Matt writes of the Midwest:

[I]t does seem to me that the political see-saw is swinging especially violently in this region, even as this broad economic story applies everywhere. I can’t help but wonder if this doesn’t reflect the fast that the Midwest seems to be in a state of pretty severe relative decline vis-à-vis the nation as a whole. Broadly speaking, we’re seeing very rapid population growth in key sunbelt areas. Then in Boston-Washington corridor and Pacific Northwest we’ve seen housing become very expensive, which has tamped down population growth but indicates that demand is present. But states like Ohio, Indiana, Michigan, and Wisconsin have neither a fast-growing metro area like Houston nor a “nobody goes there any more, it’s too crowded” city like San Francisco. You could easily imagine this kind of situation leading to an unusual degree of voter fickleness and persistent dissatisfaction with all options.

I think this is broadly right. But as I told Matt on twitter, it also overstates the universality of Midwestern decline. It’s true that the Midwest has had a very rough few decades. Its biggest cities and its economies were long based on an economic model — industrial agglomeration — that no longer works. As these cities deindustrialized, they found themselves with large populations of relatively unskilled workers, and with state organizations and budgets that were unable to handle economic decline and falling revenues. So for some time, the most talented workers tended to leave the region after college, while cheap housing attracted poor, low-skill workers. Many cities have suffered serious depopulation. Real output in Ohio is virtually unchanged from 2000. In Michigan, real output is below its 1997 level.

And yet, it would be wrong to say that the entire region is facing relative decline. There are a few obvious counter-examples, which Matt acknowledged in a tweet: Minneapolis and Chicago. But there are other bright spots, as well, even within the states Matt mentions above. Columbus, Ohio, for example. Columbus has a lot going for it — it’s the state capital and the home of one of the country’s top public universities. Over the past decade, the metro area grew by about 170,000 people. There’s a growing tech presence in the city. The contrast with Ohio’s more prominent metro areas is dramatic.

Within Indiana, Indianapolis is a notable exception. It’s also a state capital, and it’s added over 200,000 people during the last decade. The city seems to have carved out a niche for itself as a wholesale trade and distribution center. We could talk about Pittsburgh. As of 2009, its population had resumed growing after a long decline. The city has helped use its educational resources to foster a thriving tech start-up environment. And then there’s Kansas City, which also added over 200,000 in the last decade.

What’s interesting about the Midwest, to me anyway, is that it’s much more spread out than other major economic areas. It’s about 400 miles from Washington to Boston, and that 400 miles is packed with about 50 million people. It’s about 450 miles from San Diego to San Francisco, and there are 30 million people tucked along the route. Meanwhile, there’s about 800 miles between Minneapolis and Pittsburgh, and you have to be pretty generous to include 25 million people on the route. Minneapolis is as close to Denver as it is to Pittsburgh. Indianapolis is closer to Atlanta than Minneapolis.

Why is this important? Because market size is important. Economists like to play around with gravity models, which do a good job predicting (if not necessarily explaining, in a causal fashion) trade flows. What these models suggest, in a nutshell, is that trade is a function of market size and distance. This is pretty intuitive. Two big cities located near each other will trade more with each other than a big city and a small city near each other, or than two big cities far away from each other. And if there is greater trade between markets, that creates opportunities for specialization and productivity improvements and other good things. So on the east coast, improvements in individual cities reinforced improvements elsewhere, because east coast cities were very large, very close together, and very well connected.

If most large Midwestern cities were scattered between Chicago and Minneapolis, the Midwest would look a lot more like the northeast corridor. But they’re not. That doesn’t mean that individual places can’t be successful. It does mean that individual success is less likely to contribute to regional recovery.