Felix Salmon

How should bike-share services pay for themselves? Up until now, the main model has been sponsorship and advertising. But CityRyde has a bright idea: why not sell carbon offsets?

The idea’s pretty simple: as bike-share use rises, the amount of carbon-emitting vehicle use falls. So bike shares save carbon; CityRyde even has a methodology to determine exactly how much. (One thing I’d like to see, though: virtually all bike-share programs involve trucking bikes from the center of town back into the periphery, not to mention transporting broken bikes to be fixed. So somewhere in the methodology there should be an accounting for the amount of carbon emitted by the bike-share program itself.)

In any case, if the bike-share program sold carbon offsets to companies which want to claim to be carbon-neutral and to individuals wanting to offset their carbon emissions, that could raise some revenue: CityRyde co-founder Jason Meinzer told me his rule of thumb is that you could bring in between $25 and $100 per bike per year that way. For a scheme with 50,000 bikes in New York City, that would equate to between $1.25 million and $5 million per year: hardly chump change.

Meinzer didn’t share with me exactly how he got his numbers, though, so I ran a smell test. Let’s say each bike travels 15 miles per day, 350 days per year: that’s 5,250 miles per year. A lot of bike rides are simply for pleasure, and others—especially in a city like New York—replace walking or taking mass transport. Those are activities with negligible marginal carbon emissions.

But let’s say that 1/3 of bike journeys would otherwise have been taken in a car of some description. That means that each bike saves 1,750 passenger-miles in cars. If one car carries the same number of people as two bikes, on average, then that’s 875 car miles saved. At 1.2 pounds of CO2 per mile, that’s basically half a ton of CO2 emissions saved per bike per year. And while the market in carbon offsets is far from transparent, my feeling is that you’d be lucky to get $5 per ton, which would equate to $2.50 per bike per year. That’s a full order of magnitude lower than Meinzer’s lower estimate.

Meinzer’s methodology is a lot more sophisticated than that, and I’ll update this post if he wants to share his own math. But at $5 per ton, selling carbon offsets would gross only about $125,000 a year—which, by the time you subtract the cost of measuring the carbon saved and administering the sales, leaves you with little or nothing in net revenue. So while it’s an intriguing idea, I’m not yet convinced it’s a practical one.

Update: Meinzer says that he does account for the carbon costs of the program, in the “Project Emissions” and “Leakage” sections of his methodology; I don’t see it myself. And he explains that he gets his much higher estimate for total revenues from selling carbon offsets at a much higher price:

Our credits most certainly will be sold at a premium due to the novel co-benefits associated with their generation even outside of the carbon mitigated; e.g. health and social (and remember some credits sold for as high as $111 last year). This has been validated by the carbon brokers we’ve been working with over the years. Moreover, outside of the “price-point” per carbon a key angle we are taking to obtain an even higher premium on our credits is via creative bundling; by lumping the carbon credits w/ the sponsorship and advertising. Case in point – Blue Cross Blue Shield donated $1.5 million to have their name tied to the existing 1,000-bike Minneapolisbike share. Had they been able to purchase the offsets stemming from that program the donation would have been MUCH higher. This argument is reinforced by the fact that this donor in particular clearly has a key interest in health, and so the aforementioned co-benefits of our credits would prove even more attractive given the health-benefits of biking. Most big names companies are already offsetting their carbon emissions each year anyways for a variety of reasons, even outside of a regulatory mandate.

Mark Perry is convinced that the recent uptick in vehicle miles is a good sign, economically speaking; Calculated Risk is not as convinced. Both, however, are working on the assumption that vehicle miles are an excellent proxy for economic activity as a whole, and that the more they rise, the better the economy is doing.

Perry’s chart, in particular, would seem to back that up:

The way in which vehicle miles fell steadily over the course of the recession is startling. But look at CR’s chart:

And suddenly recessions don’t seem as big of a deal: vehicle miles simply tend to rise over time, except for during oil spikes.

It’s worth remembering here that the recession started in December 2007, while oil prices were still rising; they didn’t reach their all-time (nominal) high until July 2008. Given that gas prices lag oil prices, a large part of the fall in miles can probably simply be attributed to high gas prices, rather than to the recession — especially since, as Nate Silver notes, “the cost of gas twelve months ago has historically been a much better predictor of driving behavior than the cost of gas today.”

More generally, vehicle miles are a cost of economic activity, and to the extent that they can be minimized through various kinds of efficiency gains, they should be. Things which are good for a vibrant economy — mass transit systems, telecommuting, e-commerce, walkable neighborhoods — tend to mitigate against driving, while — to take the extreme counterexample — I’d guess that people who have been foreclosed upon tend to spend a lot more time in their cars.

My feeling is that what we’re seeing in the latest driving numbers is no more than the fact that gas prices were low a year ago. I do hold out some hope that we’re decreasing our national reliance on autos, if only a little bit; it would be sad if any economic recovery had to be associated with a concomitant rise in driving. As America moves back into the cities from the crumbling suburbs, is that really too much to hope?

(HT: AR)

Rachel Brown has a fantastic little 5-minute film about biking up First Avenue to work:

I love the way that she’s caught on camera all of the annoyances which drive bike commuters mad: the cars cutting across the bike lane to make left turns; the pedestrians blithely stepping out into the lane in front of you; the trucks using the lane as a parking spot; the taxis driving up it. And, of course, the Evil Bike Salmon.

At the same time, there’s more than a hint of tension, in this film, between relatively serious bike commuters, on the one hand, and slow hobbyists, on the other. And this tension, I think, is likely to get worse rather than better, even as the other problems might alleviate themselves somewhat as the number of cyclists in New York grows.

There’s safety in numbers, when it comes to cycling, and a similar phenomenon is likely to happen with regard to pedestrians and car drivers being increasingly conscious of bicyclists in their midst. Already, the First Avenue bike lane has reportedly cut injuries to all street users by 50%. But as the number of cyclists rises, the average speed of cyclists necessarily falls. Everybody thinks of northern European cities like Copenhagen as bicycling paradises — and they are. But if you’re biking around Copenhagen, you’re going to go a lot more slowly than if you’re biking the same distance in NYC.

A slow cyclist can cope with most of the dangers and obstructions that Brown complains about much more easily than a fast cyclist — and the fast cyclists, as Brown’s film shows, are now shunning the lane entirely, moving over to the right-hand side of the street, where they’re much less likely to get cut off by a car. (Cars often turn left off First Avenue, which runs up the east side of Manhattan, but much more rarely turn right.)

It’s going to be very interesting to see how fast cyclists cope with an influx of slower cyclists in Manhattan, as bike lanes continue to get built and average bike speeds continue to decline. I love to zoom down avenues at high speed, but I also love being safe. Maybe that means I’m just going to have to start going a little slower.

Many thanks to Joe Nocera for raising the issue of One World Trade Center’s finances. It’s by far the tallest and most expensive building that New York has ever seen, and it’s no thing of beauty, either. Plus, there’s not nearly enough demand for new top-grade office space to justify building so much of it at this location and at this time. So what exactly is the Port Authority thinking?

All that said, the issue is much more complicated than Nocera makes out. For one thing, the 1,776-foot tower is really the last vestige of the once-lauded Daniel Libeskind master plan for the World Trade Center site; for another, the deal that gave it to the Port Authority was a highly complex one, done with developer Larry Silverstein, and so it’s a little bit simplistic to try to view One World Trade in a vacuum, as Nocera does.

Plus, Nocera’s very vague about sourcing his numbers: he says only that “my real estate sources say they believe that the Port Authority will need to charge $130 a square foot to break even on the building”, and then adds a pro-forma Port Authority denial.

It would be very useful to learn where that number comes from. Looking at the figures in the piece, the cost of the building is $3.3 billion, with $1 billion of that coming from insurance proceeds. I’m not sure exactly what Nocera means by “break even”, but he does talk earlier on in the piece about “any shortfall between the building’s annual rental income and its carrying costs”, so let’s think about it that way.

The building will end up with 2.6 million square feet; if the breakeven rate is $130 a square foot, then that implies its carrying costs will be $338 million a year. But it doesn’t cost anything like that for the Port Authority to borrow $2.3 billion. After all, the last time the Port Authority issued bonds, it paid an average interest rate of less than 4.5%. And 4.5% of $2.3 billion is barely more than $100 million a year — less than a third of the number implied by Nocera.

Or think about it in terms of a standard residential mortgage. Let’s say you wanted to take out a 30-year fixed-rate loan on a $3.3 billion home, putting $1 billion down. Right now, mortgage rates are 4.5%, which implies a monthly repayment of $11.65 million per month, or $140 million a year. OK, you’re not going to be able to borrow that kind of money from your local credit union, and I’m pretty sure that the note would be non-conforming in the eyes of Fannie and Freddie. But still, if you want to get $140 million a year from renting out a 2.6 million square foot building, then you only need to charge $54 a square foot: a far cry from Nocera’s $130 figure.

Yes, there will be high maintenance costs, especially given all the extra security. But at the same time, the Port Authority owns the land underneath the building outright, so there are no costs associated with that. And in the early years of the project, when the building isn’t fully rented out, the Port Authority will have to carry some of the costs of the empty space.

On the other hand, there are less quantifiable costs to having empty space in that part of the New York skyline which used to be home to the Twin Towers. One World Trade Center might never be as iconic as they were, but it will still be an instant landmark, and a vast improvement on the gaping hole that we’ve been living with these past nine years. If Nocera wants to make the case that its costs will end up being borne by commuters crossing the George Washington Bridge, he’s going to have to be a lot more specific about exactly how he’s calculating them.

If you wanted proof that New Yorkers think of bicyclists more as pedestrians than as vehicles, all you need to do is look at this graphic in the NYT, which shows how Broadway is used between 59th Street and 17th Street. The lanes are labeled with only two colors: orange and green. Orange is vehicles: dotted means parked cars, while solid means they’re moving. Green is, well, pedestrians, or that conceptual combination of pedestrians-and-bicyclists: dotted means on foot, while solid means they’re moving, ie they’re on a bicycle.

