David R. Henderson  

The Great Grandson Also Rises

When ideologies change... Meant for Each Other: Open Bor...

Our former co-blogger, Arnold Kling, has an excellent review on Econlib of Gregory Clark's latest book, The Son Also Rises. The review is titled "The Heritability of Social Status."

You may have noticed that Clark, an economist historian at UC Davis, likes to come up with titles that sound almost like those of Ernest Hemingway. He does so even if he has to shoehorn the title to fit the content. A Farewell to Alms, his previous book, wasn't accurately titled because that was hardly the main message of the book. This one, The Son Also Rises, is at least closer to accurate, although a better title would have been the one I use for this post. I look forward to Clark's next book, which is rumored to be about a former Treasury Secretary in the Clinton Administration and to have the title, The Dangerous Summers. If Clark branches out to other writers for his titles, maybe we can look forward to another book titled The Class Menagerie.

Back to the more serious issues. The reason for my title is that the book is really not about whether the son of the successful father is successful by some social status measure. It's about whether subsequent generations are. And Clark shows that they are.

Arnold does a nice job of laying out Clark's argument. One key paragraph:

Clark and his researchers looked at multi-generational outcomes on a variety of measures in several countries. They concluded that under many different institutional arrangements and across many time periods, the true correlation across generations in social status is somewhere between .7 and .8, which is much higher than most conventional estimates. In short, persistence of social class is much higher than most researchers believe it to be, based on single-generation correlations that are biased downward by measurement error. [Italics his]

I have two personal stories to illustrate the point. I hasten to add they illustrate, not that two anecdotes in themselves are strong evidence.

The first is about my heritage. My grandfather, born in 1855, was a wealthy apartment owner in Winnipeg. His three sons were, from oldest to youngest: a fairly low-paid minister, a moderately paid high school teacher, and a medical doctor. The middle one was my father. I'm not as high up in the current income/wealth category as my grandfather was in his, but I'm way higher up than my father was.

The second is from Jeff Hummel and it's solely about perceived IQ, not social status. Back in the 1990s, he was invited to have Thanksgiving with his friend David Friedman and David's family, including Milton and Rose Friedman, Aaron Director [David's uncle and Rose's brother] and David's kids, Becca and Bill. I still remember Jeff's line when he came back from that event: "If there's reversion to the mean in IQ, I sure didn't see it in that family."

Back to the big picture. So what? Arnold argues that there are two so whats, one for life arrangers (my term for Arnold's term, "social engineers") and one for libertarians.

For life arrangers, the moral of the story is: "[H]is [Clark's] findings argue against extensive efforts at social engineering that try to achieve parity across groups."

For libertarians, the moral is:

[H]is findings argue against the need to create strong incentives to succeed. If some people are genetically oriented toward success, then they do not need lower tax rates to spur them on. Such people would be expected to succeed regardless.

I think Arnold goes too far on this latter. Wouldn't marginal tax rates change incentives on the margin? Can't we imagine someone who is successful at a 50% marginal tax rate (a rate that many people in the United States now face at the combined state and federal level) who would be even more successful if the combined marginal tax rate were, say, 40%? Also, of course, libertarians have not focused just on marginal tax rates. There are many other government barriers to success. Simply having to get government permission for an innovation--whether in retailing cars or introducing new pharmaceuticals--is a huge barrier to success.

But there's an even more important point to make about Clark's findings, if true (as, I suspect, they are.) It is that all of them are about where one is in the social pecking order. What if you, like me, don't care much about the pecking order? What if you want pretty much everyone in the world, except for evil dictators and a few others, to be better off?

In his review of The Son Also Rises in the Wall Street Journal, Trevor Butterworth wrote:

If policy intervention can have, at best, only a modest effect on social mobility, what policies should we pursue? If we are lucky, bold and careful thinkers will take up this challenge. If not, we are all going nowhere fast.

