Tuesday, December 29, 2009

Bruce Schneier on Security Theater

I don't usually put up blog posts that are simply links, but I think for this essay it is worth doing.

"Bad News for Housing: Prices Flattening"

The quote is the headline of a story on CNN money. The author seems to take it for granted that it is good for housing to cost more, bad for it to cost less. I don't know if it has occurred to him that every time a house is sold, it is also bought, and that while a lower price is bad news for the seller, it is good news for the buyer.

Wednesday, December 23, 2009

Toleration vs Diversity

Both toleration and diversity are viewed by many moderns as good things. I am not sure to what extent people realize that they are to some degree conflicting goals, that toleration can be a threat to diversity.

The point occurred to me most recently in my readings on Jewish law. For about two thousand years after Israel ceased to be an independent polity, Jews, scattered around the world, continued to live under Jewish law. Gentile rulers found it convenient to subcontract the job of ruling, and taxing, their Jewish subjects to the local Jewish communal authorities. One result was to preserve the differences between Jewish culture and the culture surrounding the communities of the diaspora.

What changed that was emancipation—the shift, beginning in the late 18th century, towards treating the Jewish subjects of Christian countries just like everybody else, as Italians or Germans or Frenchmen rather than as Jews living in Italy or Germany or France, as natives rather than resident aliens. Seen from some angles it was a large improvement. But from the point of view of cultural diversity as a good, it was a catastrophe. Jewish law, in particular, ceased to be a living, functioning legal system providing the legal framework for millions of people and became instead a combination of an intellectual game and a legal system applying to a limited subset of activities and enforced only by belief.

In the U.S. the process was almost total, which is why the fact that my ancestors were Jewish is only a minor element of my identity. Elsewhere it is still incomplete. I still remember, traveling in Europe as a graduate student, a conversation with a group of European strangers at (I think) a youth hostel. They wanted to know where I was from; I told them I was an American. Oddly enough, one of them asked to see my passport, so I showed it to him. At which point he told me that he was (I'm making up details--this was about forty years ago) French the same way I was American and another of the group was Italian and ... . They were all Jews, had presumably deduced from my name on my passport and other less obvious signs that I was Jewish, and to them that was a stronger identifier than nationality. And, as evidence that the emancipation was a slow process from the other side as well, I am told that it was not until sometime after the end of World War II that Swedish law changed to make it possible for a Jew to be elected to the legislature.

The same point occurred to me earlier in my studies of different legal systems. Gypsies, for about a thousand years, have maintained their very distinct cultural identity, including multiple distinct legal systems, despite being scattered as a minority through non-gypsy lands. As I interpret my readings on the subject, that identity is now under threat in the U.S. and Canada—because we are too tolerant.

The ultimate punishment under Gypsy law, in most times and places, was ostracism from the Gypsy community. That was a potent threat for people who believed that all non-gypsy were marginally human slime and (correctly) that the attitude was reciprocated by the non-gypsies surrounding them. It becomes less effective in places where gypsies, especially young gypsies, who are unhappy with the constraints of their own culture have a realistic option of merging into the surrounding culture. One result has been pressure on Gypsy institutions to relax their own constraints, under threat of losing control over their own people.

The point of this post is not to argue that it would be better if Americans hated Gypsies or Europeans saw Jews as aliens. Only that it would be different, and that one of the differences is one that many people see as good.

Tuesday, December 22, 2009

Jewish Law, Oaths, and Expensive Religious Rules

Some religions impose costs on their members—shaved heads for Hare Krishnas, keeping kosher for orthodox Jews, fasting in Ramadan for Muslims. Many years ago I heard a talk by Larry Iannacone, an economist who specializes in the economics of religion, exploring the reasons for such requirements. The puzzle he raised was why they survive in an environment like modern day America, where there is a free market in religions. Why isn't a religion that imposes such costs outcompeted by a variant that keeps everything else but leaves out the unnecessary burden?

His (interesting) answer to that puzzle is probably somewhere in his published writings; readers are invited to find it for themselves. What reminded me of that talk was an interesting feature I noticed in my study of Jewish law—the role of oaths.

There are many situations in which, due to the lack of evidence, a legal dispute comes down to "he said/he said." The usual resolution, in Jewish law as described by Maimonides, is by an oath. One of the two parties—which one depends on the details of the case—is asked to swear that his account is true. If he is the plaintiff, he gets to "swear and take." If he is the defendant, he gets to "swear and be quit." Either way, if he swears he wins the case, if he declines to swear he loses it. The nature of the oath he must swear depends on the nature of the case. Pretty clearly, some oaths are considered more serious than others.

Why would someone refuse to swear if the result is that he loses his case? The obvious answer is that the parties to the case are believers who either feel obligated to swear truthfully, fear supernatural punishment if they swear falsely, or both.

A "suspect party" is someone who is not permitted to swear and so, in such a situation, automatically loses. One reason to be suspect is that you have sworn in the past to something that later turned out to be false. Another is that you been observed in the past to violate religious rules—for example by eating food that was not kosher.

