IQ economics > behavioral economics

We use administrative and survey-based micro data to study the relationship between cognitive abilities (IQ), the formation of economic expectations, and the choices of a representative male population. Men above the median IQ (high-IQ men) display 50% lower forecast errors for inflation than other men. The inflation expectations and perceptions of high-IQ men, but not others, are positively correlated over time. High-IQ men are also less likely to round and to forecast implausible values. In terms of choice, only high-IQ men increase their propensity to consume when expecting higher inflation as the consumer Euler equation prescribes. High-IQ men are also forward-looking — they are more likely to save for retirement conditional on saving. Education levels, income, socio-economic status, and employment status, although important, do not explain the variation in expectations and choice by IQ. Our results have implications for heterogeneous-beliefs models of household consumption, saving, and investment.

That is from a new NBER working paper by Francesco D’Acunto, Daniel Hoang, Maritta Paloviita, and Michael Weber.


I knew it: I'm smarter than Thaler assumes.

Then why are you in the Philippines killing chickens?

Actually, this question is non-trivial and illustrates why behavioral economics really does trump IQ economics.

The actual distribution of life outcomes for people by IQ is very much more noisy than a structured survey of their carefully channeled opinions would lead you to believe.

This goes back to work by António Damásio. People with damage to the emotional centers of their brain can still answer these kinds of questions and give good normative advice for others to follow. They just don't follow it themselves, because emotional centers governing risk-and-reward are fouled up.

"The actual distribution of life outcomes for people by IQ is very much more noisy than a structured survey of their carefully channeled opinions would lead you to believe. "

Is it though?

"This goes back to work by António Damásio. People with damage to the emotional centers of their brain can still answer these kinds of questions and give good normative advice for others to follow. They just don't follow it themselves, because emotional centers governing risk-and-reward are fouled up."

Sounds completely irrelevant. Close to a non sequitur

It makes sense to me. IQ is about ability, and as is Damásio theme, emotion dictates how we direct that ability.

There may be, for instance, a guy with very IQ who just likes to surf in Hawaii, and maybe tends bar by the pool at the Marriott to make ends meet.

High IQ, maybe high life satisfaction, but low on both income and wealth.

He might be "low" on the chart below.

This is the first time I agree with you.

Are you the real anonamouse?

We can never know.

High IQ people don't have damage to the "emotional centers of their brain" any more than other people do

Sorry, I jumped there. Damásio's work is about emotion directing reason. One place it is seen is in a type of damage where patients pass all the tests, retain all their intellect, but tend to have bad outcomes in following years.

I remember the example of one highly intelligent man who gave his money to a shady partner because, shrug, he didn't really care anymore.

Not all people are brain damaged, but neuroscientists use such damage to piece together function.

"emotion" no more "dictates" our behavior than intelligence does.

Actually, it's perfectly rational to reside in the Philippines and slaughter chickens. Ray's cost of living is nominal and, hence, his wealth is increasing rapidly. Consider global warming. My low country home has been hit twice in recent years by hurricanes (something Ray can appreciate in the Philippines). What would a rational person do? Move to higher ground? But what if I really enjoy living in the low country and moving would provide a less satisfying life, then what should I do? When the last hurricane hit, a high IQ woman I know said she isn't leaving the low country and, instead, will limit her potential losses if more hurricanes hit by residing in a modest home (what we affectionately call shacks in the low country) with modest possessions. That's the obvious and most rational answer to global warming, and the answer of a high IQ woman. But that's not the answer to the hi IQ swells who continue to purchase high priced homes (granite counter tops and all!) in the low country because they want to duplicate their life back in the place they are fleeing.

Interglacial and Solar Output. Were they the first two hurricane since the end of the most recent ice age, 100,000 years ago the Wurm/Wisconsin glaciation?

Shocking News From The World Of Science Department: Higher IQ males make slacuial and marter choices than lower!

Plus, coldest winter temperatures in recorded history. In two short years, President Trump solved the Global Warming Crisis.

Trump 2020!

Let's hope the demoncats run on a platform that includes "Open borders and Medicare for all!"

MAGA 2020!

Run for the hills rayward, run!

Off topic slightly, a retired English friend of mine was muttering across the fence to me last week about sea levels rising 3 yards in the next 100 odd years. And he’s educated! In the sciences no less. Ok health sciences.

Is there a website where people go to be scared and grossly misinformed? al gore dot com or something

Does no one read Bjorn Lomberg?

"The likely range of sea level rise in 2100 for the highest climate change scenario is 52 to 98 centimeters (20 to 38 inches.)."

Per the IPCC.

The media is feeding the population a bunch of crap and most people can't be bothered to do any fact checking.

Love BL!

Sadly, it appears mood affiliation Trumps (sorry, npi) rational assessment of climate predictions (difficult, especially about the future).

According to the IPCC, the upper estimate of sea level rise is 3mm/yr, so

100 yrs*(3mm/yr)*(inch/25.4mm)*(1ft/12 in)*(1 yd/3ft) ~ 0.33 yards or one foot.
Ooooooo, so scawy!

Someone please check my arithmetic.

The IPCC projects that the rise will increase and thus they estimate the rise closer to 1 yard.

