Bryan Caplan blegs for a bleg

by on April 5, 2013 at 3:51 am in Uncategorized | Permalink

He blogged:

A series of queries that’s stumping most of my favorite IQ researchers:

Are there any countries where IQ testing for hiring purposes is totally legal?  Largely legal? Do we have any idea if the education premium rose less in those countries than in countries that discourage or forbid IQ testing for hiring purposes?Context.

Come on people, stand up for Cowen’s Second Law (“there is a literature on everything”)!

1 Jack Sparrow April 5, 2013 at 4:18 am

I always wonder whether there is any good reason not to have IQ testing as a legal parameter in selection of employees?

The best answer I’ve come across comes from James Flynn’s address to the APA in 1999 “Searching For Justice” :

Flynn’s argument is similar to that of a redistributive society which seems to go right in the face of what libertarians would stand for.

Two questions, I will be grateful if anyone could answer:

1) If the premise that there are innate difference amongst humans in which there are better (in terms of intelligence) humans doesn’t it make sense that the better ones will outdo the lesser ones? In this case, the whole concept of egalitarianism (in any sense of opportunity or outcome) seems bunk.

2) Can someone please tell me that in market system, how is the initial asset allocation of property rights decided? Property rights are considered sacrocant by economist, but is there a algorithm for determining initial asset allocation which can be optimal from which no one can complain about market related changes?


2 prior_approval April 5, 2013 at 5:35 am

‘in which there are better (in terms of intelligence) humans doesn’t it make sense that the better ones will outdo the lesser ones’

As noted below, that is not necessarily the case concerning the police in at least one jurisdiction – there, the ‘better (in terms of intelligence)’ are screened out. Which pretty much precludes the ‘better ones’ from outdoing the ‘lesser ones.’

3 Andrew' April 5, 2013 at 6:45 am

1. Economically speaking, this might not even matter. Warren Buffett may or may not be a genius, but he gets to command more capital because he produces returns on capital under his control, and if we are smart we want it this way as long as they are not ill-begotten gains. The problem is when they are not freely obtained by superior performance and the popular notion that rich guys are chilling on yachts with their exploitation gains, which they probably aren’t. So, the focus should always be on sanctifying the system rather than redistribution.

4 Jack Sparrow April 5, 2013 at 7:59 am

Doesn’t that beg the question of how Buffett produces returns on capital under his control? Four reasons come to mind: 1) innate ability/talent 2) skill 3) Chance 4) Some combination of 1-3. If the weightage is higher on the innate ability it is natural to expect this return. If it is skill based or chance others can be allowed to cash in on it by further investment/redistribution/competitive process.

The issue is outlined in the second question which you acknowledge superficially, “The problem is when they are not freely obtained by superior performance…” i.e. how is the initial asset allocation in the economic society justified? If it is based on some notion of ‘merit’ what is that notion? The Flynn paper tries to see whether any notion of ‘merit’ is sustainable. Tinkering or redistribution (or Keynesian injections) seem inevitable in his worldview.

5 Andrew' April 5, 2013 at 8:41 am

Buffett is a good case because (1) noone can seriously claim it is luck. (2) Most of what he does is make decisions, he’s not a manager (3) his ‘initial distribution’ of non-innate assets while not zero or negative can be assumed zero relative to his results. He went to college and met his mentor Ben Graham (hey, an example where networking actually worked!) and then borrowed much of his initial money from investors based on sales pitches. While it might be difficult in the average where results are not tallied or compounded so obviously to tease out things like soft racism, this is a very clear illustration.

6 Andrew' April 5, 2013 at 8:42 am

The only question is does capital itself confer an advantage? I think the answer is mostly no.

7 JWatts April 5, 2013 at 10:47 am

The only question is does capital itself confer an advantage? I think the answer is mostly no.

I would disagree with that. Certainly capital confers an advantage. However, it’s probably not any more of an advantage than a +1 SD of IQ confers.

8 Doug April 5, 2013 at 5:54 pm

Why can no one seriously claim that Buffet’s results are luck?

9 Pshrnk April 5, 2013 at 2:51 pm

And lucky performance/survivor bias plays what role?

