Are People Nicer in Cities? More Productive? The Socio-Economic Data Revealed

Jonah Lehrer explains in Wired the statistical underpinnings of the reasons that urbanization is one of the 21st Century's mega-trends - to say nothing of why Brooklyn, or London, or Beijing are simply cooler than wherever you're from:

"Why do cities exist? This is a surprisingly difficult question to answer. The modern metropolis, after all, can be an unpleasant, expensive and dangerous place. It’s full of rush hour traffic and panhandlers, overpriced apartments and feisty cockroaches. The air is dirty, there is litter in the streets and the public schools are falling apart. In other words, urban life isn’t easy. We cram ourselves together, but all the cramming comes with a cost.

In a recent NY Times Magazine article, I profiled the work of Geoffrey West, Luis Bettencourt, et. al. These theoretical physicists think they’ve uncovered the reason cities exist, uncovering the socioeconomic benefits that more than compensate for the high rents and vermin infestations. They found their answer by sifting through a vast array of data sets:

The scientists downloaded huge files from the Census Bureau, learned about the intricacies of German infrastructure and bought a thick and expensive almanac featuring the provincial cities of China. (Unfortunately, the book was in Mandarin.) They looked at a dizzying array of variables, from the total amount of electrical wire in Frankfurt to the number of college graduates in Boise. They amassed stats on gas stations and personal income, flu outbreaks and homicides, coffee shops and the walking speed of pedestrians.

After two years of analysis, West and Bettencourt discovered that all of these urban variables could be described by a few exquisitely simple equations. For example, if they know the population of a metropolitan area in a given country, they can estimate, with approximately 85 percent accuracy, its average income and the dimensions of its sewer system. These are the laws, they say, that automatically emerge whenever people “agglomerate,” cramming themselves into apartment buildings and subway cars. It doesn’t matter if the place is Manhattan or Manhattan, Kan.: the urban patterns remain the same. West isn’t shy about describing the magnitude of this accomplishment. “What we found are the constants that describe every city,” he says. “I can take these laws and make precise predictions about the number of violent crimes and the surface area of roads in a city in Japan with 200,000 people. I don’t know anything about this city or even where it is or its history, but I can tell you all about it. And the reason I can do that is because every city is really the same.”

What did they find? According to the equations of West and Bettencourt, every socioeconomic variable that can be measured in cities – from the production of patents to per capita income – scales to an exponent of approximately 1.15. What’s interesting is the size of the exponent, which is greater than one. This means that a person living in a metropolis of one million should generate, on average, about 15 percent more patents, and make 15 percent more money, than a person living in a city of five hundred thousand. (They should also have 15 percent more restaurants in their neighborhood and create 15 percent more trademarks.) The correlations remain even when the numbers are adjusted for levels of education, work experience and IQ scores. “This remarkable equation is why people move to the big city,” West told me. “Because you can take the same person, and if you just move them from a city of fifty thousand to a city of six million, then all of a sudden they’re going to do three times more of everything we can measure. It doesn’t matter where the city is or which cities you’re talking about.”

West and Bettencourt refer to this phenomenon as “superlinear scaling,” which is a fancy way of describing the increased output of people living in big cities. When a superlinear equation is graphed, it looks like the start of a roller coaster, climbing into the sky. The steep slope emerges from the positive feedback loop of urban life — a growing city makes everyone in that city more productive, which encourages more people to move to the city, and so on.

Here’s the bad news, though: The negative stuff of cities – murder, drug consumption, flu outbreaks, bedbugs, shoplifting, etc. – also seem to scale superlinearly. (One interesting exception is prostitution, which scales sublinearly.) This means that people don’t just become more productive and innovative in metropolitan areas – they’re also more likely to get shot and mugged. This is a tradeoff that every city dweller understands. We take the good with the bad. Other people make us smarter, but they also make us sick.

But what about “prosocial” behavior in cities? How do cities influence altruism? Do metropolitan areas make us more charitable? Or more cynical? These questions are the subject of an interesting new paper in Physica A by Samuel Arbesman and Nicholas Christakis of Harvard Medical School. (Thanks to Paul Kedrosky for the tip.) The researchers investigated a number of different prosocial behaviors for which there existed reliable data, including living and deceased organ donation, voting, and number of contributors to political campaigns.

What’s most interesting is what they did not find: prosocial behavior does not obey a single statistical pattern. Unlike the socioeconomic variables studied by West and Bettencourt, people don’t become significantly more likely to vote when living in bigger cities, and they actually become slightly less likely to donate a kidney. However, they do become much more likely to give money to a campaign. This data echoes previous work which showed similarly contradictory findings. For instance, studies show that people in big cities are more likely to return lost letters, while they’re less likely to assist random strangers on the street.

Prosocial behavior, while inherently social, like many productive and creative behaviors, appears to be in a different class [from the socioeconomic metrics studied by West and Bettencourt]. These behaviors do not scale similarly and are different from productive behaviors that have previously been examined, such as patents and economic growth across cities. Furthermore, our quantitative results confirm the somewhat contradictory initial findings related to lost letters and helping strangers: prosocial behavior is not a single category when it comes to understanding urban scaling with respect to population.

At the moment, it remains unclear what’s driving this heterogeneity. My guess is that, to truly understand the causality at work, we’ll need to examine the local differences between various urban areas. Why, for instance, does City A have such high rates of kidney donation? Why does City B have such low levels of voter turnout? And what can these outliers tell us about the influence of cities on various prosocial metrics? By examining the data, we might be able to better understand the levers of public policy, those norms and laws that make us kinder, gentler creatures. Because when it comes to encouraging prosocial behavior, it’s often the little things that count