Finance Infrastructure London Paris

An Alternative to Congestion Pricing: Roadway Traffic Restraint

» Comparing the approaches taken by Paris and London suggests that to ease traffic U.S. cities can attempt other, more politically palatable solutions than pricing.

When it comes to transportation economists, there’s pretty much one answer to every problem: Equate pricing of all modes with their greater societal impacts. In general, this means that we (in the U.S.) ought to be charging drivers more to make up for the negative effects they have on the environment and the roadway infrastructure, and that we ought to be increasing subsidies to encourage people to take transit.

This approach could be implemented in a variety of ways depending on location, but one model that has been particularly appealing to planners interested in reducing the perceived negative economic and social effects of traffic has been that of London, which in 2003 implemented a congestion charge on drivers entering its central business district. Revenues from the program went to increasing transit service. The method, unsurprisingly, has been a major success in terms of reducing traffic: Between 2002 and 2007, overall car movements in the district decreased by 39%. Meanwhile, travel on public transportation increased correspondingly over the same time period: By 24% on commuter railways, 16% on the Underground, and 18% on buses.

These are excellent results and the effects have been overwhelmingly positive for commuters and residents of London’s central areas.

But what if congestion charging is just too much of a hot topic for even progressive American cities to handle? The effort to instate a similar system in New York City in 2008 was so thoroughly brought to its feet that it is hard to imagine wanting to repeat the fight.

Yet there’s an alternative, and it may prove just as productive if the goal is to reduce traffic: Paris’ systematic engagement to make it harder to drive in the city. The French capital has proceeded in a manner far different from that of London, choosing to avoid paid penalties on drivers in order to prevent the further development of the already-existing sense that the City of Paris is attempting to isolate itself from its suburbs, which are already cut off by a ring road. 40% of drivers within the city’s borders are inhabitants of the surrounding areas.

As a result, the administration of Mayor Bertrand Delanoë has since 2001 prioritized the creation of bicycle, bus, and tramway infrastructure along with the reduction of vehicle lanes along both major boulevards and side streets. Huge sections of the city have been designated 30 km/h zones and biking is now allowed in both directions on most streets, even those that are one-way for automobiles. Free parking has been mostly eliminated. This spring, the city reinforced its efforts to commit far more street space to biking and expand that mode’s travel share.

Streetsblog’s Ben Fried provided an excellent overview of the city’s program in April 2008.

Looking back, the results have been astonishing: Even with no direct financial reason to abandon driving, the city saw a 17% decrease in driving between 2002 and 2007, a trend that is continuing (according to the most recent information, it may now be 24%). In the same time period, travel on the regional rail network increased by 16%, by 8% on the Metro, and by 2% on buses in the city. Weekend traffic has seen the most significant gains. This has reduced further the already extremely low share of overall commutes made by car or motorcycle in the city: Just 16.3% in 2008. In the near suburbs, the equivalent statistic is 40.2%, though those areas are soon to be better connected by a system of tramways and bus-only routes (and eventually by a massive circumferential metro).

Paris’ accomplishment, though not as large in percentage change as London’s, was arguably more significant since it affected the entire city of 41 square miles, versus the original eight square miles of the London congestion zone (later roughly doubled).

Moreover, these statistics fly in the face of the commonly-cited idea that “congestion pricing is the best way, and perhaps the only way, to reduce traffic congestion,” to quote transportation policy experts David King, Michael Manville, and Donald Shoup. For cities truly concerned about finding ways to limit the number of cars traveling down the street, whatever the purpose, this example demonstrates that a concerted effort to get cars off the street by limiting the space available to them can be an effective technique.

There are, of course, dissenters who make the argument that the Parisian approach limits economic productivity and results in a “decrease in mobility” because car drivers no longer are able to move as easily as they once were. That interpretation, however, is based on the fact that overall passenger-kilometers have decreased; yet that statistic favors trip distance thereby discounting the value of, say, walking to the neighborhood store — an essential trip for people living in an urban place. Also, economic discussions focused on “mobility” fail to reflect the fact that inhabitants of neighborhoods with fewer cars benefit significantly in terms of quality of life.