10 Manoel Galdino April 5, 2013 at 10:34 am

I know almost nothing on Buffet, but
1: It strikes me that Jack Sparrow didn’t consider another factor, namely that there is path dependency or the popular notion (at least here in Brazil) that “money brings money”. I’m not saying this is the case, since I don’t know the details. But it strikes me that this isn’t a factor to be considered. After all, connections, access to cheap money, information and all benefits of being rich are side effects that may help a lot to keep him make a lot of money over the years. In other words, what’s the proper counter factual here in order to establish causality? Conditional on some time period in which he was already rich, in order to properly asses the causal effect of say, innate ability (whatever this may mean), what experiment would do the job? Is it enough to imagine an experiment in which we somehow recombine his genes to estimate the causal effect of innate ability? It seems to me that any causal DAG would show that that’s not the case, since his innate ability in the pass caused not only his success at time t (of the experiment), but also the skill in time t, the connections in time t and so on.
2. It strikes me Andrew says that “no one can seriously claim it is luck”. Why not? Sure, it seems that he’s smart. But, on the other side, what’s the probability that among many people doing what he does in the whole developed world, one of them, by mere chance, wouldn’t have his track record? Even if they’re guessing?
To explain my point a bit further, it’s like people getting surprised for the same number occurring in lottery in the same year (see
Again, I’m not saying it’s luck. I’m just saying that it isn’t that hard to seriously think that it’s luck. There are a lot of books these days explaining how we’re fooled by randomness and that in hindisght, everything is obvious.

What am I missing?

11 Tracy W April 5, 2013 at 8:35 am

1) People are complex. A can outdo B on one grounds, but not on another ground (eg Tiger Woods is great at golf, not so good at marriage). We are quite free to say things like A earns more money than B, and is better at all sports than B, and is better-looking than B, and is funnier than B, and is healthier than B and so forth, but A and B still should be equal under the law (or some other ground on which we think that people should be egalitarian).

2) \To answer the sub questions in order. 2.a) Historical accident. 2.b) No. No force in the universe can stop people from complaining. (Note, also, that economists often advocate the creation of property rights, such as tradeable fishing quotas, or spectrum allocations, or land rights (see De Soto), so saying that economists consider property rights as sacrocant is wrong, perhaps some economists do, but not economists as a group.)

12 Jack Sparrow April 5, 2013 at 9:50 am

@ Andrew: I agree that capital ‘itself’ (i think you meant ‘physical’ capital) doesn’t confer an advantage, but what I intended to take was a broader view which includes human capital can confer drastic survival advantages.

13 Jack Sparrow April 5, 2013 at 9:54 am

@ Trudy: It is reasonable to assume that we still live in a darwinian world (survival of fittest/adaptable etc), but things seem to be improving for all because technological progress. But there is no inherent basis or evidence of any equity principle like equality before the law. Marriage is valuable only as a social construct. But if you look in terms of ‘value free’ observables like mate selection and number of mates, innately talented like Tiger woods out do others. There is no such thing as equality before the law. There are improvements for all, but money and power does buy a lot of material influence in law anywhere including the US. I hope I don’t sound as if I’m complaining. The reason I posted this was usually find extreme supporters of libertarianism (like friedman) or even leftists like Krugman, either intellectually dishonest (that is being too harsh i guess) or oblivious to reality (perhaps this is naive of me!)

The world maybe a positive-sum or a zero-sum stage, but in a capitalist society, we value ‘co-operate’ as a strategy of trade, but other alternatives like ‘coercive’ strategies can be beneficial as well and they are available to ones with talent/ability/luck etc. The issue is about honesty of recognizing that redistribution can serve useful purpose at times!
If you are make an investment for public goods or social goods like primary education there is some redistribution involved.

( you are correct that property rights are not held sacrosanct by all economists. i forgot there are projects like R (which is painful to learn for amatuer coders) and Linux which are supposedly without property rights… it will be interesting to see how they turn out in the future!)

14 JWatts April 5, 2013 at 10:49 am

In this case, the whole concept of egalitarianism (in any sense of opportunity or outcome) seems bunk.

Yes, Harrison Bergeron agrees:

15 Jack Sparrow April 5, 2013 at 11:52 am

+1 JWatts.

16 stats April 5, 2013 at 5:01 am

Oh, please. Is it still widely believed amongst “intellectuals” that intelligence matters anywhere in the American workforce? Maybe at Google. Everywhere else there are three, perhaps 4 criteria upon which hiring is decided: 1. Politics 2. But kissing acumen 3. Haircut … and if a public company or highly monitored private company, 4. Sex and race.

