If any young man is about to commence the world,” a reform-minded New York journalist wrote in 1838, “we say to him, publicly and privately, Go to the West.” Today, Horace Greeley’s advice comes down to us as “Go West, young man” and has been mythologized into a slogan of Manifest Destiny. But at the time Greeley wasn’t expressing any messianic desire to tame a savage continent. Rather, he was alarmed by how terrible economic conditions had become in New York.
The Panic of 1837—American history’s worst financial calamity until the Great Depression—had put one-third of the city’s labor force out of work. Greeley ran poor relief in one of the city’s wards, but available funds were meager and people were dying from lack of food and shelter. An angry mob became so enraged by the high price of bread that it rioted and looted the city’s flour merchants. Witnessing all this suffering and rage, Greeley concluded that the best solution would be for new entrants to the workforce to move someplace—anyplace—else. “Fly, scatter through the country, go to the Great West,” Greeley wrote. “Anything rather than remain here.”
Eventually, the economy improved in New York—in no small measure because so many people did scatter to other parts. Indeed, until quite recently, what most distinguished Americans from other peoples was the high percentage of us who were willing to move from anywhere to anywhere to seek a better financial toehold. The original Pilgrims and Utah’s Mormon settlers migrated for religious reasons, but they were exceptions; most American migrations have been driven by economics. Already by the mid-nineteenth century, the descendants of Puritans had abandoned New England’s played-out farms in such numbers that the forests were closing back in, prompting Herman Melville to compare the landscape to “countries depopulated by plague and war.”
Similarly, after the Civil War, huge, downtrodden populations left mid-Atlantic cities like Philadelphia or Baltimore to seek the living wages offered in then-frontier cities like St. Louis and Chicago. For others, even the prospect of busting sod on a homestead in the Oklahoma Territories beat trying to make a living in New Jersey.
In the early- and mid-twentieth century, the Great Migration of African Americans out of the South was a movement of tenant farmers fleeing Jim Crow for a chance at menial factory jobs in places like Cleveland and Detroit. More recently, the great flow in the opposite direction, from Rust Belt to Sun Belt, has seldom been motivated by any utopian political-religious dream; far more often, it’s been the desire to escape difficult economic circumstances.
This feature of American life has served the country’s economy well, if not always its culture. Writers from Henry David Thoreau to Theodore Dreiser to John Cheever may have decried the rootlessness in American life, but at whatever the price in urban anonymity or suburban anomie, the high mobility of American labor meant that a comparatively high share of the nation’s workforce migrated to wherever it could add the most economic value. As the population reduced its concentration in lower-wage areas and increased it in higher-wage areas, the effect was to gradually reduce inequality of income and opportunity, until something like an “American standard of living” emerged in the twentieth century. All told, according to a 2013 paper by Harvard economists Peter Ganong and Daniel Shoag, approximately 30 percent of the drop in hourly wage inequality that occurred in the United States between 1940 and 1980 was the result of the convergence in wage income among the different states during this period.
In our own time, though, all of that has changed. Americans are moving far less often than in the past, and when they do migrate it is typically no longer from places with low wages to places with higher wages. Rather, it’s the reverse. That helps explain why, since the 1970s, income inequality has gone up and upward mobility has (depending on who you ask) either stagnated or gone down.
When people move, it’s usually not very far—as when, for example, a young couple moves to an apartment with an extra bedroom in a nearby suburb after their first child is born. But that’s not the kind of migration Greeley was talking about. For determining how many Americans move in search of economic opportunity, the best available measure comes from data on people moving across state lines that the Census Bureau has collected since shortly after World War II.
In the early 1950s, about 3.5 percent of all American households moved from one state to another in any given year. This proportion held up through the 1970s, and then started to fall around 1980. By 2006 interstate migration had dropped to 2 percent, and by 2010 to just 1.4 percent, or less than half the rate of the early 1950s. The latest available data, for 2011-12, shows interstate migration still stuck at a mere 1.7 percent. Though it may not square with our national self-image, America today is a nation of people who tend to stay put, with a population that is no more mobile than that of Denmark or Finland.
