Size isn’t everything

This post was prompted by a very interesting article by Jonah Lehrer in which he brings together two recent papers on how a city’s population size affects the lives of its inhabitants. In the first, theoretical physicists Geoffrey West and Luis Bettencourt demonstrate that the inhabitants of a city will, on average, produce and enjoy around 15% more of almost everything – from the trademarks and patents they register, to the local restaurants they can visit, to the income they earn – than the inhabitants of another city that is half its size. They also show that it isn’t just the nice things in life that scale in this way: negative variables from murder to bedbugs also increase exponentially with population size. This isn’t a surprise in itself: as Lehrer says, city dwellers understand and accept this trade-off between the good and the bad aspects of urban life. What we didn’t necessarily appreciate is the high level of consistency between and predictability of these variations.

The second paper is by Samuel Arbesman and Nicholas Christakis of Harvard Medical School. These researchers looked at the incidence of prosocial behaviour, such as organ donation, voting and helping strangers, and how this varies with city size. In contrast to West and Bettencourt’s strikingly consistent results, they found no one statistical rule to describe variation in all types of prosocial behaviour; instead, different activities increased or decreased at different rates as city size changed.

This contrast is interesting. If people’s behaviour and experience in cities is predictable in so many ways – registering more patents and doing more shoplifting; having access to more restaurants and encountering more bedbugs – why should the same not apply to helping strangers on the street and giving money to political campaigns? What makes prosocial behaviour different and less predictable than other forms of behaviour?

One answer might be that in general the predictable factors of urban life both support and suppress a city dweller’s propensity to behave prosocially, and that more specifically an individual’s inclination to carry out a particular type of prosocial behaviour is influenced by his or her reaction to the combination of the factors he or she encounters.

Engaging in prosocial behaviour requires the opportunity and desire to cooperate with and consider the needs of others. West and Bettencourt’s research shows that larger cities certainly provide us with a greater desire and more opportunities to cooperate – for our own ends, at least. But it also shows that living in a city strengthens forces that prevent us from engaging with and caring about others, especially strangers.

Different people will react to these competing forces in different ways. A higher reported murder rate will induce varying levels of fear of crime in a local population, which in turn will deter some from going out of their way to help strangers more than others. A more cooperative environment will encourage varying levels of political discussion and debate, which will make some more interested and involved in politics than others. And that more cooperative environment may trump a fear of crime and encourage engagement with strangers, or a fear of crime may increase cynicism and disengagement from politics – or the opposite may be true.

All this would mean that, for a city as a whole, while the strength of productive and disruptive forces might be closely and broadly related to the size of the population, the combined influence of these forces on different types of prosocial behaviour will be much less consistent and predictable. Perhaps that’s why measuring and encouraging prosocial behaviour is so difficult, and why it needs more attention. It is important, after all, even if there is no convenient magic rule to describe it.