The story itself — not to mention the headline on the graphic — is very car-centric, as Aaron Naparstek has been pointing out on his Twitter feed this morning. “For the first time in New York’s modern era,” writes Michael Grynbaum, “Broadway no longer offers a continuous path from the Bronx to the Battery.” That isn’t true, of course, just as it isn’t true that Broadway is any narrower now than it was in the past. Those things are only true if you’re looking at the road from the point of view of the minority of people who navigate it by car, as opposed to the majority of people who navigate it by bike or on foot.

It’s hard to convey the overall tone of the piece with a few choice quotes; you really have to read the whole thing, with its absence of any quotes from bicyclists or pedestrians, and its framing of traffic reduction on Broadway as a war between drivers and faceless “transportation officials”. You can get a feel, though, just from the first word of the second paragraph:

It is Manhattan’s most famous thoroughfare, known around the world for its theater marquees and giant Macy’s. It has come to symbolize the outsize aspirations and swagger of New York.

But…

In Grynbaum’s world, it seems, a road with “outsize aspirations and swagger” must be full of as many cars as possible; if it’s humming with pedestrian life, that somehow diminishes it.

And it’s weird to talk about how “moving traffic is down to a trickle” on Broadway below 34th Street without pointing out that the street begins anew there: of course there’s only a trickle of traffic, because at that point it’s a local street which you can only get to by first going west on 33rd Street and then doing a very sharp left turn, almost back on yourself, onto Broadway. There’s no point in having more traffic capacity on Broadway than there is on 33rd Street, because there’s nowhere else that traffic can come from.

The graphic does a good job, though, in showing the difference between successful and unsuccessful bike lanes on Broadway. Here’s a relatively sensible style, as seen around 22nd Street:

The pedestrian zone is an extension of the sidewalk, while cyclists get their own lane alongside other vehicles.

Here, by contrast, is the unsuccessful style, as seen around 40th Street:

Here, the pedestrian zone is particularly wide and pleasant, but it’s separated from the sidewalk by a bike lane. It’s only natural for pedestrians to want to cross naturally in and out of the pedestrian zone, and they’re obviously not going to do so across the road. Instead, they’ll wander across and along the bike lane, most likely without checking for oncoming bikes first. In fact, given half a chance, they’ll even move chairs into the middle of the bike lane, and sit on them. Given that traffic on this part of Broadway is pretty light, bicyclists find it easier, and much safer, to ride in the roadway rather than in the bike lane provided for them.

It’s a shame that Grynbaum seems not to have spoken to any pedestrians or cyclists when reporting his story. He might have got a very different perspective on the successes and failures of the pedestrianization scheme, and might have at least mentioned the way in which the Broadway bike lane dumps cyclists out into very hard-to-navigate Union Square traffic the minute it hits 17th Street. Instead, we get this:

Many drivers remain hostile to what some say has amounted to a tacit decommissioning of Broadway as a major thoroughfare. The street is increasingly shunned by drivers. Compared with a year ago, the number of vehicles using Broadway between Columbus Circle and Times Square has gone down about 25 percent, the city says. And in the morning rush, traffic on Broadway passing 23rd Street has fallen 30 percent since 2008.

“I know they’re trying to beautify the city, but it’s killing the drivers,” said Gus Salcedo, 40, a daily car commuter from Queens who was parked on Broadway at 33rd Street the other day. “It’s frustrating. They don’t want you to drive into the city.”

Memo to the NYT: there’s more than one way that a road can be “a major thoroughfare”. And the current way is much more successful than the status quo ante. Even for Mr Salcedo, who truth be told probably finds it easier to find his parking spot now than he did when car traffic on Broadway ran painfully across Sixth Avenue, and the corner of Broadway and 33rd Street was a nightmare not only for bikes and pedestrians, but for car drivers too.

How to get drivers to help pay for the direct costs and negative externalities they cause? A congestion charge is one obvious way, but it’s expensive and complicated to implement. A driveway tax isn’t quite as elegant, but it’s much simpler, and it seems to be catching on even outside Oregon: it’s just been implemented by Mission, Kansas.

It’s quite simple, in theory: a set of formulas is used to work out how much traffic any given property generates. A single-family home, for example, generates about 9.5 trips per day, and will pay a $72 tax; Target generates 8,500 trips per day, and will pay $64,750. All of which works out to a tax of 2 cents per trip.

There are two problems with this scheme, beyond the entirely predictable pushback it’s getting from people who will end up paying new taxes. A tax of 2 cents per trip isn’t remotely enough to change behavior; and the tax doesn’t actually encourage people to change their behavior in any case: if you sell all your cars and go everywhere by bike or foot or public transport, you still pay the same amount of tax as the five-car family next door.

But anything which moves us away from the costly world of free parking has got to be a good thing, especially if some of the proceeds are used to pay for an express bus service. And even drivers might eventually come around to liking these taxes, if and when they start reducing traffic jams and congestion.

Beijing has a 62-mile traffic jam which is currently moving at one third of a mile per hour, and which is likely to last until mid-September. And that’s not even the really scary bit:

The mega-jam on the city outskirts comes as officials warn that downtown traffic in Beijing is steadily worsening. State media on Tuesday reported that average driving speeds in the capital could drop below nine miles an hour if residents keep buying at current rates of 2,000 new cars a day.

