A Nation of Singles
The most politically potent demographic trend is not the one everyone talked about after the election
Dec 10, 2012, Vol. 18, No. 13 • By JONATHAN V. LAST
For a brief moment last month—roughly a 72-hour span beginning at 11:00 p.m. on November 6 and concluding late in the evening of November 9—everyone in America was interested in demographics. That’s because, in addition to rewarding the just, punishing the wicked, and certifying that America was (for the moment) not racist, President Barack Obama’s victory over Mitt Romney pointed to two ineluctable demographic truths. The first was expected: that the growth of the Hispanic-American cohort is irresistible and will radically transform our country’s ethnic future. The second caught people by surprise: that the proportion of unmarried Americans was suddenly at an all-time high.
Unfortunately, by the time the window closed on the public’s demographic curiosity no one really understood either of these shifts. Or where they came from. Or whether they were even particularly true. As is often the case, people tended to fixate on a relatively small, contingent part of America’s changing demographic makeup and look past the bigger, more consequential part of the story.
So let’s begin by asking the obvious question: Hispanics are America’s demographic future—true or false? The answer is, both. Sort of.
Start with what we know. As of the 2010 census, there were 308.7 million people in America, 50.5 million of whom (16 percent) were classified as being of “Hispanic origin.” Of that 50 million, about half are foreign-born legal immigrants. Another 11 million or so are illegal immigrants. A few other facts, just to give you some texture: 63 percent of American Hispanics trace their origins to Mexico, 9.2 percent to Puerto Rico, and 3.5 percent to Cuba. And more than half of the 50 million live in just three states, California, Texas, and Florida.
But what makes people’s heads snap to attention when they talk about Hispanic demographics isn’t any of that stuff. It’s the rate of increase. From 2000 to 2010, America’s Hispanic population jumped by 43 percent, while our total population increased by just 9.7 percent. Or, to put it another way, from 2000 to 2010, America grew by 27.3 million people. Fifteen million of those faces—more than half of those new Americans—were Hispanic.
If you extrapolate those trends the numbers get even more eye-popping. In 2008, the Pew Research Center projected that, at current rates, by 2050 there would be 128 million Hispanic Americans, making the group 29 percent of the American population. The census projection is a little higher; they guess the total will be 132.8 million, 30 percent of a projected total population of 439 million.
Where do these numbers come from? It’s not rocket science. Demographers depend mainly on two variables: net migration to the United States by people from Spanish-speaking countries and the fertility rate of Hispanic Americans.
The big 130-million projections come from assumptions based on the 2000 census. Back then, immigration from south of the border was booming, with a net of about 900,000 new people—both legal and illegal—showing up every year in America. (In 2000 alone, 770,000 people came from Mexico.) Because of that trend line, demographers assumed that we’d be netting roughly 1 million new immigrants every year between now and 2050.
But trends don’t always continue to the horizon, and we’re already going in a different direction on immigration. America’s net annual immigration numbers started declining in 2006, sliding from just over 1 million in 2005 to 855,000 in 2009. We don’t have good totals for 2010 or 2011 (because the Census Bureau rejiggered its formula in 2010, making it hard to compare to previous years), but we do have numbers for Mexican immigration alone, which show—amazingly—that in the most recent years there’s been a net flow of zero immigrants from Mexico. Since Mexico has historically made up nearly two-thirds of our Hispanic immigrant pool all by itself, this would suggest that when we do get comparable data we will see that there has been a significant drop in immigration already.
Economists who have noted this sudden shift are quick to explain it as a byproduct of the recession and the bursting housing bubble, which dried up jobs—particularly in the construction industry—causing prospective immigrants to stay put and pushing many illegal immigrants already in the country to head home. The implication of this argument is that as soon as our economy goes back to “normal,” the patterns of migration will, too.
Demographers aren’t so sure. Speaking broadly, when it comes to immigration there are two kinds of countries—sending and receiving. The economic factors distinguishing the two are what you’d expect—rich vs. poor; dynamic vs. lethargic. But there are demographic markers, too. Receiving countries tend to have very low fertility rates—generally below the replacement rate of 2.1. (That is, if the average woman has 2.1 children in her lifetime then a country’s population will maintain a steady state.) In the short run, fertility rates below replacement cause labor shortages. Sending countries, on the other hand, have fertility rates well above the replacement rate, and resultant labor surpluses.
