New York Commuter Rail Rolling Stock Needs

Last night I was asked on Twitter about the equipment needs for an integrated commuter rail system in New York, with through-running from the New Jersey side to the Long Island and Connecticut side. So without further ado, let’s work this out, based on different scenarios for how much infrastructure is built and how much capacity there is.

Assumptions on speed

The baseline assumptions in all scenarios should be,

  • The rolling stock is new – this is about a combined purchase of trains, so the trains should be late-model international EMUs with the appropriate performance specs.
  • Trains are single-deck, to speed up boarding and alighting in Manhattan.
  • The entire system is electrified and equipped with high platforms, to enable rapid acceleration and limit dwell times to 30 seconds, except at Grand Central and Penn Station, where they are 2 minutes each.
  • Non-geometric speed limits (such as difficult turnouts) are lifted through better track maintenance standards and the use of track renewal machines, and geometric speed limits are based on 300 mm of total equivalent cant, or a lateral acceleration of 2 m/s^2 in the horizontal plane.
  • However, speed limits through new urban tunnels, except those used by intercity trains, are at most 130 km/h even when interstations are long.
  • Every junction that needs to be grade-separated for reliability is.
  • Peak and reverse-peak service are symmetric (asymmetric service may not even save rolling stock if the peak is long enough).
  • Urban areas have infill stations as needed to provide coverage, except where lines are parallel to the subway, such as the LIRR Main Line west of Jamaica.
  • Timetables are padded 7% over the technical travel time, and the turnaround time is set at 10 minutes per terminal.

Line trip times

With the above assumptions in mind, let’s compute end-to-end trip times by line. Note that we do not care which lines match up with which lines east and west of Penn Station – the point is not to write complete timetables, but to estimate rolling stock needs. The shortcut we can take is that trains are sufficiently frequent at the peak that artifacts coming from the question of which lines match with which likes are not going to matter. Trip times without links are directly computed for the purposes of this post, and should be viewed as somewhat less certain, within a few percent in each direction.

TerminusService patternTrip time
Great NeckLocal0:32
Port WashingtonLocal0:39
HempsteadLocal0:37
East Garden CityLocal0:37
Far RockawayLocal0:39
Long BeachLocal0:40
West HempsteadLocal0:36
West Hempstead DinkyLocal0:10
BabylonLocal0:58
MontaukLocal2:20
HuntingtonExpress west of Floral Park0:43
Port JeffersonExpress west of Floral Park1:10
RonkonkomaExpress west of Floral Park0:57
GreenportExpress west of Floral Park1:42
Oyster Bay DinkyLocal0:25
New Rochelle (via NEC)Local0:26
New Rochelle (to GCT)Local0:21
Stamford (via NEC)Local0:50
Stamford (to GCT)Local0:45
New Haven (to GCT)Express south of Stamford1:18
New Canaan (to GCT)Express south of Stamford0:43
Danbury (to GCT)Express south of Stamford1:15
Waterbury (to GCT)Express south of Stamford1:40
North White PlainsLocal0:40
SoutheastLocal1:16
WassaicLocal1:48
Yonkers (to GCT)Local0:25
Yonkers (via West Side)Local0:23
Croton-Harmon (to GCT)Local0:52
Croton-Harrmon (via West Side)Local0:50
Poughkeepsie (to GCT)Express south of Croton1:12
Poughkeepsie (via West Side)Express south of Croton1:10
Jersey AvenueLocal0:41
TrentonLocal1:01
TrentonExpress north of New Brunswick0:52
Princeton DinkyLocal0:03
Long BranchLocal1:01
Bay HeadLocal1:23
RaritanLocal0:47
High BridgeLocal1:04
Dover (via Summit)Local1:00
Dover (via Montclair)Local1:04
Hackettstown (via Summit)Local1:22
Montclair State UniversityLocal0:33
GladstoneLocal1:08
SummitLocal0:34
Suffern (via Paterson)Local0:50
Suffern (via Radburn)Local0:47
Port Jervis (via Radburn)Local1:50
Spring ValleyLocal0:50
NyackLocal0:51
Tottenville (to GCT)Local0:47
Port Ivory (to GCT)Local0:28
GCT-Penn (with dwells)Local0:04
Jamaica-FiDi adjustmentLocal0:02

The last two adjustment numbers are designed to be added to other lines: Grand Central-Penn Station with 2 minute dwell times at each stop adds 4 minutes to the total trip time, net of savings from no longer having bumper tracks at Grand Central. The Staten Island numbers are also net of such savings. The Jamaica-Lower Manhattan adjustment reflects the fact that, I believe, Jamaica-Lower Manhattan commuter trains with several infill stops would take 0:19, compared with 0:17 on local trains to Penn Station (also with infill).

The 3-line system

The 3-line system is a bare Gateway tunnel with a continuing tunnel to Grand Central (Line 2) and a realignment of the Empire Connection to permit through-service to the northern tunnel pair under the East River (Line 3); Line 1 is, throughout this post, the present-day Hudson tunnel paired with the southern tunnel pair under the East River.

With no Lower Manhattan service, the Erie lines and the Staten Island lines would not be part of this system. Long Island would need to economize by cutting the West Hempstead Branch to a shuttle train connecting to frequent Atlantic Branch and Babylon Branch trains at Valley Stream. The Harlem Line would terminate at Grand Central. Moreover, the weakest tails of the lines today, that is to say Wassaic, Waterbury, Greenport, and Montauk, would not be part of this system – they should be permanently turned into short dinkies.