Here in New York, we literally haven’t seen rush-hour driving speeds of nine miles an hour in living memory. The average speed during the morning rush between 6am and 9am is 7.03mph; in the six-hour-long afternoon rush between 2pm and 8pm, it’s just 6.78mph.

The answer in both cities, of course, is the combination of better public transport with congestion charging. In Beijing, that’s going to happen. In New York, I’m not holding my breath.

Remember the ShoreBank rescue, back in May? Well, it got lots of headlines at the time, but it didn’t pan out in the end, and now ShoreBank has failed. The FDIC’s deposit insurance fund is taking a $367.7 million loss, and the money which was going to be invested in ShoreBank by Goldman, JP Morgan, Citigroup and others is now going to be invested in ShoreBank’s successor institution, Urban Partnership Bank. UPB will have a whole new management team, led by former Bank One executive William Farrow — something which rather puts the lie to conspiracy theories which said that Goldman et al were only investing in ShoreBank because its CEO was a friend of Barack Obama’s.

This clean-sweep approach makes a certain amount of sense: it’s right that ShoreBank’s shareholders should be wiped out when it managed to lose so much money. But it’s interesting to me that the government, given the choice between losing $368 million of the Deposit Insurance Fund or investing an extra $75 million in bailout funds, chose the former option. The deposit insurance fund, I guess, isn’t really considered taxpayer money, and will ultimately (eventually, hopefully) be repaid with future insurance premiums.

I wish Urban Partnership Bank well, and look forward to it proving that community-based urban lending can not only perform a crucial social function but can also be reasonably profitable. It’s sad that ShoreBank failed, but the main reason for the failure seems to have less to do with its community lending and more to do with its overexposure to speculative commercial real estate ventures. Urban Partnership Bank, I trust, will stick to its core competency, and do well for all concerned by doing so.

SoBi, short for social bicycles, is a gutsy startup trying to make bike-sharing more accessible by reducing the up-front infrastructure costs. Don’t create dedicated parking spaces: allow bikes to be parked anywhere. Don’t commission custom bicycles: design a gizmo which attaches easily to any bike instead. And store information in the cloud, allowing what I think is the smartest new idea of all: rather than paying contractors to move hundreds of bicycles every night from where they’re not wanted to where they are wanted, simply pay the bike riders themselves a bonus of a buck or two if they leave the bike in an area where they’re in high demand.

But I still don’t think this’ll work, and not just because SoBi was silly enough to demonstrate their first real-life prototype on a fixed-gear bike with no rear brake. Something less suitable for sharing can hardly be imagined. People come in a large range of shapes and sizes, and need to be able to adjust their bikes easily when they find them: that’s easier said than done. Standardized bikes also make repairs a great deal easier and cheaper: being able to install this system on any old bike is not much of an advantage, really. And more generally, it’s pretty hard to find suitable bike parking if your only option for locking the bike is to find a rack which you can attach the bike to from one position over the real wheel.

The big problem, however, is in the ongoing costs of running the system — finding broken bikes, repairing those bikes, maintaining the information infrastructure, etc etc. Where bike sharing exists already, in cities like Paris, Munich, and now London, it’s institutionalized with a lot of corporate sponsorship and advertising: a very large chunk of money in those cities, for instance, has come from JCDecaux, Deutsche Bahn, and Barclays, respectively. Big corporations like that want standardization and predictability and on-message branding: they don’t want to be associated with users locking their bikes illegally to any old railing or pole, irritating merchants and pedestrians.

Bike sharing is something best done city-wide, at large expense, by a big corporation risking its reputation to at least some extent, with a lot of public money and support as well. Small local schemes are likely to start with great enthusiasm and then run out of steam or money pretty quickly, no matter how clever they seem. I look forward to a big public scheme paying its users for returning bikes to high-demand areas. But beyond that I’m not very optimistic about SoBi.

Update: SoBi’s founder, Ryan Rzepecki, responds in the comments:

1. We have no intention of launching a SoBi system using fixed gear bicycles. That pic is of a foam model attached to our industrial designer’s bike with a clothes hanger. It wasn’t meant to be representative of the actual bike or mount.

2. We intend to launch city-wide systems with government support and major sponsorship. This isn’t a peer-to-peer or grass-roots solution. I think sponsors will particularly appreciate the ability to connect to users with opt-in real-time promotions and advertising.

3. We are developing advanced fleet management tools for tracking not only the position of the bikes, but their repair schedule and other information to help system administrators.

4. Users must lock to approved racks – parking that is illegal or offensive will be flagged and the most recent user can be held accountable.

5. The idea is not to create a mismatched fleet of old beater bikes, but to allow each city to choose a bike that is right for their needs given the local conditions and budget.

6. Yes, suitable bike parking is a problem in most American cities. By choosing the SoBi system, a city can expand parking opportunities for all cyclists by installing new racks. These racks benefit not only bike share bikes, but private bikes as well.

