The Growing Divide in Divorce and Political Participation

Economic circumstances and education play an important role not just in people’s health and longevity but also in their family structures. Take divorce, for example. While the factors that lead to a divorce are numerous and complex, bad economic conditions are known to be an important trigger. When things are not going well for a couple, economic troubles only make the situation worse. Not surprisingly, American cities differ enormously in their divorce rates.

What is the city with the highest divorce rate in America? If you think it’s Las Vegas, think again. Using data on 8 million adults who have ever been married, I discovered that the city with the highest incidence is Flint, Michigan, where 28 percent of all adults reported being divorced in 2009. With a local economy ravaged by the closure of auto manufacturing plants, declining wages, and a disappearing middle class, Flint, together with other Rust Belt cities, has long been in a state of economic decline. Its largest employer, GM, has shrunk from a peak of 80,000 local employees to 8,000. Toledo, Ohio, another former manufacturing center, is not too far down in the divorce rankings. At the other end of the spectrum are cities like Provo, Utah, in the heart of Mormon country, where the divorce rate is low for religious reasons; State College, Pennsylvania, a university town; McAllen, Texas, with a very high density of Catholics; and Stamford, Connecticut, the best-educated and most prosperous metropolitan area in the country. San Jose is also near the bottom.

 The difference in the incidence of divorce between American communities is pronounced. Flint has about three times the per capita number of divorced people as Provo. And this gap is widening. Figure 8 shows the increase in the percentage of adults who report being divorced in the ten cities with the highest and lowest percents in each year. While many cultural and religious factors play an important role, these factors are largely fixed; for instance, Provo has always been a predominantly Mormon city. The mounting divide in the figure between cities with high divorce rates and those with low divorce rates is likely to reflect the growing divide in economic conditions.

America’s increased socioeconomic segregation is also affecting the political process in complex and far-reaching ways. At the national level, the balkanization of the electorate makes it harder and harder for Americans to come to a consensus on important issues that involve the future of the country. There are many reasons for this trend. Primaries are dominated by radical candidates more often than in earlier years; political coverage by cable TV networks is more and more polarized; representatives and senators face stronger incentives to vote along party lines. But geography is playing an increasingly important role. Geographical segregation raises the number of people who live surrounded by others like them, and this is likely to reinforce extreme political attitudes. In his book The Big Sort, Bill Bishop used data from three decades of presidential elections to reveal an explosion in the number of communities that are so politically homogeneous that they vote overwhelmingly for one candidate.

Surprisingly, the effect of balkanization may be just the opposite at the local level. Economists have long pointed out that it is far easier for more socially homogeneous communities to agree on local policies. For example, voters are more likely to come to a consensus on issues like local taxes, schools, parks, and police when they share similar income and educational levels—and therefore similar needs and tastes—than when they are economically very diverse.

The Great Divergence is also affecting voting patterns. American communities differ enormously in how much their citizens vote and therefore how politically influential they are. In the 2008 presidential election, the ten counties with the highest voter turnout cast four times as many votes per capita as the ten counties with the lowest voter turnout. This huge difference in voter participation translates into equally large differences in political clout. It is as if each resident in the top group were given four ballots while each resident in the bottom group were given only one.

 There are many different factors that determine civic engagement. One of the most important is education. In research published in 2004, based on two surveys of 3 million American citizens, two colleagues and I found schooling to be a significant predictor of registration and voting in federal elections. We also found that education had a strong effect on broader measures of political engagement, in both the United States and the United Kingdom: better-educated citizens are more likely to follow politics in the media, be informed about the issues and discuss them with others, associate with a political group, and be active in their community. Incidentally, this is one of the main reasons both liberals and conservatives support public education. In 1962 the conservative economist Milton Friedman argued that “a stable and democratic society is impossible without a minimum degree of literacy and knowledge on the part of most citizens and without widespread acceptance of some common set of values. Education can contribute to both. Most of us would probably conclude that the gains are sufficiently important to justify some government subsidy.”

Since education is such an important determinant of political participation, the increased educational polarization of the country ultimately results in increased polarization in political participation. Figure 9 shows changes in voter participation in presidential elections since 1992 for the ten counties with the highest and lowest numbers of votes per capita. The trends go up and down in accordance with nationwide turnout. The 2000 Bush versus Gore election was extremely close, and this meant that turnout was higher than in the 1996 and 2004 elections, when Clinton and Bush won by larger margins. The 2008 presidential election had one of the highest turnouts in recent history, probably because the nation’s first African American candidate was on the ballot.

 

and be active in their community. Incidentally, this is one of the main reasons both liberals and conservatives support public education. In 1962 the conservative economist Milton Friedman argued that “a stable and democratic society is impossible without a minimum degree of literacy and knowledge on the part of most citizens and without widespread acceptance of some common set of values. Education can contribute to both. Most of us would probably conclude that the gains are sufficiently important to justify some government subsidy.”

Since education is such an important determinant of political participation, the increased educational polarization of the country ultimately results in increased polarization in political participation. Figure 9 shows changes in voter participation in presidential elections since 1992 for the ten counties with the highest and lowest numbers of votes per capita. The trends go up and down in accordance with nationwide turnout. The 2000 Bush versus Gore election was extremely close, and this meant that turnout was higher than in the 1996 and 2004 elections, when Clinton and Bush won by larger margins. The 2008 presidential election had one of the highest turnouts in recent history, probably because the nation’s first African American candidate was on the ballot.  

But the more interesting part of the graph is the difference between the top counties and the bottom counties. With every passing election, the top counties gain influence in the political process while the bottom counties lose it. A similar picture emerges when we tabulate financial contributions to candidates and parties, another important measure of political engagement and influence. These trends are likely to have real effect on legislation. When crafting policies that pit the interests of some communities against those of others, presidents and governors have a clear incentive to favor the needs of the communities that are politically active and better organized.