David D. Friedman's Blog, page 5

Weight Propaganda

A recent article in the Journal of the American Medical Association reports that being overweight, as defined by body mass index, may be good for you—that people in the recommended BMI range are more likely to die ("all cause mortality") than people whose weight classifies them as overweight but not obese. What I found most interesting about the news coverage of the article was the reaction reported—people quoted as criticizing the article without offering any good reason to think it was wrong. From the USA Today story:

Walter Willett, head of the department of nutrition at the Harvard School of Public Health, says the findings are "complete rubbish" because the methodology used in the analysis seriously underestimates "the hazards of being overweight and obese."

Steven Heymsfield, one of the authors on the accompanying editorial in the journal and the executive director of the Pennington Biomedical Research Center in Baton Rouge. "We don't really know the ideal weight for a long life and optimal health. Science is still working that out. But falling in the normal, healthy weight range is still the safest place to be."

Other people offered possible ways of explaining away the result, such as the suggestion that overweight people got more medical attention, but no actual evidence. My impression was that in this case, as in the case of evidence in favor of the moderate consumption of alcohol that I discussed some time back, there is an official truth and a tendency to discount evidence against it. The Heymsfield quote is a nice example of one way of doing so, since it appears to endorse the orthodox view of what people should do while actually saying nothing: Falling in the healthy weight range is the safest place to be—and we don't know what the healthy range is.
A related point is the popularity of the body mass index, along with the use of objective sounding terms ("overweight," "obese," "obesity grade 1," ...) for arbitrary categories. The nice thing about BMI is that it is easy for anyone to calculate, since it is merely weight divided by height. The problem is that it is a poor measure, since people differ in other relevant dimensions, most notably in how wide they are. If your objective to produce accurate information relevant to health, you would want to take that into account. But doing so makes it harder to pressure other people into losing weight, since the more complicated the measure, the easier it is for people who don't want to diet to explain away their weight.
The friend I am currently visiting with tells me that he was offered, by both a doctor and a physical trainer, a simple rule for telling whether you are overweight: Take a deep breath and jump in the water. If you float you weigh too much. The theory, presumably, is that the test measures your average density, that fat is less dense than muscle which is (I'm guessing) less dense than bone, so the test is measuring the relative amount of each in your body. It doesn't work for women, since even women who are not overweight are likely to float.

The Net Value of Video Games

One of the Christmas books I received this year (from my younger son) was John Dies at the End by David Wong, an author whose blog post I recently linked to. I liked it enough to finish it, not enough to want to read the forthcoming sequel. One of the things I didn't like was the degree to which the central characters seemed to be acting irrationally, along with consuming considerable quantities of alcohol and, in one case, drugs. Another was the degree to which the whole picture did not entirely make sense and the feeling that that was not an issue the author cared much about. It occurred to me that perhaps my response was the flip side of the objection some readers make to my fiction, that everyone, and everything that happens, is too rational. Partly, I suppose, that is a disagreement about what people are like, partly about what they should be like, partly about what is interesting or entertaining about other people's behavior.
One of the throwaway lines in the book was the suggestion that violent video games are an alien plot introduced recently to human society by agents from an alternate timeline capable of making the introduction retroactive by changing our memories to fit the new reality. Which started me wondering ...  .
My natural prejudice as an economist is to assume that people act in their own interest, hence that spending time playing video games is a net benefit, at least to those who play them—a prejudice perhaps reinforced by the amount of time I have myself spent playing and enjoying video games (mostly the computer versions). But people are not entirely rational, and the designers of successful computer games, like successful creators of earlier forms of entertainment, are presumably skilled at taking advantage of the irrational elements in their customers' behavior. Which suggests several questions:
1. As compared to earlier forms of mass entertainment—novels, movies, television—are video games better or worse for their consumers? Are you more likely to end up failing out of college, losing your job, breaking up with your girlfriend, through devoting too much of your time to playing video games than from the earlier equivalents? As one very mild piece of evidence, I offer my own observation, long ago, that the way you knew a computer game was really good was that when you took a break to go to the bathroom, it was because you really, really had to go. I have never been much of a viewer of movies or television, but I have read a lot of books and do not remember a comparable effect for them. 
On the other  hand, it's easier to read in the bathroom than to play (most forms of) computer games there. 
2. The same question, with regard to the effect on others. The implication of the line in the novel was that video games were designed to coarsen sensitivities, make us more tolerant of people being killed, dismembered, tortured, generally mistreated. I am not sure that is any more true of them than of comic books and thrillers, but I suppose one could argue that the visual element, and the involvement of the player in the plot, makes a difference. You are not just watching someone else engaged in mass mayhem, you are doing it yourself.
But then, one of the reasons thrillers are thrilling is that the reader is imagining himself as the protagonist.
3. What about positive effects? Novels and films can educate (or miseducate) you about history, geography, human behavior. So can computer games. Quite a large fraction of my son's knowledge of geography and history comes from playing war games; he  taught himself to type at a young age in order to communicate with fellow players online and learned to spell so as not to look stupid while doing so. If anything, the interactive nature of computer games ought to make them more educational than earlier equivalents, since doing things wrong, failing to see the logic of the situation, sometimes results in losing, which is less fun than winning.
All of which leaves out the big question which I have discussed in the past: In what sense is doing things in virtual worlds less valuable than doing them in the real world?

