Thursday, January 22, 2009

Credit scores and IQ

GNXP's Razib alerted me to's listing of average credit scores by state*. Not surprisingly, the scores correlate strongly with a couple of the usual suspects. With the percentage of a state's population that is non-Asian minority (NAM), the correlation is an inverse .74 (p=0). With IQ, the correlation is an even stronger .77 (p=0).

Because these relationships are somberly unsurprising, they're not likely to be given any attention by Important People, anywhere, ever. (That is, unless to point out societal racism, evinced from racial disparities on the blank slatist assumption that differences in groups are entirely contingent upon external social forces, or to argue for more educational spending to combat American profligacy**).

Living around people who are not up to their eyeballs in debt makes for a better quality of life. Economic downturns are less disruptive, business is easier to conduct, neighbors are more likely to take care of homes they have equity in and cars they're not half-expecting to be repossessed, etc. A state's average credit rating could easily be included among the 44 social, cultural, and economic factors CQ Press uses to calculate its livability index. That index correlates with a state's estimated average IQ at a nearly identical .80 (p=0).

A visualization illustrates that unlike the red-blue political divide conceptualized as pitting the coasts against the nation's interior, when it comes to desirable places to live (weather excluded), a north-south divide emerges.

Steve Sailer excerpts from George Will's obituary for the late Senator Moynihan:

The Senate's Sisyphus, Moynihan was forever pushing uphill a boulder of inconvenient data. A social scientist trained to distinguish correlation from causation, and a wit, Moynihan puckishly said that a crucial determinant of the quality of American schools is proximity to the Canadian border."
Richard Lynn has suggested that a north-south IQ continuum has evolved through selection for greater intelligence in less hospitable environments. It seems at least plausible that colder environments also selected for greater resistance to and tolerance of cold in the process. Which, in addition to historical settlement patterns, might offer some explanation as to why the north-south contiuum continues to exist (not just in the US, but internationally as well^).

Speculatively, moving people closer to the Canadian border isn't going to magically make people more more intelligent or better neighbors, but warming in colder areas of the world might have a eugenic effect by opening up land and resources for populations living in them. Since cold weather isn't as threatening to humans as it used to be (especially in developed societies, which tend to be found in relatively colder places than less developed societies are), even over the long-term, assuming a secular warming trend continued, the effect presumably would not swing back in the dysgenic direction.

* It isn't clear how the averages are computed. They might be representing either the mean or the median by users from each state, or an average of all credit reports based on state of residence. As long as the method is standardized, though, it's useful, unless there are reasons to suspect the method is artificially skewing states to different extents, of which I am currently unaware.

** Educational attainment, as measured by my 'education index plus', does not correlate as strongly with credit scores as McDaniel's estimated average IQ scores by state does. More intelligent people appear to be better at managing their credit than less intelligent people appear to be. Educational attainment is probably predictive primarily because it serves as a proxy for intelligence.

^ Internationally, it's a distance-from-the-equator continuum, to be more precise.


Reason said...

This sounds verrry speculative, especially when you consider large interstate migration flows. My parents, for example, moved from Minnesota to New Jersey; I'm now in New Jersey and will move to wherever the med schools are. (And especially relevant to your thesis, don't forget the huge black migration northward in the early half of the 1900s.)

Blode032222 said...

Sounds pretty plausible. IQ correlates positively with income, and the propensity to delay gratification. It used to correlate very well with SAT scores (I suspect it still does, weakly); SAT scores correlate very negatively with dropout rates ... and I can't imagine defaulting on student loans is good for your credit rating. But I haven't connected all the dots myself, data-wise (like how many dropouts were in school because of student loans, etc.)

Audacious Epigone said...


Interesting point. I am only able to speculate on it.

More than half of the populous has never lived outside the state they were born in. The South gains the most people through internal migration, while the Northeast loses the most. Are the people leaving of lower credit quality than those who are staying? Since college graduates are more likely to move than the rest of the country, that seems unlikely. To the extent that patterns of movement skew the relationships, wouldn't they do so in a random way? If that is the case, then the actual relationships are even stronger than suggested here.


Although there was a flawed paper a couple of years ago arguing that wealth and IQ are not correlated, I suspect the truth is that wealth and IQ are correlated roughly as strongly as income and IQ are (in the .15-.30 range).