The Pulse: Disparity in income continues, but there are lots of variations

In crunching data about income inequality, I found two surprises: First, despite the growing divide, a child's odds of mobility haven't changed in the last 25 to 50 years. Second, educational progress made by women might be a contributing factor.

The divide itself is not subject to reasonable debate. Economist Emmanuel Saez from the University of California at Berkeley is one of many who have documented the growing disparity. Last September, Saez revealed that from 2009 to 2012, average real income per family in the United States grew by 6 percent, but that the gains were uneven - the top 1 percent grew by 31.4 percent while the "bottom" 99 percent grew by just 0.4 percent. In other words, the top 1 percent captured 95 percent of income gains. (And it's not just the United States. Oxfam recently detailed how the 85 richest individuals on the globe have as much collective wealth as the "poorest" 350 billion people.)

You might think that mobility would have suffered in the process. According to Saez and his colleagues, that is not the case. Nathan Hendren, an economist from Harvard, coauthored a study with Saez and told me that their research evaluated the experiences of more than 40 million children during the last 25 years, while looking at relationships between parent and child income.

Where the researchers expected to see a decline in mobility relative to the past, they instead found that children growing up in America today are just as likely to climb the economic ladder as children born here more than a half-century ago.

"The extent to which kids move up or down in relative position to everyone else in society has remained remarkably stable over the last 25 to 50 years," Hendren said.

What has changed is the difference between the top and bottom of the ladder. In other words, the distances between the rungs on the ladder have grown, but not the chance of traversing between them. Another key and troubling finding was that mobility is lower in the United States as compared with other developed countries. By way of example, kids born to the bottom quintile of income distribution in the United States have just an 8 percent chance to get to the top fifth in income distribution, whereas in Denmark those odds are 16 percent.

Of course, to listen to the public debate about income and wealth inequality is to hear about the minimum wage, the extension of unemployment benefits, the effects of globalization, and the loss of manufacturing jobs. But new research suggests that into that mix must also be placed the institution of marriage. This finding is what Matthew O'Brien recently labeled in the Atlantic magazine the "When Harry Met Sally" theory. Economists refer to it as "assortative mating."

In the National Bureau of Economic Research, four academics, including Jeremy Greenwood from the University of Pennsylvania and Nezih Guner from Markets, Organizations and Votes in Economics in Barcelona, analyzed the contribution of marriage.

Guner asked me to imagine that there are only two types of people, equal in numbers: those who went to college and those who did not. Those who went to college earn $30 and those who did not earn $10. If educated men marry uneducated women and uneducated men marry educated women, then every household will earn $40 total. So, household income is perfectly equalized. Now, imagine a world in which educated people marry only other educated people. Then, a household made up of an educated man and an educated woman will earn $60 dollars v. the $20 earned by a household of only uneducated spouses. The households at the top of the distribution would have three times the income of those at the bottom.

"We looked at the data from 1960 to 2005 and see an increasing tendency for people who have more education to get married with those who have more education as well," Guner said. "So we have this tendency for the households in some sense splitting apart and that contributes to inequality."

There is nothing new about individuals marrying within their own educational class. But where more women are educated, there is less marriage between educated men and women of less education. A boss marrying his secretary in the 1950s or '60s would not have raised an eyebrow. Today, such a union would be more of a rarity, and the boss might herself be female. The secretary is now more apt to marry within her educational class.

But could the decline of college-educated men marrying non-college-degreed women be offset by college-educated women marrying non-college-educated men? Data from the Pew Research Center suggest the answer is yes. Pew found nearly 21 percent of married women in 2012 were better educated than their spouses, a threefold jump from 1960. In contrast, a bit less than 20 percent of men had more formal education than their wives. In the more than half-century that Pew has tracked the issue, this is the first time that a higher percentage of women than men have "married down."

Guner acknowledges this development but believes it is not yet strong enough to mitigate the trend toward assortative mating.

According to Guner, when a woman with a college education marries a man with less education but who is financially well off, the result is not income inequality within the household but, rather, a "rich person is getting married with a relatively rich man."