The urban geography of transit cities and of car cities is relatively well-understood. In a transit city, there will be a strong CBD surrounded by residences with spiky secondary centers, all quite small geographically but dense, centered around train stations and junctions; because density is high throughout, minor trips are done on foot. In a car city, all trips are done by car, the core is weak, and most employment is in suburban edge cities and edgeless cities.
What I haven’t seen is an explanation of how urban geography works in mixed metro areas: there are those in which short trips are done on foot and long ones in cars, such as new urbanist developments, and those in which short trips are done by car and long ones on transit, such as park-and-ride-oriented commuter suburbs. It is the latter that I want to address in this post.
The first feature of park-and-rides is that of all combinations of modes of transportation, they are the fastest and enable suburbs to sprout the farthest from the center. This is because the segment of the trip done in a car is uncongested and so driving is faster than transit, while the segment done on a train parallels a congested road, and conversely makes few stops so that average speeds are high.
On top of this, because intra-suburban trips are done by car, the density in the suburbs is very low, comparable to proper car cities (see the lower end of the density profiles of the New York, Chicago, and Boston metro areas), and this forces sprawl to go outward. New York is the world’s most sprawling city measured in total built-up area; the only other city of comparable size that’s not a transit city or a bus/jitney city is Los Angeles, which is forced to have denser suburbs because of the mountains. Of course Houston and Dallas sprawl even more relative to size, but because they lack New York’s transit-oriented core, there’s an inherent limit to their size.
The other feature is that there’s a definite socioeconomic history to the development of the auto-oriented commuter suburbs of transit cities. First, people move to the suburbs and commute into the city, almost always by train due to road congestion (or, as in the earliest New York suburbs, because mass motorization hasn’t arrived yet). The mass exodus into these suburbs comes from cars rather than commuter rail, and so the local services for people living in those suburbs are built at automobile scale, rather than at the walkable town center scale of 1910.
In North America there’s also a definite class element here – the early movers are the rich rather than the poor. Historically this was partly because poor people couldn’t afford regular train fare, and partly because the impetus for suburbanization was idyllic country homes with access to urban jobs rather than cheap housing for the poor. If I’m not mistaken, this wasn’t the case in Australian cities’ suburbanization, leading to a more urban transit-style mode of running mainline rail. The result of this class distinction is that North American commuter rail styles itself as for the rich: agencies make an effort to ensure everyone has a seat and downplay comfortable standing space, and the expectation is that transit is a last-ditch mode of transportation for when cars just don’t have the capacity to get people downtown, and so nobody needs to take the trains in the off-peak or take a bus to the train.
The result is that the park-and-ride city will still have a strong core with high-capacity transportation, and the primary CBD will maintain its supremacy for high-income jobs. Establishing edge cities in the direction of the favored quarter can happen, but there’s still a congested city nearby, and so from many directions it’s impossible to drive, and taking transit is impossible. Thus jobs in White Plains and Stamford are not nearly as high-paying as jobs in Manhattan.
There can even be secondary CBDs, if the inner part of the metro area, where people take transit more regularly than the suburban commuters do, is large enough. But those secondary CBDs are frequently quite auto-oriented. Brooklyn’s mode share for jobs is only 42-39 in favor of transit (for residents, it’s 60-25), and all other counties in the New York region except Manhattan have more workers driving than taking transit, a situation that is not true if one looks at residents. Those secondary CBDs then have mixed characteristics: they are dense and fairly walkable, as can be expected based on their history and location, but also have plentiful parking and a large share of drivers demanding even more. They can accommodate multiple modes of transportation, but driving is more convenient, and from the suburbs the commuter rail system isn’t always geared to serve them.
One third of the MBTA’s outstanding debt, about $1.7 billion, comes from transit projects built by the state as part of a court-imposed mitigation for extra Big Dig traffic; interest on this debt is about two-thirds the agency’s total present deficit. Metra was prepared to pay for a project to rebuild rail bridges that would increase clearance below for trucks and cut the right-of-way’s width from three to two tracks. Rhode Island is spending $336 million on extending the Providence Line to Wickford Junction, with most of the money going toward building parking garages at the two new stations; Wickford Junction, in a county whose number of Boston-bound commuters is 170, is getting 1,200 parking spaces.
Supporters of transportation alternatives talk about the inequity between highway and transit funding in the US, but what they’re missing is that the transit funding bucket includes a lot of things that are manifestly not about transit. At their best, they are parking lots and other development schemes adjacent to train stations, which would’ve been cheap by themselves. At their worst, they are straight highway projects, benefiting road users only.