It would be reassuring to find out that the decline in American migration had innocuous causes. One possibility that comes to mind is the bulge in the number of middle-aged workers that occurred as some seventy-six million Baby Boomers passed through their forties, fifties, and sixties over the last three decades. Middle-aged people are less likely to move than young people, so a greater share of middle-aged people in the population should have driven down migration rates, right? But it turns out that the aging of the Boomers was only a minor factor, accounting for less than one-tenth of the decrease in interstate migration, according to a study by Raven Molloy of the Federal Reserve Board and others.
The larger reality is that Americans of all ages and stripes were becoming far less likely to move. So, for example, between the 1980s and the 2000s, the percentage of young adults (those aged eighteen to twenty-four) who migrated across state lines declined by 41 percent. Similarly, whether married or single, black or white, parent or childless, factory worker or knowledge worker, educated or not, Americans in any or all of these categories became far less likely to move than their counterparts a generation ago. Even immigrants from abroad are much less likely to move, once they get here, than were foreign immigrants in the past.
Some might suspect that the proliferation of two-earner couples is an explanation. Surely having a spouse who works can make it more difficult to pursue job opportunities in distant places. Yet the percentage of married households with two earners has hardly changed over the last thirty years. Instead, the relevant change is that today’s two-paycheck married households are about 46 percent less likely to move across state lines than were their counterparts in the 1980s.
In recent years, the bursting of the housing bubble, which has left millions of American homeowners owing more on their mortgages than their houses are worth, has surely been a factor in depressing the number of people moving. Yet it turns out that the phenomenon of “housing lock” has played surprisingly little role in driving down migration rates even during the worst years of the Great Recession. For example, rates of migration have fallen equally for homeowners and renters. And in the states with the worst housing markets, such as Nevada and Arizona, unemployed workers have been no more or less likely to move than in states with the best housing markets. Moreover, taking the longer view, rates of migration were falling year after year even when the housing boom was on and most home sellers could walk away from a settlement with a huge check.
What about the growth of the Internet? Fewer people have to move to connect with the job they want. And indeed, the percentage of workers who telecommute from home has nearly doubled since 1980. But that still comes to only 4.1 percent. (And anyway, telecommuting has mainly turned out to be a way for employees to work extra hours after they come home from the office.)
It would be heartening to discover that fewer Americans need to move these days because more are able to find great jobs right in their own communities. Yet there is abundant evidence that this is not the case. Most obviously, in recent years, unemployment has increased as migration has continued to decease. Moreover, along with everyone else, unemployed people are far less likely to move today than they used to be.
Nobody has a better reason to pick up and move than someone who can’t find a job—or at least so it would seem. But while unemployed people remain likelier to migrate than employed people, they are much less likely to migrate than in previous decades. In 1956, for example, 7.6 percent of unemployed males moved from one state to another during the previous year. Subsequently that rate fell to 7 percent (1966), 5.9 percent (1976), 5.3 percent (1986), 4.4 percent (1996), 4.3 percent (2006), and, finally, 2.7 percent (2012).
In the past, migration served to reduce inequality among states by matching workers more efficiently to the country’s best available jobs. The creation of a single automobile plant—Ford’s River Rouge complex, completed in 1928—boosted Michigan’s population by creating more than 100,000 well-paying jobs. (One of those migrants, the future United Auto Workers President Walter Reuther, made a beeline from West Virginia on hearing Ford was paying tool and die makers the equivalent, in 2013 dollars, of $16.50 an hour.) Today, the reverse appears to be happening. While migration goes down, the richest states are getting richer.
Connecticut, for example, had in 1980 a per capita income that was 21 percent above the national average. By 2011, that disparity rose to 39 percent above average. Over the same interval, New York State’s per capita income rose from 8 percent to 23 percent above average, while Massachusetts went from 6 percent to 29 percent above average.
Today, the state with the highest median household income is Maryland. If you want to do like Willie Sutton and go where the money is, migrate to Maryland, because median income there is about $70,000, or roughly 40 percent above the national median. Pack up the truck, we’re moving to Maryland! Except we aren’t. About 8,000 more people moved out of Maryland last year than moved in from other states. Indeed, Maryland experienced negative net domestic migration in every year of the past decade except two.
