Four Southern Cities

This is way off topic, but back in March when the new Census population estimates came out, I tweeted about slow growth in Alabama and negative growth in Mississippi. I’m wondering why they aren’t part of the “new south”, generally taken to be fast growing Sunbelt cities like Charlotte and Atlanta. Why does Atlanta boom, but Jackson doesn’t? Many of the responses boiled down to “because it’s Mississippi” which is sort of circular logic.

I thought it would be interesting to compare population growth since 1860 in four southern cities: Jackson MS, Birmingham AL, Atlanta GA, and Charlotte NC. Since municipal boundaries in the US are illogical, often changing, and hard to trace, I’m using county level data, comparing the populations, based on the 2015 definitions of the metropolitan statistical area (MSA) for each city.

I also added an “Atlanta Redux” where I threw out the 9 counties (out of 29) in the Atlanta MSA that still, as of 2016, have not reached population 30,000. These counties are way out near the edge of the city and were only recently sprawled into the Atlanta MSA. Their inclusion in the regional population back to 1860 probably unfairly inflates Atlanta’s size at the outset of the Civil War, when counties 40 miles out would not have had much plausible connection to the city but still had some population. I’m including both the full MSA and the reduced version to let readers compare.

Even the redux version of Atlanta was the biggest city of the four in 1860. Not much divergence is seen between Birmingham, Atlanta, and Charlotte until after World War 2. Jackson appears to have always grown more slowly, though this fate would not have been at all obvious to an observer in 1880, when Jackson was over 50% larger than Birmingham.

Plotting the population on a logarithmic scale can help us see patterns more easily, as the runaway growth of Atlanta makes the others look flat by comparison.

Plotted this way, the post-World War 2 growth of Atlanta and Charlotte doesn’t look unusual; it just looks like a continuation of previous trends. Atlanta, especially the redux version, starts growing a little faster than Charlotte after the war.

More interesting is Birmingham, which grew very quickly for decades after the Civil War, going from about 60% of Jackson’s size in 1870 to 120% of it in 1890. Birmingham even passed Charlotte for a few decades, going from 40% of its size in 1870 to larger by 1910. An observer in 1930 might have expected Birmingham to surpass Atlanta in another 10-20 years, especially the redux Atlanta. Jackson, meanwhile, experienced a strong decade from 1870-1880, but never grew very quickly again.

After 1930, Birmingham, which had been the rising star, began to fade, and was surpassed by Charlotte by 1960. So, what happened to Jackson after 1880? And what happened to Birmingham in 1930? I’ve yet to find out.

Another way to look at this data is to look at the amount of growth in the central county (where the main city is located) and compare that to the growth of suburbs. Here’s the breakdown for these four cities, with the central county on the bottom of the graph.

It is interesting that in Birmingham and Jackson, the central county stopped growing when the MSA as a whole started growing more slowly. However, it’s important to not try to read too much into this – did the MSA grow slowly because the central county was being neglected, or did the central county stop growing because there wasn’t much growth to be accommodated anyway?

One unappealing answer is that there is simply a great deal of luck in these outcomes. Obviously it helps for the city to be well run, but there are many factors making a complex result, and sometimes you just get bad luck. This rudimentary look is obviously not enough to tell us much about why these cities grew at different rates, but if anyone has any ideas, I’d love to hear them.

Advertisement

What Do You Mean By Suburb?

Sometimes I think that a lot of misunderstandings in the discussion on cities relate to inconsistent terminology. It seems to me that we have four different concepts of what a suburb is, so if we’re going to talk about suburbs, we need to be talking about the same kind of suburb. In my personal descending order of preference, they are:

• Los Angeles: typified by surprisingly dense uniform development filled in on a grid of arterial roadways, usually at half-mile or mile spacing. Development spreads uninterrupted until it runs into an insurmountable barrier, e.g. the Pacific Ocean, 10,000-foot tall mountains, or kangaroo rats. This pattern is a legacy partly of the rancho grants, partly of the US public land system. This is what people think of when they think of sprawl, but it’s actually the least sprawly.
• Northeast: typified by somewhat dense historic town centers, surrounded by low density exurban development. Subdivisions have larger lots, and there are often large relatively undeveloped areas. This is a legacy of development following the pattern of small farms. Virtually all of the characteristics that urbanists like predate the auto age.
• South: as I noted in my post on Boston and Atlanta, this is basically the same pattern as Northeast Corridor suburban growth but without the underlying pattern of historic cities and town centers. The South is what the Northeast would look like if no one had been living there to start with in 1945.
• Midwest: like the Northeast, but even more spread out. Subdivisions are built around small, historic agricultural crossroads, and there can be miles of farmland between exurban towns. Midwest sprawl is typified by an urban footprint that keeps growing quickly, despite relatively stagnant populations, as people decamp the old cities.