Really, Mr. Butterworth? Nowhere fast? As Brad DeLong laid out in his appropriately titled NBER study, "Cornucopia: The Pace of Economic Growth in the Twentieth Century," we've gone somewhere very fast.

I don't care if my daughter is in a high social class as long as she is happy and lives long. And what is most likely to assure that, to the extent institutions affect that, is freedom.

Comments and Sharing

COMMENTS (22 to date)
Joel Aaron Freeman writes:

Interesting data, David.

And I share your reluctance about his overall claim. The average South Korean earns as much income as ten North Koreans do, despite have the same genetics and culture. Clearly incentives matter.

brendan writes:

Joel, do you really think Kling and Clark attribute the North v. South divergence in Korea to genetics? Who are you refuting?

Jody writes:

On marginal tax rates and the incentive to succeed.

1) There's success and then there's success.

That varying systems result in the same rank ordering (this is approximately Clark's result) only provides insight into the ordinality of the outcomes, not the cardinality.

The inability to re-sort ordinality matters to those concerned with positional jockeying (it's nigh impossible if this is a concern for social planners). But for those that care about the cardinality of the outcomes (utilitarians, those looking to maximize the minimum and so on...) ordinality is irrelevant.

Since I think libertarians care much less about ordinal social rankings (see reactions like, "inequality? who cares?"), I don't see where the marginal tax rate supposition follows at all.

2) From a central planner's perspective in a mixed economy, in addition to incentivizing increased production, the design of marginal tax rates obviously also serve a distributional function. And this resource distribution is an input for new production (capital is loosely accumulated income).

But when allocating resources to maximize societal production, it makes sense to give the most resources to those shown to be most capable of doing more with it (assume income is well-correlated with production). In which case, a progressive marginal income tax rate is also at odds with maximizing production as it (proportionately) takes resources away from the most capable and gives them to the least capable.

Joel Aaron Freeman writes:

Brendan, I am reacting to Kling's statement:

"Such [success-oriented] people would be expected to succeed regardless [of incentives]."

I understand that my comment is about a disparity between two macroeconomic outcomes, whereas Kling's comment is about a disparity between two individuals inside the same economic system.

The reason I feel comfortable equating these two disparities (one between countries, and the other between individuals within a country) is because I'm conceptualizing a country as an aggregation of individuals. Hence, if South Korea is more economically successful, it's because more South Koreans have individually "succeeded." If North Korea is less economically successful, in my mind, it's because many North Koreans who wanted success weren't able to get it. That's my thought process. Feel free to correct me.

brendan writes:


Conventional wisdom assigns no importance to genetics in explaining group variances, and some importance to individual variances. Regarding the genetic influence on individual variances, it is assumed that mean regression and random mating remove the effects of genetics within a generation or two.

That is conventional wisdom.

Deviations to conventional wisdom take the form of, a) those who deny genetics any role, and b) people like Clark who say genes are underrated, but who still assign high weights to culture, institutions, etc.

People in camp b are routinely strawmanned as genetic determinists.

When you suppose Kling or Clark think genetics explains Korea, you're strawmanning them and implying their insanity.

You didn't mean to, but: could any sane person believe N. and S. Korea's divergence is genetically based?

Kling's point is simply this. There is a fixed amount of variance to explain. If genes explain more variance, then something else explains less.

Somehow that point is repeatedly misconstrued as genes explain everything.

English Professor writes:

There might also be a different kind of lesson here about the value of free/open societies. I grew up in a Polish neighborhood in Chicago. My grandparents came from Poland around 1910. They had been farm laborers; neither could read or write. In Chicago my grandfather got factory work, and my grandmother scrubbed floors. My father was born in Chicago, but his native language was Polish. He learned English in grammar school and completed high school. My parents expected both my brother and me to go to college. We both got advanced degrees (he got an MBA, I got a PhD). This kind of upward mobility was wide-spread in cities like Chicago. I knew dozens of kids in high school who came from backgrounds like mine who subsequently went to college and on to professional careers.