Christmas Books

First, some biased suggestions for presents:
How to Milk an Almond, Stuff an Egg, and Armor a Turnip: A Thousand Years of Recipes
My and my wife's book on medieval and renaissance cooking—more than three hundred recipes, each with the original and how we do it, plus a variety of related articles. Just the thing for the cook in your life—or the medievalist.
A Miscellany
This is the bigger book that the previous one is the cooking section of. Recipes, articles on how to make a pavilion, portable period furniture, a Germanic lyre, lots of other things, along with a good deal of my poetry and essays on historical recreation and related matters. Good if any of your friends are into historical recreation from the Middle Ages and/or Renaissance, especially through the Society for Creative Anachronism, a long term hobby of ours.
Future Imperfect: Technology and Freedom in an Uncertain World
My most recent nonfiction book, good for futurologists, science fiction readers, and the proverbial intelligent layman. For whom my 
Hidden Order: The Economics of Everyday Life
and
Law's Order: What Economics Has to Do With Law and Why it Matters
are also possibilities.
But the one I really want you to read, because I want more comments on it, is my second novel, Salamander. Extra credit for physics and math types if you can figure out what my version of magic is modeled on. 
For my unbiased recommendations, here are some of my past Christmas books—each the book that, for one year's Christmas, went to anybody I couldn't think of anything else for because I thought it was neat.
Chimpanzee Politics
Thinking, Fast and Slow
Talleyrand by Duff Cooper. Mostly for the fascinating subject, but it doesn't hurt that the author was the one member of Neville Chamberlain's cabinet who resigned over Munich.
and the book that just occurred to me for this year's Christmas book, some of our presents being given a few days late ...
John Wilkes: The Scandalous Father of Civil Liberty
And, for a final recommendation, His Majesty's Dragon by Naomi Novik, and sequels—a series I am currently rereading.

Harsh Self-Help Advice

I just came across a webbed essay by someone I have never heard of that struck me as both well written and, on the whole, sensible, so thought I would link to it.

Is Heaven Worth the Price?

While picking up a prescription in the local drug store, I noticed a book on a rack of Christian literature entitled "Heaven is real." Which started me thinking ...  . 
One possible explanation of religion is that it is wishful thinking. People do not want to die, so they want to believe in life after death. One problem with that explanation is that several of the most successful religions include both Heaven and Hell. The quiet of the grave does not sound very attractive compared to an eternity of bliss. Compared to an eternity of torture, on the other hand, there may be much to be said for it. So why are people attracted to a system of belief that offers the possibility of the former but also the risk of the latter? How high do the customers have to believe the risk is before they would prefer not to buy? Putting it in the jargon of my field, what are the relative Von Neumann utilities of Heaven and Hell?
Heaven and Hell make much more sense as an incentive system, promised reward and threatened punishment as a way of getting people to follow the dictates of a religion. That is a good reason why some people would want others to believe in them. But it does not explain why people choose to themselves believe in them. Perhaps wishful thinking is not, after all, the right explanation.
For an alternative explanation, see an old post of mine that started with the same puzzle.

