This review of the book Megaregions, edited by Georgia Institute of Technology Professor Catherine L. Ross, is the second in my three part series on megaregions. I put my cards on the table in my post with initial skepticism about the usefulness of the concept. I will follow this up with a look at potential applications of megaregionalism in the Midwest.
I was very struck by the quote at the top when reading the book. How often to do you find people questioning the very validity of the topic at hand when writing a piece for a book on it? The fact that Ross brings in people who are willing to ask tough questions about megaregions is a testament to her intellectual integrity. It would have been very easy to simply glom onto a topic that shows some early stage notions of being popular in the world at large and trying to flog it for all it was worth. Indeed, Ross is known on this topic, but here she takes an opportunity to shine a light on this emerging concept to see what she might find without excessive boosterism on the subject. As she notes herself in the book, “The quality of a new idea can be judged by the possibilities it creates, especially when such possibilities stimulate new and unbounded interpretations and allow more innovative and beneficial outcomes.” I see this book as dedicated to exploring some of those possibilities and trying to collect and develop frameworks for understanding it and applying it.
The book consists of thirteen chapters, each written by different authors, exploring some aspect of the topic, including looks at Europe and Asia. I will focus primarily on the United States, but don’t want to mislead into thinking this is a US only book.
One of the key questions to answer is, just what the heck is a megaregion? There are a few definitions, but the one I thought was best came from America 2050, a project of New York’s Regional Plan Association. They describe it as “a large, connected network of metropolitan areas that are joined together by environmental, cultural, infrastructural, and functional characteristics.” In short, it is a collection of linked metro areas in a given region. There is an entire chapter in the book devoted to ways to identify and delineate megaregions. And, of course, map them. Here’s the map America 2050 created using their approach:
America 2050 doesn’t have the only map going. Richard Florida, a leading popular exponent of megaregions who wrote a paper on the subject with Tim Gulden and Charlotta Mellander called “The Rise of the Mega-Region“, used images of light emissions from the space to draw boundaries of areas that seemed continuously developed. Here’s his map:
And here is the map that is being distributed with the Ross book’s promo materials:
Think about it, there are a huge number of metro areas in the United States. There are a bit over 50 metro areas of over one million people. The Brookings Institution Metropolitan Policy Program deals with the top 100 metros in America. These numbers are simply too high to give proper attention to each.
I’ve championed the notion that there is no one size fits all urban policy and that cities need to develop unique strategies and solutions based on their unique local context (see, “The Mayor as CEO” for instance). But think about it from the standpoint of a think tank in Washington or New York, or from that of Adolfo CarriÃ³n, director of the White House Office of Urban Affairs. How do you cope with policies for 250, 100, or even 50 metros? It would be extremely difficult. But it is certainly feasible think about 10-12 megaregions. I think that’s one reason why people so much want there to be validity to the megaregion concept. It provides a very convenient intermediate level of abstraction between the large scale United States (or Europe) and the fine grained detail of individual metro areas.
Brookings did this by positing a “Great Lakes” region to help organize a portion of its thinking. And I did too. As someone who has expressed skepticism on megaregions, I’ve got to admit that my own blog is to some extent a product of that thinking. One of the keys to its success was to pick a topic scope greater than the individual city (and thus to have more than purely parochial interest) but smaller than the nation (where I likely would never have been able to gain traction amongst long established big names). The 12 one million plus metros I focus on is conveniently similar to the total number of megaregions in the US (and the 12 Florida identifies in Europe). And I’ve been able to extrapolate out lessons from them that are relevant cross-regionally, and also to a broader audience as well. The metros of the Midwest actually have a lot of diversity. The strengths, weaknessnes, challenges, and opportunities of, say, Chicago, Detroit, and Columbus are radically different. They require very different policy approaches. Nevertheless, there seems to be some benefit in thinking about them together.
So apart from any real world manifestation megaregions might have, they are an important organizational construct in creating a hierarchy in any sort of large, multi-city geography like the United States or Europe. Megaregions enable people to conceptualize and manage these complex, fine grained territories. It is applying to metro areas the same regional aggregation concept used for functions like the Federal Reserve System (12 regional fed banks) or the federal district court system (11 appellate districts). That is, megaregions are necessary purely as a level in the hierarchy, even if they prove to be a phantom level. They can be defended purely on the basis of organizational and managerial theory even if they have no other application. Indeed, the fact that people persist in trying to find applications for them despite the lack of clear cut success to date shows that at some level they intuitively understand this organizational need.