The situation in Boston, while unique in its brazenness, is not unique in concept. In the US, where there are no pollution taxes on fuel, the only way to mitigate air pollution is by regulation and by building alternatives simultaneously. Put another way, combined highway and transit construction is in most cases not really a combined project; it’s a highway project, plus required mitigation. Requiring the transit agency to shoulder the debt and the operating subsidies is exactly requiring transit users to pay for highways. It’s equivalent to charging transit multiple dollars per gallon of gas saved from any mode shift. And it may get even worse: the proposed House transportation bill includes a provision to allow spending national air pollution control funds on regular highway widening, in addition to the current practice of spending them on carpool lanes.
Historically, the diversion of funding from transit to roads took such insidious forms. For an instructive example from Owen Gutfreund’s book, roads advocates fought to get driver’s license fees and even inspection fees for fuel trucks recognized as road user fees, whose proceeds must be diverted toward roads. For another example from the same book, in Denver, the streetcar system was required to cover 25% of the cost of road maintenance on one-way streets and 50% on two-way streets, and as car traffic rose, streetcars both became slower and had to send over more money toward roads.
Another instructive case study is grade separations. It is to my knowledge universal that expressways and high-speed railroads, both of which must be grade-separated, pay for their own grade separations. In all other cases, who pays is determined by which mode is more powerful, and in the US, this is roads. As the national highway system was built in the 1920s, interurban railroads were required to pay for grade-separations, even when the rail came first. The practice continues today: in Kentucky, the railroad has to shoulder the full cost if it’s from 1926 or newer (Statute 177.110), and half the construction cost and the full planning cost if it’s older (177.170). In contrast, in Japan, grade separations are considered primarily a road project, and so the Chuo Line track elevation project was paid 85% by the national and city governments and only 15% by JR East (page 36). The segment in question of the Chuo Line was built in 1889; I believe, but do not know, that new rail construction in Japan is always grade-separated, at the railroad’s expense.
The situation in the US today is a surreptitious underfunding of transit, and at the same time a surreptitious overfunding of roads. It is not subject to democratic debate or even to the usual lobbyist funding formulas, but, like the obscure regulations that impede good passenger rail, hidden in rules nobody thinks to question. To pay for road mitigations and for parking, transit agencies will cut weekend service and reduce frequency. It’s bad enough when done in the open, but it’s done while claiming that transit is too expensive to provide.
In her book Dark Age Ahead, Jane Jacobs complains that freeways as built are good at getting people to macrodestinations (downtown) but not microdestinations (particular addresses within city center). In her example from Toronto, this is correct, but in general, each mode of transportation will be good at serving microdestinations in an urban form that’s suited for it. Cars are not good at serving an intact city center; but equally, transit is not good at serving suburban sprawl, and regional rail that’s not integrated with urban transit is not good at serving urban destinations away from immediate train stations.
The idealized job center in an auto-oriented city is the edgeless city. Even the edge city, as explained in Lang and LeFurgy’s now-paywalled article Edgeless Cities, is too dense, and becomes congested too quickly; indeed, Tysons Corner is infamous for its lunchtime rush hour conditions. Ideally, cars drive from low-density residences to low-density office parks, primarily on freeways but with fast arterial connections at both ends; the freeway network in the auto-oriented city serves an everywhere-to-everywhere set of origins and destinations.
In such an environment, transit can’t do well. The distance between suburban attractors is too great for an easy walk, and the roads are too wide and fast for a pleasant walk. Buses and trains can serve a general macrodestination (“Warwick Mall/CCRI”), but not individual microdestinations, not without splitting and cutting frequency to each destination or detouring and raising travel time. The buses serving Warwick Mall and CCRI have hourly frequency, and are a long, uncomfortable walk from the hotel in Warwick I needed to go to. Judging by the frequency, I’m not the only person who chose not to use them, and take a taxi instead; everyone who has a car or who isn’t extremely price-sensitive does. The only way transit can serve such a destination is by concentrating development near the station – in other words, making a mini-transit city in the sea of sprawl, which generally conflicts with the goal of easy station parking.
In a city, the opposite situation exists. It’s easy to just pronounce transit more suited to dense city centers than driving, but the situation is more complicated. Transit, too, thrives on good connections to microdestinations. It can’t serve employment that’s dense but evenly dispersed in a large area – people would need too many transfers, and the result would be service that’s on paper rapid and in reality too slow. Instead, it works best when all destinations are clustered together, in an area not many subway stations in radius.
In this view, one failure of urban renewal is its failure to recognize that most people who visit city centers are going to do a lot of walking, and amenities should make it easier rather than harder. Traditional urban renewal would build cultural centers and other projects at the fringe of the CBD, to help its growth: Lincoln Center just north of Midtown, Civic Center just southwest of the San Francisco CBD, Providence Place and Providence Station just north of Downcity. In New York and San Francisco, there’s at least rapid transit serving those destinations, mitigating the effects. In Providence, no such thing exists. It’s an inconvenient walk from Kennedy Plaza to the mall and the train station – it’s not too long, but it crosses Memorial Boulevard right when it turns into a freeway on-ramp. Walking to the Westin, immediately adjacent to the mall, is practically impossible without rushing across roads without crosswalks. Even the walk between the station and the mall, which were built together and are close to each other, is much worse on the street than on a map, again involving crossing auto-centric roads.