When you look at immigration rates from Central and South America to the United States, you find that these demographic markers are fairly reliable. Over the last decade or so the high-fertility countries (Mexico, Honduras, El Salvador, Colombia) have sent larger numbers of immigrants to America while below-replacement countries (Uruguay, Chile, Brazil) have sent relatively few. Consider, for the sake of illustration, the cases of Guatemala and Costa Rica, two tiny Central American countries. With a population of 14 million, Guatemala still has a relatively robust fertility rate of 3.18. And as of 2010, there were a million people of Guatemalan descent living in the United States. Costa Rica has a population of 4.6 million and a fertility rate of 1.92. There are only 126,000 Costa Ricans in America—about 66 percent fewer than you would expect if the Guatemalan rates prevailed.
What else happened between 2006 and today, aside from the housing bubble and the Great Recession? Mexico’s fertility rate—which has been heading downward on an express elevator since the 1970s—started nearing the replacement rate. The data are slightly conflicting on how low it is—some people believe it has already dipped below 2.1, others put the number just over 2.2. But everyone agrees that the trajectory is downward still. And that the same is true of nearly every other country south of the American border.
So will America add another 38 million Hispanics by 2050 just through immigration alone, as the projections suggest? No one knows, of course. But it seems an uncertain proposition. The boom days of Hispanic immigration may already be a thing of the past.
Which leads us to the fertility rate of Hispanic Americans. As a cohort, Hispanics have the highest fertility rate of America’s racial groups, around 2.7. Much research has been done trying to figure out if, and when, the Hispanic-American fertility rate will fall toward the national average (which is closer to 2.0). Some researchers believe that by 2050, our Hispanic fertility rate will be at replacement. Others suggest sooner. Some scholars, looking at the data by cohort, suggest that Hispanic-American women currently in their childbearing years will finish them close to the replacement level. All of the research, however, indicates that in recent years the fertility rate of Hispanic Americans has been moving downward faster than it has for any other ethnic group.
Last week the Pew Center reported that from 2007 to 2010 America’s birth rate dropped by 8 percent. The decline was relatively modest for native-born Americans—only a 6 percent drop. But the immigrant birth rate dropped by 14 percent. And the birth rate for Mexican-born immigrants dropped by 23 percent. These declines were outsized, but they fit the larger trend. From 1990 to 2007, the Mexican-born birth rate had already dropped by 26 percent.
None of this is meant to predict that by such and such year there will be exactly so many Hispanic Americans. Social science has limits, and they are even nearer than you think. But when you look at the assumptions underlying the predictions for America’s Hispanic future, they’re even more uncertain than usual—and in fact are already a decade or so out of step with reality. America’s Great Hispanic Future is probably being oversold. And possibly by quite a bit.
You don’t hear nearly as much about the rise of single voters, despite the fact that they represent a much more significant trend. Only a few analysts, such as Ruy Teixera, James Carville, and Stanley Greenberg, have emphasized how important singletons were to President Obama’s reelection. Properly understood, there is far less of a “gender” gap in American politics than people think. Yes, President Obama won “women” by 11 points (55 percent to 44 percent). But Mitt Romney won married women by the exact same margin. To get a sense of how powerful the marriage effect is, not just for women but for men, too, look at the exit polls by marital status. Among nonmarried voters—people who are single and have never married, are living with a partner, or are divorced—Obama beat Romney 62-35. Among married voters Romney won the vote handily, 56-42.
Far more significant than the gender gap is the marriage gap. And what was made clear in the 2012 election was that the cohorts of unmarried women and men are now at historic highs—and are still increasing. This marriage gap—and its implications for our political, economic, and cultural future—is only dimly understood.
Americans have been wedded to marriage for a very long time. Between 1910 and 1970, the “ever-married rate”—that is, the percentage of people who marry at some point in their lives—went as high as 98.3 percent and never dipped below 92.8 percent. Beginning in 1970, the ever-married number began a gradual decline so that by 2000 it stood at 88.6 percent.
