Social Engineering! Cities That Build More Parking Get More Traffic
Build parking spaces and they will come — in cars. New research presented this week at the annual meeting of the Transportation Research Board finds a direct, causal relationship between the amount of parking in cities and car commuting rates.
University of Wisconsin researcher Chris McCahill and his team examined nine “medium-sized” cities — with relatively stable populations between 100,000 and 300,000. They compared historical parking data with car commuting rates beginning in 1960, finding “a clear, consistent association” between parking levels and car commuting that has “grown stronger” over time.
Using an epidemiological research method, McCahill’s team determined that the relationship was causal. For example, data indicated that increases in parking tended to precede growth in car commuting.
The study brings home the point that by inflating the parking supply via minimum parking mandates and other policies, cities are leading more people to drive and making conditions worse for transit, biking, and walking. It’s what you might call “social engineering.”
Researchers compared five cities with low car commuting rates (Arlington, Virginia; Berkeley, California; Silver Spring, Maryland; and Somerville and Cambridge, Massachusetts) to four cities with relatively high car commuting rates (Albany, New York; Lowell, Massachusetts; and New Haven and Hartford, Connecticut).
McCahill and his team found that for every 10 percentage point increase in parking spaces per capita, the share of workers commuting by car would be expected to increase by 7.7 percentage points. So if a city increased its per capita parking from 0.1 spaces to 0.5 spaces, car commute mode share would rise about 30 percentage points.
In Hartford, for instance, the number of parking spaces in the city increased from about 15,000 in 1960 to about 47,000 by 2000. The city’s solo car commuting rate over that time rose more than 20 percentage points.
University of Connecticut Professor Norman Garrick, who co-authored the study, said the findings indicate it’s not in the interest of cities to continue to add parking.
“Adding parking in more traditional urban settings indeed has not worked out for cities,” he said. “The cities added parking out of fear that they would be losing all their trade to the suburbs. Almost all cities did some of that between 1960 and 1985. Some cities did it less. By 1985, cities like Cambridge had realized there was no benefit to doing this and they basically capped parking. Those cities have soared in terms of economic outcomes, in almost all measures of quality of life.”
The office building where I work has exactly two parking spaces. My company occupies about half of the office building. These two spaces are reserved for the building owners, who account for a very small percent of the people in the building.
This of course means that almost nobody in my company drives to work, since street parking is basically all metered in the neighborhood. Most of us take public transit, bike, or walk. But given this is Berlin, it’s pretty easy to do all those things.
The difficult part has been trying to convince the building owners, who drive, to add more bike parking. Last summer it was completely full almost every day.
It’s nice to finally see hard data supporting what many on this site have known for a long time. I would be very interested what happens as the number of parking spaces approaches zero. Extrapolating the graph seems to indicate there would still be close to a 50% auto mode share, although maybe the dropoff gets a lot steep as the number of spots per capita approaches zero. In any case, it’s pretty clear from the last 50 years that the policy of trying to shoehorn private automobiles into dense urban areas has been a disaster on many levels. Cities trying to imitate the suburbs in the end find out they just can’t out compete them if the goal is simply facilitating a car-based lifestyle. The space for it just doesn’t exist. Cities should return to their roots.
Put your bikes in their spaces 😛
“Researchers compared five cities with low car commuting rates … to four cities with relatively high car commuting rates ”
I *almost* can’t think of anything that distinguishes the cities in the first list from the cities in the second list. Hmmm, let me think…could it be that each of the cities on the first list is actually an inner ring suburb of a major city with a heavy rail transit system, and all of the cities on the second list are not able to take advantage of their neighbor’s transit system???
You may as well compare Manhattan and Houston.
While I don’t disagree with the conclusion, the research methodology seems terrible. All the low car-use “cities” are essentially urban, and border a much larger city that also has low car usage. Not to mention Cambridge and Berkeley are essentially college towns, filled with non-car-owning students. Compare that to their high car-use cities, all of which are small and isolated. I can’t see how they can draw any meaningful conclusions from such a disparate sample.
Would comparing the same company at two different locations be more convincing?
