Emily Badger is a former staff writer at CityLab. Her work has previously appeared in Pacific Standard, GOOD, The Christian Science Monitor, and The New York Times. She lives in the Washington, D.C. area.
And counting all of the people (and jobs) who have supposedly moved there.
Last year, the U.S. Census Bureau released a report on population trends in American downtowns, a helpful step toward quantifying the claims made by many cities that residents (and jobs) are moving there in droves (you can view the original report here... whenever the federal government reopens and the Census Bureau's shuttered website comes back online). The Census' blunt definition of "downtown," though, inevitably produced some grousing about over-and under-counts of local populations. It measured “downtown,” for lack of a better universal definition, as everything within a 2-mile radius of the local city hall.
In Baltimore (at left) and New York City (at right), here is what the resulting circles look like:
You can see that in these two cases we're talking about an awful lot of water, not to mention some largely neglected neighborhoods in Baltimore. And as anyone in New York will quickly point out, this definition of "downtown" in Manhattan awkwardly includes a slice of New Jersey.
It's a little hard to blame the Census. There actually is no single definition of what "downtown" means across the country. Nowhere do the Census Bureau, the Bureau of Economic Analysis or the Bureau of Labor Statistics actually count or keep tabs on the number of jobs in American "downtowns."
All of which complicates the efforts of business improvement districts and city officials to back up what's supposed to be one of the great urban success stories of the 21st century (and a big topic at The Atlantic's CityLab summit on urban innovation in New York this week): the migration of employers and residents back downtown, redefining these places from dead zones where you one wouldn't want to be caught at night to full-service, 24-hour neighborhoods.
With this perennial problem in mind, the Center City Philadelphia business improvement district is releasing a report today prepared for the International Downtown Association that tries to offer a new way of counting jobs downtown, measuring where the people who hold them live, and enabling comparisons across cities.
The authors, Paul R. Levy and Lauren M. Gilchrist, rely on a relatively new Local Employment Dynamics dataset jointly produced by the Census Bureau and state labor market information agencies. With that tool, it's possible to create heat maps of job density, and then outline the irregularly shaped districts around them. The data also includes information on the home and work locations of employees, making it possible to determine which downtowns actually have people in them at night.
The heat maps make clear that many cities don't actually have a single downtown employment center. Seattle is an example of a city that largely does:
Cleveland has a downtown and a separate node around the "anchor institutions" near the Cleveland Clinic:
Atlanta (left) has multiple roughly equal jobs centers, while Jacksonville (at right) is decentralized with no single jobs center to speak of:
To take this data one step further, here is Baltimore's employment density, alongside a zoomed-in map of the resulting "downtown" employment node, with a one-mile walking radius drawn around it:
With this method, Levy and Gilchrist counted 231 major employment centers in America's 150 largest cities, collectively containing 14.4 percent of all of the country's jobs. Because it's now possible to compare them along various metrics, 28 of those jobs centers have more than 100 jobs per acre (the national average is 0.05 jobs per acre).
Impressively, 52.3 percent of workers who live in Chicago's downtown actually work there (in Midtown Manhattan, it's 48.2 percent). The accompanying database also counts workers living within a half mile or a mile of these jobs. If you're feeling really wonky, you can dig into all of the data and maps on these 150 U.S. cities here. Surely some people will come up finer revisions of this definition (yes, that Baltimore map above right still contains some water, and these one-mile radii don't take into account how highways get in the way of walking to work). But, for now, the technique is an improvement from measuring how crows fly from City Hall.