In the following sections, I’m going to describe each type in more detail, including why I like or dislike the pattern.

Los Angeles

For LA, let’s revisit the patch of development in Lancaster that I used as a counterexample to Boston and Atlanta.

As I said, this is what most people think of when they think of sprawl. Aerial shots of suburban tracts like this are stock images in any urbanist post about how suburbia is a monotonous, soul-crushing, doomed landscape.

But on many of the things that urbanists claim to care about, Lancaster does pretty well. It has a solid grid of arterials on half-mile spacing, and many of the arterials already have bike lanes. You could throw down bus lanes with POP fare collection and stops every half mile that would basically be Jarrett Walker’s dream grid (well, without anchoring). Add some mid-block crossings for pedestrians and boom, you’re good.

Now, depending on the whims of developers and local planners, there can be a lot of cul-de-sacs and indirect streets. You might have a circuitous path to get to one of those arterials – at least if you’re in a car. I’m always a little amused at the hand-wringing over street grids. Y’all were never kids with bikes? I grew up in a place with lots of cul-de-sacs and disconnected streets. We knew where we could cut through. That’s not to say we shouldn’t try to make new developments better by bringing back the jog, but this is a much easier problem to solve than those faced by other types of suburban development.

Then there’s that sneaky LA density. Let’s take the block in the image above bounded by J, J-8, 15^th East, and 20^th East. By my count, there are 483 SFRs and somewhere in the vicinity of 190 apartments (making some conservative assumptions), along with, very roughly, 125k SF of retail. If we assume 3 people per house (in line with Lancaster demographics) and conservatively say 1.3 people per apartment, we’ve got about 1,700 people living in a quarter of a square mile (sq mi), for a density of 6,800/sq mi.

In other words, this patch of the Antelope Valley – mostly SFRs, with a big-ass parking lot in front of the retail, the kind of place that people like James Howard Kunstler would call crudscape – already has a density higher than the weighted density of the Washington DC area, and it’s not far behind places like Philadelphia and Boston. Even if we base the calculation on the least dense sixteenth-square-mile, which has 154 SFRs, the density is 5,500/sq mi. Weighted density of Portland, for reference, is 4,373/sq mi. Oh, and it’s not even built out yet.

That last point is another secret strength of the LA suburban pattern: no one is under any delusions about what we’re doing here. Everyone expects and hopes that those vacant lots will get developed. The home builders want it. Retail and business owners want it. R Rex Parris wants it. And the folks in the apartments on the other side of J-8 aren’t going to mourn the loss of those dusty lots. If you were to liberalize the zoning, eventually you’d end up with dingbats like Palms or skinny-but-deep redeveloped Cudahy lots like you see in their namesake city and places like El Monte. This year, Lancaster changed its zoning to allow accessory dwelling units. In LA, the expectation is that more people are going to show up, and that’s a good thing – the opposite of the premise under which suburbs in New England operate.

Northeast

Now, in total contrast to LA suburbs, where people basically expect and want growth, the assumption in New England is that you have a town perfected by the descendants of Myles Standish and John Winthrop when they settled it in 16-whenever, and all this growth is irredeemably ruining the historic character of the town. Let’s have a look at The Pinehills, a recent major subdivision in Plymouth, MA.

The Pinehills is what passes for smart growth in a lot of the Northeast; the Boston Globe says the state uses it as model, in an article that proves my point by citing a resident as a 10^th-generation descendant of William Bradford. The permits allow for a little less than 3,000 houses on 3,174 acres – or in other words, a final density of about 1,800/sq mi, well down into exurban territory and on par with places like Bismarck and Pocatello. New England is the land of two-acre zoning. The expectation is that the intervening area will never be developed. Every suburban resident in Massachusetts subconsciously fancies himself an English baron, entitled to undeveloped wood lots for fox hunting or whatever.