Why Are Law Schools Expensive

There has been a lot of concern of late in the law school world over falling numbers of applicants, poor employment opportunities for graduates, high debt loads, and associated problems. I recently came across a post on another blog discussing the question, and decided that a post here would be more appropriate than a very long comment there.
From the standpoint of potential law school applicants, there are two problems—a shortage of jobs for lawyers, relative to the number of graduates, and the high cost of law school. The current administration will probably help with the former problem, since an increase in the size and intrusiveness of government is likely to lead to an increased demand for lawyers. The purpose of this post is to discuss the latter.
The fundamental problem, as I see it, is with the incentives facing the schools. Law schools are heavily dependent on their reputation to attract students. The two biggest sources of information available to the students are the American Bar Association, which accredits law schools, and the annual U.S. News and World Report ranking. Both of those are based mainly on measures of inputs, not outputs. Thus, for example, the ABA recommends a student to faculty ratio of no more than twenty and takes a ratio of thirty or more as presumptive evidence that the school does not meet the standards for accreditation. Its rules for calculating the ratio count one adjunct as one fifth of a tenure-track professor and it requires that "substantially all" of the first third of a student's coursework be taught by the full-time faculty. The standards include a lengthy list of what must be in a law school library—almost all of which is material currently available to both faculty and students online.
So a school that chose to spend less on its library, have a higher student to teacher ratio, use more inexpensive adjuncts and fewer tenure-track professors, do a variety of other things to cut costs, would risk losing its accreditation, whether or not it was doing a worse job of teaching its students. A school which provided its education in any form other than the conventional number of hours sitting in a classroom would lose its accreditation, since one of the ABA requirements is that:

 A law school shall require, as a condition for graduation, successful completion of a course of study in residence of not fewer than 58,000 minutes of instruction time, except as otherwise provided. At least 45,000 of these minutes shall be by attendance in regularly scheduled class sessions at the law school.

U.S. News and World Report does not publish the details of its ranking system, but a number of people have reverse engineered it. The four factors that together predict the ranking almost perfectly are peer reputation, fraction of graduates employed nine months after graduation, student-faculty ratio, and undergraduate GPA of the students.
The ABA includes in its requirements a measure of what fraction of students pass the bar. That, the USNWR employment measure, and the peer reputation measure, are output measures. Unfortunately, they are not very informative ones.
Start with peer reputation. A professor at one law school is unlikely to know much about how good a job other law schools do educating their students. What he is much more likely to know, and base his opinion on,  is what prominent scholars in his field are at which school—information almost entirely irrelevant to most students. And this criterion has the unfortunate side effect of giving each school an incentive to barrage faculty at all other schools with glossy pamphlets boasting the activities and accomplishments of their own faculty, an expense that does nothing to improve the education of their students.
Bar passage and (very imperfectly measured) employment rates are more relevant. The problem with both of those is that they depend on two different inputs—quality of instruction and quality of students. Top schools get very high bar passage rates not because they do a particularly good job of teaching the skills relevant to bar passage—most of their students take an additional bar preparation course before taking the exam—but because they admit only smart students.
In an old post, I proposed a simple solution to this problem. Schools should report their bar passage rates as a function of some measure of student quality such as LSAT or undergraduate GPA. That would provide the potential applicant with the information that matters to him—what the chance is that a student of his ability who goes to that school will pass the bar. A similar approach could be used for employment statistics.
There are two possible approaches to reforming legal education to lower its cost. One is to recommend specific changes, such as Judge Posner's old proposal to make the third year of law school optional. The other is to recommend changes in the incentive structure that currently prevents such specific changes from being in the interest of law schools to make. At the level of the individual law school, the first is all that can be done—but, short of crisis, mostly will not be, because it is not in the interest of law schools to do a better job for their students at the cost of risking their ABA accreditation and USNWR ranking. 
At the level of the legal education profession, I think the second approach makes more sense. It too, however, faces incentive problems. One effect of the current ABA standards is to increase law school demand for tenure-track faculty, and tenure track faculty have a substantial influence over those standards.

