Richard Florida is a co-founder and editor at large of CityLab and a senior editor at The Atlantic. He is a university professor in the University of Toronto’s School of Cities and Rotman School of Management, and a distinguished fellow at New York University’s Schack Institute of Real Estate and visiting fellow at Florida International University.
Does the number of governments in a given metro area really matter?
From the moment I began studying urbanism, I’ve heard urbanists, economists and city officials lament the so-called “fragmentation” of city government. The basic line is that our cities and metro areas are overrun with conflicting, competing, inefficient and redundant organizations. I once heard a business leader in a famously fragmented city say that the solution to his region’s proliferating governmental bodies was simple: Create a new “mother of all organizations” and fold the region’s many agencies into it. The answer to fragmentation, he and many others have argued, is amalgamation, the act of creating newer, bigger and more integrated models and mechanisms for urban governance.
Writing in the Guardian, Aaron Renn cites recent studies by the OECD and the World Bank that conclude fragmentation contributes to urban inefficiency and economic sluggishness. But while fragmentation has been blamed for a host of urban ills, from mounting fiscal problems to last summer’s tragic events in Ferguson, Missouri, Renn points out that in Toronto, it was the supposed opposite of fragmentation—amalgamation—that led to the election of dysfunctional former mayor Rob Ford.
Until now, many of these arguments have been mostly theoretical. But a just-released study by political scientists Rebecca Hendrick and Yu Shi of the University of Illinois at Chicago provides hard data on just how fragmented U.S. metro areas really are. The study outlines several key metrics of municipal fragmentation for America’s 51 largest metros (those with populations of more than one million). Their resulting “Fragmentation Index” measures the number and types of local governments in a given metro area on a per capita bases, by land area, and by other factors that reflect the distribution of population between a city and its suburbs. For the statistically minded, the resulting index values are calculated as z-scores, where positive values reflect higher levels of fragmentation, and lower and negative values indicate lower levels.
The tables below list the top and bottom ten U.S. metros on this Fragmentation Index. Chicago has the highest level of fragmentation, followed by Pittsburgh, St. Louis, New York and Philadelphia.”[F]or the most part,” the researchers write, “fragmented regions tend to be older, industrial regions in the Midwest and eastern portions of the United States.”
Top and Bottom 10 Large U.S. Metros on the Fragmentation Index
*Measured as the sum of z-scores
But what lies behind political fragmentation? Are some metros more prone to fragmentation than others? Why or why not? And, given that so many people seem to think fragmentation is bad for urban economies, to what degree is fragmentation really associated with economic performance?
To get at this, my Martin Prosperity Institute colleague Charlotta Mellander ran correlations between the Fragmentation Index and a series of key economic, social and demographic variables. The usual caveat applies: Correlation does not indicate causation, but merely points to associations between variables. Still, our results are interesting and confound the conventional wisdom on fragmentation in several ways.
Fragmentation goes along with size, as the scatterplot above above shows—bigger, denser metros are more fragmented. Fragmentation is positively correlated with population size (.41) and density (.36). This makes intuitive sense: Bigger metros will have more opportunity and more demand for public services.
Fragmentation also goes together with liberalism. The Fragmentation Index is positively associated with the share of votes for Obama in 2012 (.24), and negatively with the share of votes for Romney (-.24). This too makes intuitive sense. More liberal places also demand more government. Additionally, many more conservative metros in the South developed later, so they often have more unified structures and fewer municipalities.
One would think fragmentation would go along with the growing sorting and separation of Americans by race and class. Indeed, the Fragmentation Index is positively correlated with Hendrick and Shi’s own Population Sorting Index (.31), which measures the sorting of populations by education, income, poverty, race and ethnicity across municipalities and Census designated places. Fragmentation is also associated with income inequality (.25). “The correlation of population sorting and fragmentation supports the logical conclusion of [the economist Charles] Tiebout—that people segregate into fragmented governments,” Hendrick writes in an email.
This may well be driven by the desire and ability of more affluent groups to separate themselves and demand governmental agencies and services that reflect their needs, Hendrick adds. “Maybe it’s more like the wealthier sort themselves, leaving the less wealthy behind," she writes. "You can certainly see this in Chicago where, for instance, people with horses have several suburbs in the region to choose from that cater to their needs, including cross-walk buttons at horse height so people do not have to dismount.”
But we also find fragmentation differs along racial lines. It is positively associated with the share of population that is white (.36), but negatively associated with the share that is black (-.38). (We find no association between fragmentation and the Asian or Hispanic share of the population.) In other words, metros where whites make up a greater share of the population are more fragmented, while metros where blacks make up a greater share are less so.
This may seem counter-intuitive at first. Metros where blacks make up a greater share of the population tend to have experienced more white flight and thus would seem to be more likely to have a proliferation of separate suburban municipalities. But the three most fragmented metros—Chicago, St. Louis and Pittsburgh—are ones where blacks make up a small to modest share of metro population. St. Louis is 18.4 percent black, Chicago 18.2 percent, and Pittsburgh just 8.4 percent. Conversely, blacks comprise the largest share of the population in southern metros like Memphis, Atlanta and Birmingham. In fact, seven of the ten large metros with the least fragmentation—Memphis, Norfolk, Baltimore, D.C., Greensboro, Charlotte and Jacksonville—have sizable African-American populations. All seven number among the top 15 metros with the largest shares of blacks in the nation.There's also an interesting connection between fragmentation and the way people commute to work. The Fragmentation Index is positively associated with the share of commuters who walk (.43) and take public transit (.42). (These relationships are charted on the interactive scatterplots above and below.) Conversely, it is negatively associated with the share of people who drive to work alone (-.28). This likely reflects the fact that metros where more people walk or take transit are also bigger and denser, while those where people drive are smaller and more sprawling. As my University of Toronto colleague Dan Silver pointed out to me, it may also be that walkabilty is associated with tighter community bonds and stronger forms of local organization.
Finally, despite all the hemming and hawing about how fragmentation impairs economic growth and development, we find no evidence that this is the case. In fact, the Fragmentation Index is positively associated with high wages (.27) and even more so with the concentration of high-tech industry (.35).
That's the big takeaway: Political fragmentation may be inefficient and duplicative, but it does not appear to do damage to local economic growth. If anything, the reverse is true—it looks like more fragmented metros do a bit better.