Home > News and media > Comment and opinion > 2014 > 02 > What We Know About Income Mobility Depends on How We Define It

 

What We Know About Income Mobility Depends on How We Define It

Jonathan Hopkin is Associate Professor of Comparative Politics at LSE.

The US has long had higher inequality than other advanced democracies, although many Americans see this as part and parcel of the “American Dream” of rising living standards and social mobility.

But recent research by Harvard University’s Equality of Opportunity Project has opened up a new front in this debate. The authors establish that social mobility between generations has in fact remained quite stable in America over recent decades, by calculating the likelihood of an individual’s rank in the economic pecking order corresponding to their parents’. This means that “children entering the labour market today have the same chances of moving up in the income distribution (relative to their parents) as children born in the 1970s.” For some commentators these findings refute the connection between inequality and mobility that President Obama and liberal economists have been invoking, and others even go so far as to claim it proves the uselessness of a variety of social interventions to promote mobility, such as college grants for minorities.

So what exactly does this study show, and how reliable is it? It depends on how you define “mobility.” If you think it is about relative positions in a stratified society, it has stayed roughly the same. But if you think it is about relative incomes, it has gotten worse. A married couple where each partner earned $60,000 a year would make it into the top quintile, not exactly the stuff of riches, given America’s high cost of living. But that is still considerably higher than the bottom of the spectrum, where the top of the lowest quintile caps out at a household income of about $27,000 a year.

And the Harvard researchers are quick to underline that although they find that mobility has remained stable, the rise in inequality over the last 40 years means that the consequences of immobility are far more serious than in the past. The 8-9 per cent likelihood that a child born into the bottom twenty per cent will reach the top twenty per cent of earners has barely changed over the period studied (the research follows cohorts born between 1971-86). However, that low probability is clearly a more serious matter now because of the growing gap between the two groups: even if you exclude the very richest and the very poorest as outliers, this becomes clear. In 1980, the least-rich wealthy (the bottom of the top quintile) had three times as much wealth as the richest of the poor (the top of the bottom quintile); today, they have over four times as much. And the incomes of the top 1 per cent have grown much, much faster – from 10 times as the earnings of the bottom 90 per cent in 1980 to 33 times as much today (my own calculation, from the World Top Incomes Database). The average payoff of choosing the right parents has therefore increased.

Previous studies have found that inequality and mobility are negatively correlated, a finding aptly christened the Great Gatsby curve. This suggests that inequality may hurt mobility: in a very unequal society, the disadvantages of the poor and the advantages of the rich are compounded by the size of the gap between them. Wealthier children have access to better education, and even better nutrition and healthcare, setting them off on life’s path with advantages that secure an excellent prospect of matching their parents’ success. The poor not only have less chance of reaching college, they also face higher risks of incarceration, poor health, and drug abuse, all of which make it even less likely they will break out of poverty.

The Harvard study does however challenge some of the most pessimistic predictions about the consequences of higher inequality in America. It turns out that higher inequality does not automatically translate into lower mobility; after all, if the ‘Great Gatsby curve’ were a general law, then the United States, with its dramatic rise in inequality since the 1970s, would surely by the best laboratory to test that theory out. But does the Harvard team’s study suggest that the Great Gatsby curve is in fact spurious, and that inequality doesn’t affect mobility after all?

It’s not quite clear. The inequality-immobility relationship is certainly borne out by cross-national comparisons – more egalitarian societies have much higher mobility than the United States. But we don’t know whether changes in inequality over time within one country will affect levels of mobility, and if so, with what kind of time lag. We may simply be observing that on the whole, mobility is higher in countries with lower inequality because of a range of institutional and social conditions which may not change much over time, even when inequality does change. These ‘fixed effects’ may be driving much of the difference in mobility between countries, but do not produce much change within countries over time.

Another complication is that measuring social mobility across generations requires us to estimate a phenomenon that takes a long time to mature. To have a truly accurate measure, we would need to look at individuals’ incomes over a lifetime, but that poses insurmountable data limitations, as well as conceptual difficulties (is the mobility experienced by today’s seniors throughout their lives a measure of American society in the 1930s, or the 2010s?). The Harvard team takes cohorts born between 1971 and 1986 and measures their family incomes as children and adults, but even the oldest of these individuals are at most around halfway through their working lives, and youngest are barely out of college. They therefore make the reasonable assumptions, backed by existing data, that college graduation is a predictor of future income and that income at 30 is indicative of lifetime income. But still, this assumes that income dynamics over time will remain as they were in the recent past, which is far from certain. Even the youngest cohorts studied were born in an America rather more equal than it is today.

Even if these methodological doubts prove unfounded, there is a further reason to be resist dismissing the Great Gatsby curve out of hand. Much of the increase in inequality in recent years has been driven by the yawning chasm between the very top of the income distribution and everyone else. The Harvard study shows that it is the much slower growth in inequality amongst the bottom 99 per cent which more likely explains why social mobility and overall inequality have not moved in tandem.

The Great Gatsby curve is therefore qualified but very much alive. To the extent that rising inequality is driven in some measure by the gains of the 1 per cent, it should not surprise us that its effect on overall mobility should have been muted. But another possible conclusion is that rising inequality has cancelled out the effects of the many measures which have democratized American life since the post-war age, where incomes were less unequal, but gender and racial discrimination were rife. Greater access of women and minority groups to education and better-paying jobs have likely improved mobility, but at the same time the effects of growing poverty and the dismantling of labour protections and progressive taxes have pushed in the opposite direction. Social mobility remains caught between these two powerful forces.

The bottom line is: society is more unequal, but it’s not more mobile, so the fact that only 8-9% of the bottom fifth make it to the top fifth is now far more important than it was when inequality was lower. The stakes are getting higher, but your chances of winning haven’t budged.

This article first appeared on the Harvard Business Review Blog Network|

Share:Facebook|Twitter|LinkedIn|