Urban Revival

Immigration and America's Urban Revival

The evidence favors a hypothesis many Americans reject: Immigration has helped reduce crime and revitalize city economies. 

By Robert J. Sampson

mysterious thing happened on the way to the widely projected meltdown of American cities in the last quarter of the 20th century. Instead of collapsing, many of our largest and hardest-hit cities have rebounded and flourished, confounding critics and social scientists alike. Why did these cities experience renaissance rather than ruin? Or, to paraphrase the late University of Pennsylvania historian Michael Katz, “Why didn’t American cities burn?” More broadly, what accounts for the remarkable decline of violent crime in the United States and the return of urban vibrancy?

Although certainly not the only factor, immigration deserves attention as part of the answer for the nation’s crime decline and urban revitalization. Immigrants have gravitated to many of the urban areas that were most distressed 40 years ago and have contributed to their economic revival. Contrary to widespread beliefs, high concentrations of immigrants are also associated with lower crime rates. The evidence for the positive impact of immigration has been growing, but before I get to that evidence, let’s remember what urban life was like 40 years ago and how far we have come.

By the mid-1970s, America was experiencing an undeniable urban crisis. Cities were thought to be dying—especially older cities in the East and Midwest Rust Belt. Although the crisis had deep roots stretching back decades, analysts pointed to a variety of immediate causes: the scars of 1960s rioting, population decline, job losses, high crime, fiscal collapse, and widespread housing vacancies. Things seemed to come to a head in 1975 when President Gerald Ford allegedly told New York City to “drop dead” rather than expect a federal bailout, according to a famous headline in the New York Daily News.

The crisis continued unabated into the 1980s. William Julius Wilson’s The Truly Disadvantaged illuminated the socioeconomic processes behind the unraveling of the “inner city,” while James Q. Wilson and George L. Kelling captured the zeitgeist with their argument that even small signs of urban decay—symbolized by unrepaired “broken windows”—led to a spiral of decline. Although the debate was largely about Eastern and Midwestern industrial cities, the South and West were not immune either. At the beginning of the 1980s, Miami suffered one of the nation’s worst race riots since the civil rights era. Los Angeles witnessed rising racial tensions even before the Rodney King riots in 1992.

Spike Lee’s movie Summer of Sam, about a serial murderer on the prowl in 1977, captured the fear of New Yorkers in that era. Death Wish, a deeply disturbing series of movies starring the actor Charles Bronson, fulfilled the apparent wishes of would-be vigilantes. It was Bernhard Goetz, however, who became a real folk hero to many New Yorkers after shooting four black men on a subway in 1984 during what he described as an attempted robbery. Both the Son of Sam and Goetz cases received intense publicity, locally and nationally. At the end of the 1980s, the crack cocaine epidemic put what seemed to be the final nail in the city coffin. Violence spiraled again and New York City hit its peak in 1990, topping out at more than 2,200 murders.

The urban decay and violence of that period was a fact of life. Living in New York in the 1970s and again in the 1980s, I was the victim of three thefts and witnessed a knife-wielding junkie attempt to hold up a bodega on Broadway near 125th Street. When I moved to the University of Chicago in 1991, the first advice I received was to carry loose $20 bills for holdups—“robbing money.” Personal anecdotes, of course, are not definitive evidence of social trends, but the data tell the same story. Episodes of the kind I experienced firsthand were the subject of everyday conversation and consequently affected widely shared attitudes toward cities.

But then the world changed. Rather than the predicted or seemingly inevitable collapse, violence began to plummet, and formerly hemorrhaging cities began to grow. Today, some of the old national symbols of urban decline have become the envy of the suburbs, defying earlier predictions. Experts were caught off guard, just as they were by the fall of the Soviet Union and the Great Recession.

The magnitude of the turnaround and the scope of urban renaissance are remarkable. In 2014, for example, New York logged fewer than 330 murders despite a population larger than in 1990—the lowest count since comparable records have been kept. Across the country, violence trended down instead of culminating in a wave of violent “superpredators,” as William Bennett, John DiIulio, and John P. Walters confidently predicted in the 1996 book Body Count. The pattern for the last half-century for our most reliable measure of violence—murder—has been an “inverted U,” as shown in the figure above. We are now back to the early 1960s (and in fact the tranquil 1950s) in rates of murder.

