Category: Incompetence

Don’t Run Bilevels

For years, the RER A’s pride was that it was running 30 trains per hour through its central segment in the peak direction (and 24 in the reverse-peak direction). With two branches to the east and three to the west, it would run westbound trains every 2 minutes between 8 and 9 in the morning on the seven-station shared trunk line. Moreover, those trains are massive, unlike the trains that run on the Metro: 224 meters long, and bilevel. To allow fast boarding and alighting at the central stations, those trains were uniquely made with three very wide doors per side, and two bilevel segments per car; usually there are two doors near the ends of the car and a long bilevel segment in between. But now the RER A can no longer run this schedule, and recently announced a cut to 24 peak trains per hour. The failure of the RER A’s bilevel rolling stock, called the MI 2N or MI 09, should make it clear to every transit agency mulling high-throughput urban rail, including RER A-style regional rail, that all trains should be single-level.

On most of the high-traffic regional rail lines of the world, the trains are single-level and not bilevel. The reasoning is that the most important thing is fast egress in the CBD at rush hour. For the same reason, the highest-traffic regional rail lines tend to have multiple CBD stops, to spread the load among several stations. The Chuo Rapid Line squeezes 14 trains in the peak half-hour into Tokyo Station, its only proper CBD station, discharging single-deck trains with four pairs of doors per 20-meter-long car onto a wide island platform with excellent vertical circulation. Bilevels are almost unheard of in Japan, except on Green Cars, first-class cars that are designed to give everyone a seat at a higher price point; on these cars, there aren’t so many passengers, so they can disembark onto the platform with just two doors, one per end of the car.

Outside Japan (and Korea, where the distinction between the subway and regional rail is even fuzzier), the busiest regional rail system is the RER. The RER A runs bilevels, but the most crowded line while the RER A was running 30 tph was the RER B, which runs 20 tph, through a tunnel shared with the RER D, which runs 12 bilevel tph. Outside Paris, the busiest European regional rail systems are in London (where bilevels are impossible because of restricted clearances), and in Berlin, Madrid, and Munich, all of which run single-level trains. Berlin and Munich moreover have three door pairs per 17-to-18-meter car. Munich squeezes 30 tph through its central tunnel, with seven distinct branches. Other than the RER A, it’s the less busy regional services that use bilevels: the RER C, D, and E; the commuter trains in Stockholm; the Zurich S-Bahn and other Swiss trains; Dutch regional trains; and many low-performance French provincial TERs, such as the quarter-hourly trains in the Riviera.

Uniquely among bilevels, the RER A’s MI 2N (and later MI 09) was designed as a compromise between in-vehicle capacity and fast egress. There are three triple-width door pairs per car, allowing three people to enter or exit at once: one to the lower level, one to the upper level, one to the intermediate vestibule. The total number of door pairs per unit of train length is almost as high as on the RER B (30 in 224 meters vs. 32 in 208), and the total width of these doors is much more than on the RER B, whose doors are only double-wide.

Unfortunately, even with the extra doors, the MI 09 has ultimately not offered comparable egress times to single-level trains. Present-day peak dwell times on both the RER A and B are about 50-60 seconds at Les Halles; here, the RER B, with its prominent Gare du Nord-to-Les Halles peak in the morning, is in a more difficult urban geography than the RER A, with four stations that could plausibly lay claim to the CBD (Les Halles, Auber, Etoile, La Defense). The RER B has long had problems with maintaining the schedules, due to the 32 tph segment shared with the RER D, using traditional fixed-block signaling; the RER A in contrast has a moving-block system called SACEM. But now the RER A has problems with schedule reliability too, hence the cut in peak frequency.

The problem is that it’s not just the number of doors that determines how fast people can get in and out. It’s also how quickly passengers can get from the rest of the train’s interior to the doors. Metro systems optimize for this by having longitudinal seats, with their backs to the sides of the train, creating a large, relatively unobstructed interior compartment for people to move in; Japanese regional trains do the same. European regional trains still have transverse seating, facing forward and backward, and sometimes the corridors are so narrow that queues form on the way to the vestibules, where the doors are. The RER A actually has less obstructed corridors than the RER B. The problem is that it’s still a bilevel.

Bilevel design inherently constrains capacity on the way to the door, because the stairs from the two decks to the intermediate level, where the door is, are choke points. They are by definition only half a train wide. They are also slow, especially on the way down, for safety reasons. When the train is very crowded, people can’t just push on the way up or down the way they can on a flat train floor. If passengers get off their seats in the upper and lower levels well in advance and make their way to the intermediate-level vestibules then they can alight more quickly, but on a train as crowded as the RER A, the vestibule is already full, and people resort to sitting on the stairs at rush hour, obstructing passageways even further.

As a result, RATP is now talking about extending peak dwells at the central stations to 105 seconds, to stabilize the schedules. Relative to 60-second dwells, this is 45 seconds of padding per station; with about 3 minutes between successive stations in the central segment, this is around 25% pad (on top of the already-existing pad!), a level worthy of American commuter trains rather than of Europe’s busiest commuter rail line.

What’s more, this unique design cost the region a lot of money: Wikipedia says the MI 09’s base order was €3.06 million per 22.5-meter car, and the option went up to €4.81 million per car. In contrast, German operators have purchased the high-performance single-level Coradia Continental and Talent 2 for €1.25-1.5 million euros per 18-meter car (see orders in 2014, 2016, and 2017); these trains have a top speed of 160 km/h and the power-to-weight ratio of a high-speed train, necessary for fast acceleration on regional lines with many stops. Even vanilla bilevel trains, with two end-car door pairs, are often more expensive: at the low end the Regio 2N is €7.06 million per 94-meter trainset, at the higher end the high-performance KISS is around €3 million per 25-meter car (about 2.7 in Sweden, 3-3.5 in Azerbaijan), and the Siemens Desiro Double Deck produced for the Zurich S-Bahn in 2003 was around €3 million per 25-meter car as well.

High-traffic regional railroads that wish to improve capacity can buy bilevel trains if they’d like, but need to understand the real tradeoffs. Average bilevel trains, with a serious decrease in capacity coming from having long upper- and lower-level corridors far from the doors, can cost 50-100% more than single-level trains. They offer much more capacity within each train (the KISS offers about 30% more seats per meter of train length, with a small first-class section, than the FLIRT), but the reduction in capacity measured in trains per hour cancels most of the benefits, except in cases where peak dwells don’t matter as much, as in Zurich with its two platform tracks per approach track. In terms of capacity per unit cost, they remain deficient.

The MI 09 was supposed to offer slightly less seated capacity per unit of train length and equivalent egress capacity to single-level trains, but in practice it offers much less egress capacity, at much higher cost, around 2.5-3 times as high as single-level trains. If RATP had bought single-level trains instead of the MI 09, optimized for fast egress via less obstructed passageways, it would have had about €2.5 billion more. Since the cost of extending the RER E from Saint-Lazare to La Defense and beyond is about that high, the region would have had money to obtain far more capacity for east-west regional travel already.

The American or Canadian reader may think that this analysis is less relevant to the United States and Canada, where the entire commuter rail ridership in all cities combined is about the same as that of just the RER A and B. Moreover, with higher US construction costs, the idea of saving money on trains and then diverting it to tunnels is less applicable than in Paris. However, two important American factors make the need to stop running bilevels even more pertinent than in Europe: CBD layout, and station construction costs.

North American CBDs are higher-rise than European ones – even monocentric cities like Stockholm have few city center skyscrapers. The job density in Paris’s job-densest arrondissement (the 2nd) is about 50,000/km^2, and it’s higher in its western end but still only about comparable to Philadelphia’s job density around Suburban Station. Philadelphia has three central stations in the SEPTA commuter rail tunnel, but only Suburban is really in the middle of peak job density; Market East is just outside the highest-intensity zone, and 30th Street Station is well outside it. In Boston, only two proper CBD stations are feasible in the North-South Rail Link, South Station and Aquarium. In New York, Penn Station isn’t even in the CBD (forcing everyone to get off and connect to the subway), and only 1-2 Midtown stations are feasible in regional rail proposals, Penn and Grand Central. Some of these stations, especially Penn and Grand Central, benefit from multiple platform tracks per approach track in any plan, but in Boston this is not feasible.

The other issue is station construction costs. High construction costs in the US mean that spending more money on trains to avoid spending money on infrastructure is more economic, but conversely they also make it harder to build anything as station-rich as the RER A, the Munich S-Bahn tunnel, or Crossrail. They also make stations with multiple platform tracks harder to excavate; this is impossible to do in a large-diameter TBM. This makes getting egress capacity right even more important than in Europe.

New York and Philadelphia meandered into the correct rolling stock, because of clearance restrictions in New York and the lack of a domestic manufacturing base for bilevel EMUs. Unfortunately, they still try to get it wrong: New Jersey Transit is buying bilevel EMUs (the first FRA-compliant ones). Railroads that aren’t electrified instead got used to bilevel unpowered coaches, and get bilevel EMUs: Caltrain is getting premium-price KISSes (about the only place where this is justifiable, since there are sharp capacity limits on the line, coming from mixing local and express trains on two tracks), and the Toronto RER (with only one CBD station at Union Station) is also planning to buy bilevel EMUs once electrification is complete.

Paris’s MI 09 mistake is not deadly. The RER E extension to the west will open in a few years and relieve the RER A either way. Being large and rich can paper over a lot of problems. North American cities are much poorer than Paris when wages are deflated to tunnel construction costs, and this means that one mistake in choice of alignment or rolling stock can have long-lasting consequences for service quality. Learning from the most forward-thinking and successful public transit operators means not just imitating their successes but identifying and avoiding their failures.

