Metro Metrics

Ryan Avent’s article “One Path to Better Jobs: More Density in Cities” in the New York Times, which is a summary of his new book “The Gated City”, has caused a great deal of stirr in urbanism circles (read comments here, here, here, here, here, here and here, for example).

Avent’s basic argument is that cities facilitate interaction and idea-creation and that higher densities make for more of these interactions, which generates higher levels of economic growth. Since the overwhelming majority of the economic activity in the US, or any modern developed economy, is generated in metropolitan areas, a nation should do as much as it can to facilitate high-density living in cities. That is why both excessive urban sprawl – that spreads out the metropolitan population over a huge area – and excessive NIMBY-led building regulations that limit high-density living is bad for the economy.

Ryan Avent is right in this. A point that he could have made clearer, though, is what density means in different metros with varying levels of highly skilled residents. As Avent writes in his article, “[w]ithout a stock of skilled workers and a relatively open marketplace, density’s impact on growth and productivity will be limited”. Density is mainly beneficial if the people who work and live in high-density areas are highly skilled and educated.

As Jaison R. Abel, Ishita Dey and Todd M. Gabe, of the New York Fed, state in their research paper “Productivity and the Density of Human Capital”, metropolitan areas with a share of highly educated residents that is one standard deviation (6,4 %) below the mean (21,5 %) for all US metropolitan areas will benefit very little, or not at all, from doubling their density levels. While doubling density in a metropolitan area with a share of highly educated residents that is one standard deviation above the mean “yields productivity benefits that are about twice the average”.

This means that Detroit, as a whole, wouldn’t necessarily benefit from getting denser. But that does not mean that Detroit, and other similarly low skilled metros, could not benefit from density in the right places, i.e. if their highly skilled residents live and work in high-density parts of the metro area. QuickenLoans, and other companies that have followed in its footsteps, are therefore doing the right thing when they move their operations downtown and encourage their high-skill employees to not only work, but also live in the urban core.

Why do high skilled people both have to live and work in a high-density part of town? While it might be most important that highly skilled individuals work around other highly skilled individuals in a dense urban setting (SERC has for example showed that strictly-work-skyscrapers have economic benefits), Ed Glaeser and Matthew Kahn argued, in their paper “Decentralized Employment and the Transformation of the American City”, that “in cases where the workforce is predicted to live in suburban areas, the firms will also locate in suburban areas”.

Business suburbanized because people suburbanized, and if a city and the country is better off with increased levels of density, then you have to make sure that highly-skilled individuals also live in a dense environment – in New York and Detroit alike, since that will attract a growing number of firms to a viable, productive urban core.

Living like this is good for the economy. This picture is from The Heidelberger Papers.