Scheduling Trains in New Jersey with the Gateway Project

Devin and I have draft timetables for intercity and commuter trains on every segment of the Northeast Corridor; what is left is to merge the segments together and see how they interact, tweak based on further constraints, and look at some alternatives. The good news is that in New Jersey, the last area we looked at, sharing tracks turns out to be easy. It’s a happy accident of how the Northeast Corridor has been designed that, with 21st-century train specs, the places where fast trains need to overtake slow ones already have long sections with additional tracks. Work is still required on grade-separating some junctions (chiefly Hunter Interlocking) and fixing some curves largely within the right-of-way, but it’s rather minor. The upshot is that local commuter trains can do New York-New Brunswick in 38 minutes and would do New York-Trenton in an hour, the express commuter trains can do New York-Trenton in 51 minutes, and the intercity trains can do New York-Philadelphia in 45 minutes, all with new rolling stock but few expensive investments in infrastructure beyond what’s already funded as part of the Hudson Tunnel Project for Gateway.

Three speed classes

The Northeast Corridor near New York presents two planning difficulties. First, there is a very large volume of peak commuter traffic into Manhattan, which forces agencies to build infrastructure at the limit of track capacity. And second, there is a long stretch of suburbia from Manhattan, which means that some express commuter rail service is unavoidable. This means that both the New Haven Line and the NJ Transit Northeast Corridor Line have to be planned around three speed classes: local commuter, express commuter, and intercity; moreover, the total volume of trains across these classes must be large, to accommodate peak demand, reaching 24 peak trains per hour. This is why the Hudson Tunnel Project is being built: the existing tunnels run 24 trains per hour already split across many different commuter rail branches, and all of the trains are crowded.

The difficulties in New Jersey and in Metro-North territory are different; for a taste of what is needed for Metro-North, see here. In New Jersey, the quality of the right-of-way is high, and the outer stretches are already cleared for a maximum speed of 160 mph, and with if the Federal Railroad Administration (FRA) had more faith in the quality of rolling stock windows they could run much faster than this. The inner stretches are slower but still straight enough for fairly high speed – there are long stretches straight enough for 250 km/h and one section where trains could even briefly reach 300 km/h. Thus, the express commuter trains are noticeably slower than the intercity trains on these segments despite running nonstop from Newark to Metropark.

All trains are significantly faster than today. Little of the speedup comes from any curve modification; rather, it comes from reduced timetable padding (down to Swiss-standard 7%), plus about 1.5 minutes of speedup in the Penn Station throat from better switch geometry.

Six-track overtakes

The Northeast Corridor is largely quad-track, but two sections have six tracks, both in New Jersey: around Newark Airport, and from just south of Elizabeth to just south of Rahway, where the North Jersey Coast Line branches off. The four-track section through Elizabeth is annoying, and I was hoping that it would not be necessary to delicately schedule around it. It is fortunate that my hopes have proved correct.

Below is a rough line chart. For one, it does not have any schedule padding. For two, there are still some additional slowdowns not coming from right-of-way geometry not incorporated into it, and in particular there’s a minute of Penn Station and tunnel delay not yet depicted for the intercity train and another 30 seconds of same for the commuter trains. For three, all station dwell times are set at 30 seconds, whereas the intercity needs a minute. In total, the last two factors delay the intercity by a minute relative to all commuter trains by when they depart Newark. All of these factors figure into the trip times above, but not the line chart below.

The blue lines are intercity trains, the red lines are express commuter trains, the green lines are local commuter trains to New Brunswick or Jersey Avenue, the purple lines are local commuter trains branching to the North Jersey Coast Line, and the gold lines are SEPTA trains.

Of note, the intercity trains do not share tracks with the local commuter trains except in the tunnel to Penn Station; the current plan after the Hudson Tunnel Project is finished is for the above-depicted trains to use the old tunnel and for other lines (Morris and Essex, Montclair-Boonton, Raritan Valley) to use the new tunnel. This provides just enough separation that there isn’t much interlining to worry about. The express commuter trains are the only ones with any surface track-sharing with trains of different speed classes.

As the line chart shows, the red/green overtake occurs at Elizabeth, where the express commuter trains then need to be on the inner express tracks. Just south of Elizabeth, the line widens to six tracks, and the express commuter trains can be kept separate from both local trains and intercity trains; all that’s required is installing switches to allow this, for a very small number of millions of dollars for high-speed switches or hundreds of thousands for slower switches. By the time the intercity and express commuter trains are within the signal system’s two-minute limit of each other, the express commuter trains don’t need to return to the inner tracks again. Past Rahway, the express and local commuter trains need to use the same tracks, but are adequately separated from each other.

Robustness check

We are still looking at options for how to match this segment with other segments, in particular how this could through-run east of Penn Station. Most likely, the local trains would run through to the Port Washington Branch of the LIRR and the express commuter trains would become local commuter trains to Stamford via Penn Station Access.

The upshot is that the train most likely to be delayed from the north is the express commuter train. It can afford to be about two minutes behind schedule before it messes up the order of trains using the tunnel; the schedule padding up to Elizabeth can recover one of these two minutes, and then, with the extra minute of slowdown of intercities not depicted in the line chart, the express commuter trains are still well clear of the intercities where they share tracks at Elizabeth.

Stop Spacing on Crosstown Routes

Two different issues in New York – the bus redesign process and the Interborough Express – are making me think about optimal stop spacing again. I blogged about it in general about buses a few days ago, but crosstown routes present their own special issues, and this is noticeable on rail more than on buses. Circumferential rail routes, in particular, can justify wider stop spacing than radial routes in certain circumstances. This can explain why, over the iterations of Triboro RX leading to the current IBX proposal, the stop spacing has widened: the Third Regional Plan-era effort in the 2000s had a stop every half mile in Brooklyn and Queens, but more recent efforts proposed fewer stops, and the current one if anything has too few and misses a transfer.

Density and isotropy

The tradeoff in stop spacing on both buses and trains is that more stops reduce the amount of walking to the station but increase the in-vehicle trip time for people going through the stop without getting on or off.

Density by itself does not affect this tradeoff. A uniform increase in density along a line equally increases the costs and benefits of changing the stop spacing. However, relative density matters: stop spacing should be tighter in areas with higher density and wider in areas with lower density, both relative to other areas along the same line. This is because higher relative density means passengers are disproportionately likely to have their origin or destination in this area, and disproportionately less likely to be traveling through it, both of which argue in favor of tighter stop spacing, and lower relative density means the opposite.

This then leads to the issue of isotropy. On an isotropic network, relative density is by definition always the same; spikes in relative density make travel less isotropic. As my previous post explains with bus stop spacing formulas, also valid on rail with different parameters, less isotropic density should mean not just that there should be more stops in some places and fewer in others, but also that there should be fewer stops overall. In the simplest case of non-isotropy, assume everyone is traveling to the same distinguished node, which on a rail line can be thought of as city center (let’s say there’s just one central transfer point) and on a bus can be thought of as the connection to the subway. Then, all passengers can be guaranteed to be going to a place with a station, and therefore the cost of widening the stop spacing is halved, since only the origin walk time is increased, not the destination walk time.

Isotropy and circumferential routes

Successful circumferential routes live off of their ability to connect to the rest of the network. Over time, those connection points may grow to become large destinations in their own right – this is the story of how Ikebukuro, Shinjuku, and Shibuya, all at the intersection of the Yamanote Line with radial rail links (JR, private, or subway), became large business districts. But the connections have to come first. If passengers can’t conveniently transfer, then the route has to live off of origin-and-destination traffic just on the line, and then, because it is circumferential and by definition doesn’t go to city center, traffic will be low. This principle is why the G train in New York is so weak: it may connect the two largest non-Manhattan job centers in the region, but that’s still neither Manhattan nor service to the entire city, and with poor transfers, it has to live off of the small number of people living in Williamsburg and Bedford-Stuyvesant working in Long Island City or Downtown Brooklyn.