---
P.S. The post that inspired this one has a delightful comment by a law school student about to take his last exam, detailing what is really need to teach law. His bare bones law school ("Perhaps you feel that your students cannot survive without a cafe? Build/rent your school next to a Panera") would cost students about $10,000 a year. Combine that with Posner's proposal for a two year degree and you have the cost of legal education down to $20,000.

But I don't think the ABA would accredit it.

A Modest Request

I have just spent an hour or so on yard work, my usual form of exercise. Much of it consisted of pulling out self-seeded privet, of which my yard produces an inexhaustible supply. I also got rid of some ivy, ditto. Which suggests ...
If someone who does genetic engineering wants to make himself really useful, he should look into engineering vegetable pests to make them good for something. If only my yard grew a variant of ivy whose leaves made a tasty and nutritious lettuce substitute, I would have a lifetime's supply of salads. I am not sure what can be done with privet, other than letting the trees grow up and turning them into lumber, which would make the yard unavailable for its present function of growing fruit trees. But perhaps someone can think of something.

Unsolicited Ad for Genetic Testing

Some time ago, my elder son persuaded me to pay for genetic testing by 23andMe. They send you a test kit, you provide some saliva, they test it. They then tell you what the genetic information implies about medical problems you are more or less likely than average to have and where your distant ancestors come from, and make it possible, if you wish, to get in touch with putative relatives, people who have also been tested and whose genetic information suggests common ancestry not too far back. 
They also invite you to answer a bunch of questions about yourself designed to generate additional information about what genetic characteristics correlate with what outcomes, information that can then be used to, among other things, better inform their other customers. Thus getting tested not only provides some private benefits, it also increases the existing store of information about the results of different gene variants, which strikes me as a good thing to do.
I have two pieces of evidence that their service is real. The main one is that they correctly identified my son as my son. The minor one is that they told me I had an above average chance of a particular sort of tumor. The information was not useful, however, since it arrived after the tumor had been diagnosed and removed.
I recently got an email from them, announcing a sale—$99 for their services, which I think is what I paid but is less than their standard price. I thought some of my readers might be interested, hence this post.

Observations of Film Making

Quite a long time ago, a libertarian by the name of J. Neil Schulman wrote a novel, Alongside Night . He is currently in the process of turning it into a movie. He asked me to play a bit role, and I have just returned from doing so.
The invitation did not reflect any misguided belief in my acting ability—the role consisted of playing the King of Sweden for about ten seconds in a simulated Nobel Prize award ceremony. The reason he wanted me, pretty clearly, was that his protagonist is the son of a Nobel Prize winning free market economist. So am I—and I expect Neil believes he can get a little free publicity out of the parallel. He told me, many years ago, that the father's personality is actually based on his father, not on mine, but I do not expect that to be obvious to the random viewer or reviewer.
The main payoff for me was the opportunity to spend a day or so observing the process of movie making and chatting with the people involved. I learned a number of interesting things.
Perhaps the most interesting was from a conversation with the costume person, who was making sure that the very formal outfit he had rented for me would fit. By his account, his job is not simply providing costumes from the right date, nationality, social class and the like.  As a character moves through the plot line, different shades, textures, appearance of clothing reflect changes in his role, mood, personality.What the costumer is doing, as he sees it, is creating a work of art one of whose dimensions is time.
Another feature of the process, one which I had only partly allowed for, is how much of a patchwork it is. There is no attempt to start filming at the beginning and go on to the end;  one of the people I talked with said that the last person who made a film that way was Alfred Hitchcock and that doing so was unconventional even then. The approach instead is to shoot individual scenes, each of them many times over. The order in which they are shot is determined by considerations such as which actors are in them or what set they are being shot on. 
I was told that, for a low budget film like this one, the filming typically takes four weeks or so. Assembling the movie from the output of those weeks takes something more like four months. Most of the assembly requires only two people, while the filming seems to require a total crew of about thirty.  The finished film will be about one percent as long as the time spent filming it—and the ratio of filming time to film time is substantially higher for a higher budget production.
Another point that struck me about the experience was my own reaction to the dialogue. My natural inclination, as an author and public speaker, was to critique it, to notice places where what the character said could have been said better. In some cases I may have been right. But I suspect that in others, my critique was really of neither the scriptwriter nor the actor but of the character. What he said could have been said better—but would not have been by that character in that situation. I was reminded of my own dictum after writing my first novel: No plot survives contact with the character. For the same reasons, the author's words ought to change when put into the character's mouth, because the character is not the author and will not say things in the same way the author would.
Which may be relevant to my own writing. One of my weaknesses is a tendency for my characters to sound too similar; I have not yet got the trick of giving each of them his own distinctive voice. Part of the solution may be to remember that they will not necessarily say everything, state every argument, in the best possible way.
It was an interesting twenty-four hours or so. I look forward to seeing, sometime in the next year, how the movie turned out.

