Photographer: Jeff J Mitchell/Getty Images

So, What Do Your Parents Do? And Your Great-Grandparents?

Income mobility is more tied to past generations than we knew. New solutions may be in order.

Income mobility is a problem for everyone. Most people, in most countries, dream of a world where birth does not limit your status in society. No country has achieved this happy end, though some of them do better than others. Income mobility statistics are created, worried over, analyzed for some sign of possible cures. A new paper out of Sweden suggests that we should perhaps be worrying even more than we do, and that the cures may be harder to come by than we thought.

Most income mobility research examines how tightly people's incomes are linked to those of their parents. In "The Son Also Rises: Surnames and the History of Social Mobility," Gregory Clark suggested that this approach was too limited. When he looked at surnames over centuries rather than years, he found that socio-economic status was incredibly persistent, with the great-great grandchildren of the elites doing markedly better than those of humbler origins. His work found that it took 10-15 generations to erase the legacy of prior prosperity.

Obviously, these implications are troubling, especially because these effects persist even across major socioeconomic upheavals, even revolutions explicitly aimed at leveling the classes.

Clark's thesis has been subject to multiple critiques, but it has not been refuted. And this new paper points in the same direction, albeit less strongly. Looking at multiple family members across multiple generations, researchers found that income mobility is lower than a simple examination of parent and child incomes would suggest. And this in Sweden, that egalitarian paradise that provides income-leveling services from cradle to grave.

Why might that be? Well, every affluent family has a couple of starving artists or impecunious social workers hanging around. Those folks may have very low incomes relative to their professional siblings and cousins. But they are still marinating in a rich broth of social capital: older relatives to be role models and help them navigate the process of following in their footsteps. The child of a San Francisco hippie whose parents were doctors is probably going to have a lot easier time getting themselves into medical school than the child of a drywall installer who was the child of a drywall installer. So taking a snapshot of parent-child incomes will miss the valuable social inheritance those kids are getting, even if there isn’t much financial value in their parents’ estates.

This may also suggest that some of the apparent income mobility in places like Canada or Scandinavia is something of an illusion. Places with high taxes and a very strong social safety net somewhat lower the cost to upper-middle-class people who decide to pursue relatively unlucrative careers. But those people do not necessarily leave the class in which they were born, even if they are on the bottom end of its income distribution. Their children will still start out with a class advantage that gives them more choices than people further down the socioeconomic ladder.

That’s not an argument against the safety net, mind you. It’s simply a caution about the limits of statistics. We of course care about making sure that people have adequate income, but that is not the only reason that we care about income mobility. Income is a proxy for opportunity and choices, for the ability to become anything you want to be. If we solve the income disparity with government transfers, but leave in place a class hierarchy that limits the educational and career options of lower-class children, then we still have a problem worth worrying about.

What might we do if this turns out to be true? One implication might be that redistribution is more justified, because it’s hard to say that you earned what you have if a major component of your income was getting born to the right great-grandparents.

If family history is such a major component of financial success, that also weakens some of the evidence for present-day discrimination. If socioeconomic status really is this heritable, even in a homogeneous country like Sweden, 1 then income disparities between modern-day groups may be less attributable to invidious discrimination against individuals and more attributable to the subtle and complex systems by which societies reproduce their class structure across decades.

Of course, what we’d like to do is stop this from happening. But having just started to define the scope of the problem, it will be a long time before we can propose definitive solutions.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

  1. Sweden is getting less homogenous, but the structure of the study -- which required your ancestors to have been in Sweden decades ago -- means that the sample consists mostly of ethnic Swedes.

To contact the author of this story:
Megan McArdle at

To contact the editor responsible for this story:
Philip Gray at

Before it's here, it's on the Bloomberg Terminal.