"I started off as a libertarian economist, but I've come full circle"—Gregory Clark on social mobility

March 14, 2014
article header image
In his State of the Union address in January, President Obama observed that those at the top of the income distribution were doing better than ever before. But at the same time, he said, "average wages have barely budged. Inequality has deepened [and] upward mobility has stalled." And in the land of opportunity, that is unconscionable.

One often hears British politicians, of all stripes, saying the same thing: that social mobility has slowed, if it hasn't stalled completely, and that policymakers (especially those designing education and training policy) ought to be doing something about it. If there's a political consensus here, it's not one that is replicated among those academics who study patterns of mobility and inequality. One notable dissident from this consensus is the British sociologist John Goldthorpe, who I interviewed for Prospect last year. In Goldthorpe's view, when politicians refer to a "Golden Age" of social mobility in the decades following the Second World War, what they're actually referring to are structural changes in the labour market which saw a massive expansion of professional and managerial employment. It's not that intergenerational mobility has slowed; it never grew in the first place.

Goldthorpe is sceptical, too, about the effect that educational policy can have on mobility rates. "More advantaged families will always use their resources to maintain their children's competitive edge," he told me. "This brings home the way in which inequalities of condition serve to maintain inequalities of opportunity."

He now has an ally in Greg Clark, a Scottish-born professor of economics at the University of California, Davis—though, if anything, Clark's message is even gloomier than Goldthorpe's. In his new book "The Son Also Rises: Surnames and the History of Social Mobility", Clark argues that conventional ways of measuring social mobility between generations have prevented us from seeing that it has always been much slower than we tend (or like) to think. In the standard picture, mobility rates are also held to vary dramatically across societies, with more unequal societies, like the US or UK, having notably slower rates than, say, the Nordic countries. Clark's research tells a rather different story. This is because, rather than tracking changes over two or three generations, as most conventional studies do, he tracks status over centuries using surnames as his guide.

His conclusions are chastening: "Underlying or overall social mobility rates are much lower than those typically estimated by sociologists or economists. The intergenerational correlation in all the societies for which we construct surname estimated… is between 0.7 and 0.9, much higher than conventionally estimated. Social status is inherited as strongly as any biological trait, such as height." The political implications of that are unsettling, to say the least.

I spoke to Clark when he visited London last week.

JD: In your account, then, there was no "Golden Age" of social mobility?

GC: Right. One of the things that will make this book controversial is that it’s claiming that all the standard methods for measuring social mobility in fact miss the mark, and are likely to find differences in social mobility across societies and time periods that are in fact just spurious. People look at income correlations across generations and extrapolate from that. What this book says is that although there’s a lot of random fluctuation in terms of people’s income or occupation, there’s a much greater underlying persistence. And only by looking at things like surnames do you see that feature.

If the standard ways of measuring social mobility are faulty, how do you measure it?

If you look at England, for example, what we measure is whether you were at Oxford or Cambridge; how long you live, which is another good indicator of social status; occupational status; are you a member of parliament? Now one of the interesting findings here is that it doesn’t really matter which measure you use. For the families we’re looking at, all these things are actually highly correlated. The wealthy at any time are also the educated, members of parliament, those who live long. What the book shows is that there’s an underlying physics of social mobility which all of our political efforts seem to have no effect upon. And the startling conclusion is that we may never be able to change social mobility rates.

That sounds like a counsel of despair.

No doubt people will read this as a gloomy book. But the title, The Son Also Rises, was deliberately chosen to emphasise that there are some very positive elements in it. One of the things it emphasises is that the current data, which finds rapid social mobility in Sweden and slower social mobility in Britain and the US, and slower mobility still in South America, seems to suggest that you have massive social failures going on in a bunch of societies. The book finds no evidence of these failures because it finds very similar social mobility rates everywhere. Another implication is that if even in meritocratic Sweden you get very slow mobility, then it must be based largely on people’s abilities, aptitudes and drive. All that we’re discovering here is that we’re living in a surprisingly fair world—one in which, at birth, we could predict a surprising amount about your prospects. Is that a gloomy fact about the world?

I suppose it depends what you mean by fairness! Is the UK experience as far as social mobility is concerned similar to the US one?