But the same principle also means that non-transfer stops lose value. This doesn’t mean there shouldn’t be any of them, but it does mean that agencies can afford to be pickier about where to place them. They’re unlikely to be destinations, only origins, and even as origins their value is discounted since some passengers use the circumferential line as the second leg of a three-legged trip, between two radial lines.

The impact on IBX

I used to criticize the decision to build fewer stops on IBX. For example, here, when it was still an RPA proposal in what would later become the Fourth Regional Plan, I outlined several criticisms of the then-Triboro route. I think some of them stand, especially the section on the plan to have the route go into the Bronx and provide local commuter rail service to Coop City. However, on the matter of stop spacing, I must withdraw the criticism.

That said, a station at every connection with a radial rail line remains nonnegotiable. IBX errs in only stopping in East New York at Atlantic Avenue, connecting to the L and the LIRR, with no direct connection to Broadway Junction for the A/C and J. The distance between these two locations is only 350 meters, and it may be awkward to have two stops in short succession, but the meaning of high relative density is exactly that it’s okay to have more closely spaced stops. Alternatively, there could be one stop at a compromise location, with in-system connections at both ends, but then the walk times would be higher, which is less desirable.

Militarized and Other New Capital Cities

The news of the ongoing construction of Nusantara, Indonesia’s new capital in Borneo to replace Jakarta, got me thinking about other moves by various countries to create new capitals from scratch, to avoid having to deal with the urbanity of the existing capital. On this issue, Nusantara joins Brasilia, Islamabad, (at the subnational level) Chandigarh, and, most ominously, Naypyidaw and Egypt’s New Administrative Capital (NAC). The last two, unlike Nusantara, are built for explicitly military purposes, with the military considering the people to be its main adversary rather than any external enemy. Such capitals always waste money that could be spent on improving literally anything else – health, education, transportation, water, electricity, business climate, anti-poverty grants. In the less militarized cases, like Brasilia, these are just waste; in the more militarized cases, these showcase that the state is run by parasites.

Non-militarized cases

The move to Nusantara is being justified on the grounds that Jakarta is sinking. In truth, it’s better to view this as a continuation of the Transmigrasi program that the state has engaged in for decades. The zeitgeist was one of concerns about overpopulation leading to either forced sterilization of the poor or a program of settlement by the core population in peripheral regions; in Indonesia, this took the form of encouraging Javans to settle elsewhere in the country, where in practice they formed an overclass and sidelined the preexisting population. Nusantara, in a mostly undeveloped part of Borneo, needs to be viewed within this program, rather than as a tragic response to climate change.

Brasilia, similarly, was built on developmentalist grounds: the Brazilian elite wanted to develop the interior of the country, viewing the rainforest and savanna as low-value land to be mined and farmed. As this process predates the military dictatorship (1964-85; Brasilia was founded 1960), this is not really a matter of militarization. Rather, it’s better viewed as developmental failure – the resources invested in the new city could have gone to more productive uses, and the value of that land for farming and mining turned out not to be much.

The environmental impacts of the program of developing the interior were wholly negative. In Brazil, half of the country’s greenhouse gas emissions are land use changes and another quarter are from the agriculture sector, leading Brazil to emit 10.7 t/capita in greenhouse gases. In Germany, the figure is 8.1 t/capita, with a large industrial sector and an infamously still substantial coal sector, and on these lower emissions, Germany produces close to four times Brazil’s GDP per capita. The environmental impact of Nusantara is likely to be the same in direction as Brazil’s program, and to the extent it’s likely to be smaller in magnitude, it’s because it’s a smaller endeavor than Brazil’s entire program of developing the interior, in which Brasilia was just one component.

Militarized cases

The worst are not Brasilia, or likely Nusantara, but rather Naypyidaw and the NAC. I’ve been asked to provide some resources on social media, going over what’s involved in both projects.

Naypyidaw

The military junta that ran Myanmar between 1962 and 2011 (with significant tightening in 1988) and has run it since 2021, and exercised significant power between 2011 and 2011, built the new capital, officially since 2005, unofficially since a few years earlier, at a remote location 320 kilometers from Yangon and 240 from Mandalay. After the 1988 protests and the crackdown establishing the military government that 1990s-2000s human rights advocates knew and hated, the regime made changes to Yangon to suppress future protests, evicting 500,000 people from city center in the process. Where normally such efforts mostly target poor people in order to create the illusion of a poverty-free city center and facilitate urban renewal, in Yangon the evictions targeted the middle class, which was sympathetic to the protest movement and had communities that had sheltered protesters. But even the new Yangon was not good enough for the generals, and so they shifted to Naypyidaw.

Part of the reason was that Yangon was too multiethnic, in a part of the country that was majority-Mon until the 20th century, while Naypyidaw could be more comfortably Bamar. But the main reason was security needs. The adversary in this case is not any foreign government – a city built in 2005 by a government that might worry about American-led regime change would look at what was going on in Iraq and opt to maintain its capital in a large, dense central city to facilitate insurgency and make it easy to hide among the civilian population. Naypyidaw instead does the exact opposite – it’s easy for a superior military force to take. Rather, its threat model is a popular uprising, and thus the modernist planning with separation of uses exists to prevent the broad public from being able to stage an insurgency. There is extensive regime propaganda in the city, such as national museums telling uncritical histories, but no major religious sites, since those might shelter protesters, as happened in 1988.

The urbanism of Naypyidaw is, essentially, a giant military camp. It is designed with strict separation of uses and large roads between different complexes for movement between them; people are not expected to walk between places. The Guardian calls it a post-apocalyptic suburbia, but I don’t think that’s quite right. In an auto-oriented American city, there’s nowhere to walk, but everyone owns a car and the development forms make it convenient to drive to one’s work and regular non-work destinations. Dubai, criticized vociferously among urbanists for its tackiness and auto-oriented character, is a place where one can drive or be driven to shopping malls and towers within a close distance. Naypyidaw doesn’t have any of this and doesn’t seem to even try; it’s a collection of sites, designed for no need but that of control by the military of a population that does not want to be controlled by it. It’s a monument not to modernist urbanism, though it tries to affect that, but rather to destruction of value by an unwanted government.

New Administrative Capital

The situation in Egypt, to be clear, is a lot less brutal than in Burma. At the same time, Egypt is several times richer, which creates more value that can be extracted and given over to cronies. This can be seen in the construction of the NAC, to replace Cairo.

The immediate history of the NAC is that in 2011, Egypt famously had the Arab Spring revolution overthrowing Hosni Mubarak; in the subsequent election, the only preexisting organized political force, the Muslim Brotherhood, won the election, leading Mohamed Morsi to take office as president starting in 2012. The election was free and the state of civil liberties improved, but the Muslim Brotherhood was making moves to consolidate power, leading to fears among human rights and democracy protesters of new authoritarianism, which led to a protest movement in 2013 called Tamarod, demanding new elections; later in the same year, the military reacted to the protests by launching a coup, removing Morsi from power and restoring the military elements of the prior regime, including a pardon of Mubarak, who was on trial for corruption. Eventually, the chief of the Egyptian Armed Forces, General Abdel Fattah al-Sisi, won a rigged presidential election in 2014, and has been president ever since. Much of the focus of the protest movements was street protests in Cairo, where Tahrir Square became a global metonym for democracy protests (for example, in Israel, where people basically never draw any positive inspiration from Arab political trends). This created a need among Sisi and his inner circle for a new capital in the desert, built to forestall any future Tahrir.

