Saturday, June 27, 2015

I.Q. and the Wealth of States



One of the simplest theories of human prosperity is the idea that societal wealth comes from an intelligent populace. Obviously this is true to some degree; if you went around and forced everyone in the country to take a bunch of brain-killing drugs, economic activity would definitely decline. The question is how much this currently matters on the margin.

Some people think it matters a lot. Richard Lynn, a British psychologist, wrote a book called I.Q. and the Wealth of Nations, suggesting that average population I.Q. drives differences in national wealth. Garett Jones of George Mason University is writing a book called Hive Mind that suggests much the same thing, asserting that there are production externalities associated with high I.Q. Motivated by this hypothesis, there is a line of research in development economics dedicated to finding interventions that boost population I.Q.

Well, here is some new and relevant evidence. Eric A. Hanushek, Jens Ruhose, and Ludger Woessmann have a new NBER working paper in which they look at U.S. states. From the abstract:
In a complement to international studies of income differences, we investigate the extent to which quality-adjusted measures of human capital can explain within-country income differences. We develop detailed measures of state human capital based on school attainment from census micro data and on cognitive skills from state- and country-of-origin achievement tests. Partitioning current state workforces into state locals, interstate migrants, and immigrants, we adjust achievement scores for selective migration...We find that differences in human capital account for 20-35 percent of the current variation in per-capita GDP among states, with roughly even contributions by school attainment and cognitive skills. Similar results emerge from growth accounting analyses.
Note that the authors control for selective immigration, an oft-neglected factor in debates about I.Q.

So the upper bound for the amount of state income differences that can be explained by population I.Q. differences is about a third. If we assume that achievement scores are a good measure of I.Q. and that school attainment doesn't improve I.Q. very much, then the number goes down to about one-sixth.

Now, it's important to remember that this study, well-executed though it is, doesn't isolate causation. It doesn't show the degree to which state average I.Q. can be raised by raising state income.

What it shows is that the vast majority of differences in state income are not due to variations in state average I.Q. If we had an I.Q.-boosting device, boosting the average I.Q. of Ohioans by 1% would raise Ohio's average income by at most around around 0.17%.

Of course, that's a marginal effect. If we boosted the average I.Q. of Ohioans by 400%, we might see much more (or much less) than a 68% increase in their income. And if we gave Ohioans brain-killing drugs (insert Ohio State football joke here) that cut their I.Q. in half, we might see much more (or much less) than an 8.5% decrease in state income.

But anyway, what this really shows is that there is Something Else that is driving state income differences. My personal guess is that this Something Else is mainly "external multipliers" from trade (the Krugman/Fujita theory). Institutions probably play a substantial role as well (the Acemoglu/Robinson theory). That's certainly relevant for the debate about different models of capitalism, where we often compare the U.S. to Scandinavia and other rich places.

In any case, this result should be sobering for proponents of I.Q. as the Grand Unified Theory of economic development. Average I.Q. is not unimportant for rich countries, and we should definitely try to raise it through better nutrition, education, and (eventually) brain-boosting technologies. And it still might matter a lot for some poor countries. But for rich countries, there are things that matter a lot more.


Update

Scott Alexander seems to think that my post gives a slanted interpretation of the results of this study - that if you present the numbers in a different way, they tell a very different story, and in fact imply that "IQ is everything after all." So just in case there was any ambiguity, let me give a concrete example of what this paper says about the impact of average population IQ on GDP.

Suppose you were to take the state of Ohio, and use an IQ-boosting device to boost Ohio's average IQ by 18 points. This paper predicts that Ohio's GDP would rise by 16 percent, or about $5,600 per person.

18 IQ points is the difference between the commonly reported average IQs of Mexico and South Korea, as listed in this table. A $5600 rise in GDP would take Ohio's per capita GDP from about the level of Italy to somewhere between the levels of France and Belgium (see here for those GDP numbers).

So I think this very clearly backs up my summary of the paper's result.

20 comments:

  1. The IQ/income connection might be partially indirect. Smarter citizens might vote for smarter state policies which then increase incomes ( the converse could also operate ).

    ReplyDelete
  2. I have a working paper coming out that completely debunks the methodology of this. I can send it along if you'd like.

    ReplyDelete
    Replies
    1. Well, more accurately, it debunks the methodology of Caselli 2005 upon which this is based (they even used part of his title in their own). Also it's written as a zombie story, so there's that.

      Delete
    2. Zombie story economic theory. Sounds good so far.

      Funny, you don't look Greek. :-)

      Delete
  3. LOL! Oh yeah, *just* 1/6th, huh? Compounded generation after generation?

    ReplyDelete
    Replies
    1. Income, not growth. So it wouldn't be compounded.

      Delete
  4. "If we assume that achievement scores are a good measure of I.Q. and that school attainment doesn't improve I.Q. very much, then the number goes down to about one-sixth."

    You lost me there.

    ReplyDelete
    Replies
    1. Confused me too. What is compatible with this statement is that the power of some kind of "innate" IQ goes down to 1 sixth.

      Delete
    2. Anonymous8:08 PM

      If there is any residual error between IQ-test results and innate ability then the relationship between innate ability (which IQ-tests attempt to measure) and state income should go up not down. So Noah statement is silly.

      Delete
  5. Anonymous6:52 PM

    1) Missisippi has a gap per capita of 29k and federal spending per capita of 11k. Dumber states are subsidized by the smarter ones. Maybe doing the study on the culturally similar but fiscally independent states of Latin America would be interesting.

    ReplyDelete
  6. Kolya6:56 PM

    2) I imagine North Dakota and Alaska - oil rich but not renowned for book larnin' - are outliers. What if they are removed?

    3) there's also smart fraction theory - mean IQ could be less important than having a critical fraction of the population above a certain floor.

    ReplyDelete
  7. Anonymous11:31 PM

    Don't substances that lower I.Q. already exist and already in the environment? For instance, lead.

    Hasn't the amount of lead in the environment been decreasing in general over the last thirty years since its phaseout from gasoline?

    I'm not sure, but I would assume the rate of lead exposure has not decreased at the same rate in every state, or even neighborhood. Not all of the lead in the environment ended up coming from gasoline and due to the use of lead paint, the levels are higher in certain housing developments, as well as more heavily industrialized areas.

    Also, what about cities and states where lead abatement programs were undertaken, versus those places did not occur?

    ReplyDelete
  8. Anonymous12:53 PM

    OC, IQ is problematic across cultures, as much of the test questions depend upon cultural factors. However, there is a good substitute: reaction times. Do video games increase intelligence? Maybe so.

    ReplyDelete
    Replies
    1. Anonymous4:59 PM

      Do faster reaction times make people smarter or is reaction time just correlated with intelligence? If it's the latter, then improving reaction times won't increase intelligence.

      Delete
  9. Scott2:12 PM

    I always thought , "The closer you lived to Canada; the more intelligent you were."

    ReplyDelete
  10. Did the authors of the paper take into account the general decline in IQ over the last two centuries or so? It maybe wouldn't have much of an effect for their purposes, but it would still shed some light on the validity of the IQ/income hypothesis in general.

    ReplyDelete
    Replies
    1. Please ignore this comment. I have just now learned that apparently there is no general decline in IQ over the last two centuries or so and that I was mistaken in what I have learned. My apologies.

      Delete
    2. Anonymous12:17 PM

      General decline in IQ? I've heard of the Flynn effect, which chronicles the exact opposite of that claim. What are you talking about?

      Delete