. Let the respective probabilities be P and Q for a dollar on medicine or a dollar on faith healing. Let the proportion of the budget allocated to medicine be X. Then the patient's health utility (assuming constant units of health benefit) h(x)=PX+Q(1-X). Simple calculus will verify that so long as P>Q, the health benefit is strictly increasing in X.
And whatever else it's reasonable to think about the comparative benefits of medicine vs. faith healing, surely P>Q!
So I suppose the assumption that makes this really easy is the fixed budget. So let's drop it. We can assume that medicine is more expensive than faith healing for the same amount of "treatment" (measured somehow), perhaps because of moral hazards from insurance, etc. But then we have to ask why consumers can't be encouraged to drop ineffective medicine without replacing it with (more ineffective) faith healing. If we don't think consumers will reduce medicine use without a substitute in the form of faith healing, we must think consumers are hell-bent on consuming some fixed amount of "treatment" regardless of its efficacy. Why do we think that? Is it possible to correct for this tendency with some other social policy?
(Incidentally, on prayer as fraud, *charging* for prayer perhaps should be criminalized as fraud. We criminalize fortune-telling on much the same grounds. Free prayer doesn't defraud anyone. Perhaps only bad-faith prayer should be criminalized though -- there are good reasons not to criminalize people for things they genuinely believe -- although this would be hard to enforce by picking out the real from the false beliefs. The practical problems might be insurmountable. But in principle, sure, under some circumstances.)
Peter, I respectfully disagree. I also would like to see more posts about how to improve incentives or techniques, but this blog post does something more useful than documenting the existence of biases. It also proposes that we could save money by siding with factions that want universal health insurance expanded to not just cover dubious medicine practiced by establishment doctors, but to dubious medicine practiced by the subset of "alternative healers" that are cheaper than establishment doctors. That's an interesting, innovative idea, that might be a good idea, too. Because saving money in the realest sense can save lives, including, most importantly, my own. Although I agree that it only detracts from the good idea to express it to us in the form "Oh my god, my stupid, biased students".
Robin should give us more posts like this: they definitely reveal people's biases, and might in the long run encourage people to start to overcome themI find it hard to believe posts like this will encourage people to overcome biases. If his students, with a clear incentive (i.e. grades) to overcome their biased opinions of doctors, overwhelmingly resist, then why would blog posts (which provide virtually no incentive) do anything? It looks like the main effect of this post is to make readers more pessimistic, either about the effectiveness of pointing out biases, or the ability of this blog to identify biases. Maybe that pessimism is on average slightly more accurate, but it's hard to see much value in that increased accuracy.
So I'd like to see more posts about how to improve incentives or techniques, and only see posts documenting the existence of biases when there's some reason to think readers want to overcome those biases but aren't aware of them yet.
I'm wary of anything marketed as "natural", as I tend to associate such marketing with "Our product doesn't have any real benefits over the alternatives, we just want to manipulate gullible people who don't know any better!" "Natural is good" is one of the most pernicious associations I know of, and one of the most blatantly false. There's plenty of perfectly natural poisons around, for example. Smallpox virus is natural. Dying during a difficult childbirth is perfectly natural, while the potentially lifesaving Cesarean delivery is extraordinarily unnatural.
I could go on, but I've already done enough thread-jacking as it is.
Bertrand Russell once wrote the famous words "I wish to propose for the reader's favorable consideration a doctrine which may, I fear, appear wildly paradoxical and subversive. The doctrine in question is this: that it is undesirable to believe a proposition when there is no ground whatever for supposing it true."
Believing things that are not true can be dangerous business. Even if the beliefs in themselves appear harmless, you are likely to base other beliefs on the faulty ones.
Incidentally, I think a problem with making faith healing illegal is that people sometimes gets more interested in the forbidden and exciting. Spend the money on education instead and trust people to make good choices.
That's the proper response to discovering that doctors aren't better than faith healers - not replacing the doctors with the faith healers!"
But this is not true, Caledonian. To continue my example, the dysthymic - the worried well - do cost society in terms of absenteeism, etc. due to their anxiety, mild depression, whatever. It is in society's interest to treat them; doing nothing has a defined cost here.
