Revealed Preference

In an earlier post I mentioned Frans de Waal, author of the very interesting book Chimpanzee Politics. I just came across an online quote from him on a different topic—ending with:

"And so I usually use it as an example I don’t trust questionaires at all ... you need to pay attention to behavior, behavior is the only thing that tells us what the real preferences are."

So far as I could tell from his book, de Waal does not have a background in economics. He does, however, inhabit the same world we do, and so has reinvented the principle of revealed preference.

Which reminds me of a quote from another author I am fond of: "Any spoke can lead an ant to the hub."

Medical Prophecy

If you do a web search for information on some life threatening illness or risky operation, you are likely to come up with information in the form of survival rates: Of people who have this operation, X% are still alive five years later. Assuming I correctly understand where it comes from, such information is wrong—in a predictable direction.

My assumption is that such information is generated by looking at patients who had the operation at least five years ago—say between five and ten years—and seeing how many of them were still alive five years later. That gives a survival rate, but it is the survival rate as of five to ten years ago. Medical technology has been improving in recent decades, which means that the survival rate for someone who has the operation today is probably better, perhaps significantly better, than for someone who had it five years ago.

One way of getting a better estimate would be extrapolation. Calculate the survival rate for patients who had the operation five years ago, six years ago, seven years ago .... fifteen years ago. Assume that whatever rate of improvement you observe continued; fit the data with a straight line or simple curve. See what extrapolating that line tells you about the risk if you have the operation tomorrow.

Can someone better informed than I am about medical statistics say whether I am correct about how they are calculated, or whether something like what I have just described is current practice?

Why Are We Different?

In an earlier post, I discussed Judith Harris’s first book, The Nurture Assumption. She has now written another one, and it too is interesting.

The subject of No Two Alike is a simple and striking puzzle. Human personality, so we believe, is the product of genetics and environment. It would seem to follow that two individuals with the same genetics and the same environment should have the same personalities.

They do not. Identical twins raised together end up significantly different—about as different as identical twins raised apart. Identical twins physically attached to each other, never separated in their lives, have substantially different personalities.

About two-thirds of the book is devoted, not to solving the puzzle, but to knocking down solutions, eliminating red herrings. It is an interesting account, especially the story of one prominent book in the field that Harris pretty clearly believes—although she does not quite say so—to be fraudulent. The strongest evidence for that conclusion is its author’s response to criticism: Legal threats to try to prevent publication and a refusal to provide anyone with the data needed to check the book’s factual assertions.

The final third of Harris's book sketches a theory of how human beings, from infancy on, deal with their social environment. From that theory she derives a conjecture about the source of human differences. It is an interesting conjecture and may well be true, but those chapters would be more satisfactory if the author suggested ways of testing her conjecture and offered some evidence in its support beyond the failure of alternative explanations.

That said, it is an interesting, intelligent, thoughtful book, and I recommend it.

My New Toy

Before Palm, there was Psion--a British PDA in the form of a miniature laptop. It went through a series of models, most of which I at some point owned and used. Being hinged along the long edge plus some very good design--perhaps combined with a little black magic--gave a keyboard on which it was possible to touch type, at least if you left out your little finger.

Psion lost out in its competition with Palm, abandoned the consumer market, and I switched to a Sony Clié. It was an elegant palm style (and Palm OS) pda, but in important ways inferior to its predecessor. Typing on the Psion was easier than writing by hand, typing on the Clié harder.

Somewhat over a year ago, Nokia announced two new smart phones, the 9300 and 9500. Like my old Psion, they are in the shape of miniature laptops. Like it, they run on the Symbion operating system. Unlike it they are cell phones as well as PDA’s, capable not only of making phone calls but of browsing the web on a 600x200 pixel screen.

Both were triband phones in the European frequencies. Eventually a U.S. version of the (smaller) 9300 became available and I got one.

It is better than my Clie and more useful than any other cell phone I have seen. It is significantly smaller than the smallest Psion. But ....

Being smaller means a smaller keyboard. Whatever the magic spell Psion used to make its little keyboards work, Nokia has not licensed it. It is a very nice machine, but less useful as a portable word processor than the machine Psion brought out seven years ago.

So if anyone from Nokia is reading this, what I want next is ...

Enough increased size to make a keyboard I can (just barely) use for real typing. A 600x400 screen, about equivalent to the smallest size of ordinary computer screen. I am willing to live with something a little bigger than most cell phones. It would still be a lot smaller than a laptop.

Surely I can't be the only one.

Hardwired Tribalism

I was recently part of a rather odd Usenet exchange. Someone had suggested that if Gore had been elected in 2000 he might, like Bush, have ended up invading Iraq. A poster who appears to be a committed partisan of the Democratic party objected that that was nonsense. I offered as evidence that it wasn't nonsense Clinton's cruise missle attack on a pharmaceutical factory in the Sudan, which was a sort of miniature of the Iraq invasion—responding to a terrorist attack by attacking a nation that had no obvious connection to it, with a bogus claim of weapons of mass destruction as justification.

