Job Centralization in the City

The table below collates job centralization not by CBD as in this post but by central city. Parisian data comes from INSEE, here and here; American data comes from Wikipedia for population and OnTheMap for job counts. In general, I tried making the central city definition about 18% of the metro area to be comparable with Paris, but there is still a lot of variation, so this table should absolutely not be read as a ranking of metro areas by job centralization.

Metro area Population Jobs Central city Central pop’n Central jobs Central job share
Ile-de-France 12,082,144 5,682,048 Paris 2,206,488 1,797,745 31.6%
New York 19,979,477 8,364,410 Manhattan, Brooklyn 4,313,498 2,905,675 34.7%
Los Angeles 13,291,486 5,372,008 Downtown LA to Santa Monica ~1,500,000 1,051,648 19.6%
Chicago 9,498,716 4,142,542 Chicago ex-O’Hare 2,705,994 1,198,562 28.9%
Dallas 7,539,711 3,146,973 Dallas 1,345,047 809,077 25.7%
Houston 6,997,384 2,791,647 Inside 610 + Uptown ~650,000 749,661 26.9%
Washington 6,249,950 2,717,790 District, Arlington, Alexandria 1,100,496 859,751 31.6%
Miami 6,198,782 2,308,048 Miami, Miami Beach 563,221 324,260 14%
Philadelphia 6,096,372 2,570,460 Philadelphia 1,584,138 628,423 24.4%
Atlanta 5,949,951 2,374,233 Fulton County 1,050,114 780,259 32.9%
Boston 4,875,390 2,381,555 Boston, Cambridge 805,234 687,237 28.9%
Bay Area 4,729,484 2,121,580 San Francisco 883,305 642,375 30.3%

The Sunbelt

There appears to be a fair amount of job centralization in the Sunbelt cities, right? In Metro Atlanta, Fulton County has a slightly higher proportion of regional jobs than Paris with a slightly lower share of residential population.

But actually, no. Absolute densities matter in addition to relative centralization of jobs versus residences. In Houston and Los Angeles the central areas are drawn to encircle the downtown and near-downtown job centers – both cities preferentially annex suburban job sites so using municipal boundaries is not useful. A hefty share of area jobs are in these centers, especially in Houston. But ultimately it’s still not a lot of jobs in a very large land area, around 300 square kilometers for both, compared with 100 for the city of Paris or for San Francisco. Fulton County is vast, and the jobs are distributed all over Atlanta and its northern suburbs within the county.

Houston is a particularly good example of monocentrism with a weak center. There are not a lot of strong suburban job centers in Houston – nothing like Silicon Valley, Downtown Newark, the Route 128 corridor, La Defense, Burbank, or Tysons Corner. The city itself has about two thirds of area jobs, thanks to selective annexations. But the share of the CBD in area jobs is low, just 150,000 jobs in the 45/69/10 beltway, or 5.3% of area jobs. Outside the CBD job density plummets, as the outlying job centers making the difference between 5.3% and 26.9% are located at haphazard locations all over 610.

Older American cities

The extent of centralization in the Northeast, Chicago, and San Francisco is greater. New York in particular is a lot like Paris, with about a third of area jobs in a high-density contiguous blob consisting of less than one fifth of regional population. It has nothing like La Defense in the suburbs, but its suburban job centers, while much smaller, include some recognizably dense ones, especially Newark and the Jersey City waterfront. One needs to go well into suburbia to see the difference between Paris, where the suburbs have a structure of density with mid- and high-rise residential development as well as offices next to train stations, and New York, where the job centers in farther-out suburbia, like Central Jersey, have no such structure and are located exclusively based on auto access.