They can return to full function with the expensive essentially placebo treatment of prescribing psychiatrist + Prozac with likewise expensive side effects or with the inexpensive essentially placebo treatment of Bach's Flower Essences with no side effects.
And, biases being what they are, people will probably prefer to take a "natural" treatment from a nice homeopath than a "chemical" from the forbidding shrink, so society will probably get better compliance and faster recovery.
Thirty Three Things (v. 63)
1. P.J. O'Rourke on "fairness, idealism and other atrocities": All politics stink. Even democracy stinks. Imagine if our clothes were selected by the majority of shoppers, which would be teenage girls. I'd be standing here with my bellybutton exposed. ...
""Faith healing" should be illegal because it is fraud."
So? People feel like it helps; it boosts their utility. Faith healing is proof that ignorance is bliss. And what's wrong with having blissful idiots? That's certainly better than depressed idiots. And who's to say that they won't blow their money some other way?
The comments here have not been defending the doctors. They've been attacking the idea that faith healers would be better.
Peter, I'd suggest that it is actually more beneficial to learn that seeing the doctor is unlikely to do you much good, and that you can usually get just as much help from an alternative practitioner for much less money, than to learn that you should vote for certain universal health care policies. Learning the former can make a moderately significant difference in my life, in terms of changing my practices with regard to health care. But changing my political views isn't likely to lead to an actual difference in my circumstances, because I am not in charge of national health care policies. So to me, this kind of posting is reasonably helpful.
It's true though that not many people are open to unbiased information about health care, as Robin has pointed out in his other papers. It is a remarkably resistant bias. I find that seeing that graphically illustrated in the comments here has been quite interesting and informative.
I agree with the palm readers bit. Gambling is a bit of a weird case, but the house advantage is well known, so there's really no deception there. (I'd prefer that slot machine owners be required to disclose the odds of every prize-winning outcome, though.) Finally, alcohol also has well-known effects, some of which are desirable, so there's no issue of fraud there either and it probably doesn't count as "irrational" utility in the way that lottery tickets are. (There are other reasons why a ban on ethanol consumption may be desirable, but let's not go there.)
I wish I could ban churches, but the consequences of doing so tend to be worse than the churches. ;)
it serves as a signaling mechanism that they care deeply
Mike (and, to an extent, Paul), that's a bit needlessly negative isn't it? Are you saying there's no chance that people actually want their loved ones to get better, and are prepared to expend whatever's necessary, even to the point of irrationality?
The desire to consume a certain amount of 'treatment' is not difficult to explain, and most likely has little to do with the efficacy of the treatment in question. To put it one way, people look for treatment to feel better, not to get better. Hence all those well-documented placebo studies.
If my nearest and dearest fell seriously ill, I would probably not be at my most rational. I'd spend every penny I had on treatment in order of how likely it was to succeed. Throwing money away as a signal would not be a useful thought.
Fraud addendum: if you're banning people that sell 'irrational utility', after the faith healers, you should start on the palm readers, lottery organisers, churches, breweries....
Paul, I'm fairly certain that Robin would say, and it seems reasonable to agree, that consumers whose family and loved ones fall ill are, precisely as you say, hell-bent on consuming some fixed amount of "treatment" regardless of its efficacy because it serves as a signaling mechanism that they care deeply for the life and well-being of the one who is sick or injured.
Wait a second.
Assume someone has a fixed health care budget. They can allocate this budget to medicine or faith healing. Each dollar allocated to either source, assume, has a positive probability of producing some health benefit. That need not be very much -- there's likely to be a lot of waste in medicine. Let the respective probabilities be P and Q for a dollar on medicine or a dollar on faith healing. Let the proportion of the budget allocated to medicine be X. Then the patient's health utility (assuming constant units of health benefit) h(x)=PX+Q(1-X). Simple calculus will verify that so long as P>Q, the health benefit is strictly increasing in X.
And whatever else it's reasonable to think about the comparative benefits of medicine vs. faith healing, surely P>Q!