Robert E. Lang and Arthur C. Nelson had a chapter that hints at this as well, noting that the mere act of formalizing a construct by the government causes people to start paying attention to it. Their example is how the OMB created the construct of “micropolitan areas”. Clearly the idea is that if the federal government created official megaregion definitions, and reported data against it, the concept would take on a life of its own by virtue of that. (Data collection would likely be trivial since megaregions would no doubt be made up of counties, such as by creating a layer above Economic Area). Their idea seems more to create something called a megapolitan area rather than a megaregion, however.
Tridig Banerjee has an interesting chapter further trying to refine the megaregion concept by identifying types of megaregions along a two by two matrix (which, with my management consulting background, I of course love). The dimensions are “galaxy” vs. “corridor” and “mosaic” vs. “network” (hierarchical). The Midwest would be a galaxy-network. Scott Campbell has a chapter asking a number of useful questions, such as the one at the top of this piece.
Ross herself seems particularly interested in the transportation aspects of megaregions, and this is one where it seems to have the most direct applicability. For example, most of the various high speed rail proposals out there revolve around megaregions. There are shared corridors of interest, such as interstate highways, and other important features, such as the Great Lakes. The question is whether these are items of relevance to a megaregion properly so-called, or if they are just the focus of ad-hoc “coalitions of the willing”. I actually suspect the latter as there are many of these (think of the I-69 and I-35 NAFTA corridor coalitions for example, or California’s high speed rail proposal) that exist independently of megaregions. In my view a megaregion would need to represent some true community of interest, in the way that a metro region does, to represent some sort of truly functional element, and I haven’t seen it yet. In fact, I have argued that even things like the Midwest high speed rail network shouldn’t be thought of as a network, but rather as a series of point to point connections linking outlying areas to Chicago. Chicago will not be an HSR hub in the way that O’Hare is a hub – that is, for traffic interchange. Indeed, in we see this in Europe, where there is very little transfer traffic of this type. There seems to be something to this megaregional transportation idea, but I’m not sure what is yet.
The piece I found most compelling was the chapter by Saskia Sassen. In my original piece on megaregions I noted that lots of people talk about them, but no one says what it is we should actually do with them in order to create real value. Sassen suggests how this might happen. Here’s an excerpt:
I argue that the specific advantages of the megaregional scale consist of and arise from the coexistence in one regional space of multiple types of agglomeration economies. These types of agglomeration economies are distributed across diverse economic spaces and geographic scales: central business districts, office parks, science parks, the transportation and housing efficiencies derived from large (but not too large) commuter belts, low-cost manufacturing districts (today often offshore), tourism destinations, specialized branches of agriculture (e.g., horticulture or organically grown food), and the complex kinds of agglomeration economies evident in global cities. Each of these spaces evinces distinct agglomeration economies and, empirically at least, is found in diverse types of geographic settings, from urban to rural, from local to global.
The thesis is that a megaregion is sufficiently large and diverse to accommodate a far broader range of types of agglomeration economies and geographic settings than it typically does today. This would take the advantages of megaregional location beyond the notion of urbanization economies. A megaregion can then be seen as a scale that can benefit from the fact that our complex economies need diverse types of agglomeration economies and geographic settings, from extremely high-agglomeration economies evinced by specialized advanced corporate services to fairly modest economies evinced by suburban office parks and regional labor-intensive low-wage manufacturing. It can incorporate this diversity into a single economic megazone. Indeed, in principle, it could create conditions for the return of particular activities now outsourced to other regions or to foreign locations.
I wrote a four part series in early 2009 called “Reconnecting the Hinterland” which was all about searching for value in attempting to foster a re-created interlinked economy between Chicago and the rest of the Midwest. An answer to Sassen’s question is actually what I was looking for. The simplified idea being, to find some economic activities in which geographic proximity, though not necessarily always in a dense, face to face setting like downtown Chicago, is a source of value; to ask, is there some medium between the “spiky world” of Manhattan and the Loop and the “flat world” of China and India?
I don’t want to jump the gun and go into detail, since that is a part of the next part in this series, but if you are interested, you might want two check out two pieces in that series, “Metropolitan Linkages” (about extended labor markets) and “Onshore Outsourcing“.
One curious omission from this book was the difference between megaregional and non-megaregional locations and whether there was some benefit to being in a megaregion. I can’t help but notice that in the Midwest, Kansas City (in most maps) and Des Moines are both outside of the megaregion yet are two the of the absolute best performing metro areas. Not being part of a megaregion does not appear to have hurt them any. I’d be interested to see some analysis on this.