That being said, 1 yard is not very much and the IPCC has a history of over stating the effects of global warming.

This ...

"... IPCC has a history of over stating the effects of global warming..."

Not surprising, since some (many?) of the nations involved are seeking reparations from the US, which, of course, would be promptly stolen by the elites (thugs) of those nations.

I remember Bolton getting the axe when he said aid to Brazil (?) would be stolen. A couple decades later we discovered he was correct.

A bid contingent of the IPCC has a vested interest in the CAGW meme.

A big (not bid) contingent ...

Hi all, I got out of the chicken raising business, since the cost of chicks ate up my profits (the incubator did not work due to power outages, a chicken egg will die if left unheated for several hours). Now we raise chickens, pigeons, and other animals for fun (and profit). We gave away our monkey, since he was too much drama, a full time job, like raising a kid.

Yes, it's more fun in the Philippines for me, but frankly, as a rich 1%-er, I have properties in Greece (where Lindsay Lohan now lives, I should visit her and say hello since some relatives operate a club there) and DC (I'm in DC now, most of our money is here).

@GofT - I loved using Bjorn Lomberg when I trolled the AGW sites as a GW-denier but frankly his science is bad. @EdR - RatInPutinsMaze is correct, and the lower estimate of 11 inches to 38 inches is a big deal since geography plays a big role. Even a low vs high tide of 1 foot is a big deal (see: Do you want Florida to become Holland (Everglades is between 0 to 8 feet above sea level)? @rayward - yes counsellor, I can relate to that.

Bonus trivia: I started the meme that Al Gore invented the internet. I sent him a letter for an advocacy matter with that phrase ('father of the internet' is I think the exact phrase I used) and shortly thereafter he or his aides used something the meme was started (See: Coincidence? I think not. Ray Lopez, social media influencer.

sotto voce, savoir faire, a fortiori "seemed less to imply forgetfulness than to deplore the flatness of an old joke." --Nabokov

Why the heck do they cut off the graph at 130 and $230k? I am very suspicious of a graph that shows so many IQ 80 people with high net worth. If you look closely it seems there is a huge IQ-wealth effect anyway. Look at the zero wealth line- it's thick with dots on the left, no dots on the right. Could it be all those dots are up above the top of the graph? Based on the number of dots around the edges its clear these arbitrary cutoffs are cutting out a lot of relevant data.

That "zoom level" is sufficient to show a band of income, mostly below $80k at IQ 120, which might seem low to many.

But when you think about it, why not? I know a guy who got an ivy league law degree, and then just bummed around the world teaching tennis at hotels.

Again someone choosing a lifestyle rather than maximum marginal product.

What is there to think about? The chart is terrible but clearly shows a strong correlation between IQ and wealth.

In 1994, Phifer attended an open-casting call for director Spike Lee's Clockers, beating over a thousand others to get the lead role as a narcotics dealer embroiled in a murder cover-up.

So the entirely expected result is that someone IQ 120, making $40k working in a bookshop, with $200k net worth says "thank God for that correlation?"

No. Life choices matter much more than post hoc correlation.

And that person probably doesn't want to start over pushing funds at Fidelity.

Appreciate the halibut.

For supposedly literate people, this is hilariously stupid.

It doesn’t control for age. You morons.

Think about measuring wealth when it doesn’t control for age. Then kill yourself.

+1, I noticed that too.

One assumes that over time, People accumulate wealth. IE it wouldn't surprise anyone that a 65 year old factory worker with an IQ of 90 probably has a larger net worth than a 29 year old with an IQ of 125. However, the same 29 year old will on average have a higher net worth 30 years later.

Shrug. If there is a go-to move in comments it is to reject data and offer nothing better.

There is even a name for it.

In other words, find and link your data.

I have more data!

"How important is intelligence to financial success? Using the NLSY79, which tracks a large group of young U.S. baby boomers, this research shows that each point increase in IQ test scores raises income by between $234 and $616 per year after holding a variety of factors constant. Regression results suggest no statistically distinguishable relationship between IQ scores and wealth. Financial distress, such as problems paying bills, going bankrupt or reaching credit card limits, is related to IQ scores not linearly but instead in a quadratic relationship. This means higher IQ scores sometimes increase the probability of being in financial difficulty.'

The IQ's Corner piece you're linking to actually ridicules the idea that there's no link between IQ and wealth.

The 2007 piece is by my co-author, psychologist W. Joel Schneider, and he's ridiculing the practice of routinely drawing causal claims from multivariate regressions. Here's one of Schneider's quips from the piece:

"I Am as Tall as the Rocky Mountains! (After controlling for barometric pressure)"

This IQ/Wealth paper is a classic result of "overcontrol bias," a real danger in the age of cheap statistical software.

Concluding that higher IQ doesn't increase wealth after you control for X, Y, and Z has a lot in common with claiming that McDonald's is just as good as a Michelin 3-star restaurant after you control for service, quality of ingredients, and chef expertise.

In other words, it's a mistake, a mistake I call the McMichelin Regression.

And more!

