There has been a lot of discussion over the past few days regarding the relationship among income, genetics, and intelligence. A controversial paper by economist Bruce Sacerdote purports to show that the relationship between parental income and children's income is different for biological and adopted children. Biological kids make more money as parents' income increases whereas adopted kids have a flat income curve as parental income increases. Mike has a great writeup (follow the link) explaining the high school-caliber errors in the author's ultimate conclusion that intelligence is passed down genetically, which is stacked upon the equally fallacious claim that high income = intelligence. In other words, Prof. Sacerdote presents some very interesting if deceiving findings – I can't emphasize enough Mike's point about the four year difference in mean age between the groups of children – and then closes with some nice eugenics.
Economists are useful people to have around in the academic world. They have great quantitative skills and, in my experience, a keen eye for research design. Unfortunately most of them think they are social scientists. They aren't. They are mathematicians practicing a hard science. They should not do political science, biology, or sociology any more than practitioners of those disciplines should do economics. But economists have the misfortune of seeing the entire world in rational choice terms and a field which encourages them to make the most immoderate, speculative conclusions possible based on their findings.
Prof. Sacerdote, for example, could have written a paper in which he said "Here is a very interesting disparity I have uncovered. I hope this encourages others to study this issue and find out what's going on here." Nah. As economists like to do, he just solves the dilemma himself at the end of the paper: it's genetic. People who make more money are smarter than people who make less, and that intelligence is passed on to their children. See how much easier that was compared to doing all that messy "research" and using "logic"?
Similarly, from the are-you-fucking-kidding file we have this entry on Harvard economist Greg Mankiw's blog. This man is a very good economist. He is famous. He has achieved much in his academic and professional career. Now read this:
The NY Times Economix blog offers us the above graph, showing that kids from higher income families get higher average SAT scores.
Of course! But so what? This fact tells us nothing about the causal impact of income on test scores. (Economix does not advance a causal interpretation, but nor does it warn readers against it.)
This graph is a good example of omitted variable bias, a statistical issue discussed in Chapter 2 of my favorite textbook. The key omitted variable here is parents' IQ. Smart parents make more money and pass those good genes on to their offspring.
Suppose we were to graph average SAT scores by the number of bathrooms a student has in his or her family home. That curve would also likely slope upward. (After all, people with more money buy larger homes with more bathrooms.) But it would be a mistake to conclude that installing an extra toilet raises yours kids' SAT scores.
Is this a joke? He opens by recognizing that the data offer no evidence whatsoever about causality and closes with a warning about making spurious and unwarranted conclusions about causality when correlation is present. Which is cute, because between those two statements he grabs his ankles, reaches deep within his ass, and pulls out the uncited, unwarranted, and baseless conclusion that the real causal mechanism here is genetics. A goddamn college freshman could look at the relationship between income and SAT scores and conclude, "Hmm. Well, parents with more money can afford expensive schools and SAT prep courses." That is just about the most basic example of cause and effect one could imagine.
If you took your car to a mechanic because the alternator was shot – you can open the hood and plainly see that the alternator is burnt out and not functioning – a reasonable mechanic would say "Hey, you need a new alternator." If you took the same car to an economist, he or (rarely) she would say "The problem is obvious. Your parents have low IQs, and thus you have inherited genes which make you too dumb to pick out a car that will be free of mechanical problems." If that analogy seems ridiculous, that is exactly what Mankiw has done here. He has ignored the overwhelmingly obvious explanation and substituted his own imagined causal mechanism.
Economists may not suck at causal inferences universally, but the ones that do it well and judiciously are well-hidden. Their field provides perverse incentives to draw attention to one's work by stating the biggest, most shocking conclusion that can be wrung from the data, however tenuous the evidence. Either that or they just suck at playing social scientist. Causality is determined by hypothesis testing and supported with evidence. When hypotheses can't be thoroughly tested, statements about causality should at least adhere to basic logic. Showing a correlation and then going off half-assed on one's own preferred explanation of How the World Works is not science. And it is particularly unwelcome when that preferred explanation happens to be 19th Century eugenics and a handful of social Darwinism.