Drawing class distinctions feels almost un-American. The nation’s self-image is of a classless society, one in which every individual is of equal moral worth, regardless of his or her economic status. This is how the world sees the United States, too. Alexis de Tocqueville observed that Americans were “seen to be more equal in fortune and intelligence — more equally strong, in other words — than they were in any other country, or were at any other time in recorded history.”

So different to the countries of old Europe, still weighed down by the legacies of feudalism. British politicians have often felt the need to urge the creation of a “classless” society, looking to America for inspiration as, what historian David Cannadine once described, “the pioneering and prototypical classless society.”

European progressives have long looked enviously at social relations in the New World. George Orwell noted the lack of “servile tradition” in America; German socialist Werner Sombart noticed that “the bowing and scraping before the ‘upper classes,’ which produces such an unpleasant impression in Europe, is completely unknown.” This is one of many reasons socialist politics struggled to take root in the United States. A key attraction in socialist systems — the main one, according to Orwell — is the eradication of class distinctions. There were few to eradicate in America. I am sure that one reason Downton Abbey and The Crown so delight American audiences is their depictions of an alien world of class-based status.

One reason class distinctions are less obvious in America is that pretty much everyone defines themselves as a member of the same class: the one in the middle. Nine in ten adults select the label “middle class,” exactly the same proportion as in 1939, according to Gallup. No wonder politicians have always fallen over each other to be on their side.

But in recent decades, Americans at the top of the ladder have been entrenching their class position. The convenient fiction that the “middle” class can stretch up that far has become a difficult one to sustain. As a result, the modifications “upper” or “lower” to the general “middle-class” category have become more important.

Class is not just about money, though it is about that. The class gap can be seen from every angle: education, security, family, health, you name it. There will also be inequalities on each of these dimensions, of course. But inequality becomes class division when all these varied elements — money, education, wealth, occupation — cluster together so tightly that, in practice, almost any of them will suffice for the purposes of class definition. Class division becomes class stratification when these advantages — and, thus, status — endure across generations. In fact, as I’ll show, upper-middle-class status is passed down to the next generation more effectively than in the past and in the United States more than in other countries.

One benefit of the multidimensional nature of this separation is that it has reduced interdisciplinary bickering over how to define class. While economists typically focus on categorization by income and wealth, and sociologists tend more toward occupational status and education, and anthropologists are typically more interested in culture and norms, right now it doesn’t really matter, since all the trends are going the same way.

By now you may be experiencing a slight sense of déjà vu. After all, the separation of an affluent, well-educated class has been the subject of more than one book. Producing another volume about class and inequality might seem redundant. But I think some of the most popular efforts to date have diagnosed the class fracture incorrectly. Some analysts have let the upper-middle class off the hook (yes, that would be you) by pointing at the “super-rich” or “top 1 percent.” Take the new rock star of economic history, Thomas Piketty. For him, inequality is pretty much all about the top 1 percent.

The class gap can be seen from every angle: education, security, family, health, you name it.

Others have looked through a slightly wider lens. In Coming Apart, Charles Murray described an isolated “New Upper Class,” composed of the most successful adults (and their spouses), working in managerial positions, the professions, or with senior jobs in the media. This class, according to Murray, is defined as much by elitist culture — tastes and preferences — as by economic standing, and accounts for just 5 percent of the population.

Robert Putnam, in Our Kids, has a broader group in mind. “When I speak of kids from ‘upper class’ homes,” he writes, “I simply mean that at least one of their parents (usually both) graduated from college.” This represents, Putnam estimates, “about one third of the population.” Putnam’s concern is really with the bottom third, who he fears are being left behind.

Where these scholars agree, however, is on the right label for those at the top, whether it is 1 percent, 5 percent, or even 30 percent: the “upper class.” It is easy to see why. It is an easy idea to grasp — the upper class is simply the people at the top of the pile. To be honest, my editors would have preferred me to use “upper class,” too. But I stuck with the longer, uglier, wonkier “upper-middle class.” This is not just semantics. If people are encouraged to think inequality is an upper-class problem, something important is lost. Most of us think of the upper class as the thin slice at the very top, but the tectonic plates are separating lower down. It is not just the top 1 percent pulling away, but the top 20 percent.

In fact, only a very small proportion of U.S. adults — 1 to 2 percent — define themselves as “upper class.” A significant minority — about one in seven — adopts the “upper-middle class” description. This is quite similar to the estimates of class size generated by most sociologists, who tend to define the upper-middle class as one composed of professionals or managers, or around 15–20 percent of the working-age population.

These self-definitions are a useful starting point, providing some sense of how people see themselves on the class ladder. But for analytical purposes, we need a more objective, and measurable, yardstick. But which to choose? After all, I’ve been at pains to argue that class is made up of a subtle, shifting blend of economic, social, educational, and attitudinal factors.

Income provides the cleanest instrument with which to dissect the class distribution because it is easier to track over time and to compare between individuals and families (perhaps also because I work with a lot of economists). Income is also what philosophers call an “instrumental good,” bringing many other benefits along with it. I’ll show the growing economic divide between the top fifth and the rest and how the upper-middle class, as defined by income, is separating on other dimensions, too.

Before diving into some of the data, a big caveat: America’s growing class division does not mean that categorical inequalities on the basis of race, ethnicity, and gender have disappeared. If anything, the relative position of black Americans has worsened in recent years, as I have argued elsewhere. There are also race gaps in access to some of the mechanisms of class reproduction; class and race divisions amplify each other. The gender gap is far from being closed — although perhaps our biggest gender challenge now is the need for men to adapt. But while the barriers of race, sexuality, and gender remain in place, they have been lowered following successive victories on the identity-politics front. Meanwhile, class barriers have risen, in five areas in particular: economic fortunes, educational attainment, family formation, geography, and in terms of health and life expectancy.

— Richard V. Reeves is a senior fellow in economic studies and co-director of the Center on Children and Families at the Brookings Institution.