A few years ago, Seattle was widely known as one of the most congested cities in the country.  Annual rankings by the Texas Transportation Institute placed the metro area’s traffic as one of the nation’s 5 worst—and in some years greater Seattle was ranked among the worst 3, along with places like LA and San Francisco.

Then, a couple years ago the TTI changed its method of estimating congestion, largely to give credit to cities that are using congestion reduction strategies (metered on-ramps, quick clearing of accidents, transit service, etc.).  After the change in methodology Seattle’s congestion plummeted in the national rankings.  In last year’s report, Seattle ranked twentieth in congestion delay per traveller, about what you’d expect for a city of Seattle’s size.  Generally, more populous cities have worse traffic, so Seattle’s congestion, rather than being among the nation’s worst, now seems to be about on par with expectations.

By comparison, Portland, OR ranks 26th in congestion delay per capita among US cities, while Charlotte, NC—ranked as a "mid-sized city" rather than a "large city" like Portland and Seattle—came in at 22nd place.  That’s slightly worse than Portland, though in fairness the two are probably in a statistical dead heat.  That said, it’s telling that Charlotte, despite a smaller population, was ranked as having worse congestion than Portland.  We happened to cover Charlotte in our sprawl report covering 15 select US cities a few years back.  Of the cities we studied, Charlotte was easily the most sprawling—its average density is lowest, and it consumed more land to accomodate new residents than any other city in the study.  So if someone claims that low-density development can ease traffic congestion, tell them about Charlotte.

Of course, I think that the congestion rankings should be viewed with some skepticism; as Seattle’s plummet shows, congestion rankings may have more to do with theoretical models than with reality.

Still it seems worth noting that Portland (a city with a justified national reputation for controlling low-density sprawl) seems to have about the same level of congestion as Charlotte (which is at the opposite end of the sprawl spectrum).  Which leads me to think that "sprawl" and "congestion" are two largely unrelated phenomena—and that traffic congestion may be far more strongly affected by idiosyncracies in a city’s transportation system than by population density per se.