Quick Note: U-Shaped Lines

Most subway lines are more or less straight, in the sense of going north-south, east-west, or something in between. However, some deviate from this ideal: for example, circular lines. Circular lines play their own special role in the subway network, and the rest of this post will concern itself only with radial lines. Among the radials, lines are even more common, but some lines are kinked, shaped like an L or a U. Here’s a diagram of a subway system with a prominently U-shaped line:

Alert readers will note the similarity between this diagram and my post from two days ago about the Washington Metro; the reason I’m writing this is that Alex Block proposed what is in effect the above diagram, with the Yellow Line going toward Union Station and then east along H Street.

This is a bad idea, for two reasons. The first is that people travel in lines, not in Us. Passengers going from the west end to the east end will almost certainly just take the blue line, whereas passengers going from the northwest to the northeast will probably drive rather than taking the red line. What the U-shaped layout does it put a one-seat ride on an origin-and-destination pair on which the subway is unlikely to be competitive no matter what, while the pairs on which the subway is more useful, such as northeast to southwest, require a transfer.

The second reason is that if there are U- and L-shaped lines, it’s easy to miss transfers if subsequent lines are built:

The purple line has no connection to the yellow line in this situation. Were the yellow and red line switched at their meeting point, this would not happen: the purple line would intersect each other subway line exactly once. But with a U-shaped red line and a yellow line that’s not especially straight, passengers between the purple and yellow lines have a three-seat ride. Since those lines are parallel, origin-and-destination pairs between the west end of the purple line and east end of the yellow line or vice versa require traveling straight through the CBD, a situation in which the subway is likely to be useful, if service quality is high. This would be perfect for a one-seat or two-seat ride, but unfortunately, the network makes this a three-seat ride.

The depicted purple line is not contrived. Washington-based readers should imagine the depicted purple line as combining the Columbia Pike with some northeast-pointing route under Rhode Island Avenue, maybe with an additional detour through Georgetown not shown on the diagram. This is if anything worse than what I’m showing, because the purple/red/blue transfer point is then Farragut, the most crowded station in the city, with already long walks between the two existing lines (there isn’t even an in-system transfer between them.). Thus the only direct connection between the western end of the purple line (i.e. Columbia Pike) and what would be the eastern end of the yellow line (i.e. H Street going east to Largo) requires transferring at the most crowded point, whereas usually planners should aim to encourage transfers away from the single busiest station.

When I created my Patreon page, I drew an image of a subway network with six radial lines and one circle as my avatar. You don’t need to be a contributor to see the picture: of note, each of the two radials intersects exactly once, and no two lines are tangent. If the twelve ends of six lines are thought of as the twelve hours on a clock, then the connections are 12-6, 1-7, 2-8, 3-9, 4-10, and 5-11. As far as possible, this is what subway networks should aspire to; everything else is a compromise. Whenever there is an opportunity to build a straight line instead of a U- or L-shaped lines, planners should take it, and the same applies to opportunities to convert U- or L-shaped lines to straight ones by switching lines at intersection points.

RPA Fourth Regional Plan: LaGuardia Airport and the Astoria Line

This is the second post based on a Patreon poll about the RPA Fourth Regional Plan. See the first post, about Third Avenue, here.

The most worrisome part of the RPA Fourth Regional Plan is the LaGuardia Airport connector. The regional rail system the RPA is proposing includes some truly massive wastes of money, but what the RPA is proposing around LaGuardia showcases the worst aspects of the plan. On Curbed I explained that the plan has an unfortunate tendency to throw in every single politically-supported proposal. I’d like to expand on what I said in the article about the airport connector:

The most egregious example is another transit project favored by a political heavyweight: the LaGuardia AirTrain, championed by Governor Andrew Cuomo. Though he touts it as a one-seat ride from Midtown to LaGuardia, the vast majority of airport travelers going to Manhattan would have to go east to Willets Point (a potential redevelopment site) before they could go west. Even airport employees would have to backtrack to get to their homes in Jackson Heights and surrounding neighborhoods. As a result, it wouldn’t save airport riders any time over the existing buses.

Once again, it’s proven unpopular with transit experts and advocates: [Ben] Kabak mocked the idea as vaporware, and Yonah Freemark showed how circuitous this link would be. When Cuomo first proposed this idea, Politico cited a number of additional people who study public transportation in the region with negative reactions. Despite its unpopularity—and the lack of an official cost for the proposal—the AirTrain LaGuardia is included in the RPA’s latest plan.

But there is an alternative to Cuomo’s plan: an extension of the N/W train, proposed in the 1990s, which would provide a direct route along with additional stops within Astoria, where there is demand for subway service. Community opposition killed the original proposal, but a lot can change in 15 years; Astoria’s current residents may well be more amenable to an airport connector that would put them mere minutes from LaGuardia. Cuomo never even tried, deliberately shying away from this populated area.

And the Fourth Plan does include a number of subway extensions, some of which have long been on official and unofficial wishlists. Those include extensions under Utica and Nostrand avenues (planned together with Second Avenue Subway, going back to the 1950s), which also go under two of the top bus routes in the city, per [Jarrett] Walker’s maxim [that the best argument for an urban rail line is an overcrowded bus line, as on Utica and Nostrand].

There is also an extension of the N/W trains in Astoria—though not toward LaGuardia, but west, toward the waterfront, where it would provide a circuitous route to Manhattan. In effect, the RPA is proposing to stoke the community opposition Cuomo was afraid of, but still build the easy—and unsupported—airport connector Cuomo favors.

My views of extending the Astoria Line toward LaGuardia have evolved in the last few years, in a more positive direction. In my first crayon, which I drew in 2010, I didn’t even have that extension; I believed that the Astoria Line should be extended on Astoria Boulevard and miss the airport entirely, because Astoria Boulevard was the more important corridor. My spite map from 2010, give or take a year, connects LGA to the subway via a shuttle under Junction, and has a subway branch under Northern, a subway extension that I’ve been revising my views of negatively.

The issue, to me, is one of branching and capacity. The Astoria Line is a trunk line on the subway, feeding an entire tunnel to Manhattan, under 60th Street; the Queens Boulevard Line also feeds the same tunnel via the R train, but this is inefficient, since there are four trunk lines (Astoria, Flushing, and Queens Boulevard times two since it has four tracks), four tunnels (63rd, 60th, 53rd, Steinway/42nd), and no way to get from the Astoria Line to the other tunnels. This was one of my impetuses for writing about the problems associated with reverse-branching. Among the four trunks in Queens, the Astoria Line is the shortest and lowest-ridership, so it should be extended deeper into Queens if it is possible to do so.

The RPA is proposing to extend the Astoria Line, to its credit. But its extension goes west, to the waterfront. This isn’t really a compelling destination. Development isn’t any more intense than farther east, and for obvious reasons it isn’t possible to extend this line further; the RPA’s proposal would only add one stop to the subway. In contrast, an eastern extension toward LGA could potentially rebuild the line to turn east on Ditmars (with some takings on the interior of the curve at Ditmars and 31st), with stops at Steinway and Hazen before serving the airport. The intensity of development at Steinway is similar to that at 31st and Ditmars or at 21st, and Hazen also has some housing, albeit at lower density. Then, there is the airport, which would be about 8 minutes from Astoria, and 26 minutes from 57th and 7th in Manhattan. This is a different route from that proposed in the Giuliani administration, involving going north above 31st and then east farther out, running nonstop to the airport (or perhaps serving a station or two) through less residential areas, but I believe it is the best one despite the added impact of running elevated on Ditmars.

LGA is not a huge ridership generator; total O&D ridership according to the Consumer Airfare Report is around 55,000 per day, and 33% mode share is aspirational even with fast direct service to Manhattan hotels and an easy connection to the Upper East Side. But it still provides ridership comparable to that of Astoria Boulevard or Ditmars on the line today, and Steinway and Hazen are likely to add more demand. If the MTA closes the 11th Street Connection, taking the R from 60th Street Tunnel to the Queens Boulevard Line, in order to reduce the extent of reverse-branching, then the Astoria Line will run under capacity and need this additional demand. The total number of boardings at all stations, including Queensboro Plaza, is 80,000 per weekday today, plus some transfer volumes from the 7, which empties at Queensboro Plaza as 60th Street Tunnel provides a faster route to most Manhattan destinations than the Steinway Tunnel. An LGA extension should add maybe 40,000 or 50,000 weekday riders, without much of a peak since airport travel isn’t peaky, and make it easier to isolate the Astoria Line from the other Queens lines. This is not possible with a short extension to the waterfront as the RPA proposes.

I’ve seen someone suggest somewhere I don’t remember, perhaps on Twitter, that the reason the RPA plan involves an extension of the Astoria line to the west is to insidiously get the correct extension to LGA passed. If the RPA can propose an el in Astoria and not be killed by NIMBYs, then it will prove to Cuomo that NIMBYism is not a problem and thus he can send the subway to the airport directly, without the circuitous air train project that even less acerbic transit writers like Ben and Yonah hate.

I disagree with this line, on two different grounds. The first is that the RPA has two other reasons to support a western extension of the Astoria Line: it connects to the waterfront (which, following de Blasio and his support for the waterfront tramway, the RPA wants to develop further), and it got a station on Triboro in the Third Regional Plan, in the 1990s. I can no longer find the map with the stations on Mike Frumin’s blog, but the plan was to have a station every 800 meters, with a station to the west of Ditmar/31st still in Queens, around 21st Street; only in the more recent plan did the RPA redesign the idea as Crossboro, with much wider stop spacing.