The capital could not even be named New Cairo, because there’s already a New Cairo, an eastern suburb of Cairo built expressly in order to decentralize the capital; Cairo is a huge, dense city. For a while, it was mocked as New New Cairo, for its location even farther east of New Cairo; by now, it has the formal name New Administrative Capital, with construction having begun in 2016.

The NAC is designed around digital surveillance of the population, and showcasing that Egypt can develop the desert, and maybe decongesting Cairo. It has pretenses of being the next Dubai, but where Dubai invites global starchitects to buy prestige, the NAC is instead giving contracts to domestic elites (as was also the case for Naypyidaw); the military directly owns 51% of the agency developing the NAC and the state ministry of housing only 49%, and the contracts are designed to enrich people who are politically connected to the government.

The violence levels involved are, again, much lower than in Burma. But precisely because Egypt’s economy is solidly middle-income, it’s frustrating to see vast sums wasted on a military prestige project. Cairo’s congestion and overcrowding have a well-known solution, in the form of building a rapid transit network to facilitate non-car commutes, and connect not just the existing built-up area to the urban core but also to-be-built areas adjacent to it. It’s a metro area of 22 million people, about the same size as New York, but only has a three-line, 100 km metro network. These 100 km should be closer to 1,000.

To be clear, there is some development of the Cairo Metro. Line 3 is being extended as we speak, with a new section having opened this month. Overall, the third phase of the line, 17 km long, cost 40.7 billion Egyptian pounds per a statement in 2012, which in contemporary dollars is $2.6 billion in exchange rate terms and $9 billion in 2021 PPP terms. On Google Earth, it looks like 9 km of the line are underground and 8 km are elevated or at-grade; this more than $500 million/km cost is not even fully-underground. This is a line where the main contractor is Orascom – this isn’t a case of outsourcing the state to Japan, China, or both, and therefore building at high costs, as is common in Bangladesh, Pakistan, and Southeast Asia. Rather, this is entirely domestic waste. Most likely, the same contractors that are profiteering off of public money through the construction of the NAC are also profiteering off of public money for what little infrastructure the military government deigns to build in the capital that people actually live in.

Bus Stop Consolidation and Blocks

There are arguments over bus stop spacing in my Discord channel. As the Queens bus redesign process is being finalized, there’s a last round of community input, and as one may expect, community board members amplify the complaints of people who reject any stop consolidation on “they’re taking my stop, I’ll have to walk longer” grounds. I wrote about this in 2018, as Eric and I were releasing our proposed Brooklyn bus redesign, which included fairly aggressive consolidation, to an average interstation of almost 500 meters, up from the current value of about 260. I’d like to revisit this issue in this post, first because of its renewed relevance, and second because there’s a complication that I did not incorporate into my formula before, coming from the fact that the city comprises discrete blocks rather than perfectly isotropic distribution of residents along an avenue.

The formula for bus stop spacing

The tradeoff is that stop consolidation means people have to walk longer to the bus stop but then the bus is faster. In practice, this means the bus is also more frequent by a proportionate amount – the resources required to operate a bus depend on time rather than distance, chiefly the driver’s wage, but also maintenance and fuel, since stops incur acceleration and idling cycles that stress the engines and consume more fuel.

The time penalty of each stop can be modeled as the total of the amount of time the bus needs to pull into the stop, the minimum amount of time it takes to open and close the doors, and the time it takes to pull out. Passenger boardings are not included, because those are assumed to be redistributed to other stops if a stop is deleted. In New York and Vancouver, the difference in schedules between local and limited stop buses in the 2010s was consistent with a penalty of about 25 seconds per stop.

The optimum stop spacing can be expressed with the following formula:

\sqrt{d\cdot\frac{\mbox{walk speed}}{\mbox{walk penalty}}\cdot\mbox{stop penalty}\cdot(\mbox{average trip length} + \mbox{average bus spacing})}

To explain in more detail:

  • d is a dimensionless factor indicating how far one must walk, based on the stop spacing; the more isotropic passenger travel is, the lower d is, to a minimum of 2. The specific meaning of d is that if the stop spacing is n, then the average walk is n/d. For example, if there is perfect isotropy, then passengers’ distance from the nearest bus stop is uniformly distributed between 0 and n/2, so the average is n/4, and this needs to be repeated at the destination end, summing to n/2.
  • Walk speed and walk penalty take into account that passengers prefer spending time on a moving bus to walking to the bus. In the literature that I’ve seen, the penalty is 2. Usually the literature assumes the walk speed is around 5 km/h, or 1.4 m/s; able-bodied adults without luggage walk faster, especially in New York, but the speed for disabled people is lower, around 1 m/s for the most common cases.
  • Stop penalty, as mentioned above, can be taken to be 25 s.
  • Average trip length is unlinked; for New York City Transit in 2019, counting NYCT local buses including SBS but not express buses, the average was 3,421 meters.
  • Average bus spacing is the headway between buses on the route measured in units of distance, not speed; it’s expressed this way since the resources available can be expressed in how many buses can circulate at a given time, and then the frequency is the product of this figure with speed. In Brooklyn in the 2010s, this average was 1,830 m; our proposed network, pruning weaker routes, cut it to 1,180. The Queens figure as of 2017 appears similar to the Brooklyn figure, maybe 1,860 m. Summing the average trip length and average bus spacing indicates that passengers treat wait time as a worst-case scenario, or, equivalently, that they treat it as an average case but with a wait penalty of 2, which is consistent with estimates in the papers I’ve read.

In the most isotropic case, with d = 2, plugging in the numbers gives,

\sqrt{2\cdot\frac{1.4}{2}\cdot 25\cdot(3420+1860)} = 430 \mbox{meters}

However, isotropy is more complex than this. For one, if we’re guaranteed that all passengers are connecting to one distinguished stop, say a subway connection point, then consolidating stops will still make them walk longer at the other end, but it will not make them walk any longer at the guaranteed end, since that stop is retained. In that case, we need to set d = 4 (because the average distance to a bus stop if the interstation is n is n/4 and at the other end we’re guaranteed there’s no walk), and the same formula gives,

\sqrt{4\cdot\frac{1.4}{2}\cdot 25\cdot(3420+1860)} = 608 \mbox{meters}

The Queens bus redesign recognizes this to an extent by setting up what it calls rush routes, designed to get passengers from outlying areas in Eastern Queens to the subway connection points of Flushing and Jamaica; those are supposed to have longer interstations, but in practice this difference has shrunk in more recent revisions.

That said, even then, there’s a complication.

City blocks and isotropy

The models above assume that passengers’ origins are equally distributed along a line. For example, here is Main Street through Kew Gardens Hills, the stretch I am most likely to use a New York bus on:

I always take the bus to connect to Flushing or Jamaica, but within Kew Gardens Hills, the assumption of isotropy means that passengers are equally likely to be getting on the bus at any point along Main Street.

And this assumption does not really work in any city with blocks. In practice, neighborhood residents travel to Main Street via the side streets, which are called avenues, roads, or drives, and numbered awkwardly as seen in the picture above (72nd Avenue, then 72nd Road, then 72nd Drive, then 73rd Avenue, then 75th Avenue…). The density along each of those side streets is fairly consistent, so passengers are equally likely to be originating from any of these streets, for the most part. But they are always going to originate from a side street, and not from a point between them.

The local bus along Main, the Q20, stops every three blocks for the most part, with some interstations of only two blocks. Let’s analyze what happens if the system consolidates from a stop every three blocks, which is 240 meters, to a stop every six, which is 480. Here, we assume isotropy among the side streets, but not continuous isotropy – in other words, we assume passengers all come from a street but are equally likely to be coming from any street.

