The Social Mobility Illusion

A short summary of the most remarkable economics book I've read this year, Gregory Clark's The Son Also Rises (2014)

Across the Western world, many people earn much more than their parents, and many earn much less. Under modern capitalism, every generation gets a pretty fresh start. Or so it seems.

In the U.S., the correlation between a person’s income and their parents’ income is about .6. That’s not too high. It implies that about 36% of all the variation in earnings can be predicted by looking at parents. In the Scandinavian countries, where they try harder to promote social mobility, the correlation is about .2, which means parental earnings can only explain about 4% of the variation in the next generation’s earnings.

These data suggest that all family advantages and disadvantages get erased after about 3-5 generations.

These data represent the social mobility illusion.

In fact, about 56% of anyone’s general social status can be explained by one’s family lineage. The real correlation is .75, much higher than we typically estimate. Not only is social mobility much lower than people think, the real and lower rate of social mobility holds across a more general measure of social success. “Social status” includes earnings but also achievement in selective and prestigious occupations.

Conventional estimates of social mobility are misleading for two reasons: They focus on income, and they focus on one-generation differences. Money is more volatile than competence and ability, plus there is more randomness across one generation than there is across three generations.

When you look at general social status over many hundreds of years, the results are stunning.

Elites today disproportionately come from the same families that were elite hundreds of years ago—to a degree that is shockingly unknown to most people.

I never really grasped the illusion of social mobility until I recently read the provocative 2014 book, The Son Also Rises: Surnames and the History of Social Mobility by economist Gregory Clark. From which I’ve drawn all the data cited in this post, except for the American correlation coefficient at the top, which is more up to date.

Clark doesn’t just find that social mobility in the United States is overstated, he finds that there is little social mobility, in general, across time and geography. In England, Sweden, India, Japan, Korea, China, Taiwan, Chile, from the present to the medieval period, he was hard-pressed to find any intergenerational mobility correlations greater than .75. He even dares to call it a “social law,” and a “universal constant.”

Whereas we like to think that familial impact on social status gets washed out in 3-5 generations, Clark finds that it’s more like 15 generations (300-450 years).

Consider Sweden.

In 1626, the Swedes recorded an official list of their noble families.

About 56,000 Swedes today hold rare surnames listed in the old, official rolodex of noble families.

The average taxable income of people with noble names today is about 44% higher than that of people with a common surname such as Andersson.

A Swede who has the surname of a once-titled noble (i.e., a count or baron), is six times more likely to be a licensed lawyer. A similar pattern is found for doctors.

If this effect can be observed in one of the most equal Western welfare states, then it likely holds in other countries as well. You should read the whole book if you’re interested in the topic, but if you’re not, I can just tell you: This is what Clark finds from Chile to India to Japan to Medieval England.

On a side note, elite Chinese families before the Communist revolution were still more likely to be elite 50 years after the revolution. That was an all-encompassing totalitarian revolution explicitly designed to overturn inequality.

The underlying mechanism is genetic transmission of competence abetted by assortative mating (elite people selecting elites for mates). Chads & Staceys make babies, who sit at the Cool Table, and then make babies named Chad and Stacey.

But notice that assortative mating is crucial. If elites all started reproducing with commoners, social mobility would increase right away. And indeed, most of the social mobility we do observe comes from the fact that elites are never perfectly endogamous. Prince Harry and Meghan Markle, anyone?

Anyway, a close reading of the book brings some interesting implications, and a few fascinating surprises.

When highly successful people spend a lot of time trying to help their kids succeed, they are mostly wasting their time. Philistine mass society has convinced our elite genetic stock to worry about hundreds of things they don’t need to, then the over-achieving elite become the very best at things that don’t matter.

When people think of entrenched economic power structures, they think of a rich, reactionary CEO in a suit. That's a pervasive but misleading image. They should be thinking of a high-IQ doctor who marries a high-IQ lawyer, both of whom meticulously sifted for the most intelligent mate they could get. You can find such couples on the Left and the Right. Google "Tesla-themed wedding” for an image of economic domination that is much more realistic than the Monopoly Man.

Though it is not a main concern of the book, Clark’s data really drove home to me how the U.S. is perhaps the least White Supremacist country in the world. Due to our history of high-skill immigration policies, none of our most disproportionately successful ethnicities is white European. There are successful white Europeans but for every Gallagher who is a doctor or lawyer, there’s a Gallagher in prison, too. All the ethnicities who go most directly to the top are non-white.

Ultimately, I found the book surprising and compelling. For all intents and purposes, it is not too hyperbolic to say that social mobility is a myth. As we like to think of it, social mobility is largely an illusion that we fix upon random noise across short time intervals.

A final note.

Whereas we tend to believe equality is just around the corner and so we’re constantly disappointed, reading The Son Also Rises makes us feel like equality is far away but—for that very reason—more certain.

The African-American and the Ashkenazi Jew in the United States will eventually enjoy equal expectations for social success. It’s just going to take 300-450 years.