Arguments that suggest that bus ridership has not gained enough passengers to reflect the decrease in car traffic do have some merit, though there is no doubt that certain interventions, such as the installation of a new tramway along the southern edge of the city limits, have significantly increased public transport use.

The major failing of Paris’ approach is that it does not guarantee a new revenue source for the public transportation system. Whereas London was able to use its congestion charge to reinforce spending on its local bus system, Paris has had to continue relying on other funds to ensure the increase in services provided on increasingly packed buses and trains. Even so, that may be a compromise worth considering for other cities wanting fewer cars without the political nightmare that is congestion pricing.

Image above: A bus in Paris with policemen on bikes, from Flickr user Daniel Lobo (cc)

36 replies on “An Alternative to Congestion Pricing: Roadway Traffic Restraint”

I’m not as hopeful about the Paris model. Bloomberg’s introduction of road diets and bike lanes has been just as controversial, and is still unpopular in most of the city. In addition, the elites are against it, as the negative descriptions on the New York Times would attest, whereas they have been neutral or positive on congestion pricing.

It’s been more successful only because the city has absolute power over road diets, whereas to implement congestion pricing Bloomberg needed support from the state legislature. To the extent that Bloomberg has needed state legislative authorization for his new plans, he has not obtained it. Even something as straightforward as mounting cameras on buses to help enforce bus lanes got stuck in legislative hell; the final compromise is that bus cameras will be allowed only on the six chosen Select Bus Service routes.

When it comes to transportation economists, there’s pretty much one answer to every problem: Equate pricing of all modes with their greater societal impacts.

Isn’t this putting the cart before the horse? Should we really be trying to correct externalities when we haven’t even pegged transportation to its accounting and opportunity costs – a.k.a., put it to the market test? This is one of the reasons why conservatives and libertarians are so quick to decry pro-transit schemes – because pro-transit/density people so often speak about correcting externalities and correcting the market without even mentioning that the basic costs for roads aren’t met. Externalities are vague and debatable, so I don’t understand why urban planners always focus on them.

The cynic inside of me says that it’s because many urban planners don’t even understand the interventions in the market, and they actually think that sprawl, roads, cars, and the status quo are relatively laissez-faire outcomes. Our blog ( tries to dispell this notion, but I fear that it’s still relatively widespread.

I did notice that, and I was going to comment but for some reason didn’t. I’d actually say the tendency to look after externalities before looking at basic cost mismatches is actually more pronounced among planners than among economists. From my perspective, planners appear far more interested in externality-solving policies like inclusionary zoning and green technology than basic liberalization in terms of things like parking minimums and upzoning. I have no doubt this is more driven by political possibilities than anything else – planners would probably be more interested in parking and zoning if advocating them weren’t so unpopular, but voters are much more receptive to overt greenness and affordability than they are in the inherent greenness and affordability of market development.

Anyway, I’ll stop ranting and hogging all your comment space, but that’s just my two cents.

I would argue that congestion pricing is the easiest/simplest way to reduce car use, but that doesn’t make it the best.

While externalities are vague and debatable, they often far exceed the market price. For example: to fully fund the national, state, and local highway systems, the US would need to raise the gas tax by about 50 cents a gallon. This pales in comparison to external costs of driving. The consensus estimate of the proper carbon tax corresponds to $1 per gallon; the range of estimates for a proper pollution tax starts at 50 cents and goes up to about $5, with multiple reputable sources pegging it about $2; the cost of car accidents is $175 billion per year, which works out to $1.20 for every gallon of gas consumed.

I guess I would have two things to say to that. The first is that liberalizing zoning/parking regulations are a big part of moving to market costs. The second is that while the extra accounting costs of roads aren’t that large (although I do question where you came up with your figures for local road spending), the extra opportunity costs are enormous, and they’re highest in dense urban centers and lowest in rural areas.