Boston, Washington, and San Francisco all have varying degrees of centralization. I mentioned last year that Boston is increasingly an example of European-style job sprawl, in which jobs spill over from the CBD to nearby areas rather than to faraway office parks. New York has long had such spillover – Long Island City is such a job center, and may at this point have more jobs than Downtown Brooklyn; the Jersey City waterfront is another such example, as is the growth of the Meatpacking District around Google. In Boston the equivalents are Kendall Square and the Seaport; in San Francisco it’s SoMa; in Washington it’s jobs in Arlington around the Orange Line, where older TOD was residential.

Chicago and Philadelphia are the least centralized. Chicago has a well-defined supertall skyline with about 500,000 people working in or near the Loop. But outside that central area, job density craters. Chicago’s share of metro area jobs is about 1.5% higher than its share of metro population, and if we remove the airport, surrounded by suburbia, this difference drops to 0.5%. Philadelphia’s share of metro area jobs is actually lower than its share of metro area population by 1.5%. In these regions, if you’re not working in city center, you’re working at an office park in a middle-class-to-rich suburb built without regard for the area’s vast legacy mainline rail network.

26 comments

  1. IAN! Mitchell

    Surprising insights on Philly and Chicago.

    Houston’s not so surprising, Williams Tower is the tallest building in the world outside of a CBD.

    • Alon Levy

      Yeah, I did not expect to see Houston so monocentric-without-a-strong-center, but on second thought this seems right (cf. LA and San Diego, which are vaguely polycentric-with-weak-centers).

      • Jacob Manaker

        To my (very untrained) eye, it seems as though the Galleria area (where, e.g. the Williams tower is located) is a developing secondary CBD. Is that not the case?

    • jonahbliss

      RE Williams Tower. That’s the tallest in the US not in a CBD. But internationally, lots of other taller non-CBD towers. For example, Hong Kong’s tallest building (The ICC, 1588 feet) is in Kowloon, not Hong Kong Island.

      • michaelrjames

        That example of Hong Kong is rather atypical. The ICC is mixed-use with the top part being hotel, and the rest of the Union Square site/West Kowloon Reclamation, of which it is part, is largely residential eg. the 270m Cullinan tower. So I doubt it would rate very highly in terms of jobs, almost certainly a lot less than old Kowloon to its east. Also it sits on top of the MTR station and the Western Harbour road tunnel–very quick one or two stop connections to Central which is less than 1.5km away. So, even though there is water between them, it’s more like an extension to Central. By comparison Canary Wharf is 4km from the City, and La Defense is 6.2km from the old CBD in the 9th arrondissement.

        Not relevant to this discussion but in the context of other discussions on this blog, I can’t resist mentioning that the ICC (the 4th tallest building in the world) was jointly developed and is co-owned by MTRC (ie. majority public-owned Metro organisation) and Sun Hung Kai Properties.

      • Lee Cryer

        I’d add the tallest building in the world, Burj Khalifa in Dubai, to this list.

  2. Matt da Silva

    Philadelphia’s job sprawl problem is pretty bad, and is actually creating some very bad traffic problems now, as many young professionals prefer living in city neighborhoods but need to auto-commute to their jobs in suburban office parks. I-76 between Philly and King of Prussia/Conshohocken is particularly jammed up. Employers are wary of the city’s wage tax, and Philadelphia’s construction costs are too expensive given its market rents, which has scared office developers away to the burbs despite workers’ residential preference for the city.

    Have you read Edgeless Cities, the follow-up to Edge City that re-examines the claims made in the former book? They have some great info on Philly’s job sprawl problem. Many of Philly’s emerging “edge cities” wound up filling in with sprawl instead of building density.

      • michaelrjames

        In his Edgeless Cities, Robert Lang used commercial space to do this kind of analysis. Your analysis is similar to his Table 14-7, Typology of Metropolitan Areas 1999. Using the “percent difference between primary downtowns and edgeless cities” he divided American cities into 4 classes. I combined your table with his (last columns from both; ordered on Lang’s) as below (crosses fingers for legibility):

        Metro area……..Central job share….Diff.[core-edgeless]
        Ile-de-France……….31.60%…………………….[?]