In terms of the estimates in the book, they look identical. One of the things the book is emphasising is that in a society like the US, there’s a lot more private expenditure by people on education than in a society like Sweden, where it’s mostly provided by the state. So you’d expect, on normal accounts of social mobility, that the US would be a more rigid society with slower rates of mobility. But we don’t see any sign of that. So my interpretation here is that whatever educational system you set up, however fair and however open-access it is, there are just families that are better equipped to figure out what they need to do in the system, how you get ahead. And it’s impossible to stop those processes.

Well, impossible unless you introduce highly illiberal legislation for which there’s probably little popular appetite.

Right. There are extreme cases like India where they say, “We’re reserving a quarter of all places at university for people who the British happen to have assigned to the various scheduled castes.” Though it turns that, in India, that most of the former “untouchables” don’t benefit from [the policy]. It’s people who were misclassified and are relatively middle-class who are able to take advantage. The surprising finding is that however you construct a social system, it’s still going to be highly predictable that from three or four generations ago elite families are still be found heavily among the elite.

So when President Obama said that social mobility had "stalled" he was making a mistake?

Because America is such an unequal society there has been more emphasis on the possibilities of social mobility. How else are you going to justify the incredible inequalities in the US? So it’s going to be very unwelcome news for people in the States that there really are very slow rates of social mobility. Now what’s interesting about this book is that its message seems to be equally unwelcome to both right and left. The left loves the idea that there are slow rates of social mobility. But they want to hold on to the idea that there’s going to be a political programme that will end this problem. But the book says that there’s absolutely no sign of our ability as a society to change that. The right hates the idea that there are very slow rates of social mobility, but they love the idea that there’s nothing you can do about it. So it’s an odd book in that it’s welcome and unwelcome to both sides.

Does it follow, then, that politicians are looking in the wrong place? They’re looking at social mobility whereas perhaps they ought to be worrying more about inequality.

One of the strong conclusions of the book is that if it really is the case that more than half of people’s outcomes are predictable, then we really have to rethink the idea that it’s OK in a society to make the rewards or penalties from the accident of [who] your parents [are] as large as the market is going to throw up. I started off my life as a free-market economist. But studies such as these have these have convinced me that, in some sense, the Scandinavian or Nordic model has many attractions. If you’re going to be at the bottom of the social order, it’s better to be at the bottom in somewhere like Sweden than somewhere like the United States. That’s a decision that societies have to make. You don’t need to give that much incentive to the upper group in society to still get educated, to achieve and to maintain their status. The incentives are very low, in monetary terms, in countries like Denmark and Sweden. It doesn’t change who succeeds.

It looks to me, then, as if you’ve ended up endorsing a species of luck egalitarianism, according to which because talent is randomly distibruted, it is legitimate, and indeed a matter of justice, to correct, through redistribution, for outcomes that are the result of brute luck.

That’s right. Here’s a nice example of the role of luck. I was in Iceland last summer. We were at a place on the coast where fishermen used to gather for the annual fishing season. They had four stones there of different sizes. And which stone you could pick up determined what your wage rate was. How much you earned in pre-industrial Iceland was a function of brute physical strength. We’d think that the good society is not going to forever penalise the person who could pick up only the smallest stone—or, in my case, couldn’t pick up even the smallest stone!

You point out that, as a matter of logic, a fully meritocratic society is one in which there is very little social mobility.

If we really had a completely meritocratic society, so that people’s circumstances had no effect on their possibilities, the only thing that would be left to differentiate people would be their DNA. What I’m saying is that the circumstances of your birth—poor neighbourhood, rich neighbourhood, things like that—have surprisingly little influence once you control for lineage, family inheritance and so on.

So you're not saying that there’s nothing that policymakers can do to mitigate the effects of that on the losers in the lottery of inheritance?

You really do need to protect people because they don’t choose their lineage. When you can see that events a hundred years before a person was born are predictive of what their outcomes are going to be, you see that people are caught in this web of connections, that they arrive with certain possibilities. Any good society has to take that into account in deciding how to allocate rewards. As I said, I started off as a libertarian economist but, in a sense, I’ve come full circle.

Greg Clark's "The Son Also Rises: Surnames and the History of Social Mobility" is published by Princeton University Press (£19.95)