The second grounds for disagreement is that the RPA presented a long-term vision. If Cuomo’s flawed LGA connector is there, then it will embolden him to find money to build this connection, even though it’s slower than taking a bus to the subway today. It will not embolden anyone to look for funding for the extension of the Astoria Line to the west, since there is no force clamoring for such extension – not the neighborhood, and not even the RPA, which includes this line on a long list of proposals.

As I said on Curbed, the RPA has been around for 90 years. Cuomo is just a governor, not even the leader of a real political movement (unlike Bernie Sanders, who seems to be interested in his leftist agenda more than in himself). There is no reason for an organization so venerable to tether itself to a politician who isn’t likely to be around for more than a few more years. On the contrary, it can provide cover for Cuomo to change his plan, if it does some legwork to prove that people in Astoria actually are interested in subway expansion to the east.

Elon Musk’s Ideas About Transportation are Boring

Four years ago, I broke my comment section by declaring that Elon Musk’s Hyperloop proposal had no merit, combining technical criticism with expressions like “barf ride” and “loopy.” Since then, Musk seems to have quietly abandoned Hyperloop, while the companies attempting to build the technology, run by more serious people, are doing away with the promise of reducing construction costs to one tenth those of conventional high-speed rail. Instead, Musk has moved to a new shiny target in his quest to sell cars and compete with public transit: The Boring Company. I criticized some of what he was saying in Urbanize.LA last summer, but I’d like to go into more detail here, in light of a new fawning interview in Wired and an ensuing Twitter flamewar with Jarrett Walker. In short, Musk,

a) has little understanding of the drivers of tunneling costs,
b) promises reducing tunneling costs by a factor of 10, a feat that he himself has no chance to achieve, and
c) is unaware that the cost reduction he promises, relative to American construction costs, has already been achieved in a number of countries.

The Boring Company’s Ideas of How to Cut Costs

There is much less technical information available publicly than there was for Hyperloop. However, The Boring Company has an FAQ including an outline of how it aims to cut construction costs:

First, reduce the tunnel diameter. The current standard for a one-lane tunnel is approximately 28 feet. By placing vehicles on a stabilized electric skate, the diameter can be reduced to less than 14 feet. Reducing the diameter in half reduces tunneling costs by 3-4 times. Second, increase the speed of the Tunnel Boring Machine (TBM). TBMs are super slow. A snail is effectively 14 times faster than a soft-soil TBM.  Our goal is to defeat the snail in a race. Ways to increase TBM speed:

  • Increase TBM power. The machine’s power output can be tripled (while coupled with the appropriate upgrades in cooling systems).
  • Continuously tunnel. When building a tunnel, current soft-soil machines tunnel for 50% of the time and erect tunnel support structures the other 50%. This is inefficient. Existing technology can be modified to support continuous tunneling activity.
  • Automate the TBM. While smaller diameter tunneling machines are automated, larger ones currently require multiple human operators. By automating the larger TBMs, both safety and efficiency are increased.
  • Go electric. Current tunnel operations often include diesel locomotives. These can be replaced by electric vehicles.
  • Tunneling R&D. In the United States, there is virtually no investment in tunneling Research and Development (and in many other forms of construction).  Thus, the construction industry is one of the only sectors in our economy that has not improved its productivity in the last 50 years.

This is not the first time that Musk thinks he can save a lot of money by reducing tunnel diameter; he said the same thing in the Hyperloop paper. Unfortunately for him, there is literature on the subject, which directly contradicts what he says. In my Urbanize piece, I mention a study done for the Very Large Hadron Collider, which compares different tunnel diameters across various soil types, on PDF-p. 5. Two tunnel diameters are compared, 4.9 m (16′) and 3.9 (12′). Depending on soil type and tunnel boring machine (TBM) drive, the larger tunnel, with 1/3 larger diameter, costs 15-32% more.

Subsequent pages in the study break down the costs per item. The TBM itself has a cost that scales with cross-sectional area, but is only a small minority of the overall cost. The study assumes five drives per TBM, with the first drive accounting for 75% of the TBM’s capital cost; in the first drive the larger-diameter tunnel is 32% more expensive, since the TBM accounts for 25-40% of total cost depending on diameter and rock, but in subsequent drives the TBM accounts for about 5% of total cost. Another 6% is muck cars (item 2.05, PDF-pp. 7 and 46), whose cost rises less than linearly in tunnel diameter. The rest is dominated by labor and materials that are insensitive to tunnel width, such as interior lighting and cables.

But the actual cost is even less sensitive to tunnel width. The VLHC study only looks at the cost of tunneling itself. In addition, there must be substantial engineering. This is especially true in the places where transportation tunnels are most likely to arise: mountain crossings (for intercity rail), and urban areas (for urban rail and road tunnels). This is why there’s a trend toward bigger tunnels, as a cost saving mechanism: BART’s San Jose extension is studying different tunnel approaches, one with a large-diameter tunnel and one with twin small-diameter tunnels, and the cost turns out to be similar. In Barcelona, the large-diameter TBM actually saved money and reduced disruption in construction.

The Boring Company’s various bullet points after its point about tunnel diameter are irrelevant, too. For example, labor is a substantial portion of TBM costs, but in the VLHC study it’s about one third of the cost in easier rock and 15% in harder rock. There appears to be a lot of union featherbedding in some American cities, but this is a political rather than technological problem; without such featherbedding, labor costs are not onerous.

Tunneling Costs Aren’t Just Boring

At $10 billion for just 2.2 km of new tunnel, East Side Access is the most expensive urban rail tunnel I am aware of. The second most expensive, Second Avenue Subway’s first phase, costs $1.7 billion per km, not much more than a third as much. Is New York really spending $10 billion on just boring 2.2 km of tunnel? Of course not. The 2 km in Manhattan cost a little more than $400 million, per an MTA status report from 2012 (PDF-p. 7). The few hundred meters in Queens actually cost more, in an unnecessary tunnel under a railyard. The cavern under Grand Central cost much more, as do ancillary structures such as ventilation.

The TBM is probably the most technologically advanced portion of urban tunneling today. Even in New York, in the most expensive project ever built, the TBM itself is only responsible for about $200 million per km; more typical costs, cited in a consultant’s report for Rocky Mountain tunneling, are somewhat less than $100 million per km. This is why large-diameter TBMs are so appealing: they increase the cost of the tunneling itself, but save money everywhere else by allowing stations to be constructed within the bore.

Of course, The Boring Company is not building conventional subways. Subways already exist, and Musk likes reinventing everything from the wheel onward. Instead, the plan is to build tunnels carrying cars. This means several things. First, the capacity would be very low, especially at the proposed speed (Musk wants the cars to travel at 200 km/h – excessive speed is another of his hallmarks).

Second and more importantly, instead of having to deal with expensive subway stations, the infrastructure would have to deal with expensive ramps. Musk wants cars to be lowered into the tunnels with elevators. Underground elevators are cheap (vertical TBMs are easy), but in the proposed application they just move the problem of ramps deeper underground: the elevator (“skate” in Musk’s terminology) would carry the cars down, but then they’d need to accelerate from a standstill to line speed, in new tunnels, separate from the mainline tunnels so as to avoid slowing down through-traffic. Trains solve this problem by making the entire train stop in the tunnel and taking the hit to capacity, and compensating by running a long train with many more people than cars could possibly hold. But roads would need the same infrastructure of urban freeways, underground.

Switching between tunnel trunks poses the same problem. Flying junctions are expensive, especially underground. In New York, they were common on the IND subway, built in the late 1920s and 1930s; the IND was expensive for its time, around $150 million per route-km in today’s money, whereas the Dual Contracts from the 1910s and early 20s (with fewer junctions) were about $80 million per underground route-km. Most subway systems don’t do what the IND did, and instead of complex junctions they build independent lines, switching between them using transfer stations. With cars, this solution is impossible, forcing underground four-level interchanges; even above ground, those interchanges cost well into the 9 figures, each.

There is So Much Musk Doesn’t Know

The starting point of The Boring Company is that Los Angeles’s tunnel construction costs, which the company pegs at a billion dollars per mile, need to be reduced by a factor of ten. This means cutting them from about $600 million per km to $60 million. While there is nothing that Musk or his company has said in public that suggests he is capable of reducing construction costs, other parts of the world have substantially done so already.

In my construction costs posts, there are a few projects in the $60 million/km area. Manuel Melis Maynar, the former CEO of Madrid Metro, wrote a brief report on how he built subways cheaply; in today’s money, the underground parts of Madrid’s 1999-2003 subway expansion cost around $70 million per km, but this includes rolling stock, and without it, actual cost is likely to be where Musk wants it to be. Recent subway lines in Seoul have also been in that area, including Metro Line 9 and the Sin-Bundang Line. Going up to $100 million per km, there are more lines in Stockholm.

Melis Maynar’s writeup ignores any of the technological pizzazz Musk thinks of. Instead of trying to squeeze more power out of TBM, he emphasizes good contracting practices, and separation of design and construction. Like Musk, he believes that faster construction is cheaper, but he is aware that the limiting factor is not boring speed: even at a conservative rate of 15 meters per day, a TBM could excavate several kilometers a year, so it’s better instead to begin construction at several points along the line and work in parallel rather than in sequence. Adding TBMs does not make projects substantially more expensive: one TBM used for East Side Access cost $6-8 million, and other estimates I’ve seen only reach into the 8 figures, for multibillion dollar projects. Nor does adding staging areas raise cost underground, where there are many potential sites; underwater it’s a bigger problem, and there costs are indeed much higher, but nothing that Musk does seems designed around underwater tunnels, and his proposed map for LA road tunnels is underground.