        Core dominated:
        New York…………….34.70%…………………….26.80%
        Chicago………………..28.90%…………………….27.30%

        Balanced:
        Boston…………………28.90%…………………..-1.80%
        Washington………….31.60%…………………..-3.20%
        Los Angeles…………..19.60%…………………..-7.20%
        Bay Area………………30.30%…………………..-9.50%

        Dispersed:
        Dallas…………………..25.70%………………….-14.10%
        Houston……………….26.90%………………….-16.10%
        Atlanta…………………32.90%………………….-17.60%

        Edgeless:
        Philadelphia………….24.40%…………………-19.40%
        Miami……………………14%……………………….-52.7%

        • Alon Levy

          Yeah, so I’m not sure Paris has any recognizable edgeless cities. Suburban employment is too structured around RER stations, so the places that are too far out to be CBDs (i.e. not La Defense) are firmly edge cities rather than edgeless cities, like Cergy and MLV or various out-of-the-way unis like IHES.

          • michaelrjames

            Right, though that is residential zones. There are some sprawled industrial zones along the river to the north and south, and east along the Marne. Assuming area is a fair proxy for employment. But though they can be big, they are never far from those significant residential centres which are, as you say, dense compared to American suburban sprawl.
            Also, given Lang’s analysis is 20 years old this year, it would be good to see the same analysis updated. His prediction was that this was America’s future city form. I haven’t checked to see if there are accessible updates on the Brookings Institute site.

    • Lee Cryer

      It’s timely because it adds a new(?) metric “perceived job density” and its accompanying data set has values for this metric for several metro areas over time.

  3. RossB

    As you mentioned, the chart is misleading because cities are not all drawn the same. If they were all about the same size (physically) and radiated equally from the center, then the chart would be great. But they don’t, which is why there are a lot of exceptions (both in designating the central city in the chart, as well the text).

    I wonder if it would be more meaningful is you simply measured the distance from the center of the central business district. That introduces other issues, but I think it works fairly well. Water barriers would appear to be an issue, but they seem to be OK for the most part. For example, North Vancouver is pretty close to downtown Vancouver, yet not adjacent to it. Likewise with Brooklyn and Jersey City to Manhattan. Downtown Oakland, on the other hand, is a bit farther away, but still not as far as Silicon Valley. I would say that Downtown Bellevue is a bit close to Seattle (as the crow flies) but that is because you can’t get to Seattle that way. Downtown Bellevue is physically close, but farther away from a practical standpoint. I am sure there are other anomalies like that. That aside, I think such a measurement would work out fairly well. My guess is the sunbelt cities would represent their true selves. This is something you might want to play around with (if you feel like it).

    As far as your main points, I’ve noticed all the trends in Seattle. In the 80s you could see the emergence of office parks in the suburbs, the most famous of which was in Redmond (headquarters to Microsoft). Unlike Silicon Valley (which was influenced heavily by Stanford) there really was nothing over in Redmond prior to Microsoft. Bellevue was a bigger suburb, but it was clearly just a residential suburb, with very few office buildings. Office parks popped up all over, as they have in Chicago (but not to the same degree). But rather than just continue to sprawl, office buildings east of Lake Washington have coalesced largely in Redmond and downtown Bellevue. Even in suburbia there has been a trend towards moving towards the center.

    But Seattle itself has seen the biggest growth. The downtown area has grown to the north, to include South Lake Union, an area that previously included cheap motels and warehouses. Businesses have located in neighborhoods that are part of the city, but not connected to downtown. The University of Washington has had a big influence on all of this (and that area itself is seeing increased office development). This trend — towards an expanding downtown as well as office development in neighborhoods that are disconnected from downtown, largely residential, but relatively central — appears to be in full swing here. Perhaps the most interesting move is Weyerhaeuser, a gigantic timber company. They moved from their suburban “campus” (https://www.kuow.org/stories/dreamers-pick-over-the-bones-of-weyerhaeuser-s-old-campus) right into the center of old Seattle (part of downtown).