Today, the numbers are more striking: 23.8 percent of men, and 19 percent of women, between the ages of 35 and 44 have never been married. Tick back a cohort to the people between 20 and 34—the prime-childbearing years—and the numbers are even more startling: 67 percent of men and 57 percent of women in that group have never been married. When you total it all up, over half of the voting-age population in America—and 40 percent of the people who actually showed up to vote this time around—are single.
What does this group look like? Geographically, they tend to live in cities. As urban density increases, marriage rates (and childbearing rates) fall in nearly a straight line. Carville and Greenberg put together a Venn diagram which is highly instructive. Of the 111 million single eligible voters, 53 million are women and 58 million are men. Only 5.7 million of these women are Hispanic and 9.7 million are African American. Nearly three-quarters of all single women are white. In other words, the cohort looks a lot like the Julia character the Obama campaign rolled out to show how the president’s policies care for that plucky gal from the moment she enrolls in Head Start right through her retirement. You may recall that because of President Obama, Julia goes to college, gets free birth control, has a baby anyway, sends her lone kid to public school, and then—at age 42—starts her own business (as a web designer!). What she does not do is get married.
Singles broke decisively for Obama. Though his margins with them were lower than they were in 2008, he still won them handily: Obama was +16 among single men and +36 with single women. But the real news wasn’t how singles broke—it was that their share of the total vote increased by a whopping 6 percentage points. To put this in some perspective, the wave of Hispanic voters we’ve heard so much about increased its share of the total vote from 2008 to 2012 by a single point, roughly 1.27 million voters. Meanwhile, that 6 percentage point increase meant 7.6 million more single voters than in 2008. They provided Obama with a margin of 2.9 million votes, about two-thirds of his margin of victory. Back in 2010, Teixera noted that 47 percent of all women are now unmarried, up from 38 percent in 1970. “Their current size in the voter pool—more than a quarter of eligible voters—is nearly the size of white evangelical Protestants, who are perhaps the GOP’s largest base group,” he writes. “And since the current growth rate of the population of unmarried women is relatively high (double that of married women), the proportion of unmarried women in the voting pool should continue to increase.” In the medium run, he’s almost certainly correct.
How did we get to an America where half of the adult population isn’t married and somewhere between 10 percent and 15 percent of the population don’t get married for the first time until they’re approaching retirement? It’s a complicated story involving, among other factors, the rise of almost-universal higher education, the delay of marriage, urbanization, the invention of no-fault divorce, the legitimization of cohabitation, the increasing cost of raising children, and the creation of a government entitlement system to do for the elderly childless what grown children did for their parents through the millennia.
But all of these causes are particular. Looming beneath them are two deep shifts. The first is the waning of religion in American life. As Joel Kotkin notes in a recent report titled “The Rise of Post-Familialism,” one of the commonalities between all of the major world religions is that they elevate family and kinship to a central place in human existence. Secularism tends toward agnosticism about the family. This distinction has real-world consequences. Take any cohort of Americans—by race, income, education—and then sort them by religious belief. The more devout they are, the higher their rates of marriage and the more children they have.
The second shift is the dismantling of the iron triangle of sex, marriage, and childbearing. Beginning in roughly 1970, the mastery of contraception decoupled sex from babymaking. And with that link broken, the connections between sex and marriage—and finally between marriage and childrearing—were severed, too.
Where is this trend line headed? In a word, higher. There are no indicators to suggest when and where it will level off. Divorce rates have stabilized, but rates of cohabitation have continued to rise, leading many demographers to suspect that living together may be crowding out matrimony as a mode of family formation. And increasing levels of education continue to push the average age at first marriage higher.
Fertility rates play a role, too. Nearly one in five American women now forgo having children altogether, and without babies, marriage is less of a necessity. People’s attitudes have followed the fertility rate. The Pew Research Center frequently surveys Americans about their thoughts on what makes a successful marriage. Between the 1990 survey and the 2007 survey, there were big increases in the percentages of people who said that sharing political or religious beliefs was “important to a good marriage.” In 2007, there was a 21 percent increase in people who said it was important for a marriage that the couple have “good housing.” Thirty-seven percent fewer people said that having children was important. The other indicator to decline in importance from 1990 to 2007? “Faithfulness.”
