Nicholas Beaudrot asks:
Without citing Portland, OR as a model, what is a reasonable goal for transit adoption and transit-oriented development in cities who haven’t had serious rail or streetcar investment in 50 years? Clearly we’re not going to turn St. Louis (popn density ~5500 people/sq.mi.) into Manhattan (60,000), but can we even turn it into DC (10,000) or Baltimore (8,000) within fifteen years?
I’m not really sure why I should be prohibited from citing Portland as a model. We have a somewhat unfortunate situation in this country where the main example people are familiar with of a transit-oriented city is New York, which is one of the very biggest in the world, while obviously most of our cities are much smaller than that. So it’s worth saying that the developed world is full of examples that are neither New York nor sprawl. That’s not just Portland, Europe is full of decently walkable medium-sized cities.
On to St Louis. It’s hard to talk in too much detail about places I’m not very familiar with. But the pace at which things can change is going to be dictated, in part, but the extent to which there’s actual interest in building anything in the metro area. At the moment, clearly, nobody is going to undertake large new building projects—dense or otherwise—in St. Louis or anywhere else. And a small city in the midwest is under no particular obligation to turn itself into a particularly dense metropolis. But what you want is to avoid a situation where you’re preventing density. St Louis has a couple of decent rail transit lines and it’s important to allow dense projects to be built near those stations and along the corridors that are served by rail. These things are expensive to build, and once they’re there it’s important to utilize the served areas in the most efficient way possible. That doesn’t mean forcing people to build extremely tall projects near them, but it does mean letting such projects go through without demanding vast fields of parking to be placed around everything.
In general, I would also just note that it can get misleading to look at citywide density averages. The relevant issue for a city that (like St. Louis) has some transit is whether or not you’re achieving density at your transit nodes. Additional consideration that are important is that ideally the stations will be close enough together to create not just pockets of density but whole corridors of density, even if the corridors are surrounded by pretty traditional suburbs. The stretch of Arlington County running from Rosslyn to the Metro stations at Court House, Clarendon, Virginia Square, and Ballston are a great example of how this can look.