Musk’s Ideas: Loopy and Boring

Americans hate being behind. The form of right-wing populism that succeeded in the United States made that explicit: Make America Great Again. Culturally, this exists outside populism as well, for example in Gordon Gekko’s greed is good speech, which begins, “America has become a second-rate power.” In the late 2000s, Americans interested in transportation had to embarrassingly admit that public transit was better in Europe and East Asia, especially in its sexiest form, the high-speed trains. Musk came in and offered something Americans craved: an American way to do better, without having to learn anything about what the Europeans and Asians do. Musk himself is from South Africa, but Americans have always been more tolerant of long-settled immigrants than of foreigners.

In the era of Trump, this kind of nationalism is often characterized as the domain of the uneducated: Trump did the best among non-college-educated whites, and cut into Democratic margins with low-income whites (regardless of education). But software engineers making $120,000 a year in San Francisco or Boston are no less nationalistic – their nationalism just takes a less vulgar form. Among the tech workers themselves, technical discussions are possible; some close-mindedly respond to every criticism with “they also laughed at SpaceX,” others try to engage (e.g. Hyperloop One). But in the tech press, the response is uniformly sycophantic: Musk is a genius, offering salvation to the monolingual American, steeped in the cultural idea of the outside inventor who doesn’t need to know anything about existing technology and can substitute personal intelligence and bravery.

In reality, The Boring Company offers nothing of this sort. It is in the awkward position of being both wrong and unoriginal: unoriginal because its mission of reducing construction costs from American levels has already been achieved, and wrong because its own ideas of how to do so range from trivial to counterproductive. It has good marketing, buoyed by the tech world’s desire to believe that its internal methods and culture can solve every problem, but it has no product to speak of. What it’s selling is not just wrong, but boringly so, without any potential for salvaging its ideas for something more useful.

Agency Turf Battles and Construction Costs

This is a touched-up version of an article I tried publishing earlier this year, changed to be more relevant to regular blog readers, who know e.g. what Gateway is.

I’ve talked a lot about high rail construction costs in the US, especially in New York: see here for a master list of posts giving cost figures, and here and here for posts about things that I do not think are major reasons. In this post, I’d like to talk about one thing that I do think is relevant, but not for every project: agency turf battles.

The German/Swiss planning slogan, organization before electronics before concrete, means that transit agencies should first make sure all modes of public transit are coordinated to work together (organization) before engaging in expensive capital construction. In the US, most urban transit agencies do this reasonably well, with integrated planning between buses and trains (light rail or subway); there’s a lot of room for improvement, but basics like “don’t run buses that duplicate a subway line” and “let people take both buses and subways on one ticket” are for the most part done. Readers from the San Francisco Bay Area will object to this characterization, but you guys are the exception; New York in contrast is pretty good; Chicago, Boston, and Philadelphia are decent; and newer cities run the gamut, with Seattle’s bus reorganization for its light rail being especially good.

But then there’s mainline rail, with too many conflicting agencies and traditions. There is no place in the US that has commuter rail and successfully avoids agency turf battles, even regions where the integration of all other modes is quite good, such as New York and Boston. I have complained about this in Philadelphia, and more recently criticized the RPA’s Fourth Regional Plan for letting Long Island claim the East River Tunnels as its own fief.

But all of this pales compared with what is actually going on with the Gateway tunnel. The New York region’s political leaders have demanded funding for a $25 billion rail tunnel between New York Penn Station and New Jersey. When Donald Trump had just won the election, Schumer proposed Gateway as a project on which he could cooperate with the new president; Booker got some federal money earlier, in the Obama administration.

The circumstances leading to the Gateway announcement are themselves steeped in inter-agency intrigue. Gateway is the successor to an older scheme to build a rail tunnel under the Hudson, called ARC. In 2010, Chris Christie acquired some notoriety for canceling it as construction started.

Earlier, in 2003, Port Authority studied three ARC alternatives. Alt P would just serve Penn Station with a new cavern adding more terminal tracks; Alt G would serve Penn Station and build a new tunnel connecting to Grand Central; Alt S would serve Penn Station and build a new tunnel to Long Island, at Sunnyside. The three options each cost about $3 billion, but Alt G had the highest projected ridership. Alt G had the opportunity to unite New Jersey Transit’s operations with those of Metro-North. Instead, Alt P was chosen, and the cavern was involved in the cost escalations that led Christie to cancel the project, saying the then-current budget of $9 billion would run over to $12.5 billion.

It is hard to say why Port Authority originally chose Alt P over Alt G. Stephen Smith spent years sending freedom of information requests to the relevant agencies, but never received the full study. Agency turf battles between New Jersey Transit and Metro-North are not certain, but likely to be the reason.

I talked to Foster Nichols a few months ago, while researching my Streetsblog piece criticizing the RPA plan for kowtowing to Long Island’s political demands too much. Nichols oversaw the reconstruction of Penn Station’s LIRR turf in the 1990s, which added corridors for passenger circulation and access points to the tracks used by the LIRR; he subsequently consulted on the RPA plan for Penn Station. Nichols himself supports the current Gateway plan, which includes the $7 billion Penn Station South complex, but he admitted to me that it is not necessary, just useful for simplifying planning. The Pennsylvania Railroad designed Penn Station with provisions for a third tunnel going east under 31st Street, which Alts S and G would leverage; Alts S and G are still possible. The one caveat is that the construction of Sixth Avenue Subway, decades after Penn Station opened, may constrain the tunnel profile – the ARC documents assumed locomotive-friendly 2% grades, but with EMU-friendly 4% grades it’s certainly possible.

With this background, I believe Alt G was certainly feasible in the mid-2000s, and is still feasible today. This is why I keep pushing it in all of my plans. It’s also why I suspect that the reason Port Authority decided not to build Alt G was political: the hard numbers in the study, and the background that I got from Nichols, portray Alt G as superior to Alt P. The one complaint Nichols had, track capacity, misses the mark in one crucial way: the limiting factor is dwell times at Penn Station’s narrow platforms, and having two Midtown stations (Penn Station and Grand Central) would allow trains to dwell much less time, so if anything capacity should be higher than under any alternative in which trains only serve one of the two.

The upshot is that Christie had legitimate criticism of ARC; he just chose to cancel it instead of managing it better, which Aaron Renn called the Chainsaw Al school of government. After Christie canceled ARC, Amtrak stepped in, creating today’s Gateway project. Even without the cavern, Gateway’s estimate, $13.5 billion in 2011, was already higher than when Christie canceled ARC; it has since risen, and the highest estimate I’ve seen (by Metro, so caveat emptor) is $29 billion. This includes superfluous scope like Penn South, which at one point was supposed to cost $6 billion, but more recently Nichols told me it would be $7 billion.

While bare tunnels would provide the additional capacity required at lower cost, they would require interagency cooperation. Amtrak, New Jersey Transit, and the LIRR would need to integrate schedules and operations. Some trains from New Jersey Transit might run through to the east as LIRR trains and vice versa. This would make it easier to fit traffic within the existing station, and only add bare tunnels; the Penn Station-Grand Central section, at the southern end of the station, would keep dwell times down by having two Midtown stations, and the section connecting New Jersey Transit with Long Island (probably just Penn Station Access and one LIRR branch, probably the Port Washington Branch) would have 8 station tracks to play with, making dwell times less relevant. Unfortunately, this solution requires agencies to share turf, which they won’t – even the Penn Station concourses today are divided between Amtrak, New Jersey Transit, and LIRR zones.

Gateway is not the only rail project suffering from cost blowouts; it is merely the largest. The LIRR is building East Side Access (ESA), to connect to Grand Central; right now, it only serves Penn Station. ESA uses an underwater tunnel built in the 1960s and 70s to get to Manhattan, and is now boring a 2 km tunnel to Grand Central, at a cost of $10 billion, by far the most expensive rail tunnel in the world per unit length. But the tunnel itself is not the biggest cost driver. Instead of having the LIRR and Metro-North share tracks, ESA includes a deep cavern underneath Grand Central for the LIRR’s sole use, similar to the one in ARC that Christie canceled. About $2 billion of the cost of ESA is attributed to the cavern alone.

Agency turf wars are not unique to New York. In California, the same problem is driving up the costs of California HSR. In inflation-adjusted dollars, the project’s cost has risen from $33 billion in 2008 to $53 billion today. Most of the overrun is because the project includes more tunnels and viaducts today than it did in 2008. Much of that, in turn, is due to conflicts between different agencies, especially in the San Francisco Bay Area. The worst example is San Jose Diridon Station.

Diridon Station is named after still-living former California HSR Authority board member Rod Diridon, previously responsible for the disaster that is VTA Light Rail, setting nationwide records for low ridership and poor cost recovery. The station’s main user today is Caltrain. California HSR is planned to serve it on its way between Los Angeles and San Francisco, while Caltrain and smaller users plan to grow, each using its own turf at the station. The planned expansion of track capacity and new viaducts for high-speed rail is estimated to cost about a billion dollars. Clem Tillier calls it “Diridon Pan-galactic” and notes ways this billion-dollar cost could be eliminated, if the users of the stations shared turfs. Clem identifies $2.7 billion in potential savings in the Bay Area through better cooperation between high-speed rail, Caltrain, and other transit systems.