This is a development 45 miles out from Boston. The low density means that it is always going to be impractical to serve the area with transit. The insistence on rural character means that the arterials are unpleasant and unsafe for biking and walking. As I said in my Boston/Atlanta post, every dense neighborhood that exists in New England existed 60 years ago. Tom Menino and Joey C can conjure a few new urban districts out of semi-vacant industrial land, but that’s about it.

It’s important to note that this a fundamentally different mindset, and it affects all aspects of policy. For example, recent MBTA commuter rail extensions like Newburyport serve towns with comically low populations and population densities (Rowley, population 5,856, density 290/sq mi) that have no realistic prospects for appreciable growth. Deval Patrick gets accolades from Streetsblog for proposing “smart growth” density of 4 SFRs or 8 apartments per acre near transit stations, which will produce population densities on par with. . . Lancaster. Of course, that’s only if they actually develop an entire square mile around the station. Which they won’t, because it’s New England.

Despite all of this, the Northeast still benefits from legacy town and city centers. I’m not sure what you can do with the low-density exurbs, but the presence of these nodes at least means that people see what density looks like.

South

With the South and Midwest, we’re into territory I don’t have personal familiarity with, so I welcome any thoughts or corrections. In general, it’s harder to find “typical” suburban development outside of California, because there’s more variability. For the South, I’m going to revisit the Atlanta area: Redan, which is just outside the 285 beltway on the east side of the city. I tried to find an area of development that had some apartments, since they seem to be more common in the South than in the Northeast or Midwest. I’m looking at the area between the 278, Wellborn, Marbut, and Panola.

This part of Redan has a density of about 4,900 sq/mi, which would make it very dense by Atlanta standards, where weighted density is only 2,173 sq/mi. Part of the problem is that it’s just very hard to pick a representative plot here. The area sprawls so far that the edges are mostly undeveloped, which makes them unsuitable for measuring the pattern in the region. Here’s another shot, west of the 285, in Powder Springs. Looking at the area between Powder Springs, New MacLand, Macedonia, and Hopkins.

This area has a density of about 2,600 sq/mile, which is in line with what we expect for the region.

Looking around the South in general, using old images available in Google Earth, it does seem to me that more recent development has been build a little more densely – perhaps as developers have realized they’re running out of land? It also seems that the planning and development culture of the South is such that the region wants to keep growing in population, which is not really the case for the Northeast and Midwest. However, I’m not sure if the political and social structure of the South is ready for upzoning and density on the level of Houston or Los Angeles. The low density of the South makes it difficult to provide effective transit and more costly per capita to maintain infrastructure.

The other thing that will challenge the ability to provide effective suburban transit in the South is, like the Northeast, the mishmash incoherent network of arterials. Unlike Los Angeles and the Midwest, the South and the Northeast inherited a winding network of colonial roads that make it very hard to design transit routes that don’t have a lot of turns. Whereas Western runs over 25 miles due south from Los Feliz to San Pedro, in the South and Northeast, you’re lucky to find an arterial road that doesn’t change direction at random and dead end after a few miles. In addition, the insistence on maintaining “rural character” means that there’s often public resistance to widening arterials (even to provide transit) and building things like bike lanes and sidewalks.

Midwest

From a 10,000-foot view, the Midwest seems to have more freeways than the rest of the country, along with bigger suburban lots. That, combined with low population growth, seems to me to make this the purest form of sprawl, and the least sustainable. For our example of Midwest suburbs, I offer up Michele Bachmann’s district: Stillwater, MN. Take the area between 75^th St, Neal, McKusick, and Manning.

This area checks in at a density of about 1,200/sq mi, with 300 SFRs and 150 apartments. The weighted density of the Minneapolis MSA is 3,383 sq/mi, so this area is low and it may yet get denser. However, it’s hard to see it reaching Lancaster densities anytime soon. On the plus side, the Midwest does have a good arterial grid.

Notice that many of the subdivisions in the Midwest have large lots – what Californian planning would call “estate residential”, and relegate to a few affluent communities like Acton and mountainsides that are too steep for denser development. You won’t find any development like that in the LA Basin, the Valley, or the vast majority of Orange County. Where it still exists in the IE – for example, Fontana – the lots are being further subdivided into typical LA suburbia.

In the Midwest, though, like the Northeast, there’s no expectation that these areas will ever get any denser. With low population density, a mindset that opposes further development, and far-flung subdivisions, it’s hard to see how these areas could be served well by transit, or become very walkable. When I listen to Charles Marohn, I sometimes have to remind myself that he’s talking about places like Baxter, which, other than being called a “suburb”, has remarkably little in common with a place like Corona.

