By Gregory ClarkPrinceton
Who gets ahead is always a topic of intense interest, especially in a society like America’s, without a legally established aristocracy at the top of the pile. As Tocqueville observed after visiting the new republic, the widespread perception that anyone can get ahead creates a presumption that everyone has an obligation to at least try. Nearly two centuries later, both the perception and the sense of obligation that follows are still firmly in place. According to the World Values Survey, while 60 percent of Europeans say they think “the poor are trapped in poverty,” only 29 percent of Americans think so. Instead, 60 percent of Americans think “the poor are lazy,” compared with just 26 percent of Europeans.
But can anyone get ahead? With the gap between rich and poor widening to Gilded Age levels, if not beyond, lately the question has attracted even more attention than usual. (When was the last time a book on economics created a sensation like Thomas Piketty’s Capital in the Twenty-First Century?) Vast inequality may be acceptable to most citizens if anyone, or at least anyone’s grandchild, has a fair shot at the top. But if wealth and poverty simply perpetuate themselves within families, ever wider inequality becomes ever harder to justify. In America, the debate about inequality is, inevitably, also a debate about mobility.
Mobility is hard to measure, however, turning not just on who earns what today but on how what people earn—and what they have—relates to their parents’ income and wealth, and their grandparents’ and prior generations’ too. Economists, to their credit, are increasingly stepping up to this difficult empirical challenge. Some are tackling the politically touchy question of whether mobility is greater in America than in western-European countries. (Our traditional civic myth notwithstanding, the answer is no.) Others are investigating whether mobility in America has declined in recent years. (Contrary to President Obama’s recent statements, it apparently hasn’t.)
Gregory Clark, an economic historian at the University of California at Davis, has looked at these questions through a different lens. Clark, too, finds that mobility here is no greater than in Europe, and that U.S. mobility hasn’t declined. But he comes to a more fundamental, far more powerful conclusion. Clark argues that mobility is always the same—in all societies, and in every era. Mobility, he claims, is “a universal constant”; over time we thrive or not according to a “social law of motion,” a “social physics of intergenerational mobility.” And to make matters worse, the universal speed at which families and groups change their social position is slow—a lot slower than everyone thinks on the basis of previous research.
The implications are profound. If mobility is constant, then the ability of social institutions to affect it must be negligible. Clark points to such changes as the movement from feudalism to democracy and then the expansion of the franchise, as well as free public education and redistributive taxation. But if modern America and modern Sweden have the same rate of mobility, and that rate is the same as what prevailed in medieval England and in 19th-century China, then none of those changes mattered. And if the journey from unusually high or low status to the middle can, as Clark claims, “take ten or fifteen generations (300–450 years),” the mobility-based defense of inequality becomes strained, here and everywhere else.
Why does Clark find mobility rates to be so much slower than other economists do? To begin with, he’s measuring something different. Most economists assess intergenerational mobility by looking at what people earn, or in some cases what they own, compared with their parents’ income or wealth. Clark is after something broader, encompassing not just people’s income and wealth but also their education, their occupation, their likelihood of holding elected office or other distinguished positions, or of belonging to elite groups. He refers to the entire constellation of such attributes as “status,” or “fundamental social competence,” or “general social competence or ability”—ultimately, “an inescapable inherited substrate, looking suspiciously like social class.” Clark argues that this more comprehensive concept of mobility is what most of us really care about, and he’s probably right. The zillionaire’s son who spends all his time on philanthropy earns a lot less than his father did, but he enjoys a pretty high social standing nonetheless.
The fact that Clark is measuring something broader—or, to put it the other way around, that most economists are measuring only one element of overall socioeconomic status—helps explain why he estimates mobility to be so much slower. What’s actually being measured, by Clark and everyone else, is not mobility, but its opposite: the degree of persistence (in income, say) from parent to child. Each particular element of what makes up Clark’s “general social competence or ability” is of course subject to random variation. Any one individual, like the son-of-a-zillionaire philanthropist, may have an unusually small income but be a Princeton alum and belong to all the right clubs anyway. Conversely, a parvenu earning a bundle may have attended State Tech and be unable to join anything more exclusive than the local Y. But muddying up a statistical relationship with lots of purely random noise inevitably lowers the estimate of whatever is the object of interest, and so looking only at income (or education, or any other single feature) means finding less parent-to-child persistence. And that in turn means concluding, as many economists have done, that there is more mobility. By examining instead the whole array of attributes that constitutes socioeconomic status, Clark averages out the purely chance variation in any one factor, like income, and therefore finds greater parent-to-child persistence—in other words, less mobility.
But how does Clark get at the persistence of his all-encompassing concept of status? His method relies on the tendency, in most societies, for a son to bear his father’s surname (hence his book’s title: The Son Also Rises). Specifically, Clark measures the persistence of status by looking at what has happened over time to groups of people bearing names that, at some point in the past—generations or even centuries ago—indicated socioeconomic status either well above or well below that of the general population.