With that in mind, take a six-block stretch, starting and ending with a stop that isn’t deleted. Let’s call this stretch 0th Street through 6th Street, to avoid having to deal with the weird block numbering in Kew Gardens Hills; we need to investigate the impact of deleting a stop on 3rd Street. With that in mind: passengers originating on 0th and 1st keep going to 0th Street and suffer no additional walk, passengers originating on 5th keep going to 6th and also suffer no additional walk, passengers originating on 2nd and 4th have to walk two blocks instead of one, and passengers originating on 3rd have to walk three blocks instead of zero. The average extra walk is 5/6 of a block. This is actually more than one quarter of the increase in the stop spacing; if there is a distinguished destination at the other end (and there is), then instead of d = 4, we need to use d = 3/(5/6) = 3.6. This shrinks the optimum a bit, but still to 576 m, which is about seven blocks.

The trick here is that if the stop spacing is an even number of blocks, then we can assume continuous isotropy – passengers are equally likely to be in the best circumstance (living on a street with a stop) and in the worst (living on a street midway between stops). If it’s an odd number of blocks, we get a very small bonus from the fact that passengers are not going to live on a street midway between stops, because there isn’t one. The average walk distance, in blocks, with stops every 1, 2, 3, … blocks, is 0, 0.5, 2/3, 1, 1.2, 1.5, 12/7, 2, … Thus, ever so slightly, planners should perhaps favor a stop every five or seven blocks and not every six, in marginal cases. To be clear, the stop spacing on each stretch should be uniform, so if there are 12 blocks between two distinguished destinations, there should be one intermediate stop at the exact midpoint, but, perhaps, if there are 30 blocks with no real internal structure of more or less important streets, a stop should be placed every five and not six blocks, especially if destinations are not too concentrated.

Fares on High-Speed Trains

American journalists sometimes ask me to discuss high-speed rail fares. I’ve written from time to time about how Northeast Corridor fares are significantly higher than French, German, and Japanese high-speed rail fares, but the links to this information are never in the same place. The purpose of this post is to collect all the links together for easy retrieval, with updates to the 2020s whenever possible. Unfortunately, international high-speed rail fares connecting to France are also much higher than domestic ones, which contributes to the poor ridership of those trains relative to city size.

France

ARAFER releases statistics annually. The most recent year for which there is data is 2022; here’s the report in French, and here’s a summary in English. The relevant information is in sections 5-6. The TGV system including international trains averages fare revenue of 6,213M€ for 61 billion passenger-km; in English this is called “non-PSO,” since these are the profitable trains that SNCF runs outside the passenger service obligation system for money-losing slow trains. This works out to 0.102€/p-km. The PPP rate these days is about 1€ = $1.45, making this about $0.15/p-km.

The split between domestic and international trains is large, and the French report has the domestic trains as just 0.093€/p-km, taking a weighted average from pp. 23 and 31. Nominal fares per p-km on domestic TGVs were down 4% from 2019 to 2022, despite 7.5% cumulative inflation over this period.

The international trains, in contrast, are much more expensive: the report doesn’t give exact numbers, but from some weighted averaging and graph eyeballing it looks like it’s around 0.17€/p-km. The all-high-speed international trains – Eurostar and Thalys – are more expensive than the trains running partly at low speed in Germany and Switzerland, like Lyria; this big difference in fares helps explain their disappointing ridership. Domestic TGVs run from Paris to Lyon 28 times on the 5th of June this year, counting only trains to Lyon Part-Dieu or Perrache, which do not continue onward, and not counting trains that stop at Saint-Exupéry on their way to points south, and those trains are rather full 16-car bilevels. In contrast, on the same day I only see 16 Eurostars from London to Paris. This is despite the fact that London is a far larger city than Lyon, and the in-vehicle travel time is only moderately longer.

Germany

Germany lacks France’s neat separation of low- and high-speed trains. The intercity rail network here is treated as a single system, and increasingly all trains are ICEs even if they spend the majority of the trip on legacy lines at a top speed of 200 km/h.

Overall intercity rail passenger revenue here was 5.1 billion € in 2022; the expression to look for is “SPFV.” Ridership was 42.9 billion p-km per a DB report of 2022-3, PDF-p. 7, averaging 0.119€/p-km, which is $0.17/p-km in PPP US dollars. 2022 was still slightly below 2019 levels, when ridership was 44.7 billion p-km and fares averaged 0.112€/p-km; the three-year increase was less than the cumulative inflation over this period, which was 10.3%.

Japan

Japanese fares are higher than European fares on high-speed rail. JR East’s presentation from 2021, showing depressed ridership during the pandemic (p. 50), reports ¥189.6 billion in Shinkansen revenue on 7.95 billion p-km, or ¥23.8/p-km, and projects recovery to ¥428.9 billion/17.313 billion p-km by 2022, or ¥24.8/p-km. JR Central’s 2020 report says (p. 37) that its Shinkansen service got ¥1.2613 trillion in revenue in the year ending March 2020 on 54.009 billion p-km, or ¥23.4/p-km. JR West’s 2020 factsheets for revenue and ridership show ¥457 billion/21.338 billion p-km in 2019, or ¥21.4/p-km.

The PPP rate for 2020-1 was $1 = ¥100. Taking 9% dollar inflation from 2020 to 2022 into account, this is, in 2022 prices, around $0.25/p-km.

Northeast Corridor

Amtrak publishes monthly performance reports; the fiscal year is October-September, so the September reports, covering an entire fiscal year, are to be preferred. Here are 2022 and 2023; 2022 still shows a considerable corona depression, unlike in France and Germany. The 2023 report shows that Northeast Corridor revenue splits as $495.9 million/581.1 million p-miles Acela, or $0.53/p-km, and $768.2 million/1.6269 billion p-miles Regional, or $0.293/p-km Regional. Altogether, this is $0.356/p-km, which is nearly 50% higher than the Shinkansen, 2.1 times as expensive as the ICE, and 2.4 times as expensive as the TGV.

Discussion

High operating costs on Amtrak are the primary reason for the premium fares. The mainland JRs are all highly profitable; DB Fernverkehr is profitable, as is the TGV network (though SNCF writ large isn’t, the slow intercities falling under the PSO rubric). All five companies pay track access charges for the construction of high-speed rail infrastructure; the ARAFER report goes over these charges in France and a selection of other European countries, designed to prevent state subsidies to intercity rail operations through underpriced track access, since track construction is always done by the state but operations may be done by a private operator or a foreign state railway. The Northeast Corridor is profitable as well – Amtrak doesn’t have to pay track access charges, but the access charges for legacy 19th-century lines would not be significant. However, if Amtrak charged European fares or even Japanese ones, it wouldn’t be. Northeast Corridor rail operations in fiscal 2023 earned $1.266 billion in passenger revenue plus $28.5 million in non-ticket revenue but spent $1.0917 billion, or $0.307/p-km.

A portion of the Amtrak cost premium also comes from adversarial profit maximization, also seen on Thalys and Eurostar. The domestic TGVs and ICEs aim at making a base rate of profit while providing a service for the general public; SNCF doesn’t apply the same logic to Thalys and Eurostar and instead aims at serving only business trips to avoid the possibility of extracting less than maximum fares from international travelers. On Amtrak, the need to subsidize the rest of the system has increased Northeast Corridor fares, though to be clear, in fiscal 2023 the operating margin was small enough that this is at most a secondary factor. Performance reports from the 2000s and 10s showed a larger operating margin, but criticism from advocacy groups centering non-Northeast Corridor passengers alleging that Amtrak accounting was making the Northeast Corridor look better and the night trains look worse led to a recalculation, used in the most recent reports, in which Northeast Corridor operations still turn out to be profitable but not by a large margin.