Organic city amenities do not look like this. If they cluster at the same location (for example, 125th Street in New York, or Thayer Street in Providence), they tend to be along roads that facilitate rather than hindering pedestrian movement. And if they don’t, they are all located along a rapid transit network in its shared service area, where it is still a tight mesh rather than a network of radial lines.
In view of the recent emphasis on parking policy, due to Donald Shoup but now mirrored by other urban planning and transportation experts, the observation is that in any city center, on-site parking is difficult to find. Even in cities that make downtown parking relatively easy to get to, people can’t hope to park at every single microdestination, so instead they trip-chain, driving into the city and parking but going to multiple points within the city, all within a short and easy walking distance from one another. This is roughly the urban geography of the French Riviera, which combines easy parking with a dense, lively center in Nice and a fair amount of urbanity on some streets even in auto-oriented secondary cities such as Monaco and Menton.
The connection to regional rail is that, historically, it descends from intercity trains, and therefore the conception of connecting the suburbs to the city is very macrodestination-driven. To name two egregious American examples, the Boston’s north side lines and Caltrain both connect many suburbs to the city while also connecting people to the suburban tech job corridor, but in reality miss the biggest job centers at both ends. North Station is two subway stations north of the CBD, and as a result ridership underperforms the south side lines; 4th and King is far enough outside the Market Street CBD that it’s not close to the CBD jobs – the proposed Transbay Center site, which is, is located near more jobs than all existing Caltrain stations combined. And if microdestination-level service to an already transit-oriented CBD is bad, then service to other urban destinations is worse: urban station spacing is wide, there’s no attempt to develop near stations, and the poor integration with local urban transit ensures that even people who could be willing to make the last-mile transfer don’t.
Gendered Innovations’ charts of trip chaining and gender breakdown of public transit riders got me thinking about how different systems of transportation handle a mixture of short and long trips. Eric Jaffe at The Atlantic Cities reports this and suggests that transit agencies orient physical features such as accessibility to the needs of women who trip-chain care and work trips.
But to me, the first observation is that although women trip-chain more, it doesn’t seem to be true that women are more likely to ride transit in the US than men just because of trip-chaining features. Instead, women traditionally have been less likely to have jobs requiring commuting, and the commute gap has been shrinking more slowly than the gap in employment.
This comes from the fact that trip chaining on transit is cumbersome in most cases. Both cars and transit have to deal with the time it takes to stop for an errand, but transit tends to handle this worse, unless it’s very frequent and has practically zero access and egress times. Transit cities instead get people to take their short errand trips on foot – since their neighborhoods are denser and have more mixtures of uses, they make retail and care trips attractive on foot. In light of the fact that walking is not useful for long commute trips and transit is not useful for short errands, we can construct the following typology of cities:
|Long \ Short mode||Foot, bicycle||Car|
|Car||New urbanist, small-town, auto-oriented dense||Auto-oriented|
Auto-oriented cities are the easiest: in those places, people drive for all purposes. Trip chaining can be done on a commercial arterial road, dropping off laundry or kids or buying something on the way to work, and because of ample parking availability, the time each additional link in the chain consumes is very small, since the longest access and egress time comes from navigating from the residential cul-de-sac to the arterial and from the arterial to the office park.
Traditional suburbs, common around New York and Chicago and sometimes in other old North American cities, are similar for trip-chaining purposes. In those areas, the urban form is suburban and auto-oriented, but work trips to the city are done by commuter rail or occasionally commuter bus, since the city is not as auto-friendly as the suburbs.
Transit cities too have their long-range commuter rail, but it is built as an extension of walking rather than of driving. Neighborhoods tend to have mixed uses, and there’s a concentration of retail development near the outlying stations, sometimes forming large secondary clusters but sometimes just acting as neighborhood centers. It could take considerable time to add more trips to one chain, especially if not everything is located at the train station. But conversely, the amount of time a single short trip takes is small, unlike the case for auto-oriented cities – the supermarket is right around the corner, and within five minutes’ walk are plenty of stores. When people walk, the concept of a single trip begins to lose meaning then. Potentially, every single purchase can be considered a separate trip, in which case the chaining becomes quite long.
In many places the transit is absent and people drive outside the neighborhood, while still doing errand trips on foot. This is the typology that characterizes different environments including new urbanism, traditional cities like Providence and Tel Aviv that have been made car-oriented, and auto-oriented modernist projects such as Co-op City. Those environments all differ in how trip chaining is done. In principle, it can be done on foot, but usually people who can drive do.