    • Alon Levy

      After posting I poked around with 100 km^2 blobs around city center, same size as Paris. San Francisco holds up; New York and Boston go down by a little, New York becoming about comparable to Paris rather than a few percentage points more centralized; the other cities go down by a lot, the Sunbelt ones dipping to the low teens.

      • adirondacker12800

        Sumptin isn’t right and since I have no idea of what a blob 100 sqaure kilometers is…. sumptin is wrong. San Francisco’s metro area has a population roughly the size of Brooklyn and Queens. The combined statistical area of San Francisco and San Jose has the population of New York City and a suburban county or two. They are going to have roughly the same percentage of the population working …. sumptin is wrong. More people commute into Manhattan than live in San Francisco. BART the county’s fifth busiest mass transit system carries 416.000 fares on an average weekday. PATH is the country’s seventh busiest, tiny obscure itty bitty little PATH that only forsaken New Jerseyans use, carries 310,000 a day. Sumptin is wrong.

        • michaelrjames

          a blob 100 sqaure kilometers is…

          Paris (intramuros): 109km2, excluding the two big bois: 89km2
          Manhattan: 87km2 (everything)
          San Francisco City: 122km2.
          Staten Island NYC: 151km2

          The combined statistical area of San Francisco and San Jose has the population of New York City and a suburban county or two.

          That includes the entire Bay Area, including Oakland and everything on that side of the bay, so, at minimum, that is three proper cities each with their own CBDs. The peninsula (SF to SJ) would be ca. half that with plenty of edgeless city (with high employment) in between.

          Wiki:
          San Jose:
          • Urban 1,894,389 (29th)
          • Metro 1,998,463 (34th)
          • CSA 8,837,789 (5th)

          • adirondacker12800

            That’s very nice. I know how to ask Google to find Wikipedia pages. That doesn’t tell me what his blob is. Metro New York includes Newark and Jersey City which are served by PATH which has three quarters of the ridership of BART which someday soon will serve the whole CSA. 22 minute ride from the World Trade Center to Penn Station Newark. Or 20 minutes from Penn Station Newark to Penn Station New York on NJTransit. More people use the commuter railroads to get to Manhattan than live in San Francisco. Medium demand subway stations in the outer boroughs have more ridership than Golden Gate Transit. More people use the Times Square or Grand Central subway stations than use BART. More people live in New Jersey than live in the San Francisco CSA. Which is between Manhattan and the western edge of it’s CSA. There is something wrong the picture.

          • Alon Levy

            Yeah, so for the blob I tried doing something that was around 100 km^2 land area, vaguely centered on the CBD. I allowed for some gerrymandering to make sure I got certain job centers. In New York this is the Manhattan core plus the southern parts of the UES and UWS (enough to get Weill-Cornell, not enough to get Mount Sinai), LIC, enough of inner Brooklyn to get Downtown Brooklyn, and the Jersey City and Hoboken waterfronts. All in all this is 2.7 million jobs.

            In Boston it’s basically Boston and Cambridge, but with a bit of Brookline and Somerville thrown in and the southern parts of Boston like Hyde Park and Mattapan kept out. In Chicago I forget the exact boundaries but they don’t really matter – Hyde Park is too far to be included and the job density outside the central ~10 km^2 is so low that it doesn’t really matter how you set up your 10*10 square. In SF it’s either the entire city (minus some outer areas with few jobs, eh) or the inner 50 km^2 of it and a 50 km^2 blob around Downtown Oakland including Berkeley, which have around 700,000 jobs combined. In Houston it’s a blob designed to hit both Downtown Houston and Uptown. In LA it hits both Downtown and Century City.

