Why Crime Is Down in America's Cities

Abortion, drug laws, reduced poverty--none of these factors is the main reason our metro areas are getting safer

Yesterday, the Financial Times ran an op ed piece of mine about the mysterious downward trend in crime. While this drop is a welcome development -- and very possibly a bellwether of even more positive changes in our society -- it poses a vexing challenge to professional explainers. What follows is a longer, more detailed, and more statistics-laden exploration of why crime rates are falling.

Almost three years into the worst economic downturn since the Great Depression, with massive unemployment and pessimism rife, America's crime rates are falling and no one--not our pundits, policemen, or politicians, our professors or city planners--can tell us why. As I wrote about here, there were 5.5 percent fewer murders, forcible rapes, robberies and aggravated assaults reported in 2010 than in 2009, according to the most recent edition of the FBI's Uniform Crime Report; property crimes fell by 2.8 percent over the same period and reported arsons dropped by 8.3 percent.

And the drop was steepest in America's biggest cities--which are still popularly believed to be cauldrons of criminality. "While cities and suburbs alike are much safer today than in 1990," notes a recent report by the Brookings Institution, "central cities--the big cities that make up the hubs of the 100 largest metro areas--benefitted the most from declining crime rates. Among suburban communities, older higher-density suburbs saw crime drop at a faster pace than newer, lower-density emerging and exurban communities on the metropolitan fringe."

An essay in The Economist featured a graphic which charts the arc of American crime rates since the 1960s. Its caption poses the question that is on everyone's lips:

Some explanations evoke Sherlock Holmes' "dog that didn't bark." When crime rates first began to fall in the 1990s, Steven Levitt and John Donohue III argued that legalized abortion was responsible, since unwanted children would have been more likely to grow up to be criminals - a finding that was not only wildly controversial but has been met with substantial challenge. Research by economist Jessica Wolpaw Reyes find attributes a significant proportion of the decline in violent crime to children's reduced exposure to lead.

Others suggest that America's astronomical incarceration rate--the highest in the world--is responsible, though that begs the question why countries that don't incarcerate their citizens at anything like the same rate suffer from less crime than the US does. It's disquieting to think that what would seem to be an unalloyed social good--less crime--might have been brought about in part by what many believe to be a social disaster: America's draconian, unequally enforced drug laws.

Law enforcement must be given its due; many big city police departments have developed statistically-driven methodologies for targeting crime hot-spots, with conspicuously successful results. And we as individuals have changed our behavior as well--from taking elementary precautions like locking our doors, to investing in crime-stopping technology like burglar alarms for our homes and LoJack for our cars.

Still, it is confounding that crime would decline as economic conditions worsen. My own analysis, conducted with my Martin Prosperity Institute colleague Charlotta Mellander, turns up no statistical associations between crime and either the level or the change in the level of unemployment across metros, or between crime and the level of income inequality. There does seem to be a modest relationship between the absolute poverty rate and crime. Our analysis turned up an association of .19 between crime and the percent of population below the poverty line. But it has weakened substantially over the past two or three decades, according to the Brookings study.

So too has the relationship between crime and race. From evening news headlines to crime shows on TV, popular culture underlines the propinquity between crime and race Our analysis turned up modest correlation (.37) between crime and the share of population that is black. But again, Brookings Institute report assembles powerful evidence to show that the relationship has been weakening. "The association between crime and community characteristics--like the proportion of the population that is black, Hispanic, poor, or foreign-born--diminished considerably over time," notes the study. "The strength of the relationship between the share of black residents and property crime decreased by half between 1990 and 2008, while the association between the share of Hispanic residents and violent crime all but disappeared."

In the popular imagination, crime is frequently associated with big, densely populated cities. Here again, we can separate fact from myth.  Primary cities and older high-density suburbs exhibited the largest decreases in crime between 1990 and 2008, according to the Brookings study. And the gap between city and suburban violent crime narrowed in two-thirds of the nation's 100 largest metro areas. Our own analysis turns up no association whatsoever between metro size or metro density and the overall level of crime, though we do find a modest correlation (.25) between density and violent crime.

In a thoughtful essay in the Wall Street Journal, the distinguished political scientist and urban crime expert James Q. Wilson hit hard at strictly economistic explanations, suggesting that deeper changes in American culture can better account for the mystery. "The cultural argument" he writes, can help explain not only the current drop in crime, but also "the Great Depression's fall in crime and the explosion of crime during the sixties. In the first period, on this view, people took self-control seriously; in the second, self-expression--at society's cost--became more prevalent." My former Carnegie Mellon University colleague, the distinguished criminologist Alfred Blumstein, proposes an "Obama effect," in which young black males' increased optimism about their futures makes them more likely to refrain from violence.

But the key factor, as it turns out, lies in the growing racial, ethnic, and demographic diversity of our cities and metro areas. Our analysis found that the Hispanic share of the population is negatively associated with urban crime. Crime also fell as the percentage of the population that is non-white and the percentage that is gay increased. And of all the variables in our analysis, the one that is most consistently negatively associated with crime is a place's percentage of foreign-born residents. Not only did we find a negative correlation (-.36) between foreign-born share and crime in general, the pattern held across all of the many, various types of crime - from murder and arson to burglary and car theft. 

The Brookings study also finds evidence of a substantial shift in the connection between foreign-born residents and crime. While foreign-born share was positively associated with crime in 1990 and 2000, that relationship had disappeared by 2008. The foreign-born share of population now shows no relationship to property crime, and a negative relationship to violent crime. The pattern is most pronounced for primary cities and inner-ring suburbs, the Brookings study found, but not for lower-density suburbs and ex-urbs.

It might be hard to wrap your mind around this--especially with all the demagoguery about immigration. But the numbers tell a different story than our alarmist pundits and politicians do. "Since 1990, all types of communities within the country's largest metro areas have become more diverse," Elizabeth Kneebone, one of the authors of the Brookings report, wrote in The New Republic. "Crime fell fastest in big cities and high-density suburbs that were poorer, more minority, and had higher crime rates to begin with. At the same time, all kinds of suburbs saw their share of poor, minority, and foreign-born residents increase. As suburbia diversified, crime rates fell." Along with their entrepreneurial energy and their zeal to succeed, immigrants are good neighbors--cultural and economic factors that militate against criminal behavior, and not just in their own enclaves but in surrounding communities as well.

One additional factor bears on this. Our analysis also turns up a consistent negative correlation between crime and the overall level of city happiness. It makes intuitive sense that a low-crime city would be a happy city; still, it's worth pointing out that the happiness measure (which comes from Gallup surveys) is associated not just with overall crime but with almost every type of crime across the board. 

This is somewhat striking in an analysis where associations between crime and key social and economic variables are hard to find. More to the point, the Gallup research identifies openness to diversity as being one of the two most important factors that shape city happiness and community satisfaction across the board.

America's declining crime rates are cause for celebration, even if we can't completely explain the phenomenon. The fact that diversity appears to play such a signal role in the trend--something that most Americans regard as a moral and economic good in its own right-- makes it all-the-more satisfying.

Image Credit: Flickr / dsb nola