If my own experience is any indication, one feature of cities in this typology is that children and teenagers walk more. In Tel Aviv, my father drove me to elementary school on the way to work while (in later grades) I walked back, and I took the bus to and from middle school. Most trips my parents did in a car, but there was a reasonable number that were short enough to walk. I’d walk to farther destinations such as the cinema and the urban mall. The view of the North Tel Aviv middle and upper-middle class of the 1990s as I remember it is that the bus is fine for trips to school, but adults drive. I doubt I’d have had the same view if I’d grown up in New York, or for that matter in the Houston suburbs, where everyone drives or is driven.
Although most of the discussion about transit cities contrasts them with car-oriented cities, the other two typologies need to be examined, too. When adults and children trip-chain differently, children can get a distorted view of who transit is for (poor people, people who can’t drive yet), and the next generation will make the city auto-oriented; this is indeed what is happening in Tel Aviv, which despite population growth in the core is adding cars and spawning low-density suburbanization well outside the built-up urban areas.
Likewise, Cap’n Transit’s attacks on park-and-rides don’t quite capture what is wrong with the car/transit typology. A transit agency that wants to make it easier to trip-chain will want to concentrate development near the train stations, because that’s where it’s easiest to add minor trips without having to walk ten minutes out of one’s way. Of course in the middle of the dense city there’s development everywhere, which may well be orthogonal to where the subway is, but then trip-chaining becomes easier because each foot trip is so short.
The principle is that cars are a big one-time purchase but have a much lower marginal cost of usage. If one major class of trips can’t be done on transit – and chained trips generally can’t when they require the rider to wait for the next bus and the next bus will come in 15 minutes – then people will buy a car and then drive it even for trips they’d happily take transit to if they didn’t already own a car. The class of trips that can only be done conveniently by car needs to be kept small enough that people will use car share, take a taxi, or beg a friend who does own a car.
Thus what transit agencies and pro-transit politicians should devote more time to is appropriate development more than physical features of the transit system. Accessibility is important for so many reasons other than strollers. In contrast, the primary importance of using transit to extend the range of the pedestrian rather than provide a capacity boost for the car is precisely that transit needs minor trips to be doable on foot. A transit system that one needs to take to the supermarket may be technically successful, but it’s in a failed urban area.
Cap’n Transit is virtually alone in the transit blogosphere in opposing the Tappan Zee Bridge widening and replacement. Unfortunately, merely opposing a highway project, expensive as it is, is not enough; as we’ve seen in the failure of the ballot proposition to ban a highway tunnel in Seattle, opponents of highway expansion need to make it concrete and clear what transit alternatives there are. In the case of the Tappan Zee specifically, alternatives exist, but serve different markets, and it’s necessary to explain why the market that the Tappan Zee serves is not the most important to the region.
I propose a regional rail system instead, focusing on serving Rockland County and perhaps a few centers in Orange County. There are multiple lines crisscrossing Rockland County, with limited or no freight traffic, passing through old town centers that would make good regional rail stops and connecting to good alignments in North Jersey. For a regionwide perspective there are my original regional rail proposal and my more recent focus on connectivity from North Jersey to Lower Manhattan, but the important thing for the purposes of Rockland County is the question of which lines could be used. The Erie Main Line only goes to Suffern, but could collect passengers from the western parts of Orange County; the Northern Branch, including an abandoned northern end, goes as far north as Nyack; the Pascack Valley Line was abandoned north of Spring Valley but has an intact right-of-way as far north as Haverstraw; the West Shore Line goes north to Albany and has moderate freight traffic, easily accommodated in the off-peak if double-tracking is restored. There are so many options that the main question is which to activate just to maintain adequate frequency.
The main difference with any Tappan Zee proposal is that the existing rail lines go north-south, whereas the Tappan Zee is east-west. Fortunately, most existing movement is north-south. As can be confirmed by the 2000 census, Rockland and Orange Counties’ commute market toward Westchester and other suburbs accessed by the bridge is quite small: 18,000 to Westchester and Fairfield. The volume of commuters from those two counties to Bergen and Passaic Counties is somewhat larger (22,000), and that to New York City more so (27,000 to Manhattan, 14,000 to the other boroughs). And traffic over the bridge since 2000 has stalled.
Not only is the north-south or northwest-southeast market bigger than the east-west market, but also it uses the Tappan Zee when it could be diverted if there were alternatives. A breakdown of travel on the bridge reveals that 16% of eastbound travel is to the Bronx and another 15% is to the other four boroughs and Long Island; this could be done competitively by various transit options.
Thus, a transit option that emphasizes north-south connectivity and goes to Manhattan through Bergen and Passaic Counties is going to serve more people than adding more east-west connectivity. It could serve far more if North Jersey jobs clustered in Paterson, Hackensack, and other old city centers, but in fact they’re diffuse. It’s unreasonable to assume significant commercial transit-oriented development in North Jersey, though a few jobs in Paterson could still be captured; however, jobs in Manhattan, Brooklyn, and Queens could be served well.
