Advancing social mobility—a phrase that’s shorthand for making it more likely that children will grow up to be better off financially than their parents—elicits universal approval as a virtuous endeavor. It’s a progressive cause that most people instinctively support without giving much thought to all that it signifies.
The semantics partially inhibit critical faculties—“mobility” is preferable to inertia. Also, the idea of social mobility conveys a righteous alternative to the rigidity and oppression of countries that historically were controlled by hereditary aristocratic rulers. And even in the absence of an aristocratic class, it has a particular appeal in the United States, an immigrant society, where it wasn’t unreasonable for people arriving at Ellis Island empty-handed to hope that their children would attain middle-class prosperity. Although some people got stuck on the bottom and some ethnic groups moved up more quickly than others, the Horatio Alger myth remains a reality for many immigrants.
Economists often express the rate of social mobility in terms of the likelihood that children will rise above the economic position into which they were born. Thus, there appeared to be good reason for consternation when President Obama, speaking in 2013 to an audience at the left-leaning Center for American Progress, decried the diminishing levels of upward mobility in recent years, pointing out that an American child born into the poorest fifth of households, by income, “has less than a 1-in-20 shot of making it into the top.” Urging the country not to pit the interests of the middle class against those of the poor, but rather to try “to improve upward mobility for all people,” he observed “that it is harder today for a child born here in America to improve her station in life than it is for children in most of our wealthy allies—countries like Canada or Germany or France. They have greater mobility than we do.”
Hailed as “one of his strongest economic speeches” by The New York Times’s editorial board, not the faintest curiosity was expressed about the verity of the purported decline, the implications of the proposal to accelerate mobility, or the accuracy of the international comparisons. Such is the self-evident good attributed to social mobility that any call for improvement is cause for applause.
The desire “to improve upward mobility for all” brings to mind Garrison Keillor’s weekly musings on A Prairie Home Companion about life in the town of Lake Wobegon, where “all the women are strong, all the men good-looking, and all the children above average.” But the stubborn thing about arithmetic is that not everyone can be above average. This applies to the estimable goal of improving upward mobility for all, when it is defined by the rate at which people move from lower to higher income brackets. Judged by this relative standard, social mobility is a zero-sum game, pitting everyone in the income distribution against each other—for every winner, there must be a loser. To be more specific, the top 20 percent cannot accommodate all of society’s households; thus, any rise in the flow of households from the bottom 20 percent of the income distribution to the top is, by definition, matched by an equivalent increase in the number of those who must drop out of the higher bracket into a lower one. So, as fair and beneficial as it may sound, an appeal to increase upward social mobility is necessarily an inadvertent demand for escalating downward mobility.
This does not discredit the potential value of increasing social mobility. But it raises a critical issue entirely ignored in calls to do so: What is the socially desirable rate of upward (and correspondingly downward) mobility? In addressing this issue it is important to bear in mind that currently, no one is employed in more than 60 percent of the households in the bottom fifth of the income distribution, while 75 percent of the households in the top fifth have two or more earners. Moreover, according to Census data, less than 20 percent of the households in the bottom fifth are married-couple families, compared to more than 75 percent of the households in the top fifth; single parents represent 23 percent of the households in the bottom quintile, compared to 9 percent of those in the top; African-American households account for almost 22 percent of those in the bottom fifth of the income distribution, which is 50 percent higher than their proportion of the total population; and over 40 percent of those in the bottom quintile are at the age of retirement or just entering the labor force, which is almost three times that of those in the top income bracket. Thus any effort to discern the socially desirable rate of mobility must weigh the extent to which opportunity, discrimination, work effort, family structure, and the basic demographics of labor-force participation influence movement up and down the economic ladder. The calculation is elusive.
The absence of an agreed-upon standard for judging the just rate of mobility is typically dealt with by comparing America’s rate of mobility to those of other Western democracies. The highly publicized claim that social mobility in the U.S lags far behind that of other nations stems from a widely cited comparative analysis conducted 10 years ago, which placed the U.S.’s rate of mobility next to last among nine wealthy industrial democracies. However, in 2014 a team of economists from the U.S. Department of the Treasury, Harvard University, and the University of California, Berkeley examined almost 50 million tax returns in what is arguably the most extensive and rigorous study of social mobility to date.