Tom Vanderbilt has a great piece in Slate on the high cost of free parking:

Minimum parking requirements are based on a form of “circular logic,” in which planners estimate parking need by looking at the highest levels of parking demand at suburban locations with free parking and no transit options. As a result, the space devoted to cars often exceeds the space devoted to humans (one study found mall parking lots were 20 percent bigger than the buildings they serviced), and the country is awash in a surplus of parking supply. In Tippecanoe County, Ind., for example, a group of Purdue University researchers noted, “[I]f all of the vehicles in the county were removed from garages, driveways, and all of the roads and residential streets and they were parked in parking lots at the same time, there would still be 83,000 unused spaces throughout the county.” And as Shoup argues, there is nothing free about this parking—everyone, even those who don’t drive, pays for it in one form or another, whether the invisible parking surcharge is built into retail prices or the various costs associated with parking-lot storm-water runoff.

For me the biggest and most invidious cost of parking lots is also the most difficult to measure: the way that they kill any attempt at decent architecture, both on the level of individual buildings and on the level of city development more broadly. Your favorite buildings, your favorite cities, and your favorite vacation destinations all have one thing in common: a distinct absence of massive parking lots. So why are these things mandated by zoning regulations across the U.S.? It makes precious little sense, and it’s high time that minimum parking requirements died a long-overdue death.

What makes the whole thing even weirder is that there is no obvious powerful lobby to agitate for the perpetuation of these requirements. Developers would love to be able to determine for themselves how much parking was optimal, and it’s not like Big Parking Lot has a massive presence in Washington; even the automakers don’t really have much of a dog in this fight. So why are these regulations so very persistent?

I know a lot of people — myself included — don’t have the time or inclination to watch videos online. So for those of you who didn’t watch the full congestion pricing debate, here’s a few highlights.

The first video, at about 1:40, starts with Charles Komanoff using Times Square as an example of how a city’s residents and visitors can benefit as a group, even if some drivers are inconvenienced. New York recently pedestrianized Broadway, and the result, says Komanoff, is that “it does reduce the amount of space that cars and taxis have to operate, but considering that there are so many more people walking than there are in motor vehicles, this rebalancing has been a very positive thing for the city”.

Corey Bearak says that New York businesses need all the customers they can get right now — but Reihan Salam responds by looking at precedents elsewhere in the world, which have generally been very good for centrally-located businesses.

One of the key themes running through the conversation is that the politics of congestion charging is always local and usually anecdotal. Bearak’s opposition is grounded in the reality of where subways do and don’t reach, or whether federal funds can be found to subsidize New York City transit; it’s really hard, Komanoff’s spreadsheet notwithstanding, to return to solid empirical data.

Still, some of the anecdotes are useful. Skymeter’s Kamal Hassan, at about 8:10, notes that if you have a congestion-charge boundary, as London does, then you get boundary effects: things like cars idling on the perimeter of the boundary until the off-peak period begins, or taking circuitous routes around the boundary because they’re free. He’s right, even though he’s talking his book: a per-mile charge, which could change gradually rather than drastically according to geography, would solve all those problems. (In the second video, at about 10:30, Komanoff talks about how it might be possible to charge very low rates in Queens, and very high rates in Times Square.)

Kamal adds, at about 9:20, that “you can change traffic patterns and traffic flows just through parking charging and parking policies, and if you set up smart parking policies within a city, you can have a very large effect on what happens”. This is an important point, because it shows just how many moving parts there are in terms of congestion pricing, and how many different ways there are of getting to where we want to be.

At the beginning of the second video, Kamal goes into more detail: for instance, the cost of a taxi medallion — currently a very large fixed cost — could be turned into a per-mile charge, which would decrease the amount of empty taxicabs cruising for fares, and increase the number of stationary cabs at taxi ranks.

At around 7:00 in the second video, Bearak says that the capacity of New York’s transit system is insufficient to cope with the increase in demand that would come from a congestion charge. To which Kamal agrees that “it’s very important, if you’re going to do congestion charging of any sort, that you make sure people have alternatives” — while at the same time noting that subway ridership fell, in London, when the congestion charge was introduced, and so higher capacity on the underground wasn’t necessary. And it’s much easier to add capacity with new buses than it is to build new subway lines.

At about 12:20 in the second video, there’s an interesting terminological interlude: Bearak says that a congestion charge is “taxing folks”, Komanoff says that it’s “taxing traffic”, and Hassan says it’s “paying for convenience”. This is the point that Ryan Avent makes today:

Commuters who didn’t value their time as much as they valued a marginal dollar (and these will tend to be lower income individuals) will feel some loss from this shift. But it’s unlikely that there will be BIG losers; even the poorest automobile commuters out there (and you can’t be that poor and still be an automobile commuter) place a positive value on their time.

Meanwhile, there will be enormous utility gains elsewhere. Some subset of commuters would have preferred to use transit to driving, but found the bus system too slow and unreliable to be a reasonable option. Bus riders, current and potential, also benefit substantially from congestion pricing. Poor riders who never had the financial option to drive will experience huge gains from faster and more reliable bus transit. And just as the positive value of time ensures that there will be no big losers among drivers, it also means there will be some very big winners. This includes the rich, obviously, but also those with pressing needs. Any given day, there are many, many drivers who need to get somewhere in a hurry: for economic, or medical, or other reasons. These drivers would likely be willing to pay tens or hundreds or thousands of dollars to avoid congestion, but before congestion pricing they wouldn’t have that option. The consumer surplus generated for these drivers from an effective congestion price is simply massive.