Finally, to serve Bronx and Upper Manhattan jobs from both North Jersey and Rockland County, the trains should be combined with good bus service across the GWB. For example, bus lanes on Route 4 could be a strong start, especially if the trains are timed to connect to the buses. More speculatively, there’s a subway bellmouth allowing an extension of the C along the GWB, and relative to the cost of tunneling it should be inexpensive to extend the C as an elevated line toward Paterson over Route 4; the drawback is that the C is slow and would poorly serve the Bronx.
Although Rockland County is very sprawling, it has just enough old cities to anchor regional rail at the residential end. The effect is magnified if we can assume some TOD – for example, developing over the many parking lots currently in place in Nyack near the legacy Erie station – but as with commercial TOD, this is desirable but not very likely with the current political structure. Fortunately, American commuter rail works very well as a shuttle that extends auto-dependent commutes into cities that have no room for more cars; as a narrow alternative to constrained highways, it often succeeds, and would be a no-brainer compared to a bridge as expensive as the Tappan Zee.
The cost of reviving and electrifying the four lines proposed in my regional rail post (Erie Main, Pascack Valley, West Shore, and Northern Branch) is quite small compared to either the cost of bringing them to Manhattan or that of rebuilding the Tappan Zee Bridge. The cost of bringing the lines to Manhattan is substantial, but done right it would be much lower than the Tappan Zee Bridge’s $8.3 billion excluding any transit component.
If costs could be brought down, a new crossing, slightly farther north of the existing bridge, could work well for rail. The transit mode selection report discusses commuter rail on the new bridge, and the concept would be similar except that there should be more stations to serve local traffic better. A rail-only bridge would leave the Hudson Line north of Tarrytown, allowing west-of-Hudson commuters to access this job center and also ensuring no loss of frequency to the station, and then cross to Nyack. It would have to be underground in Nyack because the Palisades rise too steeply from the water, and would surface just west of the urban area. If all trains serving the line are EMUs, rather than diesels or even dual-mode locomotives, then the grade could be sharp enough to limit tunneling to the urban area of Nyack; the TMS report, which only considers diesels, proposes 2 miles (3.2 km) of tunneling, but EMUs climbing 4% grades could cut this by more than half.
The advantage of the east-west option is that it would serve Westchester jobs; while the commute market from Rockland and Orange Counties to Westchester is as mentioned not large, it clusters along I-287, especially in White Plains, and is thus somewhat more rail-serviceable. In addition, although the chance of commercial TOD is small everywhere in the US, it is larger in Tarrytown and White Plains than in Paterson and Hackensack.
On the other hand, if the costs could be brought down, they would be lower for everything, including highways. The same factors that cause transit construction costs to be so high in New York (namely, overstaffing, and poor contracting practices) apply to highways equally. In particular, the decision about what mode to favor should only weakly depend on cost, since relative costs both within transit modes and between cars and transit are not too different from in lower-cost countries.
To cut costs to a minimum while still providing acceptable first-phase service, the initial network could include only the lines that could be brought to Secaucus, with some track modifications near the station allowing Erie trains to terminate at the station parallel to the Northeast Corridor tracks; this still involves a fair amount of concrete pouring, but much less than a new tunnel to Manhattan, and the transfer could be made as convenient as that at Jamaica. In addition, trains could be mixed and matched: that is, to let a few of the Erie trains serve Manhattan directly, some Northeast Corridor or Morris and Essex trains could be cut to Secaucus. The main disadvantage is that no such option is possible with the West Shore Line and Northern Branch, and so it would be more useful in the western part of Rockland County than in the eastern part.
The selling point of the regional rail alternative is that, despite job sprawl, Rockland County residents are still more likely to need to travel to Manhattan than to Westchester. Thus, the promise of a one-seat ride to Manhattan on frequent train service, or at least a two-seat ride with the same quality of transfer offered to Long Islanders, could carry some political weight. One does not drive into New York out of love of driving; one drives into New York out of necessity, and making this less necessary could reduce some of the political will to spend billions more than required on widening a bridge.
Railvolution reports FTA numbers that say the average CO2 emissions of the New York City Subway are 0.17 pounds per passenger-mile (48 grams per passenger-km). That’s the equivalent of 114.6 passenger-mpg of gas, if you prefer to think in those terms. The presentation gives average seat occupancies, which we can also confirm with the NTD; it works out to about 4 car-mpg of gas. Other agencies can have somewhat different numbers, based on train efficiency and especially the local sources of power generation, e.g. BART has very low emissions coming entirely from the fact that the Bay Area has ample hydro power resources.
