Quote:
Originally Posted by glowrock
I hate to say it, but there IS another good way to reduce some of the horrible traffic congestion getting into and out of Pittsburgh especially from the south, west and east (it's really not bad coming from the north). There truly does need to be a bypass around Pittsburgh for through-travelers.
|
So ring roads and such have been tried many times, and the short version of the story is they don't work in the long run to end congestion on radial urban highways (the ones used for getting into and out of the city center).
The basic problem here is one of incentives. You put a nice highway leading into the region's most important economic center, and people are going to preferentially build up along that highway in order to use it. If there is no charge to use that highway, the number of people trying to use it at peak time will eventually exceed its maximum capacity, causing congestion. Congestion may then slow development along the radial highway, although it is possible (and common) for congestion to keep on growing as long as there are no superior and affordable alternatives to taking the highway into the center.
In other words, when you have a scarce good (more people want it than can have it), its use has to be rationed in some way. And if you don't ration use through price, it typically gets rationed by people's tolerance for delay due to congestion. This is the same basic process that led to bread lines in communist countries: give away bread for free but not enough of it, and the person who gets the bread is the person willing to wait the longest in line.
So what happens if you build more lanes, or build a ring road to divert some traffic, or so on? You might momentarily reduce congestion, but that just restores the incentive to start building up along the radial highway again. So people do that, and in not too long you are right back to bad congestion on the radial highway. And that is because intentionally or not, you have actually adopted a strategy of using congestion to ration use of your radial highways, and so you will keep getting congestion.
Two basic mistakes are being made here. First, highways are a sucky way of providing peak capacity in urban areas. That is because their peak capacity per lane is really quite low, and so the marginal cost of additional peak capacity using highways is really high in comparison to other urban transportation technologies. And that is true even when there are not a bunch of expanded tunnels and bridges you would need as well--urban land is precious, and so the cost of chewing it up for more highway lanes is inherently high.
Second, congestion is a sucky way of rationing scarce peak capacity on highways--indeed, it actually subtracts peak capacity to follow that approach.
So really, as soon as your urban highways are getting congested, you should be switching focus to technologies with much higher peak capacities per unit of cost, like passenger rail or busways, and also you should be switching to rationing the peak capacity on your highways by price rather than congestion.
As for ring roads--if there is truly an independent case to be made for them based on current needs, fine. But it is generally not going to work to rely on them to eliminate radial highway congestion, and in the real world there is almost never a current-needs case sufficient to justify the massive expense of ring roads. Instead, most of their future traffic tends to be generated by greenfield developments induced along the ring road, until they become congested too. Which is actually terrible land use policy as well as bad transportation policy, although as a matter of pure coincidence a lot of well-connected highway contractors and greenfield developers get very wealthy in the process.
Incidentally, this is why building transit in parallel ALSO doesn't work to eliminate congestion, unless you add congestion pricing as part of the policy. New transit may divert some people initially and relieve congestion for a short time--yeah!--but then people see the uncongested highway and just start piling in again until the highway becomes congested again. You can keep going through scenarios like this, and the same end result will occur: if your only method for rationing your urban highways is congestion, then congestion you will get. On the other hand, unlike with ring roads, in many U.S. cities there are in fact extremely compelling current-needs cases to be made for expanding and improving transit which do not rely on congestion relief.
Anyway, the good news is that if you have an economically important reason to be passing by but not actually into the city center, and an economically important reason to do that at peak times rather than off-peak, then you should also have the means to pay the congestion price. That could be time-sensitive freight, or it could be a commuter who has chosen to live on the other side of town from their job, or so on.
Meanwhile, the people who will shift onto public transit due to the congestion price (assuming that transit is available) likely will be the people trying to get into and out of the center specifically at peak times, which is precisely what you want to see--those people should be using alternative technologies for the reasons given above. Similarly, freight and people will shift to off-peak travel when that is not a big problem for them, which is again precisely the behavior you want to incentivize.
Note, though, the immediate shift to alternatives does not have to be very high at all. Again, that is because the peak capacity of a congested highway is actually lower than the peak capacity of an uncongested highway. So in fact in theory, if you can keep congestion from starting you can actually have more, not less, people using the highway at peak times, meaning there would be no immediate need for a shift at all. The key is to find a way to stop more people from piling in at that point such that congestion restarts, which is a matter of getting your congestion pricing just right.
But to be sure, even congestion pricing on existing highways taken alone is a short term solution--eventually a growing population is going to need more peak transportation capacity (barring something like a true revolution in remote work, or maybe driverless cars, or so on). But again, your focus at that point should be adding peak capacity through better technologies for the problem at hand. And even if you do build more highway capacity, that should be congestion priced as well.
Honestly, the idea of a free urban radial highway is just an inherently bad one, and if we had started out with congestion pricing of urban highways as the default in the United States (as paying train and bus fares is the default) people would rightly think you were crazy if you suggested making them free instead. But because they have largely been free as a matter of historic practice, people keep trying to look for another way around the very predictable problems that arise from refusing to price-ration. It just doesn't exist.
Prev
Linear Mode