American Myths of European Poverty

I occasionally have exchanges on social media or even in comments here that remind me that too many people in the American middle class believe that Europe is much poorer than the US. The GDP gap between the US and Northern Europe is small and almost entirely reducible to hours worked, but the higher inequality in the US means that the top 10-20% of the US compare themselves with their peers here and conclude that Europe is poor. Usually, it’s just social media shitposting, for example about how store managers in the US earn the same as doctors in Europe. But it becomes relevant to public transit infrastructure construction in two ways. First, Americans in positions of authority are convinced that American wages are far higher than European ones and that’s why American construction costs are higher than European ones. And second, more broadly, the fact that people in positions of authority really do earn much more in the US than here inhibits learning.

The income gap

The United States is, by a slight amount, richer than Northern Europe, which for the purposes of this post comprises the German-speaking world, the Nordic countries, and Benelux. Among the three largest countries in this area, Germany is 16.5% poorer than the US, the Netherlands 8.3% poorer, Sweden 14.3%. This is more than anything an artifact of shorter working hours – Sweden has an ever so slightly larger GDP per hour worked, the other two are 6-7% poorer per hour worked. All three countries have a much higher 15-64 labor force participation rate than the US, but they’re also older, which in the case of Germany actually gets its 15+ rate to be a hair less than the US’s. But there’s much more part-time work here, especially among women, who face large motherhood penalties in German society (see figures 5-7 in Herzberg-Druker, and Kleven et al). Germany is currently in full employment, so it’s not about hidden part-time work; it’s a combination of German-specific sexism and Europe-wide norms in which workers get around six weeks of paid vacation per year.

One implication of the small gap in income per hour is that wages for the same job are likely to be similar, if the jobs pay close to the mean wage. This is the case for tunnel miners, who are called sandhogs in the United States: the project labor agreements in New York are open – the only case in which itemized costs are publicly available – and showcase fully-laden employment costs that, as we document in our construction costs reports, work out to around $185,000/year in 2010 prices; there is a lot of overstaffing in New York and it’s disproportionately in the lower-earning positions, and stripping those, it’s $202,000/year. I was told that miners in Stockholm earn 70,000 kronor/month, or about $100,000/year in PPP terms (as of 2020-1), and the fully-laden cost is about twice that; a union report from the 2000s reports lower wages, but only to about the same extent one would expect from Sweden’s overall rate of economic growth between then and 2021. The difference at this point is second-order, lower than my uncertainty coming from the “about” element out of Sweden.

While we’re at it, it’s also the case for teachers: the OECD’s Education at a Glance report‘s indicator D3 covers teacher salaries by OECD country, and most Northern European countries pay teachers better than the US in PPP terms, much better in the case of Germany. Teacher wage scales are available in New York and Germany; the PPP rate is at this point around 1€ = $1.45, which puts starting teachers in New York with a master’s about on a par with their counterparts in the lowest-paying German state (Rhineland-Pfalz). New York is a wealthy city, with per capita income somewhat higher than in the richest German state (Bavaria), but it’s not really seen in teacher pay. I don’t know the comparative benefit rates, but whenever we interview people about European wage rates for construction, we’re repeatedly told that benefits roughly double the overall cost of employment, which is also what we see in the American public sector.

The issue of inequality

American inequality is far higher than European inequality. So high is the gap that, on LIS numbers, nearly all Western European countries today have lower disposable income inequality than the lowest recorded level for the US, 0.31 in 1980. Germany’s latest number is 0.302 as of 2021, and Dutch and Nordic levels are lower, as low as 0.26-0.27; the US is at 0.391 as of 2022. If distributions are log-normal (they only kind of are), then from a normal distribution log table lookup, this looks like the mean-to-median income ratios should be, respectively, 1.16 for Germany and 1.297 for the US.

However, top management is not at the median, and that’s the problem for comparisons like this. The average teacher or miner makes a comparable amount of money in the US and Northern Europe. The average private consultant deciding on how many teachers or miners to hire makes more money in the US. A 90th-percentile earner is somewhat wealthier in the US than here, again on LIS number; the average top-1%er is, in relative terms, 50% richer in the US than in Germany (and in absolute terms 80% richer) and nearly three times as rich in the US as in Sweden or the Netherlands, on Our World in Numbers data.

On top of that, I strongly suspect that not all 90th percentile earners are created equal, and in particular, the sort of industries that employ the mass (upper) middle class in each country are atypically productive there and therefore pay better than their counterparts abroad. So the average 90th-percentile American is noticeably but not abnormally better off than the average 90th-percentile German or Swede, but is much better off than the average German or Swede who works in the same industries as the average 90th-percentile American. Here we barely have a tech industry by American standards, for example; we have comparable biotech to the US, but that’s not usually where the Americans who noisily assert that Europe is poor work in.

Looking for things to mock

While the US is not really richer than Northern Europe, the US’s rich are much richer than Northern Europe’s. But then the statistics don’t bear out a massive difference in averages – the GDP gap is small, the GDP gap per hour worked is especially small and sometimes goes the other way, the indicators of social development rarely favor the US, immigration into Western Europe has been comparable to immigration to the US for some time now (here’s net migration, and note that this measure undercounts the 2022 Ukrainians in Germany and overcounts them in Poland).

So middle-class Americans respond by looking for creative measures that show the level of US-Europe income gap that they as 90th-percentile earners in specific industries experience (or more), often dropping the PPP adjustment, or looking at extremely specific things that are common in the US but not here. I’ve routinely seen American pundits who should know better complain that European washing machines and driers are slow; I’m writing this post during a 4.5-hour wash-and-dry cycle. Because they fixate on proving the superiority of the United States to the only part of the world that’s rich enough not to look up to it, they never look at other measures that might show the opposite; this apartment is right next to an elevated train, but between the lower noise levels of the S-Bahn, good insulation, and thick tilt-and-turn windows, I need to concentrate to even hear the train, and am never disturbed by it, whereas American homes have poor sound insulation to the point that street noise disturbs the sleep.

Learning to build infrastructure

The topline conclusion of any American infrastructure reform should be “the United States should look more like Continental Europe, Turkey, non-Anglophone East Asia, and the better-off parts of Latin America.”

If it’s written in the language of specific engineering standards, this is at times acceptable, if the standards are justified wholly internally (“we can in fact do this, here’s a drawing”). Even then, people who associate Americanness with their own career success keep thinking safety, accessibility, and similar issues are worse here, and ask “what about fire code?” and then are floored to learn that fire safety here is actually better, as Stephen Smith of Market Urbanism and the Center for Building constantly points out.

But then anything that’s about management is resisted. It’s difficult to convince an American who’s earning more than $100,000 a year in their 20s and thinks it’s not even that much money because their boss is richer that infrastructure project management is better in countries where the CEO earns as much money as they do as an American junket assistant. Such people readily learn from rich, high-inequality places that like splurging, which are not generally the most productive ones when it comes to infrastructure. Even Americans who think a lot about state capacity struggle with the idea that Singapore has almost as high construction costs as the US; in Singapore, the CEO earns an American salary, so the country must be efficient, right? Well, the MRT is approaching $1 billion/km in construction costs for the Cross-Island Line, and Germany builds 3 km of subway (or decides not to build them) on the same budget and Spain builds 6 km, but Europe is supposedly poor and Americans can’t learn from that.

The upshot is that even as we’re seeing some movement on better engineering and design standards in the United States, resulting in significant cost savings, there’s no movement for better overall management. Consultant-driven projects remain the norm, and even proposals for improving state capacity are too driven by domestic analysis without any attempt at international learning or comparativism. Nor is there any effort at better labor efficiency – management in the US hates labor, but also thinks it’s entirely about overpaid workers or union safety rules, and doesn’t stoop to learn how to build more productively.