Intelligence needs can be farmed out to contractors.

17 LemmusLemmus April 5, 2013 at 5:41 am

And the contractors hire on the basis of which criteria?

18 JWatts April 5, 2013 at 10:52 am

Intelligence & work ethic of course. That’s why we get to charge those high fees.

19 Statspotting April 5, 2013 at 5:30 am

As you move away from developed countries along the spectrum, data collection overall, and the accuracy of collected data as well, goes down significantly. Drawing conclusions based only on such data could be dangerous.

20 prior_approval April 5, 2013 at 5:32 am

Well, there is at least one county where IQ testing is used to screen police applicants – to ensure that they are not too intelligent. Further, courts have found such testing, and applying the results to reject job applicants who are deemed too intelligent, to be fully legal.

(Simply to document the reality of that country’s use of IQ testing in terms of employment opportunities – )

21 joan April 5, 2013 at 5:33 am is a study of the education premium in Europe.

However education premium in the US is mostly due to the stagnation of the wages of the non college educated which suggest changes in technology as the cause

22 Rahul April 5, 2013 at 5:58 am

Totally legal in India AFAIK. You find IQ style questions on government hiring tests too.

23 Andrew' April 5, 2013 at 6:41 am

Ahem…it is legal here, we just spread it out over 12-20+ years of free/wasted labor. I wish I had something serious.

24 Daniel April 5, 2013 at 7:21 am

What is Cowen’s First Law?

25 John Mansfield April 5, 2013 at 7:57 am

All food is ethnic food?

26 John Faben April 5, 2013 at 8:25 am

Looks like it’s the same as Cowen’s second law…

27 Thor April 5, 2013 at 12:17 pm


Very close. The answer is: “All Cowen laws are also Cowen’s Second Law”

28 anon April 5, 2013 at 7:26 am

IQ style questions are very common in germany if you apply for apprenticeships at large companies. In Germany it is called an “Einstellungstest” and there will be some questions about general knowledge (mostly history or current affairs) and some logic test.

29 Enrique April 5, 2013 at 7:31 am

Just wondering … What is Cowen’s First Law?

30 OneEyedMan April 5, 2013 at 8:00 am

The USA uses Armed Services Vocational Aptitude Battery (ASVAB) for assigning employment within the military., I believe other countries do something similar but I can’t recall which .

Exam scores on the SATs and the GRE are often explicitly asked for by high finance recruiters as a filter for screening applicants.

31 Sandman April 5, 2013 at 8:43 am

I imagine it is legal in most countries to test IQ for hiring purposes.

32 Ostap April 5, 2013 at 9:43 am

My understanding is that many software/consulting companies are fond of requiring applicants to try to solve various logical problems. That may not be “IQ testing” but it’s close enough and it’s perfectly legal.

33 jtf April 5, 2013 at 11:17 am

This. I can’t remember how many times I’ve gone through “lateral thinking” puzzles at pre-screenings for interviews at consulting companies.

34 Yancey Ward April 5, 2013 at 11:24 am

My understanding is that you can do it in the US, but if an applicant can show your test had a disparate impact among different ethnic classifications, you can be sued under civil statutes.

35 Russ R. April 5, 2013 at 12:15 pm

Cowen’s First Law is: “You don’t talk about Cowen’s First Law”.

36 Anonymous April 5, 2013 at 9:21 pm

The federal government of Canada does “General Mental Ability” testing for public service positions. Not quite IQ in that it is not averaged at 100, but still testing “g”. Information on one of these exams is here

37 FYI April 5, 2013 at 10:52 pm

This is already done in most (if not all) IT companies. Microsoft’s famous “puzzles” are nothing more than IQ tests in disguise. All other major IT companies do the same, be it in the form of complex programming exercises or be it in the form of open design questions.

Of course, you can say those are only proxy tests because you are not calculating a IQ number per se. But the end result is exactly the same so I am not sure why the tabu over allowing IQ tests. If anything,using a standard test would kind of defeat the purpose because people could practice for it much more easily than you can practice for a complex interview.

So I think the contention here is whether we have IQ tests for professions with lower intelligence requirements???

Comments on this entry are closed.

Previous post:

Next post: