Thread by @Noahpinion

1/I have some other evidence that the gender difference in racial income gaps (nytimes.com/interactive/20…) may exist across races in the U.S.

It's from this paper: nber.org/papers/w24394

2/Here is prime-age mens' income by race across generations, as a percent of whites', after controlling for education.

As you can see, for a given level of education, white men continue to out-earn men all other races, including Asian men.

(from nber.org/papers/w24394)

3/Now here's the same graph but for women instead.

Notice that by the second generation, for all races, the earnings gap with white women (conditional on education) basically disappears by the second generation.

4/Note that conditional on education, black, Asian, and Puerto Rican 2nd-generation American women out-earn their white counterparts.

The opposite is true for men.

5/There appears to be some factor specifically elevating the earnings of white American men.

What is it?

Have American institutions been set up to reward things that white men naturally do better than all other groups?

I highly doubt it.

6/My hypothesis is that it's due to human networks.

I believe labor economists have dramatically underestimated the role that human networks - or, in common parlance, Good Ol' Boys - play in determining wages and other income.

7/In other words, my hypothesis is that white men make more, even after adjusting for a bunch of other factors, because of a self-perpetuating good-ol'-boy network of white men in companies and professions, who award each other money due to affinity.

8/In the coming weeks, I will look for research into the Good-Ol'-Boy Effect, and report on my findings.

In fact, I already know a few papers, such as this one: academic.oup.com/rfs/article-ab…

9/I think this is an important area for research, and an area where economists need to borrow ideas from economic sociologists.

(end)