The Politics of QueensLink

The abandoned right-of-way of the Rockaway Cutoff, or Rockaway Beach Branch, is an attractive target for reuse by some groups. Area railfans have wanted to do something with it for years, and I was mostly negative about these plans, but more recently, QueensLink has emerged as a serious plan to extend the subway along the Rockaway Cutoff, to connect neighborhoods in southwestern Queens to one another and to Manhattan, where current connections do not exist (in the middle) or only serve Midtown Manhattan indirectly and slowly (in the south, via the A train). The right-of-way is wide enough that most of it can also accommodate additional infrastructure, in the case of this plan a bike trail.

The problem is that this more serious plan is still not getting much political play. This is not because of the typical reasons people may think of, such as high costs or NIMBYism. Rather, a competing plan for the same corridor, Queensway, wants to turn it into pure parkland, and is backed by a power broker with opinions and connections. The QueensLink advocates have asked for and been so far refused planning money for an environmental impact statement, which step is in better infrastructure environments apolitical; instead, Mayor Eric Adams is connected with a Queensway backer and thus favors Queensway.

The QueensLink project

QueensLink is a subway extension, with a short tunnel to get from the Queens Boulevard Line’s local tracks to the Rockaway Cutoff. Where there is room, a bike trail is included on the same corridor.

As this is a former rail mainline, the connections to the crossing subway lines are not great – the subway placed stations at major street intersections, leading to long walks at the Jamaica and Liberty Avenue transfers. In contrast, the connection to the LIRR is good: there is no station there today, but there used to be one, and it could be reopened, especially now that the Atlantic Branch is retooling to be more useful for urban service, with more stops, higher frequency, and perhaps integrated fares.

The subway connection makes this proposal viable. I previously criticized a proposal to run commuter rail service on the Cutoff, since it would crowd out the busy LIRR Main Line. In contrast, QueensLink has the new branch using the Queens Boulevard local tracks, which are undersubscribed even at rush hour, to the point that it may even be possible to run three rush hour services on the same tracks and not just two like today. The G doesn’t run to Forest Hills, for good reason, but the new service would vacate space at the Forest Hills terminal of the local tracks to the point that it could potentially be viable.

At the Transportation and Land Use program at Marron, we’re building tools to estimate not just costs for public transit construction but also ridership, and it’s likely that QueensLink will be next on our agenda right after the Interborough Express. I can’t give more than first-order estimates now, but it’s notable that the closest parallel bus corridor, Woodhaven, has high ridership: it carries four local or SBS routes and four express routes, with 31,000 weekday boardings among them. Then there’s the possibility of faster service to JFK Airport via QueensLink. It’s not going to be the Second Avenue Subway of Queens by ridership, but because only a short tunnel is needed, it’s not going to come close to Second Avenue Subway in costs either.

QueensLink and Queensway

The alternative to QueensLink isn’t doing nothing, as is usually the case. Some political players have eyed the corridor for a trail project called Queensway. The idea of Queensway is to create more parkland in the area, including a hiking and bike trail; as the map above shows, there is no shortage of parkland there. There’s an obsession of urbanists in the United States with linear parks, in imitation of the High Line; one attempt at imitation even wanted to build a park underground in a former streetcar terminal and called it the Lowline.

I want to dwell on the politics of Queensway, because I know I have a lot of readers in the general neoliberal and rationalist communities, including specifically in New York, and the political support for it is not what their first instinct might be.

Often, it’s the case that public transit projects are supported by broadly developmentalist interest groups, who are also fairly YIMBY, and tend to be rooted in professionals and office workers commuting to city center, and opposed by NIMBYs, who tend to be rooted in longstanding neighborhood residents and small business owners. Neighborhood NIMBYs often like parks, because they’re local improvements; when we studied the Green Line Extension in Boston, we saw some local interest groups demand money for a trail as a precondition to supporting the light rail line.

And this is not at all what is happening with the Rockaway Cutoff. Local interest groups are not consistently anti-QueensLink and pro-Queensway. Some are, but in at least one case, a local advocate came to argue on NIMBY grounds against Queensway, which would bring pedestrians to their backyard, and for QueensLink, since the passing train would not cause unwanted impact and would serve the area. On net, YIMBYer groups are more pro-QueensLink – for example much of the community at the northern end of the corridor, in Community Board 6, which due to its location on the subway has a more pro-transit and pro-development orientation. But it doesn’t boil down to these class interests pitting professional workers against small business owners, at all. Rather, it’s rather random, boiling down to individual power brokers for Queensway.

What autocrats want

I highlight the randomness of the interest groups for Queensway, because it relates to the broadly autocratic style of some leaders, who the de facto system of government in New York empowers too much. In the 2010s, Andrew Cuomo liked the idea of Queensway, and if he wanted something, anyone who wanted to stay in the good grace of the local power system had to support it. The backward air train to LaGuardia, hated by transit activists in the city from the start, is such an example – while Cuomo was in power opposition was restricted to people outside city and state politics, like the technical advocacy community or Alexandria Ocasio-Cortez. Then Cuomo left office and because nobody really wanted it, this project died. Queensway survives, I think partly because it’s not a multi-billion dollar investment, and partly because some power brokers do like it and have attached themselves to Mayor Eric Adams, such as political operative and lobbyist Travis Terry.

The point of this is not that Adams wants Queensway and therefore it will happen. Rather, it’s that, in a system with a democratic deficit like New York, professional decisions often boil down to which random advocate happens to have the ear of the autocrat. I keep contrasting this with the situation in Berlin, in which bike lanes and pedestrianization have been put on hold and even been reversed under Mayor Kai Wegner (CDU) – but Wegner ran on this agenda in the election and CDU came first, and the pivotal party, SPD, chose to enter a coalition with him and not with more left-wing parties under a known-quantity (and disgraced) leader in Franziska Giffey, whose conservative-for-SPD politics and poor relationship with the left was well-known. I dislike this outcome, but voters knew what they were getting. In contrast, in New York, Adams did not run on any explicit agenda of not building public transit, or even on support for interest groups that oppose transit (again, the Queensway support is more individualized than neighborhood-scale NIMBYs). He just makes abrupt decisions, often sniping the judgment of the city’s own civil service, based on what one favorite asks.

The way forward

In healthy infrastructure construction systems and also in Germany, the planning is not politicized; the yes/no decision on what to build must be made by politicians, but the menu of options with their costs and benefits is prepared by the civil service. In contrast, in the United States, even the choice of which projects get an environmental impact study is politicized; QueensLink advocates are asking for money for an EIS, which in the United States is where planning is done, but even that is stalled politically.

The problem is that the message this behavior by the city and state sends is that New York is too messed up to invest in. Private actors who make investment decisions need some amount of political stability and predictability. A political culture of caprice, in which everyone must constantly follow political gossip to have any idea what the autocratic mayor or governor (or in some countries president or prime minister) will do, or else be swamped by otherwise inexplicable investment decisions, screams “go elsewhere, we don’t want you.” Even turning an EIS into a big political ask screams the same thing: “we can’t do, so merely studying is an achievement by itself and you must pay fealty.”

New York’s current system deters investment, not through taxes or union empowerment, but through opacity and unpredictability. I don’t want to turn the question of one right-of-way in Queens into an existential issue that it is not, but on the margin, stonewalling on QueensLink because some politically connected actor personally wants Queensway reinforces this system that repels investment, whereas treating the EIS as an apolitical step and then based on the results of further planning building it or not based on broad interest agreement signals that New York can and wants to build things. It’s the city’s choice.