It’s fascinating how many such groups he has managed to round up. In America, his examples include people with the same last names as those who attended Ivy League universities in the early 19th century and people with identifiably Jewish names (in both cases groups with higher-than-average status); he also looks at those with names that suggest people who are black, or Native American, or the descendants of immigrants from French Canada (all lower than average). How he measures these groups’ socioeconomic status over time is quite specific as well: their educational attainment, their representation in elite professions like medicine and law, and, conversely, their representation in generally low-status occupations like farming and domestic service.
Clark applies the same method elsewhere, too. In Britain, some of the high-status groups he examines bear the names of families who arrived there with the Norman Conquest, or the names of families who owned enough land to have had their property officially recorded in the 1200s, as well as people whose last names correspond to specific geographic places. Here his subsequent measures of status include attending Oxford or Cambridge, having enough wealth at the time of death to be listed in probate records, and membership in parliament. In Asia, his high-status examples include Japanese whose names suggest that they are descendants of Meiji-era nobility or samurai, and Chinese whose names suggest that their ancestors earned top scores on the imperial examinations during the Qing dynasty. For these groups, his measures of status down the line include education, authorship of scholarly publications, and representation in elite professions or top government or business positions.
The results are remarkably consistent. In these and more than a dozen other examples, Clark again and again finds evidence of far less mobility—a much slower rate of convergence toward the society’s general population, from either above or below—than what researchers using conventional methods have concluded.
The disadvantage of Clark’s surname method is its lack of precision about the parent-child link. The name De Belcamp, for example, wasn’t common in England in Norman times, and its modern equivalents, Beauchamp and Beacham, aren’t now. Even so, according to Clark, 3,352 Brits bear those names today. He doesn’t know who is really descended from whom. This is a far cry from the directly connected information about parents’ and children’s incomes on which most other researchers today rely.
But the advantage of looking at surnames is twofold. First, it enormously increases the spans of time Clark can study—up to 900 years in some of his English examples. Second, it allows him to examine countries, like India, where the kinds of directly linked parent-and-child data other researchers use aren’t available and may not be for decades, if ever.
No doubt economists and others will debate the relative merits of Clark’s method, along with what he’s trying to measure. But the important question is whether mobility is as slow as he says—and whether it really is a “universal constant.” Those claims are well worth taking seriously, in part because his more comprehensive notion of socioeconomic status—a concept economists typically don’t try to grapple with—does seem like what really matters.
Although Clark’s story is pretty depressing, there’s at least one brighter aspect. Some time ago Alan B. Krueger, a leading labor economist at Princeton, called attention to an especially worrisome implication of the growing body of standard research on mobility: countries with wider inequality tend to have lower mobility—which undercuts the defense of wide inequality, where it’s found, on the grounds that anyone can get ahead. Krueger labeled this deeply troubling relationship “the Great Gatsby Curve,” and economists and the popular press quickly picked up on the idea.
If Clark is right, however, there is no such relationship. The point is not that countries with wider inequality have more mobility, as we used to think. But at least they don’t have less. As Clark explains, that is a statistical illusion, the result of conventional research methods focusing only on income (or wealth).
Otherwise Clark’s is indeed a discouraging story—as, inevitably, is just about any account of human existence in which heredity is the dominant factor governing our individual destinies. As he did in his earlier book, Farewell to Alms (the annoying Hemingway puns come without explanation), Clark mostly maintains a studied ambiguity about whether our “general social competence” can be traced to genetics: “By and large, social mobility has characteristics that do not rule out genetics as the dominant connection between the generations.” And elsewhere: “This is not to say that social status is determined genetically. But whatever drives it is, on the tests performed here, indistinguishable from genetic inheritance.” As the book moves on, however, it becomes ever clearer that Clark has our genes in mind.
This genetic perspective leads him to a further ambiguity. Clark concludes that pervasive economic inequality in America and elsewhere is the outcome of fair competition: what’s being rewarded in life—with income, wealth, good jobs, social prestige—is no more than the talent and energy that each individual brings to the contest. “The social world,” he writes, “is much fairer than many would expect.”
But it’s hard to see the fairness in having not just every person’s position but every person’s chances of moving up or down so dominated by whatever subcellular makeup he or she simply happens to inherit. Clark himself steps forward to acknowledge just that:
An important corollary to the finding that social outcomes are the product of a lineage lottery is that we should not create social structures that magnify the rewards of a high social position … If social position is largely a product of the blind inheritance of talent, combined with a dose of pure chance, why would we want to multiply the rewards to the lottery winners?
Clark’s book is not merely intellectually clever, it’s profoundly challenging. Especially for Americans, it calls into question our sense of ourselves as individuals, as well as our long-standing image of our society. Let’s hope he’s wrong.