In any case, for those interested in these things, this is a nice survey book to pick up. It is accessible to the general educated public, but is written in the style beloved of academics, so is likely to be very dry to all but those who are wonky about this stuff. Read the Sassen excerpt to get a sense of what is in store. For those who are in the planning or related fields, it is worthwhile to educate yourself on the megaregion concept to be able to parse a lot of the rhetoric out there about it. Reading this book would be a good way to do so.
I’ll leave you with this quote from Lewis Mumford’s The City in History, to give a perspective from one of the all time great screedmasters on this subject:
Instead of creating the Regional City, the forces that automatically pumped pumped highways and motor cars and real estate development into the open country have produced the formless exudation. Those who are using verbal magic to to turn this conglomeration into an organic entity are only fooling themelves. To call the resulting mass “Megalopolis”, or to suggest that changes in spatial scale, with swift transportation, in itself is sufficient to produce a new and better urban form, is to overlook the complex nature of the city. The actual coalescence of urban tissue now taken by many sociologists to be a final stage in city development, is not in fact a new sort of city, but an anti-city. As in the concept of anti-matter, the anti-city annihilates the city whenever it collides with it.
Don’t hold back Lewis, tell us how you really feel.
This post original appeared on December 6, 2009.
“is there some medium between the “spiky world” of Manhattan and the Loop and the “flat world” of China and India?”
Are Shanghai, Hong Kong and Bangalore any less “spiky” than Manhattan and the Loop?
What exactly is the basis of a “megaregion” – similar location, or similar characteristics? The above maps are mostly based on location, but your description focuses on characteristics. The difference is significant: Phoenix and Atlanta have similar growth patterns but are very distant from each other, while Chicago is more similar to several coastal cities than to the Midwest cities with which it is grouped.
I would suggest that a useful definition of a megaregion would be an area that functions as a single unit, with convenient transportation between the different parts. The advantages of this are the possibilities of economic competition and specialization within the region. This is effectively the definition of a traditional metropolitan region. A city needs both offices and housing. Yet it is perfectly workable to concentrate them in specific locations (downtown, suburbs), and to have multiple concentrations in direct competition with each other (office parks vs downtown for workers, different neighborhoods for house purchases), since people can travel easily between any part of the metropolitan area. Another example: Law firms seems to like concentration – in any US city the large law firms are mostly downtown. Yet rather than all law firms being located only in NYC, they are located in downtown of every metro area. Why? Because customers are generally not willing to drive/fly out of town to see a lawyer. But when there is one law office in the metro area, anyone in the area can go downtown and use it.
High-speed-rail, specifically, allows a region larger than the traditional metro area to have similar transportation connectivity and thus to form one unit. Thus, it is easy to picture Chicago as the center of a greater Midwest region – once HSR (and supporting transit infrastructure and TOD) is built. At present, though, this infrastructure not really exist (except arguably in the Northeast). So it is no surprise that we currently see no benefits to belonging to a megaregion. The megaregion is more an expression of what development will one day become possible that how cities actually function at present.
Jason Tinkey says
I like to think of regions as a city, on a macro-scale. Kevin Lynch wrote about nodes within a network…he was referring to a single city, but the idea scales up exponentially. You say that Cleveland, Detroit & Chicago have different challenges, but within a single city there are separate neighborhoods with different challenges. I’m most familiar with Chicago, so think of the contrast between Lincoln Park & Englewood. Same city, but they as well be on different planets.
Robert Hurst says
I agree with Mumford.
Jeffrey Jakucyk says
I’d agree that there’s some problems with the definition of megaregions as far as simple proximity. The Boston-New York-Philadelphia megalopolis is one that’s pretty clear cut, but you get into trouble when you start trying to connect it up with Baltimore and DC. Some of these other megaregions are a bit of a stretch too. Phoenix-Tucson isn’t very “mega”, nor is Denver-Albuquerque. When you have disconnected satellite metros as part of a megaregion, like Kansas City and Minneapolis, does that really count for anything?
Anyway, the point is that some of these regions have oddball metros that don’t really fit within their own context, even if they are close in proximity. For instance, Winston-Salem, Birmingham, Knoxville, and Memphis have much more in common with upper midwest rust belt cities like Pittsburgh, Cleveland, and Buffalo than with their “new south” sunbelt brethren like Atlanta, Charlotte, and Nashville. Winston-Salem itself might just as well be a thousand miles from Greensboro for being as different as they are.