New York’s emission number, 4 mpg, may be familiar to you as roughly the emission-efficiency of regional diesel trains. Per ton of car mass the regional diesel trains do slightly better, since the regional train in question weighs 40 tons vs. 33-39 for New York’s subway cars, but this comes from making fewer stops. At agencies with very dirty power generation, such as the Chicago L, and even ones without very dirty power, such as the energy-hungry Washington Metro, the numbers are even lower, even though they’re electric and the regional diesel trains are not.
What we see is then that railroad electrification does not add too much to fuel economy. The question is then why the situation for cars is so different. The Nissan Leaf’s EPA-rated fuel economy equivalent rating is 99 mpg – almost as good as the New York City Subway, better than nearly all subway systems in the US. But if we try to break it down based on energy consumption, we get other numbers; the EPA just massaged the numbers to make plug-in hybrids look good.
The Leaf’s energy efficiency is 0.34 kWh per vehicle-mile, pardon the mixed units; the FTA’s numbers for major US subways range from 0.186 kWh per passenger-mile in high-seat-occupancy New York to 0.388 in low-seat-occupancy Chicago. This is not 99 mpg, unless one uses a fairly clean mixture of fuels; with the New York mixture, it’s 63 vehicle-mpg. So right off the bat, the official numbers underestimate the Leaf’s CO2 emissions by 36%, and overestimate its CO2 efficiency by 57%.
But even that doesn’t take care of inefficiencies in generation. Well-to-wheels, plug-in electric cars have about the same emissions as regular hybrids. This confirms the rough numbers we’ve seen from trains. The Tesla Roadster, a very fuel-efficient car, gets even better energy-efficiency even wells-to-wheels, but it also has much lower electricity consumption, and to get the right numbers it assumes electricity is generated from natural gas rather than coal.
Bear in mind, all of this assumes certain things about the grid mix. At the current US grid mix, on average electrification does not impact carbon emissions. Of course, since people need electricity for reasons other than transportation, any regime in which carbon emissions fall is one in which electricity becomes lower-carbon, and this would tilt the field in favor of all-electric vehicles, both cars and trains.
So, why electrify, if there’s no carbon emission benefit, why electrify? Two answers: air pollution, and, for trains, performance. Electric trains outperform diesel ones, and also cost less to operate in terms of both energy and maintenance. But electrification should be sold only on grounds that are in fact correct.
I’ve been reading Earl Swift’s The Big Roads, and the early biography of Thomas MacDonald had passages that jumped at me. Unlike Owen Gutfreund, who focuses on MacDonald’s industry ties and use of astroturf, Swift portrays MacDonald as a Progressive reformist who believed in better engineering as a way to improve society, literally paving the way to the future.
While he used special interests to further his goals, he was also concerned with efficiency. He first made his name as the chief of the Iowa State Highway Commission, where he built a road system with virtually no budget; neighboring states had several times the planning budget Iowa had. At the time, the building contractors had colluded, dividing the state into regions with each enjoying a local monopoly; this drove up costs twice, first by increasing construction costs, and second by requiring more maintenance since the work was shoddy. MacDonald’s contribution was to break up the monopolies and demand that contractors compete.
MacDonald also believed in personally instructing local officials and contractors in good road construction methods. He’d often be visiting construction sites and participate in construction, partly for the photo-ops but partly for showing the locals how good engineering is done.
As a result, MacDonald became famous among road builders for his success in building roads, and was made the head of the Bureau of Public Roads. Iowa at the time had one of the highest car ownership rates in the US, about 1 per 7 people (about the same as Manhattan today). The person who became Governor toward the end of his tenure in Iowa was anti-roads, but this did not slow down highway and car growth.
The importance of this for good transit advocates is threefold. First, it shows that it is in fact possible for government officials to promote good government and increase efficiency. Of course we must not neglect broader social trends, but sometimes well-placed competent individuals can make a major difference.
Second, it reminds us that many of the rules that are currently associated with government dysfunction were passed with opposite intent and effect back in the Progressive Era. Lowest-bid contracts were an effort to stamp out corruption; civil service exams were an effort to reduce patronage; teacher tenure was meant to make teachers politically independent; the initiative process was intended to give people more control over government. All of those efforts succeeded at the time, and took decades of social learning among the corrupt and incompetent to get around. Although programs built under these rules often turned out badly, such as the Interstate network, with its severe cost and schedule overruns, this was not due to the contractor collusion seen in the 1910s or today.
And third, it’s a warning to those who hope that placing well-meaning individuals in power is enough. Every person with power thinks that his power is used for good and wants to extend it. Thus, once MacDonald became head of the Bureau of Public Roads, he made sure to maintain control over highway funding and gave himself the power to sign contracts with states, which Congress was then obligated to fund.