You don’t like the Free State? How about California? It’s got an annual GDP in excess of $1 trillion. That’s more than Saudi Arabia. Cradle of the tech revolution, California is home to more than eighty-five billionaires. But it’s been losing native-born residents for two decades.
So where have people been moving to, if not to where the money is? Generally to southern Sun Belt states, where average wages not only are lower than in the places they left behind but are also growing more slowly. So, for example, when people moved from Connecticut to Texas in 1980, they moved to a place where per capita income was 17 percent lower. By 2011, when people made the same migration from Connecticut to Texas, they wound up in a place where per capita income was 31 percent lower. And yet they kept coming.
Maybe you’re thinking that states with lower wages have a higher volume of jobs. Pay people less and you can hire more of them, right? But in fact, most Americans moving across state lines are relocating to places where they’re no more likely to find employment. As the Atlantic’s Jordan Weissman pointed out in December 2012, of the ten states with the highest rates of in-migration, more than half had unemployment rates equal to or higher than the national average.
A minor exception to this dismal trend is North Dakota. It has the country’s highest net in-migration rate right now, and the lowest unemployment rate. That’s because the state is experiencing an old-fashioned economic boom thanks to new oil-extraction technologies like fracking. But North Dakota has a total population of only about 700,000. A huge in-migration rate therefore translates into about 11,500 actual people, or about 0.005 percent of the national working-age population. That makes North Dakota’s oil-boom-driven migration—however significant for North Dakota itself—a minor economic event for the country as a whole.
The larger picture is one in which migration is not only declining but also tends to be away from places where, according to recent studies, young adults have the best chances of moving up the income scale.
So for example, in the Opportunity Index, presented here, New York State scores far better than Texas. Yet between 2010 and 2011, 26,155 people moved from New York to Texas, while only 9,151 people moved from Texas to New York. Similar results come from Harvard’s Equality of Opportunity Project, which ranks metropolitan areas according to the likelihood that young adults will rise from the bottom to the top of the income scale. Most of the metro areas at the top of this ranking are places experiencing negative net domestic migration, including Boston and San Francisco.
Why are Americans by and large moving away from economic opportunity rather than toward it? It all starts to make sense when you think about “push” rather than “pull.” One “push” factor heavily touted by conservatives is state income taxes. Raise the state income tax, conservative dogma holds, and taxpayers will make an exodus to lower-tax states. But a Reuters report in February cast doubt on this hypothesis, pointing out that the rich typically stay put when state income tax rates rise. As for working-class Americans, moving to a state with low income tax rates hardly makes sense if you have to take a bigger cut in wages. Moreover, states with low income tax rates generally have high sales taxes, which, because they are regressive, punish working-class people the most.
A far more plausible push factor is the cost of housing.
Consider the mystery of the “San Berdoo” population boom. Between 2005 and 2009 the nation’s biggest county-to-county migration was from Los Angeles County to San Bernardino County. These counties are both in Southern California, but they’re sufficiently far apart (sixty miles) that they constitute separate labor markets. Before the recession, unemployment in San Bernardino and Los Angeles Counties was about the same. After the recession hit, unemployment in San Bernardino County rose higher, topping out at about 15 percent compared to LA County’s 13 percent.
Moving from LA to San Bernardino would not have improved your chances of finding a job. We’re talking about a city that declared bankruptcy last year. Nor would San Berdoo be the place to escape big-city crime; after it laid off cops in droves, it experienced a 50 percent increase in its homicide rate. Among the white population aged sixteen to twenty-four, 18 percent are neither in school nor have a job, a number that rises to 19 percent for Latinos and 25 percent for African Americans. Yet a huge flow of migrants continues from LA. Why? Mainly because housing is cheaper. Gloria Santellan recently wrote on the Facebook page of the Victorville Daily Press, a local paper, “I was paying $750 for a studio in the San Gabriel Valley area, and out here I was in a two-bedroom for $25 less!”