Jury Nullification and the Enforcement of the Juror's Oath

In a recent post, I discussed the issue of jury nullification, the question of whether a juror ought to vote for acquittal if he believes that the defendant is guilty, but of something that ought not to be a crime. I mentioned being dismissed from a jury some time back as a result of telling the judge that I might do so.
One commenter suggested that my mistake was not that I was willing to vote for acquittal in such a situation but that I told the judge that I was, that I ought to have said I would follow the judge's instructions and then voted for acquittal if the law that the defendant was accused of violating was one I disapproved of. His point was that if I could, at no great cost to myself, keep someone from being jailed for something he did not deserve jail for, I ought to do so.
Another commenter pointed out that, as a juror, I would be required to swear to follow the judge's instructions with regard to the law. Falsely swearing would be perjury, a criminal offense. Which raises two questions ...
The first is a moral question—ought I to be willing to perjure myself under such circumstances. I think the answer is that I should. I am generally unwilling to lie to people, but this is a special case, analogous to lying to a mugger about what money I have on me. I do not regard government as a source of moral authority, so a government trying to imprison someone for (say) smoking marijuana deserves to be treated like anyone else trying to violate rights. I would be uncomfortable lying to a judge under oath and might do it badly, but I do not think doing so would be wicked.
The second is a practical question—how likely is someone who swears to vote according to the judge's instructions on the law and then deliberately fails to do so to get into legal trouble as a result. My guess is that a juror who limits himself to telling the other jurors that he is not convinced of the defendant's guilt and so unwilling to vote for conviction would be pretty safe—unless he had previously put up a blog post defending jury nullification, or in some other way provided clear evidence of what he was doing. Perhaps even then.
But all one vote for acquittal can do is produce a hung jury; if the prosecution is determined to convict, it can always try the defendant again. A more ambitious project would be to try to persuade the other jurors to vote for acquittal on the grounds that what the defendant had done ought not to be illegal. Doing that would produce evidence that the juror had perjured himself in swearing to follow the judge's instructions on the law.  
I'm curious as to whether, in practice, jurors who do that get prosecuted for it, and if so how often. 
One further point occurs to me. In discussing the risks of jury nullification in my earlier post, I took it for granted that it would be used to prevent the conviction of someone guilty of something the juror thought ought not to be a crime. It could also be used to convict someone who was innocent of the crime he was accused of but belonged to a group that the juror disliked. 
I think that is less of a problem, for two reasons.
To begin with, a single juror can't convict; the most he can do is produce a hung jury. So if only a few jurors share the dislike and the willingness to act on it, the result is not to convict the defendant but only to give the prosecution an opportunity to retry him. That might impose serious costs on the defendant, especially if he cannot offer bail, but less serious than conviction.
What about a situation where almost everyone dislikes the group the defendant is a member of and wants to use the legal system against them? In that case, jury nullification could convict the innocent defendant. But it probably isn't needed, since in that situation the government will almost certainly share the dislike and have other ways of acting on it.