Calculating three alternative measures of mobility, the researchers offered persuasive evidence that the United States has one of the highest rates of mobility in the world, ranking fourth, just behind Finland, Denmark, and Norway. Moreover, the findings showed no decline in the rate of social mobility among children born in the U.S. over the last 40 years. (Within the overall rate, however, the degree of mobility varied among 709 geographic districts throughout the country. The four characteristics most significantly related to these geographic discrepancies were an area’s racial segregation, high-school dropout rates, percentage of households with single mothers, and amounts of community involvement, as measured by factors such as voter turnout and participation in local organizations. Among these characteristics, the share of children in single-parent families was the strongest and most robust predictor of differences in social mobility.)
As for movement up the income ladder, the Harvard-Berkeley study revealed that children born to parents in the bottom fifth of the income distribution had a 9 percent chance of making it to the top quintile. That may sound dismal, but in judging this movement by international comparisons it’s important to bear in mind that the range of incomes varies among countries. When the income ladders differ in height, it takes a greater increase over a father’s earnings for his son to move up one income quintile on the taller ladder—and the U.S.’s income ladder is one of the tallest in the world.
So, rather than comparing how one’s income ranks relative to others all across the country, it would be wiser to focus on an absolute measure of social mobility, one that describes an individual’s changing level of prosperity over time. After all, average citizens are typically unaware of whether the rate of relative social mobility has gone up or down; they do, however, have a keen appreciation of whether their material standard of living is better than what they experienced as a child under their parents’ roof.
In this regard, America is doing quite well. According to the Brookings Institution, 67 percent of Americans born in 1968 had higher levels of real family income between 1995 and 2002 than their parents had a generation earlier. The overall proportion of children who were better off than their parents increased to 81 percent when incomes were adjusted for family size; most of those who were not better off than their parents were born to families with the highest incomes. When broken down into upper and lower income groups, four out of five children from the bottom fifth of the income distribution had higher family incomes than their parents. The median income for this group was twice as high as that of their parents. Moreover, the U.S. remains one of the only places in the world where the children of immigrants regularly go on to achieve a socioeconomic profile mirroring that of the general population—as adults, they have similar median incomes, college graduation rates, rates of homeownership, and poverty rates as the nation as a whole.
A recent, widely-covered analysis tracking the rates of absolute mobility over time, though, reveals that the current level of mobility is far below what is was in the mid-20th century; at the age of 30, more than 90 percent of people born in 1940 had higher household incomes than their parents did at a similar age. This does temper the Brookings finding, but it is hard to distinguish the extent to which the decreasing rate of mobility since 1940 signifies the fading of the American dream or is merely an artifact of the increasing participation of mothers in the workforce—a trend that took off between 1940 and 1970.
That is, less than 20 percent of mothers were working and contributing to the family incomes of children born in 1940. In contrast, by the time the children born in 1940 reached age 30, 40 percent of them lived in households with mothers working and contributing to their families’ incomes. Though this rate never doubled again, the share of two-earner families continued to increase slowly, leveling off by 1990, at which point comparing generations involved almost an equal proportion of two-earner families in both the children’s and parent’s groups.
Thus, it’s to be expected that the 1940 cohort of children would have higher family incomes than their parents since (beyond the general benefit from an increasing Gross Domestic Product) this group had the advantage of proportionately twice as many two-earner families as their parents. Similarly, as the relative advantage drawn from comparing two-earner family incomes to one-earner family incomes slowly declined, it would make sense that there’s something of a decrease in the remarkably high rate of absolute mobility recorded for the 1940 cohort.
When people talk about the U.S.’s social mobility, they tend to talk about relative mobility—how much of the difference between one generation’s incomes is associated with the difference between the incomes of their parents’ generation. The best evidence suggests that this has not changed since 1970. But relative mobility rates indicate very little about standards of living, which is the way ordinary citizens assess their well-being. Progressives emphasize the vague psychological discomfort of relative deprivation, which may be felt when people compare how much their income changes between generations to how much others gain or lose. This view of economic mobility discounts the tangible material comfort of an absolute gain in one’s own standard of living, regardless of how the neighbors are doing.
Most people outside of academia tend to judge their economic progress in absolute terms. By this standard, when household incomes are adjusted for family size along with the vast increase in social-welfare transfers (including the Earned Income Tax Credit, enacted in 1975), a substantial majority of couples in their 30s are living in families with incomes higher than those of their parents at that age. And most of those who are not better off than their parents were born to families in the upper income brackets; their household incomes stand to be supplemented by an ample inheritance. In this sense the American dream is alive and well: In a society of abundance, the tangible benefits of absolute mobility leave many citizens immune to the presumed sting of relative deprivation.
This post originally appeared on The Atlantic.