Bryan Caplan  

5 Miles of Economic Illiteracy

Not Groupthink... $1 for the Hands of Victory?...

Yesterday I spoke at the University of Virginia's Miller Center (video coming to local PBS in about two weeks) as well as its Department of Economics. The cost: I had to wake up at the ungodly hour of 7 AM. The unexpected benefit: As I drove west on Route 66, I witnessed the vivid fruit of economic illiteracy - 5 miles of stalled east-bound traffic.

The congestion was not the result of an accident. It's the same massive waste of time that happens every weekday while I sleep peacefully in my bed. In fact, since this was a Friday, traffic was probably better than average. Morning after morning, Monday through Friday, there are miles of commuters sitting in their cars lamenting their sad fate.

The darkly funny thing about this spectacle is that every economist in the country knows how to fix it!

In case you haven't heard, here's the solution: Raise the price when demand is high. To quote The Six Million Dollar Man, we have the technology. High-tech easy passes make it possible to collect tolls without tollbooths, or even slowing down.

Why doesn't the world embrace the obvious solution? Once again, the blame falls squarely on voters' anti-market bias. Most people think of tolls as pure transfers, devoid of incentive effects. A politician who proposed the economists' solution could kiss his career goodbye. (Well, pull it off?)

In short, congestion is a state of mind. Drivers have a slow commute because their minds are stuck in the mud of economic illiteracy. What will it take to tow them out?

Comments and Sharing

COMMENTS (18 to date)
Patrick writes:

Get someone with a pedigree as anti-capitalist as "Red" Ken's to sell it,

Maybe Bernie Sanders, except that Vermont has relatively few traffic issues. ;)

Tim Worstall writes:

"(Well, except for this guy - how did he pull it off?)"

He was rather helped by the fact that the academic proponent was Sir Alan Walters. One of Maggie Thatcher's advisors. And that people like the Adam Smith Institute (disclosure, I write for them as a freelance) were cheering the whole thing on.

So it was really rather dificult for anyone to dismiss it as trendy left wing nonsense.

superdestroyer writes:

How high would you have to make the toll to offset higher real estate prices, the price of private school school tuition, and higher insurance bills for living inside the beltway?

If you establish tolls, you raise the value of property in Arlington and Falls Church. Also, since the schools in Arlington and Alexandria are inferior to the schools in Chantilley, a rational person has to factor in the cost of tuition at Bishop Ireton or Cardinal O'Conner. The commuters will keep driving but the number of restaurant meals or charitable donations in Manassas will go down to make up the difference.

New Jersey is full of toll roads, yet traffic there is just as bad. The Dulles Toll Road causes enough traffic jams but gives local governments a reason to not build other roads or to even bother to time traffic lights.

Jeff Hallman writes:

Or you could try something really revolutionary, like building more roads.

I-66 is a great example of what happens when the if-you-don't-build-it-they-won't-come school of traffic management is followed. Inside the Beltway, inbound I-66 is HOV-only and only 3 lanes wide at the behest of the slow-growthers in Arlington County. The result has been the movement of jobs and commuters out to Tyson's Corner and Dulles Airport, otherwise known as sprawl. Further compounding the Tyson's traffic problems is the lack of bridges across the Potomac. If the Beltway's American Legion bridge has an accident or is otherwise backed up, the next closest crossing is at least a 16-mile detour.

What needs to be done around DC to alleviate the traffic problems is obvious: (i) turn US 301 into a limited-access Eastern bypass around the metro area to take I-95 interstate traffic off the Beltway, (ii) build the ICC, but also extend I-370 southwest through Potomac and Great Falls to a new river crossing and down to Route 7, (iii) new separated HOV lanes for the entire Beltway, (iv) ditto for I-66 inside the Beltway, with the current lanes opened to all nontrucks, and (v) additional lanes on I-95 South down to Fredericksburg.

It's equally obvious that most of this will never happen, as people here would rather sit in their cars blaming immigrants than take actions which would actually help.

Alex writes:

So it was really rather dificult for anyone to dismiss it as trendy left wing nonsense.

But they did; including essentially all the politicians and pundits Tim agrees with.

