A policy that couldn't possibly work no matter what, and yet is widespread

The gun buyback program is most definitely one of the stupidest ideas anyone has ever come up with. Its stated goal is to reduce the supply of illegal guns. Not only does the program not achieve this goal in its current form, but it couldn't possibly work in any other theoretically viable design.

First, let's see why it can't work when the police buy illegal guns directly from their owners. If the police buy those guns for a price lower than the market rate for illegal guns, then the owners will just get rid of guns they don't need (ones that are old, defunct or redundant); in this case the supply either stays the same or increases (because the owners who sold the guns they didn't need might use some of the revenue from those sales to buy more guns). Now if the police buy illegal guns for a price higher than the market rate, then pretty much every owner of an illegal gun will sell it to the law, then go right back to the market, replace the gun he sold with a cheaper one, and pocket the difference. In this case as well the supply either does not change or increases.

Clearly, therefore, buying back from the owners doesn't work. But what if the government bought illegal guns directly from the dealers? Could that work? Let's see. Illegal gun dealers sell their product at some market clearing price that brings them profit. In order to divert the sale of illegal guns from individual customers to the government, the government would have to outbid individual buyers; i.e., it would have to offer gun dealers to buy their product for a price higher than its market value. Then, gun dealers would simply sell all their guns to the police instead of to the gangsters, and the streets would be free of illegal guns, right? Right?

Wrong, of course. Government resources are huge, but finite. Suppose (completely unrealistically) that the government devotes all its available funds to above market price gun purchases, so that every illegal gun currently produced is being sold to the police. There are now a lot of gangsters running around who need guns but can't buy them because the government outbids the highest price they are able to offer. There is untapped demand. And because the government is (by necessity) paying prices above the market value, a newly entering dealer can tap that demand by selling guns to gangsters at precisely the market-clearing rate. The customers will want to buy from that dealer, and the dealer will make a profit. Illegal guns are back on the streets, and the government is completely broke.

It's one of the amazing paradoxes of democracy: every college kid who has taken Econ 101 knows the gun buyback program cannot possibly work. And yet it continues, and will continue for at least some time in the future. Because politicians do not care whether their policies actually work; they only care whether voters think that they're doing something about the problem.

There truly are few more depressing and infuriating sights than that of some commissioner of some police department being photographed beside a heap of "illegal guns that our law enforcement has just taken off the streets."

Inefficient agricultures

In this post I will look at a very crude measure of productivity of the agriculture sector and compare it across a completely arbitrary sample of countries. What's the measure? Suppose that in a given country every full-time worker is equally productive; that is, every worker produces the same fraction of the country's GDP. Normalize that fraction to 1; that would be our benchmark. Then we look at what fraction of that benchmark is being produced by an average agricultural worker in each country. I looked at 32 countries: all European countries with population of over 4 million plus the U.S., Canada, Turkey, Australia and New Zealand. The bar chart is below the fold:

First observation: agriculture is inefficient. Average productivity is 0.65; there are only three countries with productivity exceeding 1: Croatia (1.2), Sweden (1.45), and the U.S. with a mind-boggling 2.0 (I have no idea what that's about). The least productive countries are Mexico (0.25), Poland (0.26), Romania (0.27), Portugal (0.28), Turkey (0.3) and Greece (0.3). It's slightly surprising to me that Polish agriculture is doing worse than those of some formerly communist countries that are markedly poorer (Belarus, Russia, or Ukraine).

The most inefficient agricultures have something in common: they're all big (meaning that they employ a very large share of the labor force; the unquestionable leaders here are Turkey, Romania and Serbia, in which countries about 30% of labor force works in agriculture). As the graph below shows, returns to labor in agriculture diminish exponentially:

So it seems that one part of the problem with Polish agriculture is that there are simply too many farmers in Poland. As to what the other parts of the problem are, I don't really know.

(Note: Data source is the CIA World Factbook. Graphs were made using R. As an R beginner, I hereby apologize to any serious R users who may be reading this for the incredible crudeness of my graphs.)

Could you please maybe stop stealing?

Do you think the U.S. is better than other Western democracies in that it is truly governed by the rule of law? You probably do. And you couldn't be more wrong. Examples to the contrary are many; here's a truly outrageous one: federal and state asset forfeiture laws.

(HT: Robin Hanson.)