So can anything really be abstracted about such places and still be useful when they have such fundamental differences? It’s that level of hyper-abstraction about zoning codes, roadway engineering standards, and building development that has brought us the disastrous mess of sprawl we have to deal with. Do we really want to try to further distill and abstract our places? Maybe the more valuable tool would be to look at metro areas by type rather than by region, which has been analyzed here on The Urbanophile before.
Good points, Jeffrey. I have read through the book myself, and some of the megaregional associations definitely seem much more organic and tangible than others. The “Steel Corridor” which includes Pittsburgh, Cleveland, Youngstown and the Akron-Canton area is an obvious natural association given that region’s common historical, economic and cultural roots established through heavy steel and auto manufacturing. Although most of the factories are gone now, the region and its cities still retains a great part of its blue collar norms and ways of doing things, for better or for worse. Even demographically, it stands out as one of the country’s least diverse and most rapidly aging, recent urban core growth in Cleveland and Pittsburgh notwithstanding. On the whole, population growth here is stagnant or in decline.
The “Ohio Valley” on the other hand, consisting of Columbus, Cincinnati and Dayton seemed like much more of a stretch to me. Other than proximity on a map and being under the same state government, Columbus and Cincinnati-Dayton could each be in completely different worlds. Columbus with the bulk of its economic base in white collar service/professional jobs and government more closely resembles Indianapolis, while Dayton has more in common with rust belt cities like Toledo, Youngstown and Canton. Cincinnati doesn’t seem all that different than Columbus at first glance, economically speaking as both cities are home to several fortune 500 companies and tons of small businesses. But in terms of local culture, demographics and built environment, they are like night and day from each other. Cincinnati is aging and stagnant or very slow in terms of its growth, while Columbus is young, increasingly diverse and steadily growing. Culturally and politically, Cincinnati is famously (or infamously) conservative, whereas Columbus is becoming increasingly progressive/liberal.
I do agree that it might be helpful to look at metros across the nation for their commonalities (economic, social, political, geographic) more than simply for their proximity to others, although this can and does come into play in some places more than in others. Brookings has released a few reports that attempted to classify metros based upon their economic status. Perhaps an overlay of this data with the megaregions maps could provide a more informative, multi-dimensional picture about the cohesiveness or lack thereof between the metros within each designated region.
Alon Levy says
Jeffrey, I think the megaregion concept is still valid if you think in terms of proximity and influence rather than similarity.
By analogy, let’s consider metro areas first. The cities proper of New York, San Francisco, Boston, and Chicago have a lot in common with one another – more than each has in common with its favored quarter suburbs or working-class suburbs or edge cities. They vote similarly, they support similar kinds of culture, and so on. On an even more micro level, the Upper East Side and Back Bay have a lot in common, more than with many nearby neighborhoods in their respective cities.
The reason the megaregion concept makes sense is the same as the reason the metro area makes sense. Thanks to Amtrak, the buses, and relatively short distances, people go back and forth between New York and Boston, and this lets communities stay in touch after partial dispersal.
John Morris says
Right, so logically a city like Chicago has more potential synergy with the downtowns of Milwaukee and Madison than with it’s exurbs. It needs to build on that.
Eric O says
The claim that Winston-Salem could be one thousand miles apart from Greensboro is what I would call a seriously flawed if not spurious abstraction. Only in someone’s frame of mind is that separation significant. Neither in social fact nor in any economic sense does that statement serve reality. The Piedmont Triad exists and we consultants who have dealings in it thankfully do not have question that fact when we get up in the morning.
Just reflect a little on your work experiences, folks.
For example, I’m based in Charlotte, which generates 70% of my work. In my experience, it takes a significant city to support an architecture/planning firm which specializes in public/transportation projects. But we also serve developers and municipalities throughout the Carolinas. For the planning of a current project we consulted with an Atlanta grant-writing firm. The developer is local, one of the project architects is based in downtown Raleigh, the general contractor is based in Guildford County (between Greensboro & Winston-Salem).
To survive, we have to rely on alliances with Triad and Triangle firms to do work across the region. We agglomerate thus the talent distributed across the megaregion, except for the fact that we hardly ever step into northern Georgia. Because Atlanta has firms like ours, we tend not to compete for work in that area (apart from Savannah, in my firm’s peculiar case, which is about as far from Atlanta as it is from Charlotte). Nonetheless, other firms do. Atlanta-based developers, in fact, know Charlotte intimately.
Megaregions exist! They are not abstractions.