The Democratic poster leaped to the conclusion that I was a Bush supporter, maintained that conclusion even after I explicitly denied it, and went through various contortions in order to avoid conceding that I hadn't said what he claimed I had said—he had converted my "response to 9/11" into "justified response to 9/11." His behavior struck me as particularly odd given that "Clinton behaved just as badly as Bush" is not an argument one would usually expect a Bush supporter to make—quite aside from various other unkind things about Bush I had said in other posts.

The only sense I could make out of it was that I was encountering a tribalistic view of the world. There are two sides, everyone who isn't on my side is on the other side, hence anyone who says something negative about the Democrats must be a partisan of the Republicans and any evidence to the contrary is to be ignored as experimental error.

Not long after, I heard a radio report about the French government caving in to the demands of demonstrators that they rescind legislation making it possible for employers to fire young workers. Oddly enough, part of my reaction was a feeling of satisfaction. The news implied a further decline of the wealth, power, and status of France, France is part of Europe, Europe is at the moment the obvious status rival to America, and I am an American. Speaking as an economist, my best guess is that the decline of the French economy makes me worse off, not better off. But to some part of my mind hardwired by hundreds of thousands of years of evolution in hunter/gatherer bands, there is only us and them, and anything that is bad for them is good for us.

Immigrant Education: A Query

One of the responses to my previous post cited Samuel P. Huntington, a scholar who argued in a recent article against Mexican immigration. Looking at the article, I was struck by an oddity in his statistics. He gives figures for educational attainment of first, second, third and fourth generation Mexicans, as of 1990. The odd thing is that overall educational attainment rises from first to second and from second to third, but then drops substantially for fourth generation Mexicans.

Elsewhere in the article, the author gives figures for the foreign born population of the U.S. as of 1960. He lists the five most common ethnicities--and Mexican is not one of them. Fourth generation Mexicans would be descendants of immigrants from considerably earlier than that--sometime before 1930--which suggests that there may not be very many of them and their circumstances might be quite different from those of later immigrants. Checking an old stat abstract, there seem to have been about 1.4 million Mexicans in the U.S. in 1930, or about 1% of the population. I don't know how many were foreign born.

Does anyone have data on who fourth generation Mexicans were as of 1990 and why, assuming Huntington's figures are correct, their educational attainment is relatively low?

Welfare and Immigration: The Flip Side of the Argument

Voluntary agreements between people now in Mexico and people now in the U.S.—my renting an apartment in California to someone currently living in Mexico, or hiring him to mow my lawn—benefit both parties. Standard economic arguments suggest that although there may be negative effects on third parties, such as someone else who wants the apartment or someone else who wants to cut my lawn, the net effect is positive. The standard arguments for freedom of association, contract, and trade apply to immigration as well.

Opponents of open immigration have an obvious counterargument: not all interactions are voluntary. A Mexican who comes here in order to mow lawns benefits us as well as himself. One who comes to collect welfare benefits himself, but at our cost. A common conclusion is that free immigration may be desirable in a completely laissez-faire system, even in the relatively laissez-faire America of a hundred years ago, but not in a modern mixed economy.

The strength of this argument is in part an empirical question. Immigrants may get things they do not pay for, but they may also pay for things they do not get. On past evidence immigrants tend to be predominantly young adults, a long way from collecting Social Security or Medicare. The question is a complicated one and I am far from sure that a correct answer would support the argument against immigration, but that is not the point I want to explore in this post.

What I want to explore instead is the flip side of the argument. The existence of a welfare state may indeed make open immigration less attractive. But the existence of open immigration also makes a welfare state less attractive—which, for those who disapprove of a welfare state, is an additional argument in favor of open immigration.

Consider the analogous argument applied intrastate. Supporters of higher levels of welfare generally want them to be provided at the federal level—for a good reason. If welfare is provided and paid for by the states, high levels of income redistribution tend to pull poor people into, and drive taxpayers out of, states that provide them. That provides a potent political incentive to hold down redistribution. This is one example of a more general principle: The more mobile taxpayers are, the more governments, like businesses in a competitive market, have to provide them value for their money, and thus the less able they are to tax A in order to buy the votes of B.

The same argument applies across national borders.

Tell Me It Isn't True

In two earlier posts I discussed the puzzle of why the National Security Agency chose to violate the Foreign Intelligence Surveillance Act. If the NSA wished to intercept messages why not either get warrants from the FISA court, as the law required, or ask Congess to amend the act? The act itself provides a two week window after the start of a war, obviously intended to permit amendment, and in the weeks immediately after 9/11 it is hard to imagine any serious opposition to such a move. Senator Russ Feingold's recent proposal that the Senate censure Bush provides a possible solution. It is not one I like, and I am not at all sure I believe it, but it is, so far as I can see, consistent with the available evidence.

By instructing the NSA to wiretap in apparent violation of the law, Bush provided his opponents with bait. Arguably it was bait that at least some Democrats, concerned with their position within their own party, could not refuse. By accepting it, the Democrats give Bush the opportunity to accuse them of being soft on terrorism.

Assuming, as I think we should, that Bush and his advisors are clever politicians, we have a solution to the puzzle. Bush deliberately violated the Foreign Intelligence Surveillance Act in order to get Senator Feingold to attack him for doing so.

Tell me it isn't true.