The Future of Congestion Pricing in New York

New York just passed congestion pricing, to begin operation on June 30th. The magazine Vital City published an issue dedicated to this policy two days ago; among the articles about it is one by me, about public transportation investments. People should read the entire article; here I’d like to both give more context and discuss some of the other articles in the issue. Much of this comes from what I said to editor Josh Greenman when discussing the pitch for the piece, and how I interpret the other pieces in the same context. The most basic point, for me, is that what matters is if the overall quality of public transit in and around New York is seen to improve in the next 5-10 years. In particular, if congestion pricing is paired with one specific thing (such as a new subway line) and it improves but the rest of the system is seen to decline, then it will not help, and instead people will be cynical about government actions like this and come to oppose further programs and even call for repealing the congestion tax.

The other articles in the issue

There are 10 articles in this issue. One is my own. Another is by Josh, explaining the background to congestion pricing and setting up the other nine articles. The other eight were written by John Surico, Sam Schwartz, Becca Baird-Remba, Austin Celestin, Howard Yaruss, Nicole Gelinas, Vishaan Chakrabarti, and Henry Grabar, and I recommend that people read all of them, for different perspectives.

The general themes the nine of us have covered, not all equally, include,

  • How to use congestion pricing to improve transportation alternatives (me on transit investment, Yaruss on transit fare cuts, Nicole and Chakrabarti on active transportation, Henry on removing parking to improve pedestrian safety).
  • The unpopularity of congestion pricing and what it portends (Surico about polling, Becca about business group opposition, Schwartz on political risk, Yaruss again on why the fare cut is wise); of note, none of the authors are coming out against congestion pricing, just warning that it will need to deliver tangible benefits to remain popular, and Surico is making the point that in London and Stockholm, congestion pricing was unpopular until it took effect, after which it was popular enough that new center-right leadership did not repeal it.
  • Environmental justice issues (Becca and Celestin): my article points out that traffic levels fell within the London congestion zone but not outside it, and Becca and Celestin both point out that the projections in New York are for traffic levels outside the zone not to improve and possibly to worsen, in particular in asthma-stricken Upper Manhattan and the Bronx, Celestin going more deeply into this point and correctly lamenting that not enough transit improvements are intended to go into these areas. The only things I can add to this are that for environmental justice, two good investment targets include a 125th Street subway tunnel extending Second Avenue Subway and battery-electric buses at depots to reduce pollution.
  • Problems with toll evasion (Schwartz and Yaruss): there’s a growing trend of intentional defacement of license plates by the cars’ own drivers, to make them unreadable by traffic camera and avoid paying tolls, which could complicate revenue collection under congestion pricing.

The need for broad success

When discussing my article with Josh, before I wrote it, we talked about the idea of connecting congestion pricing to specific improvements. My lane would be specific transit improvements, like new lines, elevator access at existing stations, and so on, and similarly, Nicole, Henry, Chakrabarti, and Yaruss proposed their own points. But at the same time, it’s not possible to just make one thing work and say “this was funded by congestion pricing.” The entire system has to both be better and look better, the latter since visible revenue collection by the state like congestion pricing or new taxes are always on the chopping block for populist politicians if the state is too unpopular.

The example I gave Josh when we talked was the TGV. The TGV is a clear success as transportation; it is also, unlike congestion pricing, politically safe, in the sense that nobody seriously proposes eliminating it or slowing it down, and the only controversy is about the construction of new, financially marginal lines augmenting the core lilnes. However, the success of the TGV has not prevented populists and people who generally mistrust the state from claiming that things are actually bad; in France, they are often animated by New Left nostalgia for when they could ride slow, cheap trains everywhere, and since they were young then, the long trip times and wait times didn’t matter to them. Such nostalgics complain that regional trains, connecting city pairs where the train has not been competitive with cars since mass motorization and only survived so long as people were too poor to afford cars, are getting worse. Even though ridership in France is up, this specific use case (which by the 1980s was already moribund) is down, leading to mistrust. Unfortunately, while the TGV is politically safe in France, this corner case is used by German rail advocates to argue against the construction of a connected high-speed rail network here, as those corner case trains are better in Germany (while still not carrying much traffic).

The most important conclusion of the story of the TGV is that France needs to keep its high-speed system but adopt German operations, just as Germany needs to adopt French high-speed rail. But in the case of New York, the important lesson to extract is that if the MTA does one thing that I or Nicole or Henry or Chakrabarti or Yaruss called for while neglecting the broad system, people will not be happy. If the MTA builds subway lines with the projected $1 billion a year in revenue, politicians will say “this subway line has been built with congestion pricing revenue,” and then riders will see declines in reliability, frequency, speed, and cleanliness elsewhere and learn to be cynical of the state and oppose further support for the state’s transit operations.

The MTA could split the difference among what we propose. As I mentioned above, I find Celestin’s points about environmental justice compelling, and want to see improvements including new subways in at-risk areas, bus depot electrification to reduce pollution, and commuter rail improvements making it usable by city residents and not just suburbanites (Celestin mentions frequency; to that I’ll add fare integration). Nicole, Henry, and Chakrabarti are proposing street space reallocation, which doesn’t cost much money, but does cost political capital and requires the public to be broadly trusting of the state’s promises on transportation. The problem with doing an all-of-the-above program is that at the end of the day, projected congestion pricing revenue is $1 billion a year and the MTA capital program is $11 billion a year; the new revenue is secondary, and my usual bête noire, construction costs, is primary.

Quick Note: What the Hell is Going on in San Jose?

The BART to San Jose extension always had problems, but somehow things are getting worse. A month and a half ago, it was revealed that the projected cost of the 9.6-kilometer line had risen to $12.2 billion. Every problem that we seemed to identify in our reports about construction costs in New York and Boston appears here as far as I can tell, with the exception of labor, which at least a few years ago showed overstaffing in the Northeastern United States but not elsewhere. In particular, the station and tunnel design maximizes costs – the first link cites Adam Buchbinder on the excessive size of the digs. Unfortunately, the response by the Valley Transportation Authority (VTA) to a question just now about the station shows that not only are the stations insanely expensive, but also not even convenient for passengers (Twitter link, Nitter link).

Cost breakdown

The March 2024 agenda (link, PDF-pp. 488-489) breaks down the costs. The hard costs total $7 billion; the systems : civils ratio is 1:3.5, which is not bad. But the overall hard costs are still extreme. Then on top of them there are soft costs totaling $2.78 billion, or 40% on top of the hard costs. The same percentage for Second Avenue Subway was 21%, and the norm for third-party consultants for the Continental European projects for which we have data (in Italy, Spain, Turkey, and France) is to charge 5-10%. Soft costs should not be this high; if they are, something is deeply wrong with how the agency uses consultants.

Large-diameter tunnel boring machines

The BART to San Jose project has long had two distinct options for tunnels and station: twin bores, and single bore. The twin bore option is conventional construction of two bored tunnels, one for each track, and then stations to be built as dedicated civil construction projects outside the tunnel; this is how most subways are built today. The single bore option is a large-diameter tunnel boring machine (TBM), with the bore large enough to have not just two tracks side by side, but also platforms within the bore, eliminating the need for mined station caverns or for extensive cut-and-cover station digs. Both options cleared environmental reviews; VTA selected the single bore option, which has been controversial.

