Thursday, September 22, 2011
Both land use and transportation policy are often driven by the fear of traffic congestion. If you're interested in congestion comparisons, the most widely cited resource is the Texas Transportation Institute's Urban Mobility Report.
TTI’s most recent report revealed a startling development: congestion actually decreased in a lot of places. Out of the 15 largest urbanized areas, five (Seattle, Atlanta, San Francisco, Los Angeles, and Detroit) had decreases in congestion between 1999 and 2009, measured by yearly hours lost to congestion per driver. Most major metropolitan areas had 10-20% increases in congestion, and a few had much greater increases: in Chicago, Dallas, and Philadelphia, hours lost to congestion went up by 20-30%, and in Houston yearly hours lost increased by 38% (from 42 to 58).
What separates the winners from the losers? I’m not prepared to answer that question. But what I am prepared to do is address (and debunk) a couple of possible scapegoats. One possible argument might be that if it weren’t for those pesky environmentalists and fiscal conservatives fighting new roads, we could just build our way out of congestion. If this was true, we would find that the congestion “winners” had higher levels of road- building than the congestion “losers”- that is, the congestion-beating regions would increase freeway lane-miles at a rate higher than population growth. We would also find that the four regions that experienced the greatest growth in congestion built new lane-miles at a lower rate.
Neither proposition was consistently true. To be sure, two of the five congestion-reducing regions increased lane-mileage per person (San Francisco and Detroit). But three others failed to do so. In Atlanta, the number of lane-miles increased by 7 percent between 1999 and 2009 (from 2350 to 2520) while population increased from 3.7 million to 4.2 million (about a 13 percent increase). Yet hours lost to congestion decreased slightly (from 49 miles per year to 44). In Seattle, a 20 percent increase in population (from 2.65 million to 3.18 million) was not matched by road growth: freeway mileage increased from 1690 to 1855 miles, only a 9 percent increase. Yet congestion decreased from 52 hours to 44. In Los Angeles, freeway mileage and population were almost evenly matched: regional population increased from 12.3 million to 13 million (about a 6 percent increase) while freeway miles increased from 5400 miles to 5610 (about a 4 percent increase). But in Los Angeles, congestion plummeted more rapidly than elsewhere, from 76 miles to 63. Bottom line: the regions that beat congestion didn’t need to build more roads to do it.
What about regions that most obviously failed to beat congestion? In three of the four regions where congestion increased by over 20 percent, freeway mileage increased fastesr than population (all but Dallas). Score one for the environmentalists.
If roads failed to reduce congestion, was public transit more successful? Here the picture is equally mixed. If increased public transit ridership had a major effect upon congestion, we would find that the five successful regions would have experienced public transit ridership that increased faster than population. This was certainly the case in two such regions, Seattle and Los Angeles. According to TTI, metro Seattle’s public transit ridership increased from 142 million in 1999 to 188 million in 2009. Similarly, in Los Angeles, transit ridership increased from 562 million to 671 million. But on the other hand, transit ridership actually dipped slightly in both Detroit and Atlanta during the 2000s, and went up only slightly (from 420 million trips to 425 million, a 1 percent increase) in San Francisco. Yet congestion declined in those cities as well.
Our bottom line seems to be that the link between infrastructure and congestion is weaker than one might think.