Summary

I promise you, all of the images in this post are at the same scale. It is interesting to look at them next to each other and compare. The differences that I’ve outlined in this post explain why I think the LA development pattern is the best and why I’m essentially bullish on the future sustainability of LA.

For reference, here’s a quick tabular summary of the differences between these four types of suburbs. Suitability for walking and biking pretty much correlates with density, because if the place isn’t dense enough, you won’t be able to walk or bike to anything worthwhile, even if the infrastructure for it exists.

Forget About Boston vs Atlanta

Note: this was originally part of Sprawl and Economic Mobility but the main point of that post ended up being the continuing impact of discrimination, and this isn’t really related.

In his series of posts on sprawl and economic mobility, Krugman used a comparison of Boston and Atlanta when talking about the relationship of sprawl and economic mobility, which he said “many people have pointed out . . . as the most obvious comparison among major US metro areas”.

In a word, no. Leaving aside New York as unique among American cities, ATL vs BOS is perhaps the most obvious comparison – if you are a typical East Coast resident who thinks that the country ends at Allegheny Front.

In reality, Boston and Atlanta are similar and in the last 60 years, they have developed in similar ways. The only difference is that Boston grew to be a much larger city earlier than Atlanta did. Check out a comparison of the populations of the metro areas and the cores at the dawn of the Auto Era in 1950 and in 2010¹:

Or how about this: take a look at the two aerial images below, which are at the same scale. Which one is Boston’s beltway and which one is Atlanta’s? (Answer at the bottom of the post)

Both cities have had virtually no growth in their historic core communities. The difference is that Boston had a sizable core when the sprawl era began. Atlanta started sprawling out around very little other than a new skyscraper CBD. But every dense neighborhood that exists in Boston today existed in more or less the same form in 1950. Same goes for the denser suburban town centers. The current growth mode in Boston and Atlanta is the same: low density sprawl.

Even with the resurgence of Boston’s core cities since 1990, it’s still 200,000 below peak. Somerville – and you keep hearing about what an awesome place Somerville is, right? – has actually lost population since the Red Line Extension opened. The story of urban revival is about the core gaining population, but it is just as much about what kind of people are moving there. Somerville has more young professionals than it used to, but it also has less blue collar workers and immigrants than it used to.

Really, the same can be said for almost any East Coast city. The dense neighborhoods of Washington, Baltimore, Philly, and so on – they all predate the auto age. So does any smaller node of density like, say, New Haven or Albany. If you want a real comparison with a place like Boston or Atlanta, you need a place with a different growth mode. You need Los Angeles, or at least a city doing a decent impersonation, like San Jose or Houston.

New England is the land of two-acre residential zoning suburbs. The governor of Massachusetts wants to promote dense development around transit stations – dense being 4-8 units/acre, or in other words, about the default density allowed anywhere by the generic residential zone in far-flung LA suburbs like Lancaster and Adelanto. Let’s have a look at some recent LA growth, shall we?

I don’t expect Lancaster to be winning any awards from the CNU anytime soon (though they did get one from EPA). James Howard Kunstler probably woke up in a cold sweat when I grabbed that screenshot. But as far as I can tell, the primary problem with LA seems to be that people don’t like how it looks. That’s fine, but if we’re talking about the viability of a development pattern, we need to put our architectural preferences aside. If you want to compare Boston or Atlanta to anything, this is what you should compare it to, and fact is, it stacks up pretty well.

Now, don’t get me wrong. I used to live in Boston, and it’s a great city – they should build more of it. I’d be very happy to see the Boston region resume its pre-WW2 growth pattern so that more people could take advantage of that. Knowing the ferocity of NIMBYism, even in the near suburbs, and the strong level of local control afforded to New England towns², I’m not going to hold my breath. If the choice is between low-density East Coast sprawl like Atlanta and high-density LA style development, LA wins every time.

Boston is on the top, Atlanta on the bottom.

¹For the region, for Boston I used the MSA less the New Hampshire counties, plus Worcester and Bristol Counties; for Atlanta I used the MSA. For the core, for Boston I used Suffolk County plus Quincy, Newton, Brookline, Waltham, Watertown, Arlington, Belmont, Cambridge, Somerville, Medford, Malden, and Everett; for Atlanta I used just the city itself.

²One of the ways that not having counties hurts the Boston region.