It is not a coincidence that the worst offenders – Gateway, East Side Access, and California High-Speed Rail – involve mainline rail. American and Canadian passenger railroads tend to be technologically and managerially conservative. Most still involve conductors punching commuter tickets as they did in the 1930s; for my NYU presentation, I found this picture from 1934.

I suspect that this comes from a Make Railroading Great Again attitude. Old-time railroaders intimately understand the decline of mainline rail in the United States in the middle third of the 20th century, turning giants like the Pennsylvania Railroad into bankrupt firms in need of federal bailouts. This means that they think that what needs to be done is in line with what the railroads wanted in the 1920s, 30s, 40s, and 50s. Back then, people lived in the suburbs and commuted downtown at rush hour, so there was no need for intra-suburban service, for in-city stops (those were for working- and middle-class city residents, not rich suburbanites in Westchester), or for high off-peak frequency. There was no need for cooperation between different railroads then, since commuters would rarely need to make an onward connection, which led to a culture encouraging competition over cooperation.

Among all the explanations for high construction costs, turf battles is the single most optimistic. But Americans should be optimistic about building cost-effective passenger rail. If this is the main culprit – and it is in the Bay Area, and one of several big culprits in New York – then all it takes to fix the cost problem is bringing organizational practices to the 21st century, which is cheap. It is too late for East Side Access, but it is possible to drastically reduce the cost of Gateway by removing unnecessary items such as Penn Station South. This can be repeated for smaller projects in the San Francisco Bay Area and everywhere in the US where two separate transit agencies fight over station space.

Am I optimistic that Americans will actually do this? I am not. Even outfits that should know better (again, the RPA) seem too conservative and too politically constrained; the RPA is proposing systemwide integration in its Fourth Plan, but in a way that incorporates each player’s wishlist rather than in a way that uses integration to reduce capital investment needs. In California, the HSR Authority seems to be responding to demands for value engineering by procrastinating difficult decisions, and it comes down to whether in the moment of truth it will have politicians in the state and federal governments who are willing to pay billions of dollars of extra money.

However, I do think that a few places might be interested in running public transit better. Americans are not incorrigible, and can learn to adapt best industry practices from other countries, given enough pressure. From time to time, there is enough pressure, it’s just not consistent enough to ensure the entire country (or at least the most important transit cities, led by New York) modernizes.

Suburban Transit-Oriented Development

Here’s a Google Maps image of Southport, a section of Fairfield, Connecticut with its own Metro-North commuter rail station:

Here’s an image at the same scale of Bourg-la-Reine, an inner suburb of Paris on the RER B, at the junction between the line’s two southern branches:

At Bourg-la-Reine, the buildings just east of the station are high-rise. There are local community amenities, including walkable schools, supermarkets, and pharmacies, and people can comfortably live in this suburb without a car. This generates significant RER traffic at all hours of day: outbound trains are often standing-room only until they reach this station even in midday, outside rush hour.

At Southport, there are a few townhouses near the station. But the roads are wide and hostile to pedestrians, and the nearest supermarket closes at 6 pm, too late for commuters returning from the city. Car ownership approaches 100%, and nobody rides the trains except to get to office jobs at the traditional peak hour in Manhattan (or perhaps Stamford).

The difference between the two places is so stark that they can barely be compared. Southport has 317 inbound boardings per weekday. Of those, 263, or 83%, are in the morning rush hour; the Metro-North-wide average is 63%, and the average on the SNCF-operated parts of the RER and Transilien is about 46%. Bourg-la-Reine has 4.5 million annual riders, about 16,000 on an ordinary working day.

A huge part of the difference is about service provision – Bourg-la-Reine has a train every five minutes midday, Southport a train every hour. But it’s not just about service. The RER has stations farther out, with somewhat less intense service, such as a train every 15 minutes, with comparable ridership. And the LIRR and Metro-North have little off-peak ridership even at stations with more frequent service, such as Mineola and Hicksville. Transit-oriented development (TOD) is as important as good service in such cases.

I bring up Southport because the RPA just dropped a study about suburban TOD that grades every New York commuter rail station between 0 and 3, and gives Southport the highest mark, 3. The RPA study looks at zoning within 800 meters of each station and considers whether there’s a parcel of land that permits multifamily housing with a floor are ratio higher than 1.25. Southport has such lots, supporting some townhouses, so according to the RPA it gets full marks, even though, by RER standards, it is like every other American car-oriented suburb.

Based on this methodology, the RPA identifies a number of good suburbs, and even comes to policy conclusions. It proposes more TOD in the mold of existing exurban New York examples, such as Patchogue. The model for the program is the real reason the RPA study is so weak: rather than calling into attention the big differences between land use at suburban stations in New York versus in Paris (or any number of big European cities with suburban rapid transit), it overfocuses on small differences within auto-oriented suburbia.

Some of the ultimate conclusions are not terrible. For example, the RPA is proposing linking federal infrastructure development to permitting more multifamily housing. This would improve things. However, the problem with this is twofold. First, it is unrealistic – the federal government gave up decades ago on enforcing fair housing laws, and has no interest in attempting to make exclusionary suburbs behave. Were I to propose this, hordes of American commenters would yell at me for not understanding American politics. And second, it misunderstands the nature of the problem, and ends up proposing something that, while unrealistic, is still low-impact.

The best way to understand the problem with the study is what author Moses Gates told me on Twitter when I started attacking it. He said that the RPA was looking at zoning rather than actual development. Since there is zoning permitting multifamily development within the prescribed radius at Southport, it gets full marks. With my understanding of what good TOD looks like, I would be able to say that this is clearly so bad the methodology must be changed; on Twitter I suggested looking at zoning within 300 meters of the station rather than 800, since the highest-intensity development should be right next to the station. I also suggested looking at supportive nonresidential uses, especially supermarkets. A development that isn’t walkable to retail at reasonable hours is not TOD.

The RPA does not think in this language. It thinks in terms of internal differences within the US. Occasionally it deigns to learn from London, but London’s suburban development is auto-oriented by European standards (transit mode share in the London commuter belt is at best in the teens, often in the single digits). Learning from anywhere else in the world, especially places that don’t speak English, is too difficult. This means that the RPA could not reach the correct conclusion, namely, that there is no such thing as an American suburb with TOD. The only exception I can come up with in the United States involves Arlington, on the Washington Metro, and Arlington is no longer considered a suburb, but really a full-fledged city in a different state, like Jersey City.

The other thing the RPA missed is that it drew too large a radius. TOD at a train station should include townhouses 800 meters out – but it’s more important to include high-rise residential construction next to the train station and mid-rise apartment buildings 500 meters out. Giving American suburbs latitude to place TOD so far from the station means they will act like Southport and allow small amounts of multifamily housing out of the way, while surrounding the station itself with parking, a tennis court, and large single-family houses with private swimming pools. This is not hypothetical: suburbs in New Jersey have reacted to court rulings mandating affordable housing by permitting apartments at the edge of town, far from supporting retail and jobs, and keeping the town core single-family.

Because the RPA missed the vast differences in outcomes between the US and France, it missed some useful lessons:

  • States should centralize land use decisionmaking rather than give every small suburb full autonomy.
  • TOD doesn’t need to be fully mixed-use, but there should be some local retail right next to housing.
  • Housing should be high-density right next to the station. A floor area ratio of 1.25 is not enough.
  • Publicly-funded social housing should be next to train stations, in the city as well as in the suburbs, and this is especially important in expensive suburbs, which aren’t building enough affordable housing.

Without suburban TOD, any regional rail system is incomplete. I wish I could have covered it at my talk, but I didn’t have time. Good service needs to run to dense suburbs, or at least suburbs with dense development within walking distance of the station. It needs to extend the transit city deep into suburbia, rather than using peak-only commuter rail to extend the auto-oriented suburbs into the city.

There’s More Redundancy Than You Think

I was visiting Boston last week, and am in New York this week; you can see me at NYU on Thursday tomorrow. Last week, I met with TransitMatters activists talking about bus and rail improvements in Boston, and on the way saw something that made me understand two things. First, the MBTA is run by incompetent people. And second, even two subway lines that are perpendicular and serve completely different areas can be redundant with each other.

Two and a half years ago, I said redundancy is overrated. In this post, I’d like to argue from the opposite direction: transit networks have more redundancy than they appear to. One implication is identical to that of my older post: transit agencies should build subway lines without regard for redundant service, since not only is redundancy overrated, but also a new subway line is redundant with old lines even if they serve completely different areas. But the other implication concerns service interruptions and shutdowns.

The issue in Boston is that, although there are nighttime shutdowns, there are also occasional weekend shutdowns, as in New York, for major capital projects. The Red Line is being closed on weekends for two months on the segment between Boston proper and Cambridge. But the Orange Line is also being closed on weekends on segments, after deferred maintenance led to a meltdown in the last two months, with frequent delays and slow zones. Last weekend, I found myself having to go between Davis Square (on the Red Line, just off the edge of the map) and Jamaica Plain (near the bottom of the Orange Line) to visit Sandy Johnston, with the highlit segments shut down:

Shuttle buses replaced the subway on both segments. On the Red Line, the MBTA contracted it out to a private company that used wheelchair-inaccessible high-floor buses; there were not enough MBTA bus drivers to run the shuttles on both segments, and by union rules the MBTA could not use contract drivers on its own buses even though it did have the equipment, forcing it to use inferior private-sector buses. I am able-bodied enough to climb high-floor buses, but I would not use the shuttle buses replacing the Red Line for another reason: as can be seen in the map, there is no continuous street grid between Charles/MGH and Park Street. If there were a crossover right east of Charles/MGH then only the Kendall-MGH segment would be bustituted, and there, the buses would go on Longfellow Bridge, with a serious but not fatal slowdown. But between Kendall and Park Street the buses have to swerve through side streets that were not designed for fast traffic; in 2012, I was on such a shuttle and as I recall the trip took 15 or 20 minutes, where the subway does it in about 5.