            Yes, absolute size matters, so the central 100 km^2 should, at the same level of centralization, have a higher job share in SF-ex-S than in NY. But then in LA the job share of the central 100 km^2 is maybe 13%, same as Dallas and Houston. If I had centralization data for Berlin (e.g. inside the Ring) and for Munich I’d include it as a sanity check, but I unfortunately don’t.

    • RossB

      I was thinking about this some more. I think it would be really interesting to see a graph like so: On the Y axis is percentage of employment. On the X axis is distance from the core. You would have various colors for each city (just like this graph — https://pedestrianobservations.files.wordpress.com/2018/11/densitygraph.jpg). You might have to play around with the X axis, so you could show more detail close to the CBD (the difference between 1 and 5 KM is more interesting than 60 and 65 KM).

      Cities that have a very strong CBD will start out pretty high. If the downtown area spreads outward, then you will see a continuous increase. If not, then you might see a flat line for a while, followed by a jump (especially for secondary employment areas).

      I think the MSA data would be more interesting and representative than the CSA data. That means, for example, that “San Fransisco” includes Oakland, but not San Jose (http://en.wikipedia.org/wiki/File:San_Jose-San_Francisco-Oakland_CSA.svg) while New York City includes Newark, but not Trenton. I suppose there is nothing wrong with putting out one that includes CSA data, although it seems like all it would do is show that CSA areas are really big (from a distance perspective), and sometimes include large, largely independent cities (by their very definition).

  4. Henry

    Hah classic Alon. “Here are some painstakingly collected numbers…but they don’t show what I like so ignore them”

    🙂

  5. kingledion

    I think you need better data. Paris can be approximated as a circle with a 6 km radius. So why not draw a 6 km radius around the center of major US cities and see what is in them? Turns out I have done this, using python geo-tools and Monte Carlo sampling at the zip code level; using population and jobs data from the US Census. Here are the results of a 6 km radius around major US cities. Note, given the size of the 6km radius and the sprawl of several US cities, you can have multiple. The best example of this is Downtown LA, Burbank, and Century City; and Santa Ana and Irvine in Orange County California. The sampling error is roughly +- 5%; different between cities based on the shape sizes of the local geographies.

    New York………….1585000….2273000
    Chicago……………..504000……790000
    Boston……………….567000……659000
    San Francisco…….569000……566900
    Los Angeles……….656000……363000
    Philadelphia……….609000……393000
    Washington DC…..460000……515000
    Century City……….399000……378000
    Burbank…………….258000……406000
    Santa Ana………….445000……198000
    Minneapolis……….298000……297000
    San Jose…………..375000……209000
    Seattle………………249000……334000
    Las Vegas…………210000……347000
    Irvine………………..291000……251000
    Baltimore………….323000…….218000
    Denver……………..285000……255000
    Miami……………….322000……186000
    Pittsburgh…………227000……268000
    Houston……………235000……246000

    Dallas………………196371……223000
    Atlanta……………..199482……199000

    An important note is that government jobs are not counted in the data set, so DC in particular is heavily under counted on the jobs front; cities that are state capitals (Boston, Denver, Atlanta) or have a very large state university (UCLA in Century City; Minnesota, Georgia Tech etc.) probably get a relatively greater jobs boost than those that are not. Still, it is evident that the Sunbelt cities are barely cities at all in the traditional sense. Dallas and Atlanta’s “cores” are out of the top 20, sandwiched between Providence and New Orleans.

    Last note, this is distance based, not area based, so not a fully fair comparison. NY, Chicago, Boston, SF, Seattle and Miami have about 75% the land area of other cities within a 6 km radius.

    If you are interested in more information, or how it is generated; it is all replicable, let me know; kingledion (at) gmail . com

    • Alon Levy

      I played with Paris-size blobs (100 km^2), trying to gerrymander to include job centers when possible, and the results I came up with are pretty similar to yours. The Sunbelt is extremely dispersed; New York, Boston, and San Francisco are roughly comparable to Paris.

Leave a Reply to kingledion Cancel reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.