It was not just crime that fell. Today, the Big Apple is thriving and as exciting as ever, as are Los Angeles, Boston, Seattle, and San Francisco. Chicago and Miami also appear to be back from the brink. These and other central cities are magnets not only for the young but also for empty nesters and families with children. Construction cranes are seemingly everywhere in the race to meet the demand for city living.

Teenage pregnancy came down too, as did mortality rates, especially among the most disadvantaged groups. For example, from 1990 to 2010, life expectancy increased 7.3 years for black males, compared to 3.8 years for white males. Teen pregnancy declined 56 percent for blacks, 45 percent for whites. While poverty is typically associated with inner cities, poverty rates actually increased faster in the suburbs from 2000 to 2010. Indeed, the foreclosure crisis of the Great Recession hit suburbs harder. Meanwhile, gentrification, population, and housing prices in coastal cities have pushed upward.

To be sure, the urban renaissance has not unfolded evenly. While creativity, diversity, lower crime, and bustling streets define cities on the move, the picture is far different elsewhere. Detroit and St. Louis, for example, continue to struggle, and Baltimore has recently witnessed considerable turmoil. Meanwhile, smaller cities like Stockton, California, are the new face of today’s urban fiscal crisis.

What explains the trajectory of metropolitan revitalization—both the winners and the losers? There is no single answer. As of about ten years ago, law enforcement officials, politicians, and social scientists had advanced various explanations for the unexpected drop in crime rates. The primary suspects were a decline in crack use, aggressive policing, increased incarceration, a relatively strong economy in the late 1990s, aging of the population, and the availability of legalized abortion beginning in the 1970s. (Declining childhood lead exposure from the 1970s on may also have had a lagged impact on crime.) While each of these had probably played some role, I thought the discussion was missing a critical factor.

Immigration was notably absent from the discussion ten years ago. That omission seemed odd to me because the transformed vitality of cities was most visible in the places that had seen the greatest increases in immigration. New York, a leading magnet for immigration, had for a decade already ranked as one of America’s safest major cities. Violent crime in Los Angeles dropped considerably in the late 1990s (more than 45 percent for homicides), as it did in other cities with large Hispanic populations such as San Jose, Dallas, Chicago, and Phoenix. The same can be said for cities along the border, such as El Paso and San Diego, which have ranked as low-crime areas for some time. Moreover, in the 1990s alone, foreign immigration had increased more than 50 percent, so the changes were very large and affected much of the country. The United States had become more ethnically diverse not only in our nation’s cities but in suburban and rural areas as well.

Based on these changes, and given a long line of criminological research showing that first-generation immigrants were less crime-prone than their second- and third- generation counterparts, I advanced the hypothesis that increased immigration contributed to falling crime rates.

I naïvely thought that this was a plausible and largely uncontroversial hypothesis, one that might have been overlooked but that was nonetheless rooted in firm logic and prior evidence. The spew of hate mail I received suggested otherwise. It included not just the anger and bile that is typical online, but also assertions that the idea was crazy—“lunacy,” as one response put it. Yet in the ensuing years, the thesis that immigration has led to lower rates of violent crime has become increasingly debated, and a number of empirical tests have emerged. The change in reception is refreshing and is supported on several fronts, to which I now turn.

Immigration may affect crime rates, first of all, because of who chooses to immigrate—a factor that social scientists refer to as “selection bias.” Although there are exceptions, it is widely recognized that most immigrants, Mexicans in particular, selectively migrate to the United States based on characteristics that predispose them to low crime, such as motivation to work and ambition. First-generation immigrants (those born outside the United States) may also be more law-abiding because of their interest in not being deported. The population of the United States now includes more than 40 million foreign-born people. If they are, as the evidence suggests, predisposed to lower crime, they increase the denominator of the crime rate while rarely appearing in the numerator.