Here's the point. My grandparents came from a society in which the bulk of the population got little or no education. There was obviously a lot of talent in those Polish peasants, but they had no means of expressing it. It was only in a free society that their grandchildren could achieve something.

brendan writes:

English Prof:

Same story for my grandparents, but substitute NY for Chicago and Italy for Poland. Same story for millions more.

But we're misled. I used to think, "hey, if a sicilian shoe shiner can do it, anyone can."

But poor 1st gen european immigrants are not the same as poor 5th gen american's. The immigrants poverty reflects circumstance and tells you little about his talent. The american's poverty is much more likely to correlate with low talent.

The world is much harsher than either the "education everybody" Left or "inculcate virtue + free markets" Right believes.

Steve Sailer writes:

For Whom the Bell Curve Tolls

The Haves and the Have-Nots

Joel Aaron Freeman writes:


I see your model. It says that some people have untapped talent, and only in a good system can that talent be expressed. However, if the system is ideal and that person still isn't succeeding, it might be due to innate characteristics. That is, there are SOME limits to how much policy can compensate for innate characteristics.

I believe there are no limits, and that social mobility can be high for the entire population. I'm not straw-manning the opposing argument, I just have a different understanding of causality.

I think about individual variation like how plants need water. Some plants, like the cactus, grow fine with barely any water. Some plants, like tropical flowers, need a lot of water or they die. A very talented person is a cactus, and they need only a weak incentive structure to flourish. However, someone else may need a well-designed incentive structure to flourish. Under this view, there is no innate characteristic that can thwart social mobility. We need only choose the correct policies, and social mobility will be available to everyone.

I consider America to be a C at providing incentives for good outcomes, and therefore bad outcomes don't surprise me. School vouchers would be a big improvement. Creating a welfare system that NEVER punishes someone for increasing their income would be a big improvement. Eliminating black markets through legalizing recreational drugs would be a big improvement. If you did all those things and didn't see a dramatic increase in social mobility, I would reconsider my views.

brendan writes:

Aesop's Fables: Libertarian Edition.

MikeP writes:

Steve Sailer, your proposed titles are truly masterful.

But I notice you didn't do Across the River and into America.

David R. Henderson writes:

@Steve Sailer,
Good ones. I was working on those two titles and didn’t think of yours. Well done.

Foobarista writes:

The type of "success" that society allows is hugely important. If you don't allow smart people to get super-rich, these people may seek power and status in other ways, such as by going into the bureaucracy.

This is what you have in places like Russia, China, and North Korea - which often have quite brilliant people in their governments - and is not something we'd want, as they use their "smarts" to screw each other and prey upon the populace.

At least someone who gets super-rich by building a good company creates vast social value. A corrupt, predatory bureaucrat, not so much.

Tracy W writes:
The first is about my heritage. My grandfather, born in 1855, was a wealthy apartment owner in Winnipeg. His three sons were, from oldest to youngest: a fairly low-paid minister, a moderately paid high school teacher, and a medical doctor. The middle one was my father. I'm not as high up in the current income/wealth category as my grandfather was in his, but I'm way higher up than my father was.

You only state your paternal heritage. But your paternal grandfather only supplies on average 1/4 of your genes.

Tracy W writes:
If some people are genetically oriented toward success, then they do not need lower tax rates to spur them on. Such people would be expected to succeed regardless.

I think though the question matters is what sort of success? And where?
If tax rates are high, someone has the choice of moving to another country (as successful British people did in the 1960s and 1970s), or putting efforts into forms of success that aren't taxed, such as leisure time or fame, or putting effort into tax avoidance, which is a social negative. In Sweden:

Even in the later 1980s, Swedish doctors worked only some 57 percent of the hours that American doctors worked. And as early as the 1970s, doctors, dentists, lawyers, and so on worked only a few months each year—to avoid astronomical tax rates.


David R. Henderson writes:

@Tracy W,
You only state your paternal heritage. But your paternal grandfather only supplies on average 1/4 of your genes.
I’m working within Clark’s framework. That’s all he’ll look at by looking only at surnames. So your criticism is a criticism of his methodology.