Then at about 13:00, Reihan Salam has a great rant. “The idea that you should pay $2 or $9 to drive in to New York when other folks have to pay some amount to take a subway into New York and have a much smaller impact, in terms of traffic congestion, seems pretty fair”.

And at 16:40, Komanoff makes one of those points which is only obvious once it’s made: that the places with the worst congestion are invariably the same as the places with the best transit access. Moving people from cars to transit is not, in principle, all that hard.

And at about 17:00, Hassan raises the prospect of charging people little or nothing for driving in their own neighborhoods, especially in the suburbs — again, something that’s very easy with Skymeter technology. At about 19:30, Komanoff pans out a bit, talking about a tax on miles traveled as an extension of the congestion charge for the rest of the country:

“Even though it’s true that congestion pricing needs to have subways and very dense cities and preferably large cities, the idea of per-mile road pricing, in urban areas and in metropolitan areas, is extremely viable. The average person who has a car is making 1,500 trips a year. It’s not as though every one of those 1,500 trips has equal value to that person. So as we impose the per-mile charges, everyone will prioritize his or her own trips.”

Matt DeBord watched the whole thing, and he’s worried about the implications for “individual liberty” of a Skymeter solution. I’ve addressed this before: in theory, if you make your Skymeter payments in advance, you never need to have your movements monitored or recorded anywhere: the payments just get deducted from your account. Certainly the Singaporean government, if it implements this kind of a solution, is going to be prone to keeping all that extra data. But I’m convinced that DeBord’s “serious privacy issues” can be addressed as and when they arise: they’re not a reason to object ex ante to charging for certain types of driving.

DeBord also raises the issue of electric cars:

Many electric car startups, and even well-established automakers getting into EVs, are aiming to market their limited-range vehicles to city dwellers. You can image what the congestion-charge crowd thinks when it thinks about that.

Actually, as someone who’s spent quite a lot of time around the congestion-charge crowd of late, I can’t imagine what they think on this issue. But I know what I think, and I’m pretty sure that Komanoff thinks similarly: traffic congestion is traffic congestion, no matter what the car engines look like.

Finally, it’s worth pointing out that Frank McArdle (Megan’s dad) has left two interesting comments on this subject, here and here. A snippet:

Congestion is actually a good thing. It tells us that many people see great value in being someplace at the same time.We really don’t want it to go away, since it also reflects the higher values in the economy of places where there is congestion. We know that when there is no congestion at the beaches it is probably not a good beach day for most. That lesson applies very broadly.

What we really want is there to be congestion, but not enough to upset our private benefit. But we all know that the roads we take will have congestion at some points and some days. We factor it into our lives. The proof is in the changes in average commuting times over the last four decades,which have changed significantly only in those places where everyone wants to be because they can make more money than elsewhere.

I think this elides the distinction between congestion and density: the holy grail is to get the latter without the former. New York already has high density; its next job is to lower the amount of congestion that density is prone to create. One way of doing that is through public transit. But another way is through a congestion charge. It’s silly not to use all the tools at one’s disposal in the service of making a city as livable as possible.

Here’s the videos:

Many thanks to Megan McArdle for giving me something to push back against with respect to congestion pricing. Megan, like me, favors congestion pricing, but she does see some problems with it.

The first, I think, is a bit of a straw man:

There’s a real tendency to tell drivers that congestion pricing is great for them: less traffic! I can kind of buy that argument, and then I notice something: almost no one making it commutes by car…

All the people commuting by car seem to think they will hate it. And that makes me think that they probably will.

Yes, this is absolutely true. Congestion pricing will hurt people who commute by car. That’s the whole point. If you currently commute by car, then you’ll either end up spending more money, or else you’ll use public transit. Neither is an obvious improvement. Some richer commuters will like their faster commute, but car commuters who aren’t wealthy will absolutely be the biggest losers here. We tax what we don’t want, and what we don’t want is car commuters. Most advocates of congestion pricing are pretty clear on that point.

That said, it’s worth pointing out a couple of mitigating factors here, beyond the simple idea that the cost of the toll might be balanced by the benefit of faster traffic. First, car commuters already pay for commuting into the central business district of New York, which has no on-street parking: they have to pay to park somewhere, and parking in the CBD is expensive. Megan talks about a car commuter paying $200 a month in congestion charging, and she’s right that that is “a lot of money for most people”. But on the other hand, we’re talking here about people who are already paying much more than that just for parking.

Secondly, where the congestion charge makes the biggest difference isn’t with commuters, so much as it is with through traffic and trucks. Insofar as commuters end up driving more quickly, it’s not mainly because there are fewer car commuters. Instead, it’s because truck traffic will end up moving from peak times to off-peak times, and because through traffic — cars which have no business in Manhattan but just drive through on the way somewhere else — will take alternate routes. Charging is pretty much the only way to make these things happen, and they’re clearly necessary.

