Replying to 2 blog posts Yourself and Kevin Corcoran On skepticism towards Pigouvian taxes Scott Sumner gives the example of Orange County’s congestion charge. It was a huge success. Scott writes:
This example shows that not all Pigouvian taxes have failed: congestion charges in cities such as Singapore, London and Stockholm are other successful examples.
Scott could have added Fairfax County, Virginia to the list. Interstate 66 runs from Washington DC through Fairfax County into West Virginia. It is a regulated highway, and only carpoolers are allowed on it during rush hour (violators are fined heavily). Even so, 66 is always jammed. In 2017, the Department of Transportation chose to remove the HOV restrictions and switch to dynamic tolling during rush hour. The tolls go up and down depending on traffic, with the goal of keeping the average speed on I-66 at 55 (the speed limit). This was a huge success. With the introduction of dynamic tolling, traffic congestion has virtually disappeared. This dynamic tolling is a tolling system that is a type of Pigouvian tax.
This is a Pigouvian tax, and I wholeheartedly welcome it. In grad school, I often rode the 66. God knows how many hours I wasted sitting in traffic. When dynamic tolling was introduced, I could easily determine whether the value of my time was worth the toll.
In our posts, Kevin and I talk about skepticism. I am a skeptic. Market failure Modifications are necessary, but I’m not entirely against them. Modifications can be effective in some cases. As always, the devil is in the details.
One reason I am skeptical of Pigouvian taxes is that it can take a long time for governments to adjust tax rates (increase or decrease). Various political factors get in the way, special interests get involved, and the political process itself is just generally slow. So tax rates can be too high or too low for long periods of time, leading to suboptimal results. One of the great things about congestion taxes, especially congestion taxes like the I-66 toll, is that they are easy to adjust. The I-66 toll is fully automated. It increases and decreases fairly quickly to adjust traffic. It doesn’t require voting, lobbying, or other expensive procedures. The adjustment costs are low.
Moreover, the negative (or positive) effects of a congestion tax are immediate: we immediately see traffic increase or decrease depending on the level of the tax. The cost of analysing the tax is low. Conversely, the effects of a carbon tax are very slow to emerge: it can take years for the negative (or positive) effects of carbon emissions to emerge. The cost of analysing a carbon tax is quite high.
Finally, at least in the case of I-66, the tax is collected via transponders installed in many vehicles (or, if you don’t have a transponder, a bill is mailed based on the registration linked to your vehicle’s license plate). The tax is collected instantly and does not require measurements or audits. The cost of administering the tax is low.
I support congestion fees. Having spent hours sitting in Boston traffic (as I’m sure Scott has), I’d like to see my home state do something similar on its highways. Congestion fees can overcome my skepticism.
The point of my post on correcting market failures is not that such actions should never be taken, or that they are doomed to failure. Rather, it is to remind economists and readers of a most fundamental lesson: nothing comes for free. Market intervention is not cost-free. Yet most economists treat it as if it were. Standard economics textbooks provide only a casual summary of market failures, along the lines of “markets work well, but when transaction costs become too high, markets break down and the government can/should intervene.” I try to inject an economist’s natural skepticism into these decidedly non-economic ideas. We must compare real-world alternatives. Most market interventionists fail to do that. They look at market failures and Assuming Government intervention will improve the situation. As economists and scientists, we must be skeptical. all You have to be realistic about the plan. You have to scrutinize the details, because that’s where the devil is. You can overcome that skepticism, but it needs to be there.
Interestingly, here Public Choice Analysis comes into play. Without public choice, the benefits of dynamic or automatic tolling are not obvious. But public choice teaches us to study politics without the romance: that the incentives faced by politicians are not the same as those faced by people who interact directly in price-guided markets. To a standard economist, all other things being equal, a tax set by a committee and a tax set by automatic dynamic tolling would be equally preferable. Public choice economists recognize that these two methods are not equally preferable; one produces much worse outcomes than the other.
John Murphy is an assistant professor of economics at Nicholls State University.
