Cities With Denser Cores Do Better
Density — the close clustering of people together in communities — is a big factor in the technological and economic progress of cities and nations. Economists, urbanists, and place makers have found density to be associated with everything from greater energy efficiency to higher levels of skilled and talented people, higher rates of innovation, and higher income.
Most studies of the effects of density measure it rather crudely, as I noted recently, and simply average out the number of people per unit of land area. By this standard measure, Los Angeles is denser than New York. But density is far, far more concentrated in the center of Manhattan than anywhere in L.A.
That raises a big question: Do cities with more concentrated density — where people and economic activity are concentrated and spiky near the center — see even better economic performance?
Recent data from the U.S. Census enable us to take a look, providing a measure of density that is based on the actual distribution of people within and across cities and metro areas. This measure of 'population-weighted density' is based on the average density of the small local areas — the Census tracts — that comprise metros. The Census Bureau also measures the density of cities and metros at the one-mile and five-mile radii of the city center.
With the help of my Martin Prosperity Institute colleague Charlotta Mellander, I looked at how these different types of density are associated with a wide range of regional economic outcomes. Mellander ran simple correlations between these density measures and variables like regional income, wages, economic output, college grads, the creative class, high tech industry, home prices, commuting patterns, and even happiness. The table below summarizes the key results. As usual, I note that correlation does not equal causation. Still, the findings are interesting and help shed light on the role of density — and especially cities where population is more densely concentrated in and around their cores — on regional economic development.
The first thing one notices is that metros which rank highly on the average measure of density — those that average a lot of people per square mile — are not necessarily those where people and economic activity are most concentrated near the center, as the scatter-graph below shows. While average density and population-weighted density are correlated (.46), they are clearly not the same thing.
Not surprisingly, large metros are the ones that tend to have density more concentrated near their cores. The correlation between population-weighted density and population is .64.
Ever since Jane Jacobs, urban thinkers and economists have argued that clusters of talented and ambitious people increase one another’s productivity and the productivity of the broader community, spurring economic growth. So, what about economic growth: Is it higher in metros where density is more concentrated? The short answer is yes.
Economic growth and development, according to several key measures, is higher in metros that are not just dense, but where density is more concentrated. This is true for productivity, measured as economic output per person, as well as both income and wages.
Talent levels are also higher where density is more concentrated. This holds for both the share of college grads and the share of knowledge, professional, and creative workers. Conversely, working class jobs are more likely to be found in metros that are less densely concentrated.
The same pattern holds when we look at underlying skill. Studies find that returns to physical skill are declining while those to cognitive and social intelligence skill — the two skills that underpin knowledge work — are rising. Physical skill is negatively associated with density, while cognitive and social intelligence skills are positively associated with it.
What about specific occupations: Are some more connected to density than others? We find that three types of occupations — arts, media and entertainment, science and technology, and business and management are strongly associated with density, and especially in metros that are more densely concentrated at their cores. Although much has been made of "meds and eds" as a source of revival in lagging cites and urban areas, we find no significant statistical association between them and any measure of density.
Despite the prevailing wisdom that high-tech industries tend to cluster in sprawling suburban "nerdistans," high-tech industry tends to follow density as well, according to our analysis.
Diversity is more likely in places where density is more concentrated. This is true of immigrants and foreign-born people (where the correlation for population-weighted density is more than double that for average density) as well as for gays and lesbians. It might be the case that diverse groups of people are simply attracted to denser metros rather than being instrumental in causing them to be that way, but diversity and density are clearly related.
Density also plays a role in how we commute to work. A study [PDF] that Gary Barnes conducted for the Minnesota Department of Transportation in 2001 found close associations between population-weighted density and the way people commute. Our analysis supports his findings. There are many more bike commuters and far less people who drive alone in metros where density is more concentrated. The share of bike commuters is positively associated with population-weighted density and negatively associated with the share who drive to work alone.
Housing prices are higher in places that are denser and spikier at the center. This is true of both median housing prices and the share of income people devote to housing. This may be the consequence of several factors: the higher incomes and wages that denser cities and metros offer, the greater demand for living closer to the center in such places, or restrictions that keep housing in short supply — natural ones like coastlines or mountain ranges, or those imposed by polices, like height limits, strict building codes, landmark designations, and zoning ordinances.
There is a long tradition in America dating back at least to Thomas Jefferson that suggests people will be happier in more pastoral, less congested settings with their own space. There is also a long line of thinking that sees crowded, dense urban centers as the source of anxiety, agitation, unhappiness, even pathology. But that is not at all what we find. Instead, happiness levels are modestly associated with density that is more concentrated at the center of the city.
Density is a key factor in both the growth of cities, the happiness of cities, and the wealth of nations. And cities and regions where density is more concentrated near their urban cores — appear to gain the biggest economic advantage. I'll be writing more on this in future posts.