Instead of relying on shuttles, I took a bus north of the river to get to Lechmere and use the Green Line to reach Chinatown on a chain trip. From Chinatown the options were all bad, and I rode the 39 bus, which parallels the Green Line E Branch (the southernmost one) and continues south to Forest Hills, where the Green Line once ran as well. The way back was not a chain trip, and with a bus-bus-Red Line trip and no 39 bus in sight (the online bus tracker was down), I gave up and took a taxi.

The Red Line and Orange Line look like they go in different directions, so shutting down one does not affect the other. But in reality, in a city with buses, taking the bus to a different line is a common strategy to deal with shutdowns – hence, using the Green Line to get between Davis and Chinatown, taking a bus in a place where the buses are less slow than between Charles/MGH and Park Street.

If any city in North America did not use buses at all, it would be Boston. It has legendarily narrow and twisted streets, and crawling buses. It has higher rail-to-bus ridership ratio than any other American city except possibly New York, and far higher ratio than the major English Canadian cities with their bus grids. Its transit network, inherited from midcentury, uses the buses to feed the subway, and has no bus service through downtown, where even before mass motorization there were traffic jams of streetcars.

But even in Boston, using the bus outside the core to get to a better subway line is possible, and normal when there are service interruptions. This means that any pair of subway lines could potentially be redundant with each other. This means that it is bad practice to shut down more than one line at once for repairs. The reason the Orange Line needs emergency repairs in the first place is that the MBTA maintained it poorly and wouldn’t act when it was less urgent, such as six months ago (Sandy reports noticing a consistent deterioration in service since January). Today, the shutdowns are probably unavoidable. But the Red Line shutdowns, for a capital construction project involving the Longfellow Bridge, can be delayed. The MBTA should do that in the future in order to both avoid having to use inaccessible buses and allow passengers to take a circumferential bus to a functioning subway line.

Fix DeKalb Avenue

In New York, there are two dedicated subway tracks on the Manhattan Bridge offering a bypass of Lower Manhattan. Between DeKalb Avenue in Brooklyn and Canal Street in Chinatown in Manhattan, Q trains run nonstop for 3.5 km, while the R train goes the long way, taking 5.5 km and making 2 intermediate stops in Downtown Brooklyn and 4 in Lower Manhattan. The N skips DeKalb Avenue, with a 4.5 km nonstop segment between Canal Street and the Atlantic/Pacific/Barclays station complex.

The Q and N should be immense time savers. Instead, the Q does the trip in 8 minutes and the N in 10, both of which average 26-27 km/h. The subway’s overall average speed, weighed down by local trains stopping every 700 meters, is 29 km/h. The Q and N are still time savers, though, because the R does the 5.5 km in 18 minutes, an average speed of 16 km/h – far less than the systemwide average, and even less than the slowest Paris Metro line, Line 4 with its 500-meter interstations and 20 km/h average speed. Between DeKalb and Pacific, about 800 meters, the R takes 3 minutes. Unfortunately, New York City Transit is not taking any measures that would fix this, and when I asked about one possibility, I got excuses.

There are two reasons why this part of the subway is so slow. The first is something called signal timers. Timers are devices installed at frequent intervals on long interstations, such as the bridges and tunnels connecting Manhattan with Brooklyn and Queens, limiting train speed. These timers have always been around, but after fatal accidents in the 1990s, New York City Transit tightened them, reducing speed further; for some more background, see my Vox piece from last summer. The timers are more safety theater than safety. The biggest conclusion I reached from looking at the accident postmortem on the NTSB and some NYCT information was “make sure your trains’ brakes work as intended”; NYCT derated the trains’ service and emergency braking rates later in the 90s, which marginally reduces maintenance costs but is bad for safety and brutal for train speed.

The second reason is the switches at DeKalb Avenue. DeKalb is a six-track station, with four tracks feeding the Manhattan Bridge and two feeding the tunnel through Lower Manhattan. The two tunnel tracks then continue to the south as local tracks on the Fourth Avenue Line, carrying the R; this is the least used of all subway trunk lines into Manhattan, because the detour and low speed make it useless for most Midtown-bound passengers. The four bridge tracks include two express tracks at DeKalb going to the Brighton Line, and two super-express tracks skipping DeKalb continuing to the south as express Fourth Avenue tracks. Today, there is a splitting and recombining of branches. The B and D run together from Sixth Avenue to the Manhattan Bridge, and the N and Q run together from Broadway, but just north of DeKalb they recombine as B and Q running to Brighton, and D and N running super-express down Fourth Avenue.

This recombination at DeKalb slows down trains considerably, in two ways. First, the interlocking is complex. You can see it on this map on NYCSubway.org; in addition to splitting and recombining the B, D, N, and Q, it also has a non-revenue connection allowing R trains to serve the Brighton Line. Trains on diverging turnouts go at glacial speeds. And second, trains from four lines influence one another’s schedules, and delays propagate. Supervising train movements is thus difficult, and control center has to have a camera watching the trains enter the interlocking to ensure they adhere to schedule; timetables have to take the resulting delays into account.

When I first complained about reverse-branching in New York, I talked about capacity limits imposed by having more trunk lines than branches, a situation that is still to some extent true going north and east of Midtown. At DeKalb, there are six tracks going in and six going out, but the recombination makes things slower, and should be removed. NYCT should make a decision between having B and D trains run on the Brighton Line and the N and Q on Fourth Avenue, or the reverse. The interlocking permits either option, with entirely grade-separated junctions, allowing the trains on the two lines to no longer interfere with each other’s operations.

I in fact asked NYCT about it by proxy. NYCT dismissed the idea, on the grounds that transfer volumes between the B/D and N/Q would be too big. At Atlantic/Pacific, the Pacific side has a cross-platform transfer between the local R and express D/N, but going between the Pacific side and the Atlantic side (the B/Q, and separately the 2/3/4/5) involves a lot of walking. NYCT believes that passengers would flood the corridors looking for a train to their preferred destination, and the transfer volumes would require trains to have long dwell times. NYCT said nothing about whether the overall speed would actually fall, but I believe that based on the large transfer volumes NYCT predicts, passenger trip times (including transfer times) would rise. The only problem: I don’t believe NYCT’s prediction is true at all.

The B and D trains go express up Sixth Avenue, making stops at Grand Street in Chinatown, Broadway-Lafayette on Houston Street, West Fourth Street in the Village, and Herald Square. The N and Q trains go express up Broadway, serving Canal Street in Chinatown, Union Square, and Herald Square. North of Herald Square the two lines are never more than one long block apart until they leave Midtown. Passengers going toward Midtown are unlikely to have strong opinions about which of the two lines they would prefer.

Passengers going to destinations between Manhattan Bridge and Midtown might register stronger preferences. Union Square is the fourth busiest subway station in New York, and is quite far from the B and D. The closest alternative using the B and D is to change cross-platform to the M or F at West Fourth, and get off at 14th Street and Sixth Avenue, two long blocks from Union Square. Three more stations are potential concerns: Canal Street ranks 18th, West Fourth ranks 21st, and Broadway-Lafayette ranks 25th. Getting to Broadway-Lafayette from the N or Q is easy: the station and Canal Street are both on the 6, and passengers can transfer to the 6 at Canal.

West Fourth and Canal remain concerns, but they are not huge ones; they are secondary destinations. Canal is only a major destination for Chinese-New Yorkers, and in Brooklyn they cluster in Sunset Park along Fourth Avenue, suggesting that the Fourth Avenue express tracks should carry the N and Q and the Brighton tracks should carry the B and D. The urban geography of Chinese-New Yorkers is changing due to the combination of fast immigration and fast integration and migration to the suburbs, but this is a service decision, not an infrastructure investment; it can be reversed if demographics change.

Moreover, as a destination, West Fourth is predominantly used for NYU. The Village is a dense residential neighborhood, and West Fourth allows its residents to easily reach Lower Manhattan, Downtown Brooklyn, and two different four-track trunk lines through Midtown. But it has few jobs, outside NYU, which lies mostly between Sixth Avenue and Broadway. Union Square can adequately serve people going toward NYU, and stations on the R and 6 to the south can serve people going to NYU even better. The one problem is that the transfer between the R and the N/Q at Canal Street is not cross-platform; the cross-platform transfers start at Union Square. But with coverage of multiple stations walkable to NYU, the loss of the one-seat ride to West Fourth is not fatal. Even the transfer to the A, C, and E trains at West Fourth has alternative options: passengers from the N or Q going to the E can transfer to the F or M at Herald Square and reach the same stations, and passengers going to the A or C can transfer to the 1 at Times Square and to the A or C at Columbus Circle, both of which transfers are not much harder than climbing two flights of stairs at West Fourth.