So I suppose the assumption that makes this really easy is the fixed budget. So let's drop it. We can assume that medicine is more expensive than faith healing for the same amount of "treatment" (measured somehow), perhaps because of moral hazards from insurance, etc. But then we have to ask why consumers can't be encouraged to drop ineffective medicine without replacing it with (more ineffective) faith healing. If we don't think consumers will reduce medicine use without a substitute in the form of faith healing, we must think consumers are hell-bent on consuming some fixed amount of "treatment" regardless of its efficacy. Why do we think that? Is it possible to correct for this tendency with some other social policy?
(Incidentally, on prayer as fraud, *charging* for prayer perhaps should be criminalized as fraud. We criminalize fortune-telling on much the same grounds. Free prayer doesn't defraud anyone. Perhaps only bad-faith prayer should be criminalized though -- there are good reasons not to criminalize people for things they genuinely believe -- although this would be hard to enforce by picking out the real from the false beliefs. The practical problems might be insurmountable. But in principle, sure, under some circumstances.)
So Robin, who does your faith healing?
Peter, I respectfully disagree. I also would like to see more posts about how to improve incentives or techniques, but this blog post does something more useful than documenting the existence of biases. It also proposes that we could save money by siding with factions that want universal health insurance expanded to not just cover dubious medicine practiced by establishment doctors, but to dubious medicine practiced by the subset of "alternative healers" that are cheaper than establishment doctors. That's an interesting, innovative idea, that might be a good idea, too. Because saving money in the realest sense can save lives, including, most importantly, my own. Although I agree that it only detracts from the good idea to express it to us in the form "Oh my god, my stupid, biased students".
Robin should give us more posts like this: they definitely reveal people's biases, and might in the long run encourage people to start to overcome themI find it hard to believe posts like this will encourage people to overcome biases. If his students, with a clear incentive (i.e. grades) to overcome their biased opinions of doctors, overwhelmingly resist, then why would blog posts (which provide virtually no incentive) do anything? It looks like the main effect of this post is to make readers more pessimistic, either about the effectiveness of pointing out biases, or the ability of this blog to identify biases. Maybe that pessimism is on average slightly more accurate, but it's hard to see much value in that increased accuracy.
So I'd like to see more posts about how to improve incentives or techniques, and only see posts documenting the existence of biases when there's some reason to think readers want to overcome those biases but aren't aware of them yet.
Personally, I'll take the chemical. ;)
I'm wary of anything marketed as "natural", as I tend to associate such marketing with "Our product doesn't have any real benefits over the alternatives, we just want to manipulate gullible people who don't know any better!" "Natural is good" is one of the most pernicious associations I know of, and one of the most blatantly false. There's plenty of perfectly natural poisons around, for example. Smallpox virus is natural. Dying during a difficult childbirth is perfectly natural, while the potentially lifesaving Cesarean delivery is extraordinarily unnatural.
I could go on, but I've already done enough thread-jacking as it is.
Bertrand Russell once wrote the famous words "I wish to propose for the reader's favorable consideration a doctrine which may, I fear, appear wildly paradoxical and subversive. The doctrine in question is this: that it is undesirable to believe a proposition when there is no ground whatever for supposing it true."
Believing things that are not true can be dangerous business. Even if the beliefs in themselves appear harmless, you are likely to base other beliefs on the faulty ones.
Incidentally, I think a problem with making faith healing illegal is that people sometimes gets more interested in the forbidden and exciting. Spend the money on education instead and trust people to make good choices.
I just recently learned via La Rana that the placebo effect may be largely a myth.
@Caledonian:
"Not buying anything.
That's the proper response to discovering that doctors aren't better than faith healers - not replacing the doctors with the faith healers!"
But this is not true, Caledonian. To continue my example, the dysthymic - the worried well - do cost society in terms of absenteeism, etc. due to their anxiety, mild depression, whatever. It is in society's interest to treat them; doing nothing has a defined cost here.
They can return to full function with the expensive essentially placebo treatment of prescribing psychiatrist + Prozac with likewise expensive side effects or with the inexpensive essentially placebo treatment of Bach's Flower Essences with no side effects.
And, biases being what they are, people will probably prefer to take a "natural" treatment from a nice homeopath than a "chemical" from the forbidding shrink, so society will probably get better compliance and faster recovery.
The random-number paper, of course. Know what's an even better deal than that? Not buying anything.
That's the proper response to discovering that doctors aren't better than faith healers - not replacing the doctors with the faith healers!