Jim writes:

Or you could provide a half-decent mass-transit option and watch as people switch from cars to a more sensible means of transport. One of the main reasons Ken Livingstone was able to sell his congestion charge was because its revenues are ploughed into improving public transport. Contra Tim, the support of the Adam Smith Institute didn't help, in fact probably the opposite if anyone actually noticed.

Sigve Indregard writes:

How did Livingstone pull it off? Well, he is the mayor of a city where many do not own a car (and even fewer use it daily), because the net effect of using one is negative. The tube plus the occassional cab gets you from A to B faster and cheaper for the average Londoner.

Scott Scheule writes:

So you're saying poor people shouldn't be allowed on the roads?

Troy Camplin writes:

I recommend that every university have a sign over the front door that reads: NOE MAY ENTER WHO HAVE NOT MASTERED ECONOMICS (with pseudo-apologies to Plato).

CVD writes:

decent response here

Taimyoboi writes:

Or you could just impose really high taxes on businesses of all forms, which would then lay off lots of employees, thereby reducing congestion on roads from commuters...

Eric writes:

In Houston, there is an easy-tag only road that costs commuters up to $6.00 1-way. However, these costs have not substantially affected the level of congestion. Should the price be increased further?

What is the empirical evidence that tolls have a substantial impact on traffic congestion? Have you modelled the demand curve?

Dan Weber writes:

"(Well, except for this guy - how did he pull it off?)"

Bryan, you've got an error in your HTML that prevents a significant part of that sentence from appearing in Firefox. Put a quote-mark after "nofollow" in your anchor tag.


More on subject, the UK has been dealing with forged and stolen license tags in their anti-congestion efforts. Well, the state hasn't been dealing with it, since they just mail fines to the people whose plates were forged, and they are the ones who have to deal with it.

As my motorcycle & I zip by, I get a bird's eye view of the problem and there's already a repsonse to the loss of productivity: tons of drivers are blackberrying or on the phone and maintaining their productivity. Also, where professionals work long hours and devote considerabel effort to their families, those who spend an hour of quiet time in the car probably benefit on a personal level.

But, I agree with other posters, more roads is the best approach.

Broadcast guy writes:

I just moved to LA from DC. No one believes me when I say the traffic is better here!

1) Most of the jobs in DC could be done as well with telecommuting and video conferencing if we could break that social taboo

2) Another Potomac crossing north of the beltway, another south of the beltway (maybe Ft. Washington), as well as improving 301 as a DC bypass.

In fact, we should go back to the original 1950's proposals of three concentric beltways (The ICC is step one...)

Mark Seecof writes:

Prof. Caplan, you've got to consider people's varying dollars-to-hours exchange rates. You could levy tolls to maintain optimum throughput during rush hour. That would be good for people with more money than time, but not so good for people in the opposite position. I think it would be better than the current scheme (using congestion to impose a "time toll"), but I also think you should recognize that part of the opposition to cash tolls may come from rational self interest, not simple economic ignorance.

Or to rephrase, if a commuter would have to pay more in cash tolls than he would earn in the time those would save him, he might prefer the congestion.

John Fast writes:

If Professor Caplan's book is correct, then "dismissing it as trendy left wing nonsense" may be a good way to sell it!

Claim (truthfully) that a congestion charge is a way to penalize/reduce pollution and global warming.

One might also promise that all tolls will be used to pay for mass transit programs, or environmental programs.

(Of course I think that economic literacy programs count, because they will improve environmentalism.)

Mark Seecof writes:

Only an economic illiterate would promise to use road tolls to pay for mass-transit programs. Mass transit is much less efficient than personal automobile transportation. If you want to follow "economic best practices" then use road-toll moneys to pay for roads-- help everyone internalize the costs of the services they consume. If mass transit can't recover nearly all of its capital and operating costs through the farebox then it is not worth building or operating. (Note that in America today, road users pay for both roads and transit, whereas transit users consume truly gigantic subsidies. The most wasteful thing you can imagine would be to subsidize transit more.)

Details for those of you who haven't realized that I don't make stuff up:

Mass transit only makes sense for workers in a few very dense cities, and for people too poor to afford a car. It is terribly inefficient for most workers in the USA. That is why politicians force (road-using) taxpayers to subsidize transit-- otherwise it would not exist outside of NYC, Chicago, and a few other cities. In most cities mass transit is a welfare program.

