Spillovers: Knowledge, Beer, and Technology

Why Technology Hasn't Yet Overcome Geography

In my last article, I wrote about how an economically and culturally vital city is able, without central planning, to generate two things that are essential to the city’s success: diversity and cohesion. I argued that when lots of people who reflect a huge range of skills and tastes can live together peacefully (diversity) and when they feel free to associate to any degree they choose (cohesion), the result tends to be rich, complex, and dynamic social networks that channel information from person to person rapidly and effectively. For many who hear this argument, however, how this happens remains a mystery. And I don’t blame them for feeling that way.

Some urban economists and urban sociologists explain this in terms of “knowledge spillovers.” Proximity enables people who are socially distant to mingle and exchange ideas; the image of chance meetings and chatting up over coffee or a beer comes to mind. Sometimes it does happen that way. Many of the innovators in Silicon Valley in the 1960s, for example, informally traded ideas at the famed Wagon Wheel bar. The term “spillover” evokes this informality over the occasional sloshed mug of beer (or sloshed patron).


Social Media:  Complement or Substitute?

I listened to an interview recently with urban economist Edward Glaeser of Harvard University on Russ Roberts’ podcast, Econtalk. The main topic was the recent declaration of bankruptcy by the Detroit city government, but the discussion also touched on more basic issues such as why people still need to live in cities.

Some of my colleagues argue, for example, that face-to-face interactions are becoming obsolete with technical advances such as Facebook, Skype, and email. Back in the early twentieth century, as the telephone spread to more and more homes, pundits at the time—among them the architect Frank Lloyd Wright—claimed that science would “obliterate distance” to such an extent that densely populated cities would become a thing of the past. (Since then, of course, the number of cities just in the United States with a population of over one million has grown from five in 1950 to nine in 2012.)

And recently one of my Facebook friends told me that at a conference, one of speakers asserted that the Internet has “overcome geography.” Now I wouldn’t dispute this assertion if “overcome” here means merely allowing more interactions between acquaintances and maintaining relationships over great distances, in the vast majority of cases complementing personal relationships. But if it means that the Internet can eventually substitute for large cities, and that relationships over the Internet have pretty much eliminated the need for face-to-face contact, then I strongly disagree.

Before you start telling me about your 3,000+ Facebook friends, let me say two things:

First, people who have rigorously studied social media, such as Nicholas Christakis and James Fowler, point out that “online networks … do not appear to expand the number of people with whom we feel truly close, nor do they necessarily enhance our relationships within our core groups.” Second, the anthropologist Robin Dunbar, after studying various cultures around the world, discovered that the number of people with whom a person can maintain a stable, social relationship is surprisingly constant: about 150. That includes relationships of all kinds. And when Christakis and Fowler examined Facebook, they found that not only did each user have about 110 friends on average, which conforms to Dunbar’s observation, but that “on average just 6.6 were close friends.” These are people who you tend to see regularly face-to-face. And those of you who have thousands of Facebook friends? Well, as the saying goes, some people make friends, other people collect them.


What Does This Have to Do with Cities and Spillovers?

I think the Wagon Wheel model of knowledge spillovers, while true of certain industries, is just one example of how diversity and cohesion spur entrepreneurial discovery and economic development. As Russ Roberts notes toward the end of the interview with Edward Glaeser, people outside the knowledge industries—such as high tech, academia, and the arts—don’t usually exchange information so freely. In fact, they probably like to keep things pretty secret. I agree. But the point is that some important kinds of secrets are harder to keep in the big city.

So it would be helpful to think of “knowledge spillovers” as including things like “noticing what others are doing.” Again, “spillover” suggests that the knowledge is being conveyed accidentally. Someone starting a new business doesn’t want competitors to know what she’s doing, just as Oscar de la Renta wouldn’t want to tip his hand to Donna Karan. But the density of big cities like New York or London makes three things possible.

First, because there’s so much going on, on the streets and in the theaters, it’s much easier to notice what other people are doing, often whether we want to or not. But if someone in a competing or a complementary line of activity does something new, that newness gets diffused through a city (via social networks or by sheer accident) very quickly. By the same token, the possibility of our new ideas leaking out to competitors is also higher. All of which fuels competition.

Second, the cohesiveness of a big city, its “ability-to-get-together-with-socially-distant-peopleness,” means making those novel connections and then following through on them is also much easier. Those 6.6 people we frequently meet in a city are more likely to have connections with more diverse and relevant talent than if they were in a smaller town.

Third, cities, especially big ones, don’t just have a wide range of knowledge and skills on the supply side. On the demand side, there’s a much broader range of tastes and an ethos of greater tolerance and experimentation than you’re likely to find elsewhere. Of course, you can sell globally via eBay or online stores. But with a few exceptions those places don’t start global. Rather, they tend to start small and cater to local demand (even though they have a website); then, if they’re successful locally, they may expand globally.

I don't think anyone could reasonably contest that technology has reduced the need for some kinds of proximity. It's the tired hyperbole that "science obliterates distance" and the like that gets my goat. The gulf between "reducing" and "eliminating" is too vast.