Erica Barnett has a different takeaway from the study that showed a VMT reduction I linked to last week. She says that nothing can reduce congestion, and she is absolutely right that we cannot build our way out of congestion. However, in reality, congestion pricing can definitely reduce congestion, and get much needed revenue for transportation improvements. A win-win in my mind.
(Yes, I know the photo is of HOT lanes, which is not even close to the same thing as congestion pricing, but it’s the closest look we’ve got at the sort of infrastructure required.)
I really liked the phrase in one of the comments,”the annoyance equilibrium”. I really think this is what it’s all about. Build more lanes you compress the congestion (for a while). People put up with what they’re willing to put up with. Make transit more annoying (longer times, more transfers, obnoxious riders) and ridership goes down. Make driving more expensive, take longer, more aggravating, drivers opt out. I don’t know why you say HOT lanes are different from congestion pricing (they should be dynamic but averages work). There is nothing demonic about driving a car, nothing noble about riding a bus and nothing sacred about rail.
I’ll state for the record that I’ve never described driving a car as “demonic”.
HOT is taking the HOV lane and making it available to SOVs. Congestion pricing is charging everyone (including HOVs) to drive on congested roads. As Carless in Seattle used to say “Excess demand for roadways during peak hours is the real problem, to which congestion is the most feasible solution”.
Driving on a road is free, but it has real value. I lot of people realize this and take advantage of the value, but since it’s free, demand is way higher than capacity. Instead of paying with money, you pay with your time. Congestion pricing is the idea of charging the price you would pay based on demand. It’s got nothing to do with demonizing driving or punishing drivers, it’s an idea instead to help drivers by eliminating congestion on the roadways.
I want to add, even if you dramatically increase capacity on some road that is congested, you lower the time-price of driving on that road, and more drivers show up (It’s called induced demand). Those new drivers who pick up the slack come from somewhere, and you end up worsening congestion at that somewhere. If you add a lane to 520, you end up with congestion at the I-5 and I-405 interchanges. Add a lane there and you get congestion in downtown Bellevue, Mercer Street, 50th st, etc. Adding lanes just makes people drive more and ultimately makes congestion worse somewhere.
And that’s why I-5 is one of the most congested roads in the city.
“Yes, I know the photo is of HOT lanes, which is not even close to the same thing as congestion pricing…”
How is HOT not congestion pricing? Prices go up when congestion goes up. Prices go down when congestion goes down. Is this not congestion pricing?
HOT lanes do not create a reduction in overall congestion. Rather, they create an increase in capacity.
Congestion pricing charges every single person to drive, which reduces congestion. See my comments above.
I’m still having a hard time with the concept that increased capacity doesn’t reduce congestion. Is the reasoning that it simply moves it to another point? That’s often the result but it’s not always the case.
I do think I’m understanding what you mean by congestion pricing being along the lines of London charging a toll to enter the city during peak hours (essentially all day and night for London). I guess it’s easy for me to say, since I don’t live or work in Seattle, that the problem for downtown isn’t how to get more cars into the city; it’s how to get some of the cars out of the city. The first big battle is where to draw the line. Maybe somewhere around the recently purchased (think it was a done deal) postal parking garage for the south end?
Basically induced demand. You should browse the now dead Carless in Seattle blog when you have some time: http://blog.carlessinseattle.us/2007/04/520_bridge_mode.html
So you can fit more cars, but more people will use the free capacity if you’ve build and you won’t significantly affect traffic. (1% increase in capacity = 0.9% increase in cars.) And of course it gets pretty expensive fast. It’s a billions cheaper to add another train run than to add another lane — and it’s an increase in capacity. The only way to increase capacity for roads is more lanes.
Building whole new corridors of light rail is like building an entirely new highway, except it takes up less space which can mean a lot in already developed areas. And the nature of rail doesn’t encourage sprawl nearly as much as highways (and sprawl can lead to expensive new highways and lanes). Of course it’s pretty expensive to build a new light rail line, but it often is cheaper than a major widening or highway construction project.
OK, John. After much reluctance I looked at Carless in Seattle via the link you provided. I got as far as, “The broad conclusion (according to the summaries I read): for every 10% increase in road capacity, there is a concomitant 9% increase in vehicles using those roads after a few years[emphasis mine]. In other words, building more roads induces people to drive more, keeping congestion effectively constant.” This is silliness. The increase in demand is not because the same people are induced to drive more because there are new roads. There is greater demand and frankly I’m amazed that the increase in vehicles using those roads hasn’t increased at more than 10% (which would indicate a decrease in congestion).
> Building whole new corridors of light rail is like
> building an entirely new highway
Yes, it has the same effect of dividing neighborhoods and driving development. In the greater Seattle area I don’t see a new freeway in our future. For the same reasons I believe new rail corridors should follow as much as practical the existing corridors and NOT hold up the creation of new development as a positive.
I mean it’s sort of obvious that traffic wouldn’t spring up over night right when a new lane opens, but having that capacity would induce people into moving further away (sprawl) or working further from where they live (sprawl), or perhaps working closer to where they live along a faster route (temporary). And of course, our population will continue to grow. I don’t get your point — are you saying that the demand is simply latent? That makes no sense to me since that would mean there is another mode serving that trip now. And an increase of more than 10% of vehicles with an increase of just 10% in capacity would indicate an increase in congestion.
To dismiss a congestion study in three sentences leaves a bit to be desired. Especially when we’re talking about billions of dollars of public expenditures. Perhaps this merits deeper analysis from you, since that blog has dozens of posts on the subject.
Crude analogy: if I give away 100 candy bars a day, then eventually only 110 or so people are going to show up to take them since I obviously don’t have enough capacity to handle everyone showing up. If I up my ante to 400 candy bars a day, obviously many more people are going to show up to eat my free commodity. Eventually, more than 400 will show up daily. The problem isn’t the candy bars, it’s the cost ($0). With an actual a real cost to eating the candy bar, the market would find an equilibrium. Perhaps at that point sugar taxes wouldn’t be needed to fund my candy bar distribution system, but maybe I wouldn’t want candy bar purchases to fund the entire system and only a fraction of it. And when I began to get too many people wanting to pay for candy bars, I could reduce demand by raising their cost.
The same argument holds true for transit, except we don’t save as much money by reducing peak demand for transit [yet]. But that is an argument for peak fares vs. non-peak fares. Obviously highways “operate” 24/7 so they’re frequently built to handle peak conditions with excess supply outside of that. Transit is much different in that it operate with variable capacity and thus taxpayer cost throughout the day.
Since light rail has a fraction of the size requirement as something like a modern highway, you cannot simply assert that it has the same effect of dividing neighborhoods and expect that for that to be taken as a matter of fact. The footprint of at-grade light rail is about two lanes of road. Move that to be elevated and you get some shadows but nothing like the Viaduct. Move it to a tunnel and you have no footprint at all.
The problem with building all of your alignment near current highway ROW is that the best neighborhoods and destinations are not *right* off the freeway, with some exceptions. Highways will influence development, of course, and so will rail. There’s no reason to confuse that development — the type of development most suited to depend on highways is least suited to depend on rail. Just like you wouldn’t necessarily build your highway infrastructure to match existing rail infrastructure, you shouldn’t do the opposite. But in either case, you can take slightly different routes and still hit a lot of the obvious places that highways do serve and effective rail should serve. Bellevue wouldn’t exist without I-405 and perhaps SR-520, and light rail is going to serve that city with many stops even though it won’t follow the alignment of I-405 or SR-520 for more than a quarter mile.
Heh “Erica Barnett has a different takeaway” amused me.