You’re right about extra opportunity costs, of course, as well as about parking. But just road construction and maintenance runs at an okay-by-US-standards recovery ratio, of between 51% and 67%, depending on what you count. The underlying data is here; the denominator you should think of is total receipts for road spending plus gas tax collection costs, and the numerator is either total gas tax and toll receipts or total gas tax and toll receipts minus gas tax diversions to non-highway uses.

The number for local governments is severely understated in that chart, because it’s only talking about highways (which in the US includes with “Main Street,” but not the very large amount of smaller streets) – what about all the local roads and police enforcement? For a while I actually thought these figures included that, but then Randal O’Toole admitted to me in an email correspondence that they don’t cover local roads, which are basically financed almost entirely out of general revenues.

I should also add, in your argument’s favor, that the total price of gas and automobiles should by all rights be factored into the equation as well. It seems unfair to leave them out. In the end, if you want to be totally honest, you have to admit that computing the various numerical taxes and subsidies on the various modes of transportation is a fool’s errand. In the end, you just have to learn as much as you can about it and go with your gut.

Okay, that I didn’t know. Thanks for the information. I suspected so at first, but then thought that “local roads” covered all local roads. Do you have a link for it I could reference in the future?

I don’t actually think the base price of cars should be included; we’re talking about public costs, not private costs. By analogy, if riding the subway requires tapping a credit or debit card, as New York is proposing, then the cost of getting a credit card should not count toward either user fees or expenses.

On the other hand, the private costs, especially the base cost of gas, could be a useful indicator of what effect a subsidy elimination would have.

First, I have to echo Alon in questioning the idea that taking away space from cars is all that politically palatable.

Otherwise, I think you’re attacking a hell of a straw man. Congestion charging is, to my mind (and probably to King, Manville, & Shoup), about using prices to decrease congestion on roads. This is in order to make them usable as transportation, to eliminate the waste of time stuck in traffic, and perhaps to provide a revenue source for (other) transportation infrastructure. But car traffic is not tantamount to congestion.

If you view the purpose as decreasing the single-occupancy-vehicle modal share, then just call it a driving charge. Viewing it that way, sure, you can make driving harder in various ways besides price and you can also improve the other options. But that’s not going to do much about congestion on the remaining pieces of road — because that’s not the problem you’re concerning yourself with.

I don’t think it is terribly helpful to view approaches as being more or less successfully anti-car. What do they do for people trying to get somewhere? Congestion charges can be win-win: they make the road passable for driving, and they improve the other options so that you don’t have to rely on the road. Redistribution of road space would similarly need to be explained on the grounds of improvement in transportation choices. If it’s just a way of decreasing traffic, it’s not ever going to be palatable.

given that manhattan has bridges and tunnels on all 4 sides of the island it would seem easier to toll all of the bridges that arent already tolled, it would almost be the same as congestion pricing.

i’ve been a fan of having dedicated lanes on all the avenues and major cross streets in manhattan exclusively for buses/emergency vehicles plus separated bike lanes elsewhere on the street and let the private vehicles fight each other over the remaining lanes.

The avenues in Manhattan already have dedicated emergency vehicle lanes. Almost no one pays attention to them unless there’s an emergency vehicle with a siren blaring on the block and then not always.

but combine them with bus lanes then physically separate them with medians and for gods sake have some strictly enforced fines for other vehicles using them

Choose your battle, I suppose: with congestion pricing you face anti-tax (or simply anti-city) politicians; with willful vehicle capacity reduction you face the civil engineering staff of local governments. The latter may not make direct policy and funding decisions, but they tend to capture the ears of the former and are usually really good at grandstanding.

Keep in mind that I am excepting the outlier cities in the US– New York, San Francisco, Portland and Washington aren’t quite like everywhere else, and thanks to the still-dominant mid-century culture of automotive freedom nearly every other city is beholden to an ideal of suburban-standard mobility preservation at seemingly any cost.