David writes:

One quick point:

If some people are genetically oriented toward success, then they do not need lower tax rates to spur them on.

I often see this type of thinking, which I believe is backwards. The true point, I believe, is that if you give a random person an extra $100, that random person consumes it. But if you give someone that creates wealth $100, he will probably use it to create even more wealth.

I know my efforts to create valuable things for society are severely limited by my resources - and my personal resources are far larger than average on this planet.

English Professor writes:

On the question of incentives, I think Foobarista nailed it. We can have incentives that reward innovation and creativity, or we can reward successful place seeking and political manipulation. I suspect that the progressive vision, in which the state tries to guide the economy (whether through subsidizing green energy or excessively regulating the financial sector), draws the brightest people towards rent seeking and crony capitalism, thus discouraging innovation.

Hazel Meade writes:

I wonder if, to some extent, people's economic success depends upon their own perceptions of their own social class.

That is, if you know that you have a wealthy or famous ancestor, maybe that would encourage one to strive harder. It would be proof that success is possible. There might also be family stories about how so-and-so got his start.

I bet one could devise an experiment over maybe a year's time scale, where you told various children that you had discovered they had a certain wealthy great-great-uncle, even though the wealth was inherited down a different family line or something. Concoct an elaborate story about it and then observe their academic progress over the course of the next year, and compare it to a control group.

RickC writes:

Hazel Meade, I believe this sort experiment is carried out every day in the U.S. My oldest friend's family were, at one time, large landholders in Georgia with roots that led directly to the Randolph family of Virginia. They lost it all after the Civil War and hadn't recovered even by the time my friend was born in the early 1960s. My friend's dad was a small engine repair man with only rudimentary education. The family (three boys, one girl) would easily fit into the poor, working class category - very poor. My friend related to me that his grandmother would constantly remind them not to buy into their current situation, that they came from better stock.

My friend and one brother have made themselves successful. The brother is an independent contractor specializing in HVAC applications for homes and businesses. My friend is currently a project manager for the largest elevator company in the world, has had five books published on interrogation/body language and related subjects using skills he acquired in the military. He's also appeared on Fox, CNN and other news outlets as an expert on body language during coverage of high profile criminal cases.

The other two siblings are, by any measure, social and economic basket cases completely dependent on welfare, food stamps, Medicaid etc.

Clearly, some mixture of genetics and environment is at work here but I don't see any way to quantify or predict these outcomes, much less develop policy or an economic model which would have everyone succeed.

Troy Camplin writes:

I don't need lower tax rates, but I most certainly could use elimination of a variety of barriers to entry, most notably licensing and other regulations.

My mom came from a middle class family. My dad came from the poorest of the poor in western Kentucky. I was raised working middle class -- my dad was a coal miner. My brother and I are the only two offspring of our parents. He has a MFA and I have a Ph.D. I'm living more or less the same lifestyle my father did -- and part of it is due to regulations blocking me from doing the sorts of things I would like to do (thus raising the cost of starting businesses too high for me), while another part of it is institutional prejudice against people with Asperger's (bureaucracies, especially).

Vicky Smith writes:

I don't know if some of the commentators recall that before Woman's Lib, before about 1970, US parents obtained jewelry or other goods in expectation that their Daughters Would Marry.

And to sons of other similarly situated people.

Therefore, it didn't matter if one's grandmothers were uneducated farm girls, shop assistants, or others: it mattered that the grandfathers Did The Right Thing and married before the birthing.

And stuck with it.

And had the drive to put their children within reach of the children of other driven men.

I don't know if the emergence of "I can have a career AND kid(s)" women will alter this to "my grandmom is a bank VP, and my parents met at (fancy college x)". Maybe by 2050.

In the interim, perhaps:

My grandma and aunt put me through college, so I got to do x and met my love and I got to be a mom before I'm 30!!!(smiley)

Comments for this entry have been closed
Return to top