One charge often leveled at planners is that they deliberately foster congestion in order to force people to live in dense, walkable developments. I'm generally critical of planning, and I certainly don't believe one group should foist its preferences on another.
The fact, though, is that some congestion is good. Not because compact, centrally located development is more virtuous than single-family homes in the suburbs. Rather, drivers are willing to pay only so much to avoid congestion -- which means they are willing to pay only so much for a given amount of road capacity. Building more capacity than necessary is a wasteful overinvestment.
This follows from the result I cited in a recent post:
One important feature of congestion prices is that they not only discourage usage when congestion is present, but they also generate revenue for capacity expansion. Indeed, it has long been recognized that under certain conditions the optimal congestion prices for a fixed amount of capacity will automatically generate the appropriate amount of revenue to finance capacity expansion.*
This really is a remarkable result. We can pay all the costs of new road construction simply by calibrating road capacity and pricing the roads.
This has two other implications:
1. We can build new, unsubsidized toll roads without charging drivers when the road is uncongested. The roads could be free most of the time; we just need "enough" congestion to pay the bills.
2. We should not build so much capacity that roads are never congested. Capacity costs money. The only purpose of extra capacity is to relieve congestion. When congestion pricing generates just enough revenue to cover the cost of construction we have, by definition, built exactly the amount of capacity drivers are willing to pay for. Building more capacity is as much a waste of money as building a three-story school when one story would do, or buying 20 fire trucks when the town only needs 10.
Note that the optimal capacity does not change merely because we refuse to price the road properly. The hypothetical stream of revenue tells us how much drivers would be willing to pay, whether they are charged for it or not. Congestion pricing simply monetizes the waste from sitting in traffic. Another way of looking at it is that when we refuse to price congestion, we "pay" for the roads twice. Drivers pay with lost time. Taxpayers pay again with cash.
*I've attempted an intuitive explanation of this result below the jump.
Imagine we plan to build a new road, and we have a black box that perfectly predicts the cost of a given amount of road capacity as well as the revenue congestion pricing would generate from that capacity.
Our black box has a dial on it. Turning the dial to the left decreases the road capacity. Turning the dial to the right increases the road capacity.
Our black box also has two counters on it. One displays the congestion revenue that a given amount of capacity would generate. When we turn the dial to the left, road capacity decreases, which means congestion -- and hence congestion revenue -- increases. Conversely, turning the dial to the right increases capacity, thereby decreasing congestion revenue.
Our other counter displays the cost of building the given amount of capacity. The dial has the opposite effect on the construction cost: When we turn the dial to the left, we decrease capacity and thus decrease the cost of construction. When we turn the dial to the right, we increase the cost of construction. Thus, whichever way we turn the dial, one of our counters goes up and the other goes down.
The result I cited says that there is a position on the dial where the two counters display the same number. There's a mathematical proof, but it makes intuitive sense. If we start the dial at a very high capacity, the congestion revenue will be zero but the cost of construction will be high. Slowly turning the dial to the left will gradually increase congestion revenue and gradually reduce construction cost. Eventually, we should reach a point where the numbers meet.