And thirdly, car commuters are in the minority in every district in the New York metropolitan area: not just in the boroughs generally, but even in congressional districts in eastern Queens or in Westchester or even in New Jersey. Today, without congestion charging, the number of people who commute by car from any given area into New York’s CBD is smaller than the number of people who commute by public transit from the same area. It’s not like these people have no choice in their mode of transportation.

Megan continues:

I still think that congestion pricing is a good idea for a lot of reasons, but let’s not kid ourselves that it makes everyone better off. It makes affluent people who can afford taxis and congestion fees better off, and poorer folks who can commute by bicycle.

Well, under the Komanoff plan, there’s a hefty 33% taxi surcharge, so let’s not jump to conclusions about whether people taking taxis will be better off. And most of the bicycle commuters I know are actually pretty affluent — for starters, it’s generally much easier to commute by bike if you live in Manhattan than if you live in Queens or the Bronx. And living in Manhattan, for the most part, ain’t cheap.

Certainly every city with congestion pricing has used some of the freed-up road space for bicycles, and the rate of bike usage in New York is rising fast. The more bicyclists there are, the safer and more pleasant biking becomes, and the more everybody wins: as someone who commutes either by bike or by subway, depending on the weather, I can assure you that biking is much more pleasant. And London is just one of many examples of cities where a large number of people have switched from the subway to a bicycle, not because biking is cheaper, but just because it’s a much more efficient way of getting from A to B. In New York, a congestion charge would be extraordinarily helpful in terms of increasing the number of bicyclists.

Megan then says that reducing subway fares “is a terrible idea”, on the grounds that rush-hour subways are already at capacity. But again, if you look at London, subway ridership went down after the congestion charge was introduced, because it made buses so much more attractive. Subways are capacity-constrained, but buses really aren’t.

But more generally, New York’s subways are woefully underused. For a couple of hours a day they’re crammed to capacity, and for most of the rest of the days there’s huge amounts of space. If you reduced the fare at off-peak times, that would do wonders in terms of maximizing the utility provided by the subways — while at the same time reducing congestion on the streets above.

What’s more, rush-hour subway traffic is very price inelastic. The people who brave the subway at 8am are the people who have to brave the subway at 8am, and that’s true whether it costs $4 or $2 or $0. Reducing the fare at peak times might increase peak-time ridership a little, but it wouldn’t increase it a lot. And in fact, if it were done in conjunction with putting a lot of new free buses on the road, there’s a very good chance that, just as in London, subway ridership at peak times would actually fall.

Finally, Megan says that “if you want to reduce auto traffic, you need more capacity on the [transit] system–and you cannot build that capacity if the system has no reliable source of revenue other than the largesse of the city government.” Well, yes. That’s exactly why congestion-charging revenues should be earmarked for transit.

Of course there are some losers with congestion charging. But everywhere that it’s been implemented, it’s been a bit like the smoking ban: people hated it before it was enacted, and then actually kinda liked it afterwards. There’s no reason that the U.S. should be any different.

I’m having quite a lot of difficulty coming up with someone to take the anti-congestion-pricing stance in the debate that I’m taping at Reuters in Times Square tomorrow. Policy wonks from left to right seem to like the idea with remarkable unanimity, at least in theory if not in the exact form that any given proposal might take. Libertarians, for example, quite like the idea of pricing externalities to avoid the tragedy of the commons — they tend to the econogeeky that way.

For me, the strongest argument against a congestion charge is that there’s a decent chance that it will be a very expensive way of achieving not very much. After all, there are lots of urban planning ideas which work in theory but not in practice. Mark Ambinder has a good interview with Joe Flood, who wrote a whole book about how a bunch of geeks from the RAND Corporation managed to persuade New York authorities in the 1970s that shutting down fire stations wouldn’t result in more fires. They were disastrously wrong.

More recently, notes Flood, the Bloomberg administration ran all manner of studies designed to demonstrate that the Atlantic Yards project in Brooklyn would create jobs and tax revenue, rather than lose money for the city.

If I have any skepticism about the wonders of congestion pricing, it’s in the gap between theory and reality: to talk to someone like Charles Komanoff, it’s easy to come away believing that just about everybody will win. Drivers will get faster commutes, bus riders won’t pay fares and will travel faster, subway riders will save money — what’s not to like?

But as my previous article for Wired showed, there’s always model risk, and it’s impossible to price. Empirical evidence from London is mixed: the quantity of traffic is down, and revenues are up, but the speed of traffic doesn’t seem to have increased very much, and the initial gains seem to be eroding over time. Even Komanoff expects something similar to happen in New York: in order to keep congestion constant, the congestion charge is going to have to rise at a pretty substantial rate, pretty much in perpetuity. No matter where it starts, be it $8 or $16 or something else, it’s certain to get higher pretty quickly.

What’s more, there’s no doubt that a congestion charge is politically unpopular, especially in the outer boroughs: people see only what they would pay, and not what they would save — especially when, as with the Bloomberg plan, there’s no decrease in transit fares.

But if someone has stronger arguments against the congestion charge, let me know — especially if you’re free at noon on Wednesday.

