It is fair to conclude this double trilogy on income inequality with a brief review of some of new research on the future of social mobility. I warn you gentle reader, this topic might be unnerving.

A century ago, most Americans married young. Their associations were local and marriages tended to be of people who were alike in race, religion, culture and geography. This necessarily smaller pool of eligible mates (or marriage market in unromantic economic jargon) meant that other factors such as intelligence and education played a smaller role than race and religion in choosing a partner.

Among the great achievements of this republic was that in the decades following World War II, we began to marry with less regard for race, religion and culture. We married later in life and with far fewer geographic restrictions.

One result of this is that the marriage market shifted from small towns to colleges and workplaces. So, educational attainment, not race and religion, became a more important factor. We plainly see this all around us. It is happily unremarkable to see mixed race families, or marriages between Methodist and Catholic that were difficult even a half century ago and nearly impossible a century ago (I use examples from my own family). Yet, it is unusual to see college grads and high school dropouts marry each other.

Here is where I get a bit unsettled. You see, this research comes to economics from sociology, anthropology and evolutionary biology. Their interest is in the social and biological changes wrought by these changes to mating patterns. For you see, factors such as race and religion are uncorrelated with intelligence. Not so with educational attainment. So, if humans are now more likely to mate with people who are more like them in intelligence, there are evolutionary implications which matter greatly to income mobility.

There are several important studies floating around today that address this issue. From what I can tell, they all come to the same broad conclusion. Changes to assortive mating may, within a few generations, make evolutionary changes to intelligence in our society, splitting the bell curve of smarts into two separate populations.

One of the more interesting papers on the subject attribute almost all the increase in income inequality over the past fifty years to assortive mating, but not through human evolution. As it turns out, better-educated people are considerably more likely to marry before having children, far more likely to be employed and stay married. The math here is simple. Assortive mating leads to higher levels of income inequality between households because individuals with higher income are more likely to marry other high earners, and vice versa.

So it gets back to basics. Choices determine how much schooling we get, where we live, who we marry and when we bear children. These choices largely decide household earnings. While it is alluring to wish away the power of choice by blaming foreign competition, racism, or bad tax policy for income inequality, those theories do not fit the facts.­