"New research has found that people who score higher on intelligence tests end up with the same net worth as others when lifestyle factors are taken into account. And the study confirms that you don’t have to be smart to be wealthy."

Article: “IQ is clearly overwhelmed or trumped by the cultural imperative to consume,” says economist Richard Wolff at the University of Massachusetts-Amherst, US. “People with higher IQs are acutely aware of all the goods and services that they can consume,” he says.

Wolff believes that smart people often have high expectations for what they deserve. “It’s a notion of ‘That’s what I’m entitled to as an American – that’s what I get for working hard.'”

Obvious, but almost never mentioned due to halo effect on places like this, where it's assumed that only "bad" people consume at high levels relative to incomes, and that smart people are "good" who'd rein in their spending if they had lower income and only low income who are poor are like that because they are dumb.

Steve Sailer is shocked, shocked!

This is similar to the point you made about the same finding earlier this month:

I would like to see a DAG written for this model. It is my understanding that one causal assumption is that past values of covariates do not have direct effect on outcome, which it is probably not true. It is easy to think how past values affected directly their forecasts, say if they got a loan or not in the past, or if they worked in finance or not in the past, or if they studied economics in the past etc.
Thus, a lot of confounders, which will bias the estimative.

individual IQ => ability to understand compounding effects => inflation predictions

It's probably not that simple, but not much more complicated either, confounders being present but not fatal. My takeaway from Superforecasting is the correlation between IQ and predicting real outcomes breaks down at higher IQs in the one to two std dev range, so I wonder what if anything IQ or behavioral economics says about people with IQs higher than say 120.

I'm skeptical about this claim. Even among very smart people, higher IQ correlates with better performance at almost everything. It would be quite surprising if forecasting (which is an exercise in careful abstract thought and probability calculations) was an exception.

"High-IQ men are also less likely to round": only high IQ men who are innumerate or lack a sense of proportion. What sort of chump would predict an inflation rate of 3.4567%?

One performing calculations in a fact/prediction landscape wherein that is the native level of precision?

Oh, if only we could predict future outcomes to four significant digits.

When predicting inflation, rounding to single digits is probably over-stating the accuracy of prediction. In fact, a better answer would be, "We think there will probably be inflation of some kind, and it will likely be higher than what it is now, but we aren't certain, and in fact we could be completely wrong not just in magnitude but in sign as well. Because the future is pretty damned unpredictable."

Question: If you have a model which seems reasonable, and you predict inflation to three decimal points because the 'fact/prediction landscape' creates inputs of similar precision, how much should you discount that precision when adjusting for the fact that similar predictions using similar models in the past produced results barely discernable from chance?

If your best estimate is 3.14159% inflation, and you got that by some kind of calculation, then overall, you should estimate exactly that--you'll get (very slightly) better results on average that way than by rounding.

However, you should assume that the error bars on anyone's estimate of inflation are very wide, and that doesn't depend on how many significant figures they include.

If you keep coming up with highly precise numbers for what is certainly an extremely imprecise prediction, you are doing it wrong. Significant digita mean something. They are supposed to be an actual representation of the known uncertainty of a result. Just because the numbers come from 'some kind of calculation' does not mean they are accurate.

If you keep getting highly precise yet consistently wrong numbers, you are either missing terms you don't know about, or you have previously assigned certainties to numbers for which there can be no such certainty.

Speaking of which: just how many high-IQ males regularly consult "the consumer Euler equation" to make informed guesses about rates of consumption and inflation?

I have stop looking at Pew Research data because they rounded everything to 1 descimal place, even when the expected data are much less than 1. Useless as stepping stones to calculate something else. To me Pew data are just useless.

Would like to hear Nassir talebs thoughts on this one...

They are all fooled by randomness!

I would too, actually. Or rather, hear his thoughts explicated by someone other than himself, as he presents them so incoherently: a couple charts, terse remark about distributions and tails, "gabish"? I would love a robust explanation of why the definition of "success" should not vary as you go from the merely above-average-IQ to the very high IQ.

Yet another paywalled article?

From NBER:

"Access to NBER Papers
You are eligible for a free download if you are a subscriber, a corporate associate of the NBER, a journalist, an employee of the U.S. federal government with a ".GOV" domain name, or a resident of nearly any developing country or transition economy."

To the rest of us, NBER says "F*CK YOU!!!"

Whey Men only for the paper?

Why men only?

It's the patriarchy.

Because women are stupider than men.

Your mom was kind of an idiot.

In what decade or century might behavioral economists begin documenting and assessing the conduct of business leaders?

If the spectrum of "lying" is as intrinsic to contemporary business and corporate practice as our tech tyrants tell us practically every day now when will metrics be devised with which to measure the volume or intensity or commitment to lying that an executive or set of executives might model?

--or should "lying" properly be construed simply as an aspect of "ruthlessness" (another time-honored business practice that must admit correlates that some algorithm or metric could measure at least approximately)?

Lying and ruthlessness are part-and-parcel of American business conduct: when might behavioral economists begin taking these behaviors seriously?

Has any behavioral economist yet devised an accurate metric for the prevalence or incidence of "trust" in the economy, given the egregious lying to which business leaders today routinely resort?

Comments for this post are closed