Suppose you want to measure how inherent intelligence is distributed amongst the population.  That is to say, you want to get a sense for how many are born with the potential to be very smart, and how many are not, and what the capabilities of each are.  How would you do it?  The most trusted attempts to measure such quantities probably come from standardized testing, such as IQ tests.  More common but perhaps cruder metrics include academic degrees (perhaps weighted by relevant factors such as the type of college), and income level.  Given the distribution of intelligence measured by one of these means, what could we then conclude about the distribution of innate ability?  

IQ tests are intended to test for problem-solving capabilities not dependent on previously acquired knowledge.  This kind of intelligence, as measured past some critical young age, is popularly believed to be innate and immutable.  In 2008, psychologist Suzanne Jeiggi of the University of Michigan, and colleagues, endeavored to test this immutability hypothesis [1].   Inspired by studies showing a link between IQ and working memory, they developed exercises intended to improve working memory.  

Subjects were tested for the types of problems found on IQ tests, then trained to improve their working memory, after which they were retested.  Meanwhile a control group took and retook the tests at the same times but without the training in between.  The trained group showed dramatic improvement, while the control group did not.  Moreover, the amount of improvement depended on the amount of training -- the more the subjects trained, the more they improved.  No upper limit  to this trend was found during the course of the study.  And strong improvement was found regardless of the initial test performance, although the weaker performers on the first test benefitted more from the training than the stronger first-testers.  Overall, the study suggests, IQ can improve significantly with appropriate training, for anyone.  Therefore, IQ is not an adequate indicator of innate ability.

What about educational achievement, or income level?  Beginning in 1976, a social welfare program referred to as the Gautreaux program facilitated the relocation of families from public housing in Chicago to higher quality housing elsewhere.  The families, who all started out at comparable income and educational levels, were arbitrarily relocated to either urban areas or slightly more affluent suburban areas.  These suburban areas had some advantages over the urban areas, most notably: less crime, and better schools (better paid teachers, higher average ACT scores, etc.).  The random nature of this relocation, together with the fact that the program kept tabs on its participants for more than ten years, afforded a unique opportunity to compare the effects of these disparate environments on the children growing up in these families.

Sociologist James Rosenbaum of Northwest University gathered data from hundreds of the families involved in the Gautreaux program, over a decade after the relocation [2].   He found that, in contrast to the urban children, the suburban children were  2.6 times as likely to attend college, and of those attending college, they were 2.5 times as likely to attend a 4-year college.  Further, they were 1.8 times as likely to be employed, and of those employed, their average income was higher.   Clearly, these differences had nothing to do with innate ability, since the initial relocation had nothing to do with innate ability.  The implication is that, had all the families been relocated to these better suburban neighborhoods, then all of their children would have been more likely to go to college, etc.

In summary, high intelligence, or the potential for such, may be much more broadly distributed than popular metrics would indicate.

[1]  S. Jaeggi, M. Buschkuel, J. Jonides, W. Perrig, "Improving fluid intelligence with training on working memory", Proceedings of the National Academy of Sciences 105, 6829 (2008).

[2]  J. Rosenbaum, "Black pioneers - do their moves to the suburbs increase economic opportunity for mothers and children?", Housing Policy Debate 2, 1179 (1991).