As Kotkin explains, comparatively speaking, America is still doing pretty well when it comes to singletons. In Europe, Asia, and most advanced countries, people are running away from marriage, children, and family life at an amazing rate. To pick just a smattering of data points from the highlight reel: Thirty percent of German women today say that they do not intend to have children. In Japan in 1960, 20 percent of women between 25 and 29 had never married. Today the number is more than 60 percent. Gavin Jones of the National University of Singapore estimates that “up to a quarter of all East Asian women will remain single by age 50, and up to a third will remain childless.”
The question, then, is whether America will continue following its glidepath to the destination the rest of the First World is already nearing. Most experts believe that it will. As the Austrian demographer Wolfgang Lutz puts it, once a society begins veering away from marriage and childbearing, it becomes a “self-reinforcing mechanism” in which the cult of the individual holds greater and greater allure.
What then? Culturally speaking, it’s anybody’s guess. The more singletons we have, the more densely urban our living patterns are likely to be. Sociologist Eric Klinenberg believes that the masses of city-dwelling singles will sort themselves into “urban tribes,” based not on kinship, but rather on shared interests. The hipsters, the foodies, the dog people, and so on. Klinenberg teaches at NYU, so he would know. As a result, cities will gradually transform from centers of economic and cultural foment into what urban theorist Terry Nichols Clark calls “the city as entertainment machine.”
The urban tribes may be insipid, but they’re reasonably benign. Kotkin sees larger cultural problems down the road. “[A] society that is increasingly single and childless is likely to be more concerned with serving current needs than addressing the future,” he writes. “We could tilt more into a ‘now’ society, geared towards consuming or recreating today, as opposed to nurturing and sacrificing for tomorrow.”
The economic effects are similarly unclear. On the one hand, judging from the booming economic progress in highly single countries such as Singapore and Taiwan, singletons can work longer hours and move more easily for jobs. Which would make a single society good for the economy. (At least in the short term, until the entitlement systems break because there aren’t enough new taxpayers being born.) There is, however, an alternative economic theory. Last summer demographers Patrick Fagan and Henry Potrykus published a paper examining the effect of nonmarriage on the labor participation rate. Fagan and Potrykus were able to identify a clear statistical effect of marriage on men’s labor participation. What they found is that without the responsibility of families to provide for, unmarried American males have historically tended to drop out of the labor force, exacerbating recessionary tendencies in the economy. We’ll soon find out which view is correct.
And as for politics, the Democratic party clearly believes that single Americans will support policies that grow the government leviathan while rolling back the institutions that have long shaped civil society. The Obama campaign targeted these voters by offering them Planned Parenthood and Julia.
That the Republican party hasn’t figured out how to court singles may partly be a function of failing to notice their rapid growth. But before the GOP starts working on schemes to pander to singletons, it’s worth considering an alternative path.
Rather than entering a bidding war with the Democratic party for the votes of Julias, perhaps the GOP should try to convince them to get married, instead. At the individual level, there’s nothing wrong with forgoing marriage. But at scale, it is a dangerous proposition for a society. That’s because marriage, as an institution, is helpful to all involved. Survey after survey has shown that married people are happier, wealthier, and healthier than their single counterparts. All of the research suggests that having married parents dramatically improves the well-being of children, both in their youth and later as adults.
As Robert George put it after the election, limited government “cannot be maintained where the marriage culture collapses and families fail to form or easily dissolve. Where these things happen, the health, education, and welfare functions of the family will have to be undertaken by someone, or some institution, and that will sooner or later be the government.” Marriage is what makes the entire Western project—liberalism, the dignity of the human person, the free market, and the limited, democratic state—possible. George continues, “The two greatest institutions ever devised for lifting people out of poverty and enabling them to live in dignity are the market economy and the institution of marriage. These institutions will, in the end, stand or fall together.”
Instead of trying to bribe single America into voting Republican, Republicans might do better by making the argument—to all Americans—that marriage is a pillar of both freedom and liberalism. That it is an arrangement which ought to be celebrated, nurtured, and defended because its health is integral to the success of our grand national experiment. And that Julia and her boyfriend ought to go ahead and tie the knot.
Jonathan V. Last is a senior writer at The Weekly Standard. His book What to Expect When No One’s Expecting: America’s Coming Demographic Disaster (Encounter) is forthcoming in February.
Recent Blog Posts