Second, the composition of the immigrant population may also be a critical determinant of community welfare and public health. Latinos tend to do better on various indicators of well-being than do other socioeconomically disadvantaged groups. In Chicago, my colleagues and I found a significantly lower rate of violence among Mexican Americans compared to blacks and whites. Moreover, first-generation immigrants were 45 percent less likely to commit violence than third-generation Americans, adjusting for individual, family, and neighborhood background. Second-generation immigrants were 22 percent less likely to commit violence than the third generation. This pattern held true for non-Hispanic whites and blacks as well, and took into account poverty and other relevant social characteristics such as income, marital status, and even individual “IQ.” And when we controlled for immigrant status, Mexican Americans no longer had lower rates of violence than blacks and whites. Immigrant status thus appears to result in lower levels of violence among all groups—blacks, whites, and Latinos—but many more Latinos are foreign born.

Third, we showed that living in a neighborhood of concentrated immigration was associated with lower violence (again, after taking into account a host of correlated factors, including poverty and an individual’s immigrant status). Rather than generating crime, high concentrations of immigrants appear to reduce it. An average male is almost 25 percent more likely to engage in violence if he lives in a high-risk neighborhood without many immigrants than if he lives in a high-risk immigrant neighborhood.

These findings are broadly consistent with other research in criminology and provide a potential explanation of the sweeping crime drop. As foreign immigration to the United States, especially from Mexico, rose sharply in the 1990s, the proportion of first-generation immigrants increased, and that change helped reduce overall crime and improve well-being.

The pattern of immigration and lower crime nonetheless goes against popular stereotypes, which may explain the angry initial reaction to the idea. Among the public, policy-makers, and many academics, a common expectation is that the concentration of immigrants and the influx of foreigners drive up disorder and crime because of the assumed propensities of these groups to commit crimes and settle in poor, presumably disorganized communities. This belief is so pervasive that the concentration of Latinos in a neighborhood strongly predicts perceptions of disorder, regardless of the neighborhood’s actual amount of disorder or the rate of reported violence. Nonetheless, whatever people think, increases in immigration are correlated with less violence, and first-generation immigrants tend to be less violent than those born in America, particularly when they live in neighborhoods with high numbers of other immigrants.

This conflict between perceptions of immigrants and objective evidence about them is not entirely new. In the early 20th century, many Americans believed that immigration and greater ethnic diversity increased crime, although first-generation immigrants did not, in fact, bear out stereotypes popular at the time that linked immigration to higher crime rates. Today’s immigrants come from different parts of the world, but the gap between perception and reality is much the same as it was a century ago.

The immigration thesis becomes more interesting when we broaden the discussion to include aspects of urban life beyond crime. As Jacob Vigdor has recently shown, immigration to New York City is linked to population growth, lower rates of vacant and abandoned buildings, and economic revitalization. Elsewhere in the country, many cities gained population in the 1990s because of an influx of immigrants who brought new vitality to previously decaying inner-city neighborhoods. In Chicago, the West 26th Street corridor in Little Village, a large immigrant enclave, is one of the most economically dynamic areas of the city.

Immigrants aren’t the only beneficiaries of these economic and demographic changes: Native-born blacks, whites, and other groups have also gained from the improved economic and civic health of central cities. In New York City, the income of blacks in Queens surpasses that of whites, with the surge in the black middle class driven largely by the successes of black immigrants from the West Indies. From Bushwick in Brooklyn to Downtown Miami, to large swaths of South Central Los Angeles and the rural South, to pockets of the north and south sides of Chicago, immigrants are revitalizing areas that once seemed unlikely ever to emerge from poverty.