I’ve written positively about large-diameter TBMs before, and I don’t think I’ve written a full post walking this back. I’ve written about how large-diameter TBMs are inappropriate for San Jose, but the truth is that the method is not treated as a success elsewhere in urban rail, either. This is controversial, and serious engineers still think it works and point to successes in intercity rail, but in urban rail, the problems with building settlement are too serious. The main example of a large-diameter TBM is Barcelona L9/10, which uses the method to avoid having to open up streets under multiple older metro tunnels in Barcelona; it also has high construction costs by Spanish standards (and low costs by non-Spanish ones). In Italy, whose construction costs are also fairly low if not as low as in Madrid, engineers considered using large-diameter TBMs for the sensitive parts of Rome Metro Line C but then rejected that solution as too risky, going for conventional high-cost mined stations instead.

Regardless of the wisdom of doing this in Southern Europe, in San Jose it is stupid. There are wide streets to dig up for cut-and-cover stations. Then, the implementation is bad – the station entryways are too big, whereas Barcelona’s are small elevator banks, and the tunnel bore is wide enough for a platform and two tracks on the same level whereas Barcelona has a narrower bore with stacked platforms.

Thankfully, it is administratively possible to cancel the single bore option, since the twin bore option cleared the environmental reviews as well, and in 2007 was already complete to 65% design (link, PDF-p. 7). Unfortunately, there isn’t much appetite among officials for it. Journalists and advocates are more interested, but the agency seems to stick to its current plans even as their costs are setting non-New York construction cost records.

Is it at least good?

No. Somehow, for this cost, using a method whose primary advantage is that it makes it possible to build a station anywhere at the cost of massively more expensive tunneling, the station at the city’s main train station, named after still-alive Rod Diridon, will not be easily accessible from mainline rail. The walking distance is 400 meters, which has been justified on the grounds that “The decision had to do with impacts and entitlements. It’s also beneficial for the future intermodal station.”

It is, to be clear, not at all beneficial for a future intermodal BART-Caltrain station to require such a long walking distance, provided we take “beneficial” to mean “beneficial for passengers.” It may be beneficial for a Hollywood action sequence to depict characters running through such a space. It is not beneficial for the ordinary users of the station who might be interested in connecting between the two systems. There are 300 meter walks at some transfers in New York, and passengers do whatever they can to avoid them; I’ve taken three-seat rides with shorter transfers to avoid a two-seat ride with a long block transfer, and my behavior is typical of the subway users I know. Transfer corridors of such length are common in Shanghai and are disliked by the system’s users. It’s not the end of the world, but for $1.3 billion/km, I expect better and so should the people who have to pay for this project.

Subway Expansion and Bus Redesign

The ongoing designs for the Interborough Express are making me think about bus redesign again. Before the Transit Costs Project, Eric and I worked on a proposal for a bus redesign in Brooklyn, which sadly was not adopted. The redesign was based on the reality of 2017 – the ridership patterns, the bus speeds, the extent of the system, etc. Since then the subway map has not changed, but IBX stands to change the map, and with it, the buses should change as well.

With our program having produced both the bus redesign proposal and soon a comprehensive proposal for how to change the city’s built-up layout to take advantage of the new line, I should probably say something about how the buses should change. I say I and not we, because so far we don’t have a project under our program for this; for now, this is just a blog post, though one informed by past work on the subject.

Parallel and orthogonal buses

In general, when a new line opens, it reduces demand on parallel bus routes, which it outcompetes, and increases it on orthogonal ones, which feed it. However, what counts as parallel and what counts as orthogonal are not always obvious.

Case in point: when Second Avenue Subway opened at the end of 2016, ridership on the east-west buses between the Upper East and West Sides fell. The new line in theory runs north-south, but it undulates from the Upper East Side to Times Square, where passengers can connect to trains to the Upper West Side and points north; when I lived at 72nd and York and commuted to Columbia by bus and subway in 2009-10, I calculated that if Second Avenue Subway had been open already, a two-seat subway ride with a Times Square connection would have cut my one-way commute from 50 to 37 minutes.

This means that to understand how a new rail line will impact buses, it’s necessary to look beyond just the line itself, and think what it connects to.

For example, note on the above map that the increase in job access at the Flatbush Avenue station, intersecting the Nostrand Avenue Line, is relatively small, and doesn’t have a big north-south footprint along Nostrand. This is because the location already has subway service connecting to Manhattan, a much larger job center than anything IBX would connect to; the buses at the station, the B41 on Flatbush and B44 on Nostrand, already function as connectors to the subway at this point, and are unlikely to acquire more ridership as a result.

In contrast, the stations at Myrtle and Metropolitan are both seen to have a large increase in job access, and in particular a large increase in job access along those two avenues even somewhat away from the stations. On Myrtle, the current buses are the B54 and Q55; the B54 connects to the M train, but it’s one branch, and then the bus continues to Downtown Brooklyn, to which there’s no good subway connection from the future IBX station. The B54 is likely to lose ridership to Downtown Brooklyn but gain it to the new IBX station, and the Q55 is likely to gain in general, as they ferry passengers to a station where they can quickly and with one change go to any number of express lines. Metropolitan has a similar issue – the Q54 already connects to the M, but at least from points west, nobody has any reason to make that connection since it would just double back, whereas with IBX, the Q54 would efficiently connect people to Jackson Heights, and with an additional change to anywhere on the Queens Boulevard and Flushing Lines.

New nodes

Public transit lines serve two functions: to run along a corridor, and to connect nodes. New York usually thinks in terms of corridors, and indeed names nearly all subway lines after the streets they run on (such as a Manhattan avenue) rather than after where they go. But nodes are important as well. Some of that is reflected in the above analysis of the Flatbush-Nostrand Avenue station, currently Brooklyn College on the Nostrand Avenue Line: it really needs to be thought of as a node, and IBX will strengthen it, but not by enough to require running more B41 and B44 buses. In contrast, other nodes will be strengthened enough that bus service increases are warranted.

East New York/Broadway Junction is the biggest standout. East New York’s bus network today is not much of a grid – instead, buses connect outlying areas to the nearest subway station; the bus redesign we did for Brooklyn would make it more of a grid but still follow the logic of feeding the subway wherever it is closest. However, IBX makes Broadway Junction and the Atlantic Avenue station more interesting, which should leads to some changes, turning the new station into more of a node for buses. Buses avoiding this node should instead make sure to stop not just at the subway but also at a new IBX stop, such as Linden.

Jackson Heights is the other. It is a node to some extent today, served by the Q32, Q33, Q47, Q49, Q53, and Q70. But in that general area, the intersection of Woodhaven and Queens Boulevard is an even larger node, and in Queens writ large, the ends of the subway in Jamaica and Flushing are far and away the biggest ones. With IBX, more buses should run to Jackson Heights; for example, all Woodhaven buses, and not just the Q53, should continue along Queens Boulevard and Broadway to reach the station.

Substitutions

In Queens, the street network connecting Jackson Heights with the neighborhoods near the borough line with Brooklyn is not at all conducive for good transit. Buses are usually a good indicator of relative demand along a corridor, but sometimes they aren’t; the situation of IBX is generally one in which they are not, but this is especially bad in Queens. This means that the question of which buses would see demand fall as IBX substitutes for them is even harder than on Second Avenue Subway, the north-south line that efficiently substitutes for east-west buses.

In Brooklyn, I think the answer is relatively straightforward, in that the main crosstown routes, like the B35 on Church, exhibit substitutability. In Queens, it’s harder, and I don’t have concrete answers, only general thoughts that we can turn into a report if there turns out to be demand for it:

  • If a bus has sections along the corridor but also away from it, like the Q18 or Q47, then it should be cut to just connect to the line, in these two case at Jackson Heights.
  • If a bus runs directly between two nodes that could get faster service via a subway-IBX connection, and it doesn’t serve much along the way, then it’s likely to be analogous to the east-west buses across Central Park, and see reduced ridership demand.
  • In general, the routes in Central Queens zigzag so much that IBX is likely to represent a massive improvement in trip times, making such buses less useful.