Read what the US Department of Transportation reports:

The private vehicle, especially driven alone to work, is the mode of choice for most Americans living in the large metropolitan areas. In every major metro area, workers who drove alone to work increased in numbers and share in the last 40 years. However, transit as a commute mode is critical in the large metro areas. Nearly 7.5 percent of commuters in large metro areas use transit, compared to less than one percent in the rest of the country. Workers are less likely to drive alone to work in the large MSAs (73 percent compared to 79 percent in the rest of the country), but the percent of workers carpooling is about the same (12 percent). Exhibits 4.10 and 4.11 show percent of workers using different modes of transportation for 1990 and 2000.
Ninety-two percent of the nation's commuters who use public transit live in the large metropolitan areas. In most areas, the travel time by public transportation is almost twice the travel time by driving alone. For 2000, the average travel time by workers who drove alone was 24.1 minutes, while transit travel time was 47.7 minutes. Carpool travel times are close to drove alone at 28.5 minutes, on average. Exhibit 4.2 shows the mean travel time by mode for the 6 large MSAs with the highest number of transit commuters.

Everyone in America depends on cars and trucks (directly or indirectly-- how do you think food gets to your favorite grocery store?), and all roads are funded by road users. Only about 4% of American workers take transit, and many of those folk use cars for other travel (e.g., shopping). Effectively, there are no non-road users subsidizing road users.

You can argue about the efficiency and transparency of paying for roads through general taxes[1], but not whether non-road-users are subsidizing roads.

(What do you think a subsidy is? Consider persons A who buys a good X at price P, and B who buys good Y at price Q. If A and B each pay the full price for their goods, neither subsidizes the other. Suppose we impose tax T on purchases of X. If A pays (P + T) for X, and B pays (Q - T) for Y, then A subsidizes B.

(In the USA, A pays (P + T) for roads, and B pays (Q - T) for transit. (Of course, A and B both pay general taxes for local streets, but, they both use them, so neither one subsidizes the other.)

If you look at the numbers (US FHWA, 2005 Table HF-10), you will find that around 65% of all road spending (the chart title says highways, but it includes all public streets, roads, etc.) comes from fuel and vehicle taxes.[2] Only about 26% is collected by local governments using non-road taxes. Only around 30% of all expenditures are for "locally administered roads," which matches the local government contribution fairly well. So we spend general taxes mainly on local streets, which, please note, we would pay for even if there were no private automobiles. It is quite proper to finance local streets by local/general taxation-- after all, people without cars get as much service from, e.g., firetrucks and paramedic ambulances as drivers do.[3]

Note that $21,285 million (~14%) of highway taxes are diverted to mass transit and other non-road uses.[4] If we quit diverting highway taxes to transit boondoggles and so-forth, we could cut local taxes spent on roads by more than half.

Or we could spend the money reducing congestion. If we did that, fewer people would find transit attractive, neatly reducing demand for transit at the same time we cut subsidies to it. Look, if transit cannot pay for itself through the farebox (mainly), then it deserves to die. Road users pay for their travel already. Why should they subsidize transit (and transit unions, and grafting politicians, etc.)?

[1] Improving efficiency and transparency through, e.g., congestion- related road tolls might be a good idea, but that has little to do with the phony claim that some beleaguered class of utterly immobile people is being forced to "subsidize" roads.

[2] 2005 numbers. 65.6% is what you get by dividing the total Highway User Fee (tax/toll) revenue actually used for roads ($90,343 million), by the total of all tax money devoted to roads ($137,668 million). Some more highway money comes from selling bonds, but those will be paid back by the same sort of tax money as we spend directly.

[3] Not to mention motor trucks delivering goods. Are you from New York, too? New York paved its streets in the horse-and-wagon era. Besides, if we don't use cars we'll need a lot of buses. Buses need paved streets, you'll note.

[4] The numbers in this table greatly understate transit subsidies by omitting general-tax revenues spent on transit. Many states and localities use non-road taxes to fund transit (e.g., extra 0.5% general sales tax in Los Angeles).

Comments for this entry have been closed
Return to top