Ryan Avent weighs in on my Wired article about congestion pricing with a question I’m going to be putting to some experts on the subject tomorrow. I’ve invited Reihan Salam, John Avlon, and Skymeter CEO Kamal Hassan to chat with Charles Komanoff in the swanky Reuters TV studio overlooking Times Square; the video should be up online later this week.

Ryan’s point I think will echo with Reihan, and indeed with Komanoff’s patron Ted Kheel: they all think there’s a lot to be said for a single flat congestion charge, set at $16, which would allow all transit — not just buses but subways too — to be free. Komanoff’s plan, which has many different fees per mode of transport, depending on the time of day and the day of the week, is hard for people — including legislators — to understand:

Any charge is going to be sent through the policy grinder en route to enactment, and so it doesn’t make a ton of sense to fine tune pricing before that process. [And] for now, these prices have to be processed by human drivers, who are going to want simplicity and certainty. That $16 to get into the city is easy to understand and plan around.

My feeling is that there’s a barbell solution here, and that the Komanoff plan lies somewhere in the unhappy middle. There are definitely problems with a crude flat congestion fee, including the fact that at many times of the day and week there really isn’t much congestion to mitigate. (Even Ryan, at the end of his post, starts talking about introducing an off-peak charge.) On top of that are the problems associated with making subways free at peak times: they’re crowded enough as it is, and although today’s peak-time riders are pretty price-inelastic, ridership would surely increase by some significant amount if the fare was brought down to zero.

At the other end of the spectrum is a Skymeter solution, which is granular not only on the time-of-day question but also on question of which exact part of the city you’re congesting. Komanoff’s plan makes no distinction between a car driving up Avenue D, on the one hand, and a car driving around Times Square, on the other; that’s silly. It also makes no distinction between a car driving in the CBD for 5 minutes and a car driving around the CBD for 5 hours. That’s even sillier. Those failures are failures of technology, and can be solved by Skymeter at a stroke.

If you’re going to set up a complex system, then, I think the Skymeter system is the way to go. And if you’re going to set up a simple congestion charge, then there’s a strong case for making it as simple as possible.

At tomorrow’s discussion, I’m looking forward to talking to four people with different answers to this question. Hassan thinks the complex Skymeter system is the way to go; Komanoff prefers his middle-of-the-barbell plan; Salam is fond of keeping things as simple as possible and just having one flat fee; and Avlon thinks there shouldn’t be any congestion charge at all. Should be a fun talk!

Matthew DeBord thinks that my Wired article about Charles Komanoff is really all about turning Manhattan into “biketopia”. He couldn’t be more wrong. Yes, Komanoff himself is a big advocate of biking. And yes, at the margin, having less traffic in midtown would help in terms of being able to create more bike lanes. But biking is not a big part of Komanoff’s spreadsheet at all: instead, it’s all about public transit. Buses become free, subways become cheaper, and the overwhelming majority of New Yorkers — the people who get into Manhattan by some means other than private car — become better off.

DeBord thinks that New York is already “transitopia” — but the fact is that its transit system can be improved greatly through the implementation of a congestion charge, which would unsnarl traffic for buses and would provide new money for buses, subways, and local trains.

Debord says that New Yorkers don’t want to get rid of gridlock: “many people have decided that they’d rather live with it than have, for example, Michael Bloomberg tell them when they can and can’t affordably drive into Midtown”, he writes. But check that link: the “many people” in question are, essentially, Shelly Silver and his small band of dysfunctional Albany politicians. It’s worth remembering that it’s already pretty much impossible to affordably drive into Midtown: here, for instance, are the rates posted at my favorite parking garage, across the street from Reuters. All of them are higher than Bloomberg’s proposed congestion charge.

What’s more, there were two big problems with the congestion charge as proposed by Bloomberg. The first was that Manhattanites benefited unduly: since they have many fewer cars, they got all the benefits of faster traffic while bearing only a small proportion of the total fees. Under Komanoff’s plan, Manhattanites — who take the overwhelming majority of yellow-cab journeys — would pay a taxi surcharge, making things more equitable.

The second problem with Bloomberg’s congestion charge is that it didn’t directly benefit most New Yorkers. If they ever drove into Manhattan, they would pay more, but they would never save anything, since the plan didn’t reduce transit fares. Komanoff’s plan of course is very different, and especially benefits residents of Queens and the Bronx who take a lot of buses, which would always be free. (The huge advantage of making buses free is that it eliminates long lines and waits at the farebox.)

DeBord accuses Komanoff, me, and “the other sons of Jane Jacobs” of wanting a “place where we just won’t be able to do the car thing anymore”. That simply isn’t true: Komanoff likes to say that he doesn’t want a car-free New York, he wants a car-fee New York. Yes, we would like to see fewer cars making through-trips through Manhattan without even stopping, and many fewer trucks doing that. Right now, such through trips account for 40% of all car trips in the central business district, which is crazy.

But Komanoff’s plan is most emphatically not about pedestrianizing New York or turning it into some kind of Dutch-style biketopia where cars are all but absent from the city center. It’s just about making life better for all New Yorkers — even the drivers, who no longer need to suffer the frustrations of gridlock, and who can get to their destination in a much more predictable amount of time. There’s a lot of people who would pay good money for that.