Why Some States Are More Economically Mobile Than Others
Economic mobility - the quintessentially American idea (ideal, really) that any one, no matter how humble their origins, can become wealthy - has taken some terrible hits in the last few years. Writing in The New Republic, Timothy Noah notes that income heritability ("a measure of how determinative one generation’s relative income status - what we used to call ‘station in life’ - will be of the next generation’s relative income status') is much higher in the U.S. than in many of the countries that people once emigrated to America from, in search of greater opportunities. “Mobility in the United States has fallen dramatically behind mobility in other comparably developed democracies," he writes.
A 2007 study by the Organisation for Economic Cooperation and Development combined a number of previous estimates and found income heritability to be greater in the United States than in Denmark, Australia, Norway, Finland, Canada, Sweden, Germany, Spain, and France. The United Kingdom, which had been far less mobile than the United States during the late nineteenth century, brought up the rear, but this time it was just a bit less mobile than the United States. Thanks to a 2012 recalculation by Miles Corak, an economist at the University of Ottawa, we can now add Switzerland, Japan, New Zealand, Singapore, and Pakistan to the list of societies that are more mobile than the United States.
According to an important new study from the Economic Mobility Project at the Pew Center on the States, economic mobility varies by geography within the United States as well. The map above shows that economic mobility is highest in the New England and the mid-atlantic states, especially New York, New Jersey, and Maryland. The states where residents experience the least economic mobility are all in the south, with Louisiana, Oklahoma, and South Carolina scoring at the bottom.
The data, developed by Bhashkar Mazumder of the Federal Reserve Bank of Chicago, is based on the Survey of Income and Program Participation and the Social Security Administration. Americans’ earnings between the ages of 35 and 39 were averaged during the period from 1978 to 1997 and then again 10 years later, when the same people were between the ages of 45 and 49. Absolute mobility measures their earnings’ change over time. Relative upward and relative downward mobility are measures of people’s ranks on the earnings ladder relative to their peers and their own movements up or down the ladder.
I contacted Mazumder and asked him if he and his colleagues had run any systematic comparisons of their mobility statistics with other demographic, cultural, and economic data. Mazmuder connected me with Erin Currier at Pew, who manages the Economic Mobility Program and who has been thinking about these issues, and Scott Winship of the Brookings Institution (who has done some guest-blogging for Megan McArdle recently) who had run some correlations of his own at the prodding of Reihan Salam. Salam called his attention to a Family Research Council study conducted by childhood researcher Nicholas Zill. Using Zill’s data, Winship looked at child poverty, high school graduation rates, government education spending, teen birth rates, and family stability (the percent of teens with continuously married parents).
I enlisted my Martin Prosperity Institute colleague Charlotta Mellander to help me examine the possible effects on economic mobility at the state level of factors like economic output and income, poverty, inequality, knowledge/ high-tech economies versus working class structures, college grads, religion, openness to immigrants, artistic creative and gays, and political affiliation.
As usual, I note that correlation is not causation; other factors that we haven’t looked at could play equal or greater roles. Nonetheless, our findings were intriguing and worth talking about.
It makes intuitive sense that upwardly mobile people would be attracted to higher income states. And that is exactly what we found. State income is associated with both absolute national upward mobility (.47) and relative national upward mobility (.65). This is clearly visible on the scatter graph above, with the least mobile southern states clustered at the bottom left and the most mobile New England and mid-Atlantic states at the upper right.
Conversely, poverty was negatively associated with both national absolute upward mobility (-.51) and national relative upward mobility (-.49). Winship similarly found child poverty to be negatively associated with upward relative mobility (-.62). Catherine Rampell at The New York Times’ Economix blog compared Pew’s state mobility measures to the Gini index of income inequality and found a very weak negative association.
Mobility was lower in more working class states. We found a negative correlation between the share of the labor force in working class jobs and national relative upward mobility (-.46). This is dramatically represented on the scatter graph above, where the states with the highest mobility cluster at the top left. The states with the least mobility are clumped at the bottom right.
Education is a key driver of economic development. Incomes and wages are higher in states with more highly educated people. So it stands to reason that mobility would favor states with more highly educated populations. This is borne out by our analysis. The high school graduation rate is positively associated with relative national upward mobility (.63), according to Winship‘s analysis. He also found the absolute national mobility to be positively associated with public education spending per student (.48) and with grade reading levels (.57). The share of adults that are college grads, according to our analysis, is associated with both absolute national upward mobility (.39) and national relative upward mobility (.52).
Openness matters to mobility. Places that are more open to outsiders appear to have more mobility. Immigrants, for example, signal openness. And there are positive correlations between immigrants and national relative upward mobility (.57) and regional relative upward mobility (.44). However, we found little, if any, little correlations between mobility and race, either black or Hispanic.
Despite President Obama’s embrace of same sex marriage, gays and lesbian remain among the most discriminated-against groups in society. Recent successful moves to ban same-sex marriage in a number of states reflect this. We found positive correlations between the Gay Index and national relative upward mobility (.37) and regional relative upward mobility (.21).
So-called artistic and cultural “bohemians” also signal places that are open to the new, the creative and the different. Here again we found positive correlations between artistic and cultural creatives and both national absolute mobility (.49) and national relative upward mobility (.55).
Family structure may also play a role. Winship’s analysis looked at the effect of two factors – births to teen mothers and the share of teens that live with continuously married parents. He found the former to be negatively associated with national relative upward mobility (-.70), while the latter is positively associated with national absolute upward mobility (.73).
Religion factors in as well. The percentage of adults who say they are "very religious” according to Gallup polls is negatively associated with national relative upward mobility (-.51)—a pattern that is graphically represented on the above scatter graph.
Much has been made of America’s sorting across political lines, into the proverbial “red” versus “blue” states. Economic mobility maps onto this divide. Mobility is higher in blue states and lower in red ones. Both absolute mobility and national relative upward mobility are correlated with whether or not a state voted for Obama (.38 and .53 respectively). Conversely, both are negatively associated with whether or not a state voted for McCain (-.40 and -.55). There is a positive correlation between national relative downward mobility (.46) and whether or not a state voted for McCain.
Mobility is also connected to happiness, according to our analysis. State happiness is positively correlated with both national absolute upward mobility (.33) and national relative upward mobility (.30).
It is depressing to contemplate that class-related fault lines are not just dividing us today, but appear to play a large role in determining our own and our descendants’ future. Class, it seems, is increasingly destiny.