Exclusionary zoning at the local level is one way people like Santellan get priced out of those parts of the country where jobs are most lucrative and plentiful. This is the theme of two recent e-books, The Gated City by the Economist’s Ryan Avent and The Rent Is Too Damn High by Slate’s Matthew Yglesias. Here’s a brief rundown by Avent:
In some cases, there are explicit zoning limits. Buildings can only be so tall or can only be used for commercial or industrial purposes.… In many cases, neighbors opposed to new developments in their neighborhood lobby the government to change either the zoning rules or historical designations in order to block development projects. And in some situations, no actual law or regulation is necessary to limit redevelopment—community opposition is sufficient to do the work of curtailing supply.
San Jose, the de facto capital of Silicon Valley, has a metropolitan-area population of 1.7 million inhabiting about 1,300 square miles. But in 2005 the metro area approved permits for only 5,700 new units. The city’s tallest building, Yglesias notes, is a mere twenty-two stories high. Average earnings in the valley grew by nearly 40 percent between 1997 and 2000. That’s a lot. But home prices in nearby San Francisco grew even faster, doubling during the same period.
Tight restrictions on housing construction don’t keep out the affluent (who continue to migrate to places like Boston and New York and San Francisco). But they do reduce available housing for working people. In its most extreme form, zoning has, in certain gaudily affluent neighborhoods, become a tool for maximizing the size of single-family McMansions. In Dallas’s luxury gated community of Enchanted Hill, for example, you aren’t allowed to build a house that’s less than 4,000 square feet, according to Affluent magazine. Bungalows need not apply.
Another factor driving up the cost of real estate in at least some high-income cities is underinvestment in transportation infrastructure. In metro areas such as Washington, D.C., for example, suburban commuting, particularly by car, has become such an ordeal that proximity to a Metro stop will send home prices into the stratosphere. An answer to that problem is more Metro lines.
The underlying problem, however, isn’t the price of housing per se so much as its relationship to income. Since 2009, when the recession ended, the median price of a new house in the United States has risen 13 percent, even as median household income has fallen by about 4 percent. That doesn’t pose much of a problem for a migrating architect whose income is already well above the median, and who is likelier to have existing home equity that he can transfer to another state. But for construction workers, for example, it’s likely to be a big problem, and a reason why they can’t easily move to where the best-paying jobs are.
A construction worker can generally make more money in San Francisco than in suburban Fresno. But it won’t likely be enough more to make up the difference in the relative cost of living. Indeed, few working-class people earn enough money to live anywhere near San Francisco anymore, to the point that there is now a severe shortage of construction workers in the Bay Area. David Hayes, CEO of Skyline Construction, recently told the local business press that his biggest challenge is “[r]ecruiting, recruiting, recruiting. We have started relocating candidates from Southern California and recruiting out-of-state candidates, along with asking retired workers to return.”
If labor markets were operating efficiently, construction workers, along with electricians, plumbers, nurses, nannies, elementary school teachers, and other working-class Americans, would receive enough compensation to live near the places where their work is most needed. But our labor markets are not efficient; rather, they are rigged and skewed, offering too much compensation to people with some skill sets (merging companies and writing derivatives, for example) and not enough to others whose skills are often just as hard to learn (e.g., brick laying and teaching children to read) and often more vital to society.
Ganong and Shoag provide a data series that captures the bottom line. In 1940, the income of “lower-skilled” workers captured 88 cents of every dollar increase in state per capita income. That share began to decline in the 1970s, and by 2010 it was down to 36 cents. Put another way, working-class people in the richest regions of the country have a much lower share of the income around them than they once did. That, more than any other reason, is why they have such a hard time moving to where incomes are highest. Incomes aren’t high for them.
Yogi Berra supposedly once said, “Nobody goes to that restaurant anymore. It’s too crowded.” We might similarly observe, “Nobody moves to that state anymore. It offers too much economic opportunity.” It doesn’t make any sense, but that’s life in our present post-migration era. For all his historic foresight, Greeley could never have imagined an outcome so undemocratic and economically perverse.