Simulate thy enemy

Imagine you're playing chess with someone, and your opponent makes a puzzling move. You didn't expect it, and cannot immediately see the reason for it. Even though you can't see the reason, however, you probably do assume there is one, so you think to yourself "What is she up to? What is she seeing that I'm not?" and then try to figure out why she made that move, changing your strategy if the conclusions you draw from the new situation call for it. In other words, you try to simulate the mind of your opponent. You certainly would not simply dismiss this new information thinking "Whatever, she did it because she's evil and she hates me" or "I don't see any rhyme or reason in this move, therefore it must be a stupid move and I'll just ignore it." In order to benefit in strategic situations we have to simulate our enemies' minds as accurately as possible. It's not an easy task (and perhaps that is one of the reasons we came up with all these games that help us train that skill, like chess, go or poker), but we do it reasonably well.

However, we only do it well when it comes to our own personal adversaries. When we think about the enemies of a larger group that we belong to (such as nation, religious denomination, political party and the like), we fail miserably at this task. Shortly after 9/11 happened and we were thinking about why the hijackers did what they did, all we could come up with was "because they are evil and they hate us and they want to inflict pain and suffering on innocent people." How accurate can that be? No one is a villain in their own narrative about themselves. No one (except maybe bona fide clinical psychopaths) ever thinks they're evil and that what they want is to bring pain upon innocent people. And yet when we think and talk about the enemies of our nation (religion, party, whatever), that's how we see them. We think they're evil, or stupid, or both. We think of them as some sort of monsters, and that's why we refuse to try to simulate their minds--after all, they're nothing like us. And don't think this mechanism is present only in extreme circumstances such as wars or terrorist attacks. The mundane, everyday political debate is ripe with such thinking. Take healthcare. When Republican supporters are discussing Democrats' proposals, and vice versa, they are almost never discussing the actual proposals but instead some haphazard straw man constructions that make the other side look ignorant and callous.

Depending on context, we can be either reasonably good, or completely and utterly disastrous, at the task of simulating other people's minds. The reason for it is fairly simple. In the context of strategic situations you find yourself with other people, what those people do matters to you directly. Therefore, being able to think like they think gives you an advantage, because it helps you predict their actions, or deceive them, or assure them of your good intentions, whatever the need may be. In the context of large group conflict, what the members of the other group are thinking does not matter to you directly. What matters to you is signalling loyalty to other members of your group, and this can be done by showing bias against the other group. If you start trying to really understand how terrorists think, your fellow group members will not like it very much. After all, if you're trying to understand them you must assume they're at least a little bit similar to us, and that's a dangerous thought if you want to be perceived as a group member in good standing.

Last thought: this bias is of course only present in ordinary group members; group leaders are by and large free of it. That's because leaders have direct stakes in group conflict; to them, group conflict is really personal conflict. Donald Rumsfeld may have been saying that terrorists attack us because they're just evildoers who hate freedom and want to destroy the American way of life, but I'm certain that was not what he actually thought. That was just talk for the masses.

Too bad for reality

This is a follow-up to my yesterday's post about extrapolating long-term growth of Soviet economy in 1961. In his famous 1961 economics textbook, Paul Samuelson projected that the output of Soviet Union will converge to the output of the United States sometime between mid-eighties and mid-nineties. Today this is just a curiosity illustrating how wrong economic projections can be; but even at the time Samuelson first published this projection, he was accused of displaying a pro-Soviet bias.

Just yesterday, economist Bradford DeLong defended Samuelson saying that it's wrong to accuse him of bias because his conclusions were simply a consequence of the Solow growth model, which at the time was the workhorse of growth economics.

The details of the model are not important here; what's important is subjecting predictions of the model to reality checks. Suppose for the sake of the argument that DeLong is right: in 1961 any economist, biased or not, who studied the Solow model and the data that were available then, could not escape the conclusion that in about 25-30 years Soviet economic output will equal the U.S. economic output.

The problem with this line of defense is that a non-biased economist would then conclude that something must be terribly wrong either with the Solow model, or the data that were fed to it, instead of simply taking this ludicrous projection at face value.

Simple extrapolation

Here's xkcd on the dangers of extrapolation:


...and here's some extrapolation in action:

(Click on the graph to make it bigger.)

The graph (taken from the famous 1961 economics textbook by the late Nobel prize winner Paul Samuleson) projects that the real GNP of Soviet Union will have caught up to the real GNP of the United States sometime between 1984 and 1997.

(HT: Alex Tabarrok.)

Why not pick on TSA some more

(HT: Cafe Hayek).

Non sequitur of the month: personal communication

I'll describe this month's entry in abstract terms first, then provide two examples. Suppose you have two sets A and B. They're both non-empty and they overlap; other than that, no additional assumptions about them are made. Our non sequitur of the month is the following implication:

If the number of elements that belong to the intersection of A and B constitutes x percent of the number of all elements of A, then it also constitutes x percent of the number of all elements of B.