With so many options, not many riders would be connecting at Atlantic/Pacific, and trains could keep dwell times short. If anything, dwell times might be shorter, because missing a train would be less fatal: the next train on the same track would serve the same destinations in Midtown, so riders would only need to wait about 3 minutes at rush hour, and 5 minutes off-peak. The gain in speed would be substantial, with the interlocking imposing fewer operational constraints.

NYCT might need to slightly rework the switches, to make sure the chosen matching of the lines in Manhattan and Brooklyn takes the straight and not the diverging direction at the turnouts; typically, the straight direction imposes no speed limit (up to full line speed on high-speed rail lines), but the diverging direction is slow. A matching in which the B and D go on Brighton and the N and Q on Fourth Avenue express to my understanding already involves only one diverging move, if I am reading the track map linked on NYCSubway.org correctly. At the same time, NYCT could fix the switches leading to the R: there was through-service from the Brighton Line to the tunnel tracks the R uses today, but there no longer is, so this out-of-service connection should get diverging and not straight moves. But even with the R, the capital investment involved is minimal.

I do not know the potential travel time gains between DeKalb and Canal Street (or Grand Street) with no timers or reverse-branching. With straight tracks across Manhattan Bridge, and wide curves toward Grand Street, 3.5-minute trips are aspirational, 4-minute trips are still possible, and 5-minute trips should be easy. From Pacific Street, add one more minute, corresponding to cruising at 50 km/h, a speed limit the subway routinely attains even on local tracks. This saves passengers from DeKalb about 4 minutes, and passengers from Pacific about 5. The average trip across the system is about 21 minutes, and the average delay (“excess journey time“) is 3 minutes. The saving would be immense, and contribute to both more casual ridership between Brooklyn and Manhattan, and lower operating costs coming from faster trips.

NYCT should not make excuses for this. The timers may have been originally justified as a safety improvement, but reducing train braking rates had the opposite effect. And, uniquely among the various reverse-branch points in New York, DeKalb feeds two Manhattan trunks that are very close to each other, especially in Midtown, to the point that one-seat rides to every stop have limited value. It should make a decision about whether to run the B/D together on Fourth Avenue and the N/Q on Brighton (switching the Q and D) or the reverse (switching the B and N), based on origin-and-destination data. Some passengers might bemoan the loss of one-seat rides, but most would cheer seeing their trips sped up by 4-5 minutes.

Meme Weeding: Land Value Capture

Last month’s Patreon poll was about meme weeding – that is, which popular meme in public transit I should take apart. The options were fare caps on the model of London, popular among some US reformers; wait assessment, a schedule adherence metric for trains I briefly complained about on Vox as used in New York; and land value capture/tax increment financing/the Hong Kong model. The last option won.

Good public transit creates substantial value to its users, who get better commutes. It’s an amenity, much like good schools, access to good health care, and clean air. As such, it creates value in the surrounding community, even for non-users: store owners who get better sales when there’s better transportation access to their business, workers who can take local jobs created by commuters to city center, and landowners who can sell real estate at a higher price. All of these positive externalities give reason to subsidize public transit. But in the last case, the positive impact on property values, it’s tempting to directly use the higher land values to fund transit operations; in some cases, this is bundled into a deal creating transit-oriented development to boost ridership. In either case, this is a bad way of funding transit, offering easy opportunities for corruption.

Value capture comes in several flavors:

  • In Japan, most urban private railroads develop the areas they serve, with department stores at the city end and housing at the suburban end.
  • In Hong Kong, the government sells undeveloped land to the now-privatized subway operator, the MTR, for high-density redevelopment.
  • In the US and increasingly Canada, local governments use tax increment funding (TIF), in which they build value-enhancing public infrastructure either by levying impact fees on development that benefits from it or by programming bonds against expected growth in property taxes.

In both Hong Kong and the major cities of Japan, urban rail operations are profitable. It is not the case that value capture subsidizes otherwise-money losing transit in either country, nor anywhere I know of; this did not prevent Jay Walder, then the head of New York’s MTA, from plugging the MTR model as a way of funding transit in New York. What’s true is that the real estate schemes have higher margins than rail operations, which is why JR East, the most urban of the remnants of Japan National Railways, aims to get into the game as well and develop shopping centers near its main stations. However, rail operations alone in these countries are profitable, due to a combination of high crowding levels and low operating costs.

The Japanese use case is entirely private, and does not to my knowledge involve corruption. But the Hong Kong use case is public, and does. For all the crowing about it in Anglo-American media (the Atlantic called it a “unique genius” and the Guardian said it supported subsidy-free operations), it’s a hidden subsidy. The state sells the land to the MTR, and the MTR alone, at the rate of undeveloped outlying land. Then the MTR develops it, raising its value. Other developers would be willing to pay much better, since they can expect to build high-density housing and have the MTR connect it to Central. This way, the government would pocket the profits coming from higher value on its land. Instead, it surreptitiously hands over these profits to the MTR.

While Western media crows about Hong Kong as an example of success, local media excoriates the corruption involves. Here’s the South China Morning Post on the MTR model:

The rail and property model was never anything but a delusion to which only Hong Kong bureaucrats could be subject. It traded on the odd notion that you cannot assign a value to property until you actually dispose of it.

Thus if you give the MTR the land above its stations, these sites suddenly and magically acquire value and the proceeds cover the cost of building the railway lines. Ain’t magic wonderful? We got the MTR for free.

Stephen Smith dealt with this issue in 2013, when he was still writing for NextCity. He explained the local corruption angle, the fact that MTR rail operations are profitable on their own, and the lack of undeveloped land for the state to sell in most first-world cities. (Conversely, one of his arguments, about construction costs, doesn’t seem too relevant: Hong Kong’s construction costs are probably similar to London’s and certainly higher than Paris’s, and doing value capture in Paris would be an urban renewal disaster.)

Stephen also tackles American examples of value capture. With no state-owned land to sell to the public transit agency at below-market prices, American cities instead rely on expected property taxes, or sometimes levy special fees on developers for letting them build TOD. Stephen talks about scale issues with the TIF-funded 7 extension in New York, but there are multiple other problems. For one, the 7 extension’s Hudson Yards terminus turned out to be less desirable than initially thought, requiring the city to give tax breaks. See for examples stories here, here, and here.

But there are more fundamental problems with the approach. The biggest one is the quality of governance. TIF is an attractive-looking option in American jurisdictions that recoil at raising direct taxes to pay for service. This means that as happened in New York, it is tempting for cities to promise property tax windfall, issue bonds, and then let successor governments raise taxes or cut services to pay interest. This opaqueness makes it easier to build bad projects. When the government promises especially high benefit-cost ratios, it can also keep issuing new bonds if there are budget overruns, which means there is no incentive for cost control.

TIF also requires the city to use zoning to create a shortage of land in order to entice developers to pay extra to build where it wants them to. Stephen complains that New York reamed problems on upzoning in Midtown East, one of the few locations in Manhattan where developers are willing to build supertall office towers without any tax breaks; the new zoning plan, in the works since he was writing for NextCity in 2013, only just passed. Another such location is probably the Meatpacking District, near the Google building at 14th and 8th, now the city’s tech hub – there is no tall office construction there due to the power of high-income residential NIMBYs. Were the city to loosen zoning in these areas and permit companies that need a prime location to set up offices in these areas, it would find it even harder to entice developers to build in a lower-demand area like Hudson Yards. Midtown East and the Meatpacking District are replete with subway lines, but there are no new plans for construction there, so the city wouldn’t do a TIF there.

The same problem, of TOD-reliant funding requiring the city to restrict development away from targeted investment areas, also works in reverse: it encourages development-oriented transit. In 2007, Dan Doctoroff, then a deputy mayor and now head of Google’s Sidewalk Labs, opposed Second Avenue Subway, on the grounds that the area is already developed. Second Avenue Subway was eventually built, but the 7 extension omitted a stop in an already-developed area amidst cost overruns, as Bloomberg prioritized Hudson Yards. This is not restricted to New York: San Francisco is more interested in a subway to Parkmerced than in a subway under Geary, the busiest bus route, busier than the subway-surface light rail branch serving Parkmerced today. Smaller American cities propose core connectors, aiming promoting redevelopment in and around city center. This in turn means ignoring low-income neighborhoods, where there is no developer interest in new buildings except as part of a gentrification process.

These problems are for targeted investments. But when there is more widespread TOD, TIF ends up being a tax on transit users. Cities build roads without levying special taxes on sprawling development, whether it sprawls by virtue of being near the highway or by virtue of being far from public transit. When they build transit, they sometimes tax TOD, which means they are giving developers and residents tax incentives to locate away from public transit.

Hong Kong is not the right model for any TOD scheme; its corruption problems are immense. It’s a shiny object for Americans (and other Anglophone Westerners), who are attracted to the allure of the exotic foreigner, like a premodern illiterate attributing magic to the written word. Instead of replicating its most questionable aspect, it’s better to look at models that are attractive even to local corruption watchdogs.

This means funding public transit and other services out of transparent, broad-based taxes. Paris uses a payroll tax, varying the rate so as to be higher in the city (2.95%) than in the outer suburbs (1.6%). Everyone will hate them, especially people who don’t use transit and don’t view it as directly necessary for their lives. This is why they work. They compel the transit agency to run efficient service, to stave off opposition from aggrieved center-right middle-class voters, and to run it well, to stave off opposition from populists (“why am I being taxed for trains that break down?”). They leave no room for waste, for cronyism, or for slush funds for favored causes, precisely because they’re hard to pass.