This suggests a function for religious rules that impose costs. They act as a filter, a signal, a way of distinguishing people who really believe in the religion from ones who don't. Knowing whether someone really believes in the religion can be valuable information. It tells you, for instance, whether to believe his oaths.

Status and Evolutionary Biology

Economists tend to judge in absolute terms. If my real income doubles that's a big win for me—even if yours triples. We find it odd and annoying that other people often prefer to look at relative measures. If the income of poor people doubles and the income of rich people triples, many will see that as the poor losing out, or at least falling behind. In Choosing the Right Pond Robert Frank, an original and interesting economist, explored the implication for economics of the fact that people care about relative as well as absolute outcomes. That fact suggests an obvious question: Why do we care about relative outcomes? To explain why humans are as they are, the obvious tool is evolutionary biology.

Humans, like other living creatures, are "as if designed" for reproductive success. Reproduction requires two sorts of inputs—resources and a mate. If you are a better hunter than I am you will get more resources than I will, making you better able to feed your offspring. But that doesn't make me any less able to feed mine. From our standpoint as members of a hunter/gatherer tribe, the environment in which humans spent almost all of their evolutionary history, game is out there to be hunted in effectively unlimited supply.

If you are a better hunter than I am, more generally if you have more resources, status, whatever matters in our society than I do, you will also be better able to attract a mate. Mates, unlike game, are in strictly limited—for practical purposes, perfectly inelastic—supply. The better you are at attracting one or more, the worse my chances of doing so are. That is a good reason for me to be concerned about relative as well as absolute results, to wish not only that I should succeed in the hunt but that you should fail. It might even be a reason for me to put some efforts into increasing the chance of your failing if suitable opportunities arise.

The late George Stigler taught me an important lesson when he rejected the original version of what was to become my first published journal article in economics. He told me that in order to be publishable, the article required not only a theory—in my case of the size and shape of nations—but also some way of testing that theory. In revising to meet that requirement I not only found evidence in support of my theory, I also, and perhaps more important, was forced to think through more carefully and precisely what the theory said.

I have done no testing of my theory of why we care about relative status, but I do have predictions. The first is that males should be mainly concerned about their status vis a vis other males, females about theirs vis a vis other females, since males are competing with males for mates, females with females.

The second is that males should be more concerned with relative outcomes than females. Reproductively speaking, wombs are a scarce resource, sperm is not. Even a not very successful female can expect to reproduce, although her success in mate search may determine how much help she gets raising her children. An unsuccessful male is likely to have no children at all, a successful one many. From the standpoint of reproduction, being male is a high risk gamble.

The third prediction is that people should be most concerned about relative outcomes in a range near their own level. If Bill Gates increases his wealth from twenty billion dollars to thirty billion, that has no effect on his ability to compete with me for mates; insofar as wealth is the relevant criterion, at twenty billion I've already lost, although that would be less true in a polygynous society, where his extra wealth might result in his bidding a few more potential wives away from me. If a homeless man finds a job at MacDonalds, that has no effect on his ability to compete with me for mates either. The people I ought to worry about—supposing that I am a male in the mate market, as most males were for most of their adult lives in the environment where we evolved—are the men at about my level, the ones who might beat me out in courtship if they were a little richer, or a little handsomer, or ... .

This is a blog post not a journal article; I haven't actually done the research to test these predictions, although I wouldn't be surprised if someone else has. If any readers know of such ... .

(It has been many years since I read Robert Frank's book and I don't have a copy ready at hand to check; it's possible that he came up with some or all of my explanation first.)

Tuesday, December 08, 2009

An Entrepreneurial Proposal

Many museums, such as the Metropolitan Museum in New York, sell replicas of some of the historical jewelery in their collection. Typically the quality of the replica is significantly lower than the quality of the original—cast when the original was constructed, sometimes using glass instead of the original gemstones. Typically the replicas are expensive.

For quite a long time, I have been seeing imported jewelery, usually in silver, coming from places such as Bali and India, with a quality of execution comparable to that in historical pieces—precise filigree, some of it possibly done by the fusion/colloidal hard soldering technique developed in antiquity to do fine filigree and granulation without having the details blurred by solder. Such jewelery is, materials aside, better than the museum replicas—and much less expensive.

This suggests an interesting possibility for an entrepreneur with an interest in historical jewelery and suitable contacts somewhere in the third world. Put together, and web, a collection of pictures of pieces of historical jewelery. Locate craftsmen willing and able to make copies of those pieces. Offer to make, for online customers, any piece in the collection, at a suitable price. For a somewhat higher price, guarantee never to make another copy of the same piece.

This particular example occurred to me because I happen to be interested in historical jewelery. But there must be many other market niches of the same sort, categories of goods for which the combination of online marketing and hand-craft technology would make it possible for customers to get unique items of special interest to them, while providing profitable work for craftsmen in low income parts of the world.

More Fun with Jewish Law

I've been reading Maimonides and came across two things that I found interesting.

Part I

Suppose you kill someone who is dying of a lethal disease. Maimonides concludes that that isn't really murder, since he would have died anyway—while pointing out that you have to be really sure he was dying of a lethal disease.