This is true if and only if A and B are exactly equal in size (or, as a special case, if they are one and the same set). But no such assumption was made, so it is definitely not true; in fact, it is a stunning non sequitur.

I've encountered this baffling mistake in two real-life contexts; describing those may help highlight how incredibly wrong this reasoning is. First context: There are studies trying to estimate the percentage of homosexual men among pedophiles. Any study that comes up with such an estimate (say 10 percent) will invariably be quoted somewhere in the media as saying that 10 percent of homosexual men are pedophiles--as if it were the same thing. Second context is a guy I knew named Bob (not his actual name). Bob is a very successful individual. He holds a Ph.D. in one of the social sciences, and among many things he does for a living is teaching a course in quantitative research methods at an Ivy League university. One of Bob's research interests is the connection between homelessness and foster care. There's research out there estimating that roughly 20 percent of youth that "age out" of foster care experience a significantly long spell of homelessness. Bob is convinced that this means that 20 percent of homeless people have aged out of foster care. I am not making this up; both myself and several other people have asked him to articulate his conviction as clearly as possible, and it is indeed how I've just described it. All attempts to show him the absurdity of said conviction have failed miserably.

Think about what this logic implies. Suppose 5 percent of smokers get lung cancer; this means you should believe that 5 percent of people with lung cancer are smokers. If 95 percent of men have a high school diploma, then 95 percent of people with a high school diploma are men. If half of the cats are black then half of all black things are black cats. Etc., etc. I think the sheer stupidity of this particular non sequitur is clear enough by now.

There are errors in reasoning that are subtle: it's easy to make them if one is not careful enough, and even if avoided, it's easy to understand where the mistake originates. None of the non sequiturs of the month fall into that category. Try as I might, I cannot come up with a plausible line of reasoning that could lead to them.

TSA: Intensified universal screening cannot possibly work

Q.E.D.

A frog and mouse battle

Back in the day (meaning the first half of the twentieth century), most mathematicians were deeply engaged in a dispute between two philosophies: Platonism and intuitionism. The details of the conflicting views are not important here; the main point is that both sides were arguing as if their respective positions were the true state of affairs, seemingly not realizing that the debate they were engaged in was of the sort which allows for more than one point of view to be correct. Upon hearing the details of the disagreement from one of his mathematician friends, Einstein reportedly called it "A frog and mouse battle." I have no idea what the thought process behind choosing this particular phrase could be, but nonetheless it seems to capture the essence quite nicely.

Economists have their own frog and mouse battle going on right now, and one that is substantially sillier than the original. It's the debate over whether or not GDP is the correct measure of quality of life. Fairly recently an international commission of economists led by Joseph Stiglitz (and created by the French President Sarkozy) released a report which suggested a measure of aggregate quality of life alternative to GDP. Numerous economists have then spoken up in defense of GDP as a better measure; one example can be found here.

I think there are two things that render this debate quite silly. First, I don't think there is a law of the universe saying that there has to be just one correct measure of economic performance. Quality of life is a multidimensional phenomenon, and different measures can and will place different weights on these dimensions, so I fail to see how a choice between those measures could be anything other than arbitrary. Second, GDP by its very definition is not meant to be a measure of quality of life. It's supposed to measure aggregate output, or, in other words, aggregate value added.

I really can't see anything other than a frog and a mouse in this debate. Am I missing something?

The mother of all fallacies

Well, not really. Nothing is ever the mother of all anything. But, whenever politics and/or economics is being discussed, there is one fallacy that seems to be much more common than all the others: it's the Nirvana fallacy, i.e. belief that if some policy solution isn't perfect, it's useless. Or, to put it another way, comparing a policy proposal not to its best viable alternative, but to a perfect state of affairs.

Healthcare debate provides an example. Most of its participants, regardless of what they're arguing for, seem convinced that it's possible to build a system in which everyone is taken care of; they just disagree on what that system should look like. But, alas, it's not possible; at least not now, not with the income levels our economies generate as of today. In any system we can have right now, there will be people dying due to lack of adequate healthcare. In a purely market system, this will happen because providing healthcare is very expensive, and some people will not be able to afford the market rates (we see this happening right now in the U.S.). In a purely government-run system, this will happen because when the government controls prices of medical services, those prices will be set either below or above market clearing rates (most likely way below)--and both those situations generate shortages of supply and decrease of quality (we see this happening right now in Poland).

Scarcity is a fact of life. There are three ways of dealing with that fact. First, we can keep increasing our income and wealth, thereby decreasing scarcity in some areas of life by making things cheaper and cheaper to buy. Second, we can complain to whoever it is we think invented the world. Third, we can simply pretend it doesn't exist--and this is precisely what most of the political debate does.