Secaucus Junction location (suburban, free parking): 88% drive alone
Downtown Newark location (major transit hub, market rate parking, transit pass subsidy): 36% drive alone
This is the change Panasonic North America’s HQ made in 2013. http://www.citylab.com/cityfixer/2015/10/how-panasonic-turned-car-commuters-into-transit-riders/412417/
If you have a chance I would encourage you to please read the paper so you can understand the methodology including the criteria used to select the cities. I will just mention one point – many of our immediate reaction and believes about these cities are often wrong. To give just one example, Cambridge has a higher car ownership rate than Hartford, New Haven and Albany. In other words this issue is a bit more complicated than how we talk about it.
Do you have a link to the paper?
I agree it is (much) more complicated, so I’m curious as to how you controlled for those factors. I just find it a little implausible that Albany could have simply *chosen* to build less parking, and would look like Cambridge today.
Um, no.
All that this study shows is that in places w/o transit access, most people drive to work, and in places with good transit access, fewer people drive to work.
What the study is essentially saying is that Panasonic could have simply deleted parking at its Seacaucus Junction location, without making any other changes, and gotten an instant reduction in car usage with no consequences.
The point I’m making is, the two locations are so different that it’s impossible to attribute the reduction in car usage to any one factor.
“To give just one example, Cambridge has a higher car ownership rate than Hartford, New Haven and Albany”
Is it higher within comparable wealth cohorts? Cambridge is a *much* wealthier community than those other three.
Also, quick and dirty look at the ACS (maybe you have a better data source) shows that Albany and New Haven have notable higher ‘vehicle available’ rates than Cambridge, and Hartford is higher among the ‘owner’ cohort.
The poverty rate in Hartford is 3x the rate in Cambridge (safe to believe that most are in the renter cohort in both cities), thus most likely more than explaining the less than 2 percentage point higher ‘vehicle available’ rate among Cambridge renters compared to Hartford renters–poor people are less likely to own cars.
Panasonic management didn’t think that, as they cited the “need” (yes, exaggeration, but still) to drive in deciding to move:
“Consider one of the biggest leases of the past year: the decision in April by electronics giant Panasonic to move its North American headquarters from Secaucus to Newark. The move, which is taking advantage of the state tax-credit program, is driven partly by Panasonic’s desire to be near mass transit.
“We have literally 1,000 people driving cars every day,” says Peter Fannon, Panasonic’s vice president for technology policy. “The key element for us, which really brought the focus back to Newark, were the environmental benefits, specifically the ability to be in an urban center where there are housing, restaurants, hotels, and most importantly, mass transit facilities, all within a three- or four-block radius of our new location.””
http://www.wsj.com/articles/SB10001424052702303848104576381873749553788
Just because parking availability and price isn’t the *sole* factor in mode choice doesn’t mean it’s not *an important* one. Transit service matters; parking does too.
And so does the context of the location you are talking about. Burning Man is going to have nigh on 100% automotive mode share, and a similar-sized event in Central Park or on the National Mall very low.
It’s far more complex than “supply of parking drives [heh] solo car use”, especially when comparing quite disparate locations.
If Cambridge, Brookline and Somerville had all adopted statistically different approaches to parking, and there was correlation bt more parking and more solo car use, it would be *far* more interesting,.
But this is exactly the point. Cambridge is full of rich people who own cars AND THEY WALK AND TAKE THE BUS ANYWAY.
Hartford could get more people on the bus and more people walking if they bothered to design their city right.
Albany absolutely could have chosen to build less parking. And fewer expressways. And more pedestrian bridges across the Hudson. And it could have kept its train station in downtown instead of moving it across the river. Which doesn’t have any good pedestrian bridges either….
Albany made a huge number of bad car-centric choices.
The short answer is that the number of parking spaces never approaches zero. At a certain point, with a small enough parking supply, the price of parking rises high enough that starts becoming profitable for private industry to build profitable, privately funded, standalone parking garages.
That parking supply — the one where private profitable parking garages get built — is a *lot lower* than most of the US has. You can see what the rates might be by looking at Manhattan private garage rates…
I agree. I simply challenge the assertion that Albany (or New Haven, or Lowell MA) is at all comparable to Cambridge (or Berkeley, or Arlington). Albany could have done all of what you said and more, and would still be nowhere near Cambridge in car usage.
Designed it right by having it close enough to piggy-back on a larger city’s commuter rail system?
That’s not the point of the study at all. The point is that *supposedly* if you have more parking spaces, you have more driving.