To me, making real progress on this issue relies on a logical link to fiscal sustainability, especially for the local street obligations of local governments. Prosperous cities better themselves through growth and expansion of the tax base, but they cannot sustain steadily higher costs of right-of-way acquisition, road construction and increases to their maintenance obligation to accommodate it in the long term. There must be an acknowledgement of a lower-cost, more balanced approach to increasing system capacity and that transit and bike/ped infrastructure need appropriate levels of funding commitment to achieve it.

Yonah, thank you for posting. I wonder if ideas like scheduling street and road work for weekday daytimes (when alternate bus and rail services are most frequent), removing peak-direction lane flexing, and ceasing to deploy traffic agents to control traffic near bridges (I am thinking of the platoon-sized element of traffic cops that is deployed every afternoon to Delancey Street, NYC), could be added to your list of capacity-reducing interventions.

There are platoons of Brownies out because before there were Brownies Manhattan frequently gridlocked. Too bad for the private vehicles but it stopped the emergency vehicles from getting through. Brownies are a necessity, unless you want fires to burn out of control and people to die waiting for an ambulance.

Jonathan, the purpose of road diets isn’t just to cut auto capacity. It’s to increase livability by promoting alternatives. Taking lanes from roadways allows installing bike lanes, wider sidewalks, or pedestrian plazas. Removing freeways allows people access to the waterfront or to severed neighborhoods. Rearrangement of signal optimization, which has so far not been done but should be, allows buses to stop at red lights less.

In contrast, just making traffic worse without a gain elsewhere does not work too well. It has the same effect on car travel as congestion pricing but does not generate revenue, correct the externalities, or reduce congestion for people who are in a hurry.

Let’s be sure during this conversation to not equate reducing speed with reducing road capacity. When I cross a street like 1st Avenue in Manhattan, my first thought is, wow, this street is completely empty. That is, until a group of cars rides the green wave down the street. Then the street is generally devoid of traffic (besides those turning from side streets) until the next wave.

What traffic calming does is more evenly distribute cars along the road, so that you can have narrower roads, slower roads. It does not reduce travel capacity, per se. Obstacles like traffic lights do a much better job of reducing capacity on a road than do bus lanes.

Road diets are increasingly the rage across US cities. Projects often involve reducing lanes for parking, bike lanes, refuge islands and curb extensions. They work well, when located in areas with strong connectivity, transit options, and walkable land uses.

Even fairly auto-oriented cities like Charlotte have seen a huge increase in road-diet projects. Not surprisingly, Charlotte’s projects are largely located within its relatively older urban core areas.

One of the advantage of road diets is that they are not all-or-nothing in the way congestion charging is. If people kick up a fuss about a particular street, you can drop it or scale it back without really affectign the tesrt of the project. Plus it can be rolled incrementally, starting with the easy wins, which is easier on the budget.

The major failing of Paris’ approach is that it does not guarantee a new revenue source for the public transportation system. Whereas London was able to use its congestion charge to reinforce spending on its local bus system, Paris has had to continue relying on other funds to ensure the increase in services provided on increasingly packed buses and trains. Even so, that may be a compromise worth considering for other cities wanting fewer cars without the political nightmare that is congestion pricing.

This does not compute. If ridership is increasing (as you say, regional rail by 16%, Metro 8% Buses 2%), then funding is increasing. If your statement is true, then the failure is the pricing system for transit…not the lack of a “new” funding source.

Unfortunately, using Paris for a comparison with virtually any American city is a non-starter. Many of the ideas are sound, but the street life there is vastly different from what we experience in daily life over on this side of the Atlantic. Paris is able to absorb a lot more trips into walking, so it’s okay to have the car VMT drop by 24% and only have transit trips rise half as much. Transit in the U.S. is much more limited in flexibility and our zoning methods have destroyed the walkability of many old neighborhoods and made many newer towns completely messed up.