Good engineering can improve engineering standards, but it cannot improve society. Although the decisions to tear apart neighborhoods were made by local officials more, of whom Robert Moses is the most infamous, the idea that a cadre of technocrats who look at cities on maps and in models know what cities ought to look like more than the people living in them was an inherent part of this attitude. Indeed, the 19th century impetus for suburbanization, using rapid transit rather than roads, came from the same class of reformists. The Interstate system was simply when they had enough money and power to impose their modernist vision nationwide.
The Tappan Zee Bridge is about to fall down. As a result, the replacement and widening project is in spare-no-expense mode. Ordinarily, widening a bridge from seven lanes to ten would be judged in terms of costs and benefits, after which the costs would be ignored as they always are for US road projects. But now everyone thinks New York needs this project, to the point that even transit and livable streets advocates are more worried about commuter rail tracks on the new bridge than about the costs of the entire project.
Cap’n Transit cribbed study numbers before they disappeared from the official website. The budget of the project, without the transit component, was about $7 billion, and is now up to $8.3 billion; this includes highway widenings at both ends. The transit component people are fretting about is another $1 billion for BRT and $6.7 billion for commuter rail.
To put things in perspective, consider the Øresund Bridge-Tunnel complex. Whereas the Tappan Zee is 5 kilometers of bridge, Øresund consists of 8 kilometers of bridge, an artificial island with 4 additional kilometers of road, and 4 kilometers of tunnel. The cost, including landworks on both sides, was a little more than €3 billion in 2000, which works out to $5.5 billion in 2010. The bridge-tunnel is narrower than the Tappan Zee replacement – four lanes of traffic plus two tracks of rail – but it’s also three times as long, and more complex because of the tunnel.
More importantly, if the Tappan Zee really needs that capacity, and width is such a constraint, they should build rail first, BRT second, and car lanes last. Roads will never beat mass transit on capacity per unit width of right-of-way. With all traffic from Rockland to Westchester County funneled through one chokepoint, and some centralization of employment (in Manhattan, White Plains, and Tarrytown), rail could work if it were given the chance. So the only environment in which a bridge with so many traffic lanes is justified is one in which the cost of ten lanes is not much more than the cost of four.
To be completely fair to irate Rockland County residents, more people use the Tappan Zee than Øresund, since the tolls are lower and it’s a commuter route. But not enough. The bridge is crossed by 138,000 vehicles per day. This means the replacement and widening project, excluding all transit improvements, is $60,000 per car. With normal commuter seat occupancy, it’s perhaps $50,000 per person. Transit projects in the US routinely go over this, but those are for the most part very low-ridership commuter rail projects. Second Avenue Subway, the most expensive urban subway in the world per kilometer, is about $25,000 per expected weekday rider.
Given the high cost, the only correct response is a true no-build: dismantle the bridge, and tell people to ride ferries or live on the same side of the Hudson as their workplace. Given expected ridership and Øresund costs, I believe the Tappan Zee replacement would make sense at $3 billion, with the transit components; without, make it a flat $2 billion. Go much above it and it’s just too cost-ineffective. Not all travel justifies a fixed link at any cost.
As a followup to my previous post about the TTI’s new congestion report, I finally did a multivariate regression analysis, with the dependent variable being cost and the independent variables being size and freeway lane-miles per capita. Such an analysis reduces the regression coefficient between freeways and congestion even more, to -42.5 from the uncontrolled -233. More interestingly, if we log all numbers (population, congestion cost, and freeways), the regression coefficient becomes a positive 0.02 – that is, adding freeways is correlated with making congestion a little worse.
Of course, it’s not literally true that adding freeways makes congestion worse. There’s a correlation if we look at the variables in some way, but it’s not going to have any statistical significance. Therefore tweaking variables slightly can make a correlation go from weakly positive to weakly negative.
In univariate regression, we can think about the square of the correlation as the percentage of the variance that is explained by the regression line. Freeway lane-miles per capita explain 3.8% of the variance in congestion (and logging either variable makes this number smaller); with 101 urban areas surveyed, it’s statistically significant, but barely so. But after controlling for population, this proportion drops to 0.7%. Thus, any sentence of the form “adding one freeway lane-mile per thousand people only cuts $42.5 from the annual congestion cost per capita” is inherently misleading: the correlation is so weak that some cities can reduce congestion without building the requisite amount of roads, or building any roads at all (for example, nearly all American cities in the last five years, congestion having crashed in the oil price boom and the recession), while others can keep building but see congestion increase (for example, Houston since the 1980s, and even today).
It goes without saying that such analysis is not going to appear in the TTI report itself. The TTI gets funding from APTA and the American Road and Transportation Builders Association. It pays lip service to congestion pricing as a solution to congestion, and instead talks a lot about building public transportation and even more about building freeways to keep up with demand. American cities may be building freeways faster than their population growth, but cities that enact no traffic restraint and just pour concrete can expect demand to grow faster than population as people become more hypermobile.