Now suppose someone who is dying of a lethal disease kills someone else. If, being a helpful sort, he commits the crime in the presence of the court, he has committed murder and can be convicted of doing so. If, however, he only commits the murder in the presence of witnesses, there is a problem.

Witnesses, in this case or others, might lie. In other cases, one thing discouraging them from perjury is that if it is discovered that their false testimony led to the execution of an innocent defendant, they will be found guilty of murder and themselves executed. But if their testimony leads to the execution of an innocent defendant who is himself dying of a lethal disease, they won't be executed, because killing someone who is dying of a lethal disease isn't murder.

Since the witnesses are not at risk of execution for perjury, they might commit it, so their testimony can not be trusted—cannot be taken as sufficient evidence to convict someone of murder. So if someone who is himself dying of a lethal disease commits murder, and doesn't do it in the presence of the court, he cannot be convicted.

There is a certain beautiful logic to this very screwy result.

Part II

In Maimonides' discussion of what we would call tort law, he considers a number of borderline cases—cases where it is not clear whether the tortfeasor owes the victim a damage payment equal to half the damage or a quarter of the damage done. His conclusion in such cases is that the court can only award the plaintiff quarter damages. If, however, the plaintiff has seized property of the defendant amounting to half damages, the court will not make him give it back.

Part of what is going on here seems to be a rule holding that the court will not transfer property unless it has good reason to do so. It can't award half damages, because it isn't sure that more than quarter damages are owed. But it can't make the plaintiff who has acted on his own to collect half damages give part of the money back, because it isn't sure that half damages aren't owed.

A different way of looking at this is that it represents a hybrid of a conventional legal system, with action by the state or analogous authorities, and a feud system, in which parties act on their own, within some set or explicit or implicit rules, to enforce their rights. I get the same impression looking at the legal rules applied to killing. Under some circumstances, a killer cannot be convicted and punished by the court. But the "avenger of blood," the kinsman of the victim who, in a feud system, would be expected to avenge the killing, can kill the killer with impunity. His right to do so is complicated by various rules, in particular the existence of cities of refuge; once the killer gets to one of those he is in theory safe.

I should probably add that Maimonides is writing at a time when there are no cities of refuge and have been none for a thousand years or so. Substantial parts of his legal code describe what the rules were back when the kingdom of Israel was a going concern and the Temple still standing. One possible explanation is that he believed that that situation was going to be reestablished in the not too distant future—so legal scholars ought to be prepared.

Sunday, December 06, 2009

AIDS in Africa: Disturbing Evidence

In the U.S. and Europe, AIDS transmission via vaginal intercourse seems to be very low, with sexual transmission occurring mainly via anal intercourse—one reason why the infection rate is much higher among male homosexuals than in the general population. It is widely believed that this is not true in Africa, that, due perhaps to the prevalence of genital sores, vaginal transmission rates are high enough to provide much of the explanation of the very high rates of AIDS infection.

As a result of references in an online discussion, I recently came across two published articles which offer evidence that this explanation is wrong, that vaginal transmission rates in Africa are not substantially higher than elsewhere. They go on to suggest that what is really going on may be iatrogenic, doctor caused, disease, that much of the transmission may be due to sloppy medical procedures, in particular the reuse of needles for injections. The evidence is in part from the pattern of infection—rates are apparently much too high among young people who have not had sex and whose mothers are not HIV positive, suggesting a non-sexual transmission mechanism. In part it is from studies that try to measure the transmission rate via vaginal intercourse. In part it is from regional patterns that don't fit the patterns of the supposed causes.

The articles are:

Evidence of iatrogenic HIV transmission in children in South Africa

and

HIV infections in sub-Saharan Africa not explained by sexual or vertical transmission

My non-expert impression is that they represent serious scientific work, and that the evidence presented is pretty convincing. The implication is that this is a case of people trying to do good and doing harm instead, always disturbing. Does any reader know of later work either confirming or rebutting the argument?

Dishwasher Woes

A while back, we replaced our dishwasher, which came with the house when we bought it some fourteen years ago. The new one, selected on the basis of a positive online discussion of the previous model in the (Bosch) line, turned out to be in almost all ways worse than the old. It held fewer dishes, cleaned less well, dried much less well. It's only significant advantage, so far as we could see, was that it was quieter. It is bad enough so that we are considering simply throwing out our new dishwasher and replacing it with another, after doing a more thorough job of research.

It occurred to me to wonder whether part of the problem had to do with pressure, either from the market or from regulation, for energy efficiency. The external dimensions of a built-in dishwasher are fixed. One way of making it more energy efficient is by putting on more insulation to make it easier to keep things hot while they are being washed—which also makes it quieter. More insulation is likely to mean thicker insulation, which means less space for dishes. Along similar lines, the new dishwasher, unlike the old, doesn't have the option of hot air drying—dishes are dried (or not dried) only by the residual heat from the washing. That saves energy, but makes the dishwasher a good deal less useful.

Does anyone reading this know enough about dishwasher engineering to say whether new dishwashers are, typically, worse than old for these reasons? Whether, if so, the problem is energy efficiency standards set by regulation, or merely the advantage of being able to advertise energy efficiency and low noise?