Correlation does not prove causation, and much of the data I have discussed so far is only correlational. Yet some recent evidence is consistent with the notion that immigration has a causal impact on community welfare. In Chicago, for example, I found that increases in immigration and language diversity over the decade of the 1990s predicted decreases in neighborhood homicide rates in the late 1990s and up to 2006, adjusting for a host of internal characteristics. In Los Angeles, John MacDonald and colleagues report that Los Angeles neighborhoods where immigrants were most likely to settle had significant reductions in crime. Using data for all U.S. metropolitan areas over the 1994–2004 period, Jacob Stowell and colleagues examined the impact of changes in immigration on changes in violent crime rates. After adjusting for confounding factors, their research indicated that violent crime rates tended to decrease as metropolitan areas experienced gains in concentration of immigrants. These results support the hypothesis that the broad reductions in violent crime during recent years are partially attributable to increases in immigration.

One way immigration may spur revitalization is through the reduction of housing vacancies. According to the “broken windows” thesis, abandoned buildings are a signal of urban disorder and decline. Vacant or boarded-up housing sends a strong message to prospective or current residents about a neighborhood’s viability. Immigration may have contributed to America’s urban turnaround, therefore, partly by filling up housing left vacant in the urban crisis of the 1970s and 1980s.

The evidence I have gathered supports this analysis. Increases in immigration from 2000 to 2010 are directly linked to decreases in housing vacancies in that decade across the more than 70,000 neighborhoods, or census tracts, in the United States. This pattern holds in the nation’s three largest cities—New York City, Los Angeles, and Chicago. The changes over two decades, from 1990 to 2010, also indicate that rising immigration helped cut vacancies. For example, tracts that were on an upward trajectory of immigration in the 1990s saw reductions in vacancies in the next decade, net of both prior and concurrent changes in racial composition, poverty, homeownership, and education (a proxy for gentrifiers).

For a 2012 issue on immigration of the Annals of the American Academy of Political and Social Science, MacDonald and I commissioned some of the most meticulous research to date about the effects of immigration on a cross section of American communities—urban, suburban, and rural. Not all the work found positive effects of immigration, but the scholars who participated were in agreement that while new immigrants are poorer than the general population and face considerable hardship, there is no evidence that they have reshaped the social fabric in harmful ways. A subsequent study by Aaron Chalfin estimating the effect of Mexican immigration on U.S. crime rates finds no effect one way or the other.

At a minimum, therefore, immigration appears to have a benign impact on the social fabric. The bulk of the evidence, however, favors the hypothesis that immigration leads to significant reductions in violence in neighborhoods and cities and at the national level. Recent immigrants—whether white, black, or Latino—have been less violent than native-born Americans. Concentrations of immigrants also appear to have a “neighborhood effect” on the overall level of violent crime in an area. Taken together, my findings in Chicago and the research of scholars such as Vigdor and MacDonald suggest that the large influx of first-generation immigrants has had spillover effects on local communities, such as economic renewal in formerly poor areas, reduction in vacancies, population growth, and possibly the diffusion of nonviolent social mores.

If I am right about the evidence, the rise in immigration has had net positive effects on a wide swath of urban social life. Reduced violence, in particular, has helped turn around urban areas long characterized by deep poverty.

But not all changes from immigration are necessarily positive. I have shown that residents perceive more disorder when there are more immigrants, and according to sobering research by Robert Putnam, neighbors are also less trusting the more diverse their neighborhoods. Internationally, governments tend to cut social welfare provision as societies become more diverse; in the Netherlands, France, and other European countries, political conflict has risen in response to immigration and increased diversity. Moreover, there is considerable heterogeneity among immigrant groups—refugees from war-torn countries leave their homeland for very different reasons than economically motivated immigrants. These patterns urge caution about drawing strong conclusions about the role of immigration in the revival of American cities.

Social and demographic changes are also turning cities inside out, calling into question long-held models of urban society. For example, poverty is moving to the suburbs, wealth is increasingly concentrating in the center cities, and gentrification is reshaping many formerly working-class and poor areas—often with considerable conflict, as in San Francisco’s Mission District recently. The conception of “inner city” as a ghetto no longer applies. American cities are becoming more like many European cities, where low-income and marginalized groups have been pushed to the periphery. On balance, however, the evidence suggests that immigration has improved urban life and is responsible, at least in part, for the remarkable turnaround of the American metropolis.