The Texas Transportation Institute has just released the latest version of its much-criticized Urban Mobility Report. Although the conclusions and recommendations made by the TTI tend to reflect its funding sources (APTA, American Road and Transportation Builders Association), the underlying data seems sound, and suggests conclusions orthogonal to those made by the report. In addition, looking at the correlations more closely suggests obvious hazards coming from any simplistic analysis of linear regression. It even showcases how we could use data dishonestly and lie with statistics. So let’s take the data that’s relevant right now and see what we can conclude ourselves.
First, the size of an urban area is a very strong correlate of its level of congestion. The linear correlation between size and per capita congestion cost is 0.71. The correlation increases to 0.8 if we take the log of population and the log of congestion, or if we consider congestion in the absence of public transportation; in both cases, it comes from the fact that New York is far below the population-congestion regression line.
Now, more freeways do not really lead to congestion reduction. There’s some correlation between freeway miles per capita and congestion per capita, going in the expected direction, but it’s weak, -0.2, and while it’s statistically significant, the p-value is an uninspiring one-tailed 0.025. Looking at a scattergram doesn’t make any nonlinear relationship obvious.
Moreover, size is a correlate of both congestion (0.71 as above) and freeways (-0.23). This is fully expected: literature on cities’ economies of scale (here is a story of one controversial example) suggests that congestion and the economic activity causing it grow faster than linearly in city size while the amount of required energy and infrastructure grows slower than linearly. I open the floor to anyone with more powerful tools than OpenOffice Calc to do multiple regression; again, the sanitized data is here.
Even without controlling for population, freeways are not a very strong correlate. The regression coefficient is -233: increasing the number of freeway miles per thousand people by 1 (the range is 0.13-1.4, with few large metros above 1 or below 0.35) reduces the congestion cost per capita by $233 per year, also uninspiring.
The regression number alone can be used as a dishonest trick when arguing on the Internet. If we overinterpret weak correlations, we can declare that the only way to decrease congestion is to build an unrealistic number of freeways, and thus declare the problem unsolvable. Of course, for most cities we can find other cities of comparable size with much less congestion and without enormous amounts of asphalt – this is why the correlation is so weak. But a good hack should not bother himself with such caveats to talking points.
So if making an urban area larger makes it more congested, independently of and much more strongly than all else, should we give up on cities? Well, no. Assuming no change in traffic policy, congestion results from more economic activity. It then becomes straightforward to institute congestion pricing. It’s no different from how big cities can use their resources to hire more cops to deal with the crime that could result from extra interactions between people. On top of this, in very large cities, mass transit becomes a serious option: this not only reduces the amount of congestion per capita, but also removes many people from the highways to the point that congestion becomes irrelevant to their daily lives, except perhaps through higher transportation prices, which they can fully afford given the extra wealth.
Another thing to consider is that most American cities have added more freeways than people since 1982, the first year for which TTI data is available, while also becoming much more congested. If a simple relationship between freeway miles per capita and congestion held, it would be robust to these changes over time. Of course, traffic has grown even faster, leading the main report to showcase on PDF-page 21 how congestion increased the fastest in regions where road demand outgrew supply the most. But this raises the question of whether the main issue is one of demand, rather than one of supply. This is not just an issue of size: the log-log regression coefficients with cost and time is 0.42, i.e. doubling an urban area’s population will raise its per-driver congestion cost and travel delay by a factor of 2^0.42; since 1982, the average urban area on the list has seen its population grow by a factor of 1.46 and its travel delay per driver grow by a factor of 2.85 = 1.46^2.77. Cost has grown even faster, because of higher value of time.
That said, quantity of freeways does not equal quality (from the drivers’ perspective, of course, rather than the city’s). On paper, Greater New York has added freeway lanes about 9% faster than people over the last thirty years. In practice, none has addressed the major chokepoints within and into the city itself, where traffic is worst. Of course, commutes involving Manhattan are overwhelmingly likely to be done on public transportation, but diagonal commutes within the city are more likely to be done by car than on transit.
On a parenthetical note, the units of comparison here are TTI-defined urban areas. TTI’s belief about urban area population growth trends is sometimes at odds with that of the Census Bureau, but the raw population numbers are close enough. More important is the question of what to do about urban areas that are really exurbs of larger areas, such as Poughkeepsie-Newburgh and the Inland Empire. My first instinct was to lump them in with their core metro areas, but their congestion level per capita is not high. Their commutes are long, but not very congested for their size. Finally, although most correlations here are with congestion cost, the correlation numbers with travel delay and excess fuel consumptions are very similar; the one exception I’ve checked, for which I have no explanation, is log-log congestion-fuel correlation (0.84, with regression coefficient 0.73).