The table below makes some implicit assumptions about which lines run through and which do not; those that do only require one turnaround as they are paired at the Manhattan end. Overall this does not impact the regionwide fleet requirement.

Total peak service under this is likely to be,

TerminusTrip timeTphFleet size
Great Neck0:3268
Port Washington0:3969
Hempstead0:371217
Far Rockaway0:39610
Long Beach0:40610
West Hempstead Dinky0:1064
Babylon0:581228
Huntington0:43611
Port Jefferson1:10616
Ronkonkoma0:571227
Oyster Bay Dinky0:2534
Stamford (via NEC)0:50611
Stamford (to GCT, via Alt G)0:49611
New Haven (via Alt G)1:22618
New Canaan (via Alt G)0:4736
Danbury (via Alt G)1:1939
North White Plains0:401220
Southeast1:161235
Yonkers (to GCT, via Alt G)0:2967
Croton-Harmon (via West Side)0:50611
Poughkeepsie (via West Side)1:10615
Jersey Avenue0:41610
Trenton1:01614
Long Branch1:0137
Bay Head1:2339
Raritan0:4736
High Bridge1:0437
Dover (via Summit)1:0037
Dover (via Montclair)1:0437
Hackettstown1:2239
Montclair State U0:3334
Gladstone1:0838
Summit0:3434

This totals 379 trainsets; most should be 12 cars long, and only a minority should be as short as 8 cars; only the dinkies should be shorter than that. Off-peak, service is likely to be much less frequent – perhaps half as frequent on most lines, with some less frequent lines reduced to dinkies with timed connections to maintain base 20-minute frequencies – but the peak determines the capital needs, not the off-peak.

The 5-line system

The Lower Manhattan tunnels connecting Jersey City (or Hoboken) with Downtown Brooklyn and Grand Central with Staten Island make for a Line 4 (Harlem-Grand Central-Staten Island) and a Line 5 (Erie-Atlantic Branch). With such a system in place, more service can be run. The Babylon Branch no longer needs to use the Main Line west of Jamaica, making room for very frequent service on the Hempstead Line, with very high frequency to East Garden City.

In addition to the 379 trainsets for the 3-line system, rolling stock needs to be procured for Staten Island, the Erie lines, and incremental service for extra LIRR trains. In the table below, trip times for the Erie lines absorb the 2-minute adjustment for the LIRR trains they connect to; Staten Island lines are already reckoned from Grand Central. Dwell times for such lines are not included at all, as they are already included in the 3-line table.

The table also omits Port Jervis, as a tail of the Erie Main Line.

TerminusTrip timeTphFleet size
East Garden City0:371219
Suffern (via Paterson)0:5266
Suffern (via Radburn)0:4965
Spring Valley0:5266
Nyack0:5366
Tottenville0:471219
Port Ivory0:281212

This is an extra 73 trainsets, for a total of 452.

Further lines

Most of my maps also depict a Line 6 through-tunnel, connecting East Side Access with Hoboken and completely separating the Morris and Essex system from the Northeast Corridor. This only adds trains in New Jersey, including 6 on the M&E system (say, all turning at Summit, roughly at the outer end of high-density suburbanization), and presumably 6 on the Raritan Valley Line (all turning at Raritan or even closer in, such as at Westfield) and 12 on the Northeast Corridor and North Jersey Coast Line (say, 6 to Jersey Avenue, 3 to Long Branch, and 3 to Bay Head). This adds a total of 37 trainsets. As a sanity check, this is really half a line – all timetables, including the 3-line one, assume East Side Access exists – and the 5-line system with its extra 73 trainsets really only adds 2.5 half-lines (the Harlem Line and 5-minute Atlantic Branch service preexist) and those lines are shorter than average.

More speculative is a Line 7, connecting the Lower Montauk Line with an entirely new route through Manhattan to add capacity to New Jersey; this is justified by high commuter volumes from the Erie lines, which under the 6-line system have the highest present-day commute volume to New York divided by peak service. On the Long Island side, it entails restoring through-service to the West Hempstead Branch instead of reducing it to a dinky, changing a 4-trainset shuttle line into a 19-trainset ((36+10)*2*12/60) through-line, and also doubling service on the Far Rockaway and Long Beach Branches, adding a total of 20 trainsets, a total of 35 for the half-line. On the New Jersey side, it depends on what the service plan is for the Erie Lines and on what is done with the West Shore Line and the Susquehanna; the number of extra trainsets is likely about 40, making the 7-line system require about 600 trainsets.

If ridership grows to the point that outer tails like Wassaic, Waterbury, Greenport, and Montauk justify through-service, then this adds a handful of trains to each. Every hourly train to Southeast that extends to Wassaic requires slightly more than one extra trainset; every hourly train to Greenport requires 1.5 (thus, half-hourly requires 3); every hourly train to Montauk requires three. Direct service to Waterbury, displacing trains going to New Haven, is slightly less than one trainset per hourly train; the most likely schedule that fits everything else is a peak train every 20 minutes, which requires 2 extra trainsets.

What Does Pete Buttigieg Think the US has to Teