One of the most controversial posts I wrote this year was the one where I doubted that work from home will become normal and people will hardly ever come back to their offices and meet in person.
One of the interesting observations was that younger people were much more enthusiastic about working from home and didn’t think that personal encounters and interactions are that important. Older people, on the other hand, valued in person meetings much more. It could be that we are gradually shifting toward more work from home with younger people adopting a different style of work, but as one of my readers said: “If they want to get a promotion or a raise, they better show up in the office.” Good point.
Another response that I encountered on social media was harder to address. A fellow Twitterati wondered if the assumption that chance encounters lead to more creative solutions to problems and increase productivity is correct? I stated that chance encounters with colleagues and the chitchat that comes from it leads to out-of-the-box solutions but I didn’t quote any references. Well, the good thing is that Gilles Duranton and Diego Puga have published a fantastic article in the Journal of Economic Perspectives on “The Economics of Urban Density”. It couldn’t have come at a better time because it neatly summarizes all we know about the benefits (and costs) of living in a dense urban environment. In short, the benefits come in the form of higher productivity, lower logistics costs, and higher wages, while the costs come in the form of higher rents and house prices, higher transport costs, and increased congestion and pollution.
What I want to stress here are the results for productivity and innovation. Using the results of another literature review, Duranton and Puga confirm that the elasticity of labour productivity to population density in cities is about 0.04. That may not mean much to most readers but it means that a 1% increase in population density leads on average to a 0.04% increase in productivity. For context, Chicago has an average population density of 4,582 people/km2 while New York City has an average population density of 10,431 people/km2. All else being equal, this higher density and closer proximity to other businesses and people should lead to 5% higher productivity in New York City vs. Chicago. Similarly, Paris should have a 10% higher productivity rate than a city like Grenoble, for example.
So being closer to other people makes us more productive. But it also makes us more innovative. Using patents as a measure of innovation, research has shown that a 1% increase in density leads to a 0.21% increase in accepted patents. This can explain why there are naturally evolving “innovation hubs” like Silicon Valley near Stanford University or the high concentration of pharmaceutical companies in Basel (Roche, Novartis, Lonza, Givaudan, etc.) and further down the Rhine in Ludwigshafen (BASF) and Ingelheim (Boehringer Ingelheim).
These innovation hubs have started in these places for idiosyncratic reasons like access to university talent or freshwater, yet, once established, they started to attract more and more talent and became an international hub that has distinct competitive advantages.
In recent years, the research could even become more granular and use anonymised phone records to establish who communicates with whom and how distance leads to changes in social and professional networks. Two researchers from the University in Bern used changes in the timetables of the Swiss Railway network to measure the induced changes in social interactions. They found that even small increases in distance led to a significant reduction in social interactions. And this was using data from a country where trains are famously punctual! It seems as if every minute counts and smartphones could not overcome the loss of connection from longer train rides (please also note that this study was done in 2017 so it was made in a time when smartphones and modern utility apps like slack were readily available and widely used in many companies).
But they also found something interesting. People in cities do not have larger social networks than people in less urban environments. Higher density doesn’t mean you interact with more people. It simply means it is easier to find more people who are working on similar issues as you are. The quality of interactions gets better with higher density and increased in-person interactions, not the quantity. Duranton and Puga cite an unpublished paper (that I couldn’t find on the web, unfortunately, so I have to take their word for it) that looked at the smartphone records of people who work in the same building in Silicon Valley. They isolated chance encounters where two phones were at the same place at the same time (as opposed to meetings) and found that “chance meetings result in more patent citations across firms in different sectors whose workers had met by chance more often.”
So, there you have it. The most direct proof I know of that chance encounters lead to innovation and higher productivity.