Knowing the average IQ of rich folks is important because money is one of the most obvious and important signs of intelligence and adaptability, so for IQ to have validity and relevance, rich people should score high.

And yet the correlation between IQ and networth is only 0.16 (Zagorsky 2007).

In his book, The Millionaire Mind, Thomas Stanley interviewed hundreds of millionaires and asked them about their SAT scores and grades.

At the time the median millionaire was around the top 1% of wealth. If there were a perfect correlation between IQ and wealth, we’d expect they’d have an average IQ of 135 (35 points above the U.S. mean of 100) but given a 0.16 correlation between IQ and wealth, we’d expect they’d average only 0.16(35) + 100 = IQ 106

And yet, after adjusting for self-reporting bias, Stanley estimated they averaged 1100 on their SATs. Stanley argued this was mediocre and proved IQ has little to do with money, however for their era, an 1100 equates to an IQ of 120 (14 points higher than predicted from the IQ-wealth correlation). And while 1100 may sound low compared to the stratospheric SATs of Ivy League students, we must understand that Ivy League students are selected for their score on the SAT and thus regress precipitously to the mean on all other tests. By contrast millionaires are selected by life’s IQ test, and thus regress precipitously to the mean on the SAT. In other words, the SAT overestimates Ivy League students, but correctly estimates millionaires.

But why did the 0.16 correlation underestimate the IQs of millionaires? In my opinion it’s because for most people, net worth is largely a function of how much you spend, and since smart people realize money is useless unless you spend it, a lot of smart people will end up with low net worth, thus driving down the correlation.

However there’s a limit on how much you can spend on yourself, and so the more prosperous you are, the less spending habits correlate with net worth.

To avoid this complication, it’s better to think of millionaires not as people who have a lot of money, but as people who have made a lot of money (or their heirs). This allows us to use the correlation between IQ and income instead of the correlation between IQ and net worth.

The correlation between IQ and income is about 0.30 (Zagorsky 2007) , but increases to almost 0.4 when you look at permanent income (which is most relevant for predicting wealth) and increases to about 0.5 when you look just at white men (the group the vast majority of millionaires belong to). Using these correlations would have made the predicted IQ more accurate.