It’s easy to see why politicians avoid such funding sources. The democratic deficit of local governance in the US is immense, and that of Canada is only somewhat better. Nobody wants to lose an election over raising taxes, even in cities where the political spectrum runs from the center leftward. Value capture sounds like a good, innovative idea to fund government without hated taxation, and its abuses are hidden from sight. Even as it forces city residents to endure opaque fees (never call them taxes!), it wins accolades to politicians who propose it. No wonder it continues despite its failures.

Anti-Infill on Surface Transit

I wrote about infill stops on commuter rail two weeks ago, and said I cannot think of any example of anti-infill on that mode. But looking at Muni Metro reminded me that there is need for anti-infill on surface transit. This is called stop consolidation normally, and I only use the term anti-infill to contrast with the strategy of adding more stops on commuter trains.

The root of the problem is that in North America, transit agencies have standardized on 200-250 meters as the typical spacing between bus stops. In Europe, Australasia, and East Asia, the standard is instead 400-500 meters. Even without off-board fare collection, the difference in speed is noticeable. In Vancouver, the difference between the local 4 and the express 84 is substantial: on the shared segment between Burrard and Tolmie, a distance of 4.8 km, the 84 makes 5 stops and takes 10 minutes, the 4 makes 18 stops and takes 16 minutes. A bus with the normal first-world stop spacing would make 10-12 stops and take, linearly, 12-13 minutes. 23 km/h versus 18 km/h.

With off-board fare collection, the impact of stop spacing on speed grows. The reason is that a bus’s stop penalty consists of the time taken to stop and open its doors, plus the time it takes each passenger to board. The former time is independent of the fare collection method but depends on stop spacing. The latter time is the exact opposite: if the stop spacing widens, then there are more passengers per bus stop, and unless the change in stop spacing triggers changes in ridership, overall passenger boarding and alighting time remains the same. Another way to think about it is that judging by Vancouver data, there appears to be a 30-second stop penalty, independent of ridership. Off-board fare collection increases bus speed, so the 30-second stop penalty becomes more important relative to overall travel time; the same is true of other treatments that increase bus speed, such as dedicated lanes and signal priority.

In New York, there aren’t a lot of places with local and limited-stop buses side by side in which the limited-stop bus has on-board fare collection. One such example is the M4, meandering from Washington Heights down the 5th/Madison one-way-pair, over 15.3 km. At rush hour, the local takes 1:45, the limited-stop takes 1:30: 9 vs. 10 km/h. But the limited-stop bus runs local for 6 km, and over the other 9.3 km it skips 26 local stops if I’ve counted right. The B41 has a limited-stop version over 8.3 km (the rest is local), skipping about 17 stops; the time difference is 10 minutes.

One possible explanation for why the stop penalty in New York seems a little higher than in Vancouver is that the M4 and B41 routes are busier than the 4/84 in Vancouver, so every stop has at least one passenger, whereas the 4 in Vancouver often skips a few stops if there are no passengers waiting. Conversely, the higher passenger traffic on buses in New York comes from higher density and more traffic in general, which slows down the buses independently of stopping distance.

On subways, there’s reason to have more densely-spaced stops in denser areas, chief of which is the CBD. On surface transit, it’s less relevant. The reason is that absolute density doesn’t matter for stop spacing, except when expected ridership at once station is so high it would stress the egress points. What really matters is relative density. Putting more stops in an area means slowing down everyone riding through it in order to offer shorter station access times to people within it. On surface transit, relative density gradients aren’t likely to lead to variations in stop spacing, for the following reasons:

  1. Historically, surface transit stop spacing was always shorter than rapid transit stop spacing because of its lower top speed and the faster braking capabilities of horses vs. steam trains; often people could get off at any street corner they chose. So it induced linear development, of roughly constant density along the corridor, rather than clusters of high density near stations.
  2. If there is considerable variation in density along a surface transit line, then either density is medium with a few pockets of high density, which would probably make the line a good candidate for a subway, or density is low with a few pockets of higher density, and the bus would probably skip a lot of the low-density stops anyway.

Most importantly, the 400-meter standard is almost Pareto-faster than the 200-meter standard. In the worst case, it adds about 4 minutes of combined walking time at both the start and the end of the trip, for an able-bodied, healthy person not carrying obscene amounts of luggage. The breakeven time on 4 minutes is 8 skipped stops, so 3.2 km compared with the 200-meter standard. Bus trips tend to be longer than this, except in a few edge cases. In New York the average unlinked bus trip is 3.4 km (compare boardings and passenger-km on the NTD), but many trips involve a transfer to another bus or the subway, probably half judging by fare revenue, and transfer stations would never be deleted. If the destination is a subway station, guaranteed to have a stop, then the breakeven distance is 1.6 km.

This also suggests that different routes may have different stop spacing. Very short routes should have shorter stop spacing, for example the 5 and 6 buses in Vancouver. Those routes compete with walking anyway. This may create a spurious relationship with density: the 5 and 6 buses serve the very dense West End, but the real reason to keep stop spacing on them short is that they are short routes, about 2 km each. Of course, West End density over a longer stretch would justify a subway, so in a way there’s a reason short optimal stop spacing correlates with high bus stop density.

The situation on subways is murkier. The stop penalty is slightly higher, maybe 45 seconds away from CBD stations with long dwell times. But the range of stop distances is such that more people lose out from having fewer stops. Paris has a Metro stop every 600 meters, give or take. Some of the busiest systems in countries that were never communist, such as Tokyo, Mexico City, and London, average 1.2 km; in former communist bloc countries, including Russia and China, the average is higher, 1.7 km in Moscow. The difference between 600 meters and 1.2 km is, in the worst case, another 1.2 km of walking, about 12 minutes; breakeven is 16 deleted stops, or 20 km, on the long side for subway commutes.

One mitigating factor is that subway-oriented development clusters more, so the worst case is less likely to be realized, especially since stops are usually closer together in the CBD. But on the other hand, at 1.2 km between stations it’s easy for transfers to be awkward or for lines to cross without a transfer. London and Tokyo both have many locations where this happens, if not so many as New York; Mexico City doesn’t (it’s the biggest subway network in which every pair of intersecting lines has a transfer), but it has a less dense network in its center. Paris only has three such intersections, two of them involving the express Metro Line 14. Even when transfers do exist, they may be awkward in ways they wouldn’t have been if stop spacing had been closer (then again, Paris is notorious for long transfers at Chatelet and Montparnasse).

In all discussions of subway stop spacing, New York is sui generis since the lines have four tracks. On paper its subway lines stop every 600-700 meters when not crossing water, but many trains run express and stop every 2 km or even more. Average speed is almost the same as in Tokyo and London, which have very little express service, and it used to be on a par until recent subway slowdowns. This distinction, between longer stop spacing and shorter stop spacing with express runs, also ports to buses. Buses outside the US and Canada stop every 400-500 meters and have no need for limited-stop runs – they really split the difference between local and limited buses in North America.

On a subway, the main advantage of the international system over the New York system is obvious: only two tracks are required rather than four, reducing construction costs. On a bus line, the advantages are really the same, provided the city gives the buses enough space. A physically separated bus lane cannot easily accommodate buses of different speeds. In New York, this is the excuse I’ve heard in comments for why the bus lanes are only painted, not physically separated as in Paris. Mixing buses of different speeds also makes it hard to give buses signal priority: it is easy for buses to conflict, since the same intersection might see two buses spaced a minute apart.

Buses also benefit from having a single speed class because of the importance of frequency. In Vancouver, the off-peak weekday frequency on 4th Avenue is an 84 rapid bus every 12 minutes, a 44 rapid bus every 20 minutes, and a local 4 every 15 minutes. The 84 keeps going on 4th Avenue whereas the 4 and 44 divert to Downtown, but the 4 and 44 could still be consolidated into a bus coming every 10 minutes. If there were enough savings to boost the 84 to 10 minutes the three routes could vaguely be scheduled to come every 5 minutes on the common section, but without dedicated lanes it’s probably impossible to run a scheduled service at that frequency (pure headway management and branching don’t mix).

The example of 4th Avenue gets back to my original impetus for this post, Muni Metro. Only diesel buses can really run in regular surface mode mixing different speed classes. Trolleys can’t. Vancouver runs trolleys on the local routes and diesels on the limited routes. At UBC, it has different bus loops for diesels and trolleys, so people leaving campus have to choose which type of bus to take – they can’t stand at one stop and take whatever comes first.

On rail, this is of course completely impossible. As a result, American subway-surface trolleys – the Boston Green Line, SEPTA’s Subway-Surface Lines, and Muni Metro – all run at glacial speed on the surface, even when they have dedicated lanes as in Boston. In Boston there has been some effort toward stop consolidation on the Green Line’s busiest branch, the B, serving Boston University. This is bundled with accessibility – it costs money to make a trolley stop wheelchair-accessible and it’s cheaper to have fewer stops. Muni Metro instead makes one stop every 3-5 accessible (on paper), but keeps stopping at all the other stops. It would be better to just prune the surface stops down to one every 400-500 meters, which should be accessible.

If you view rail as inherently better than bus, which I do, then it fits into the general framework: anti-infill on surface transit has the highest impact on the routes with the best service quality. Higher speed makes the speed gain of stop consolidation more important relative to travel time; trolleywire makes it impossible to compensate for the low speed of routes with 200-meter interstations by running limited-stop service. Even on local buses, there is never a reason for such short stop spacing, and it’s important for North American cities to adopt best industry practice on this issue. But it’s the most important on the highest-end routes, where the gains are especially large.