Slow down cars, speed up transit. Narrow car lanes but add bike lanes. Add parking to streets where it doesn’t exist, but get rid of extra surface lots (and maybe a few parking ramps too). These changes need to be backed up by rezoning in better ways to put people closer to the businesses they need, and by improving the street grid when land is redeveloped rather than taking roads away. The pedestrian experience also needs to be improved — I’m amazed by the permeability of some old cities, where narrow alleys and walkways can appear every 100 feet or even less. Going less than that is likely overkill, but many of the buildings we have these days are way too monolithic.

Looking around my region, the area I enjoyed walking around the most was probably the University of Minnesota campus. They’ve got plenty of screwed up stuff mostly done after 1960, but its core was designed by people who had to walk everywhere. With a highly mobile daytime population over 100,000 crammed into a few square miles, it’s no surprise they’ve got the second-largest bus system in the state, carrying 3.9 million riders last year — and for only about $1.25 a head (compare with $3.20 per head for the main regional provider, Metro Transit). It’s basically too cheap to bother with fares, so they don’t.

Of course, a college suffers from having huge population shifts each year (and each day, for that matter). I’d like to see cities built more like our colleges — if we can fit people together without having them commute long distances or moving in and out every year, it’d work out great. I hope, anyway…

In Toronto, there’s been a lot of kneejerk pushback against such measures. The introduction of bike lanes on Church St and the attempted introduction of bike lanes on University Ave triggered a ridiculous amount of irrational backlash.

In the current mayoral campaign, there are major candidates that actually want to get rid of all bike lanes on major streets (i.e. the ones that go to useful places). There’s another one that wants to abolish speed bumps on residential streets.

As a result, I am more optimistic about the prospects for congestion pricing than for this.

In Los Angeles, a significant percentage of traffic comes from student commuters to the universities. If they just started classes at noon instead of 8am, that would eliminate a good 30% of the traffic. Traffic always gets much worse when school is in session. Then provide significant incentives for all businesses to encourage employees to telecommute. It seems so easy and obvious.

Children growing up in suburbs are clueless about the ecological/cultural advantages of cities. Perhaps an introduction to a more critical analysis of the possibilities of city life should be a part of the curriculum at the appropriate level of the students. Even a study of world urban geography would provide the external references which might permit a more realistic view of possibilities. Americans are unimaginative when it comes to alternatives to automobile culture and its true social, cultural and economic costs.

Except for the Americans who live car free or in one car households….. All those people living in places like New York, Chicago and San Francisco are Americans.

“. Americans are unimaginative when it comes to alternatives to automobile culture and its true social, cultural and economic costs.”

Ah yes, the classic “they don’t agree with me because they’re dumb” argument.

Its more often ignorance than stupidity, and the much larger position of the oil industry in the US than in Europe allows for the investment in quite a bit of useful ignorance on the part of Americans.

Safety as opposed to livability. As the nice guys at have cried, What metrics do you use for livability? Indeed, we can assume that the streets of Paris have become much safer for all users since the retrofit, and much safer relative to benefits from London’s Congestion Charge.

I lived in Paris twice for 6-month periods, once in 2005 and later in 2008. The difference was astounding. Paris now has QualityBus/BRT on nearly ever route, and it has bike lanes. Often the bus and bike travel in the same, wider-than-normal-bus-lane lane. Bike travel has exploded in Paris, and this deserves to be compared with London. It may explain the lower-than-expected increase in bus use.

Isn’t eliminating free parking an indirect form of congestion pricing? With paid parking, the people the drive in for the morning peak and leave in the evening peak ought to be paying more for parking than people driving in for a shorter period or in the evening.

Although Paris still relying on other funds to ensure the increase in services provided on increasingly packed buses and trains their economy has boosting along the way and stay competitive.

Leave a Reply to Alon Levy Cancel reply