August 30, 2007

Can Congestion Pricing Be Fair and Efficient?

Congestion pricing aims to cut urban traffic congestion and pollution by charging drivers entering downtown cores an access fee. Nicolas Stier, Nachum Sicherman and Eric Johnson weigh in on the process.



Nicolás Stier
Assistant Professor, Decision, Risk and Operations

Congestion pricing aims to encourage commuting decisions that maximize efficiency. But an optimized system isn’t everything. Equity should also be a top priority for congestion-pricing schemes. If such systems are well thought out, social welfare should also improve.

For example, a system focused overwhelmingly on efficiency might make the poor worse off, since they would have to pay more than they currently do (and can afford). It might also make life easier for the rich, since the fee is so little to them, and they can thus drive faster downtown.

The poor, on the other hand, who live far away, may not have any other option than driving into the city if their lower-rent neighborhoods offer few mass transit options. Or driving may be too expensive and they must make a time-consuming combined bus-subway commute.

Thus, when designing a congestion-pricing system it is important to quantify who can pay what, and analyze their alternatives. Ideally, over time, some solutions to improve equity could in part be financed from congestion-pricing proceeds.

If congestion pricing were done well, with toll revenue used for better mass transit in areas that are poorly served by subways or buses, the overall system would improve, while improving equity.

Reinvesting in roads rather than mass transit is the wrong approach. It gives a mixed signal to commuters. Tolls motivate drivers to switch to public transportation; road improvements encourage driving.

Correlating the amount paid in tolls to income could also improve equity. If the collection scheme records total payments per year, low-income drivers could be reimbursed after filing a tax return.

One should look to London — the textbook case for congestion charges — for lessons on fairness. People who live in London’s central business district get a 90 percent discount on the congestion tax just for living in “the zone.” This encourages people with cars to move into that area. For that reason, urban residents should be charged the same rate as those who live outside the city. On the other hand, if city residents are driving outside the zone against traffic, they shouldn’t have to pay a fee, because their presence in a less dense area imposes less of a cost.

Critical success factors to meet a congestion charge’s dual goals of improving efficiency and equity include the amount to be charged and the manner of implementation. If the charges are too high, everyone will be angry and boycott the system — and the poor, who have few options other than driving, will suffer the most.

On the other hand, setting the price too low means not enough people will switch from cars to mass transit, and the public transportation system will not improve. So, though drivers would pay, little congestion relief would follow. This scenario may be the most dangerous because a second opportunity to implement a congestion charge may not emerge if a scheme does not work in its first trial.

Nachum Sicherman
Professor, Finance and Economics

Congestion pricing makes sound economic sense. It builds on two economic concepts: peak load pricing, which means, “spread the load,” and negative externalities.

Peak load pricing is similar to lower-cost calls at certain hours. Just as phone calls may be more expensive between 9 a.m. and 5 p.m., thus encouraging more calls in the evening, higher peak-load pricing fees during rush hour might motivate drivers to shift their driving patterns accordingly. Deliveries, for example, might happen more frequently in the early mornings or after 7 p.m. because they would be less costly.

As for externalities, with cars come costs, such as pollution, noise and traffic jams. Drivers don’t take those into account. Congestion pricing fixes this by imposing an additional price that would correct a market failure. That price forces decision makers to consider the full consequences of their actions, and thus improves efficiency.

Congestion pricing is an old idea.  Nobel Prize–winning economics professor William Vickery of Columbia applied the concept to New York City, its suburbs and the subway as early as the 1950s. He suggested different prices, depending on location and time of day. He also proposed a sophisticated plan for an EZ Pass–type system years before others. His vision didn’t materialize then; perhaps it was too complex in a computer-free world. New York City Mayor Michael Bloomberg’s congestion-pricing proposal builds on these ideas. But it would charge just two prices: zero and one fixed rate.

Of course, the practice demands efficiency and practicality tradeoffs. And questions remain about the appropriate fee, and how to collect it. For example, since fewer people drive on a holiday or in bad weather, crossing beyond the toll border imposes no or little cost — and should mean no charge. What about moving a car from your home just one block over the border at 11 a.m.? Should you pay? Many “border” issues exist, and the border must be defined. But changing fees based on traffic flow would be too complicated. While the idea is good, the devil is in the details.

Eric J. Johnson
Norman Eig Professor of Business, Marketing

So, it remains unclear how or when — perhaps even if — New York City will adopt congestion pricing.

Amid heated debate and partisan politics, politicians in Albany and elsewhere may want to turn to behavioral economics for clues on how the new practice might pass muster and actually work. That’s because behavioral economics suggests two things about congestion pricing that standard economics might miss.

The first is that congestion pricing will run into some roadblocks. The second is that if it becomes a reality, people will get used to congestion pricing very quickly and might, as in London, end up liking it more than they thought.

Behavioral economics emphasizes fairness (or at least the perception of fairness) and studies how it weighs heavily in peoples’ economic reactions. For example, research shows that people will actually make themselves worse off to punish someone who was unfair. If congestion pricing is seen as unfair to a large number of people, it will have a hard time being accepted. In New York City, the Bloomberg administration anticipated this in its proposal and had already wisely excluded some people, like cabbies, from the charge, even though an $8 congestion charge might well hurt them much less than a 10-cent increase in the price of gas. 

But in New York City, this has been the argument most used by the opposition, which perhaps led to the proposal’s recent setback: people were worried about charges in cases where it would seem unfair, like paying the charge for trips to the hospital. The major argument advanced by many opponents was that the plan would be unfair to most major neighborhoods brushing Manhattan, including Queens and the Bronx. But the Bloomberg administration should also emphasize how the plan, through improvements in mass transit, would help ordinary New Yorkers. Otherwise people could be offended by the image of a limousine-driven Manhattan executive paying the same charges as cleaning staff driving in from the outer boroughs. Emphasizing how mass transit will be improved outside Manhattan might help win over those initially opposed to the program.

Why would people be surprised by how much they like congestion charges? Another basic difference between traditional economics, with its rational expectations, and behavioral economics is that in many cases people have a hard time predicting how they will feel about future situations. For example, research shows that people adapt to negative events (for academics, one dire event is not getting tenure) much more quickly than they predict. Think about the ban on smoking in restaurants and bars: pundits offered dire predictions of the end of New York nightlife, but these predictions did not come to pass. Similar arguments about economic outcomes are made for congestion pricing.

In introducing the smoking ban, the Bloomberg administration ignored polls and did what most people found out was the right thing. Polls might not always be good predictors of what people really want. Similarly, I suspect people will adapt to the congestion charge and learn to enjoy its perhaps unanticipated benefits: fewer clogged roads, a cleaner atmosphere, quicker entry and exit into the city and more funds for mass transit. If this is true, ignoring the polls and correctly anticipating public reactions might be called a new kind of political leadership.

Have You Read Columbia Business?

Columbia Business School’s alumni magazine connects alumni with each other and the School; celebrates alumni milestones and accomplishments; and chronicles the impact of Columbia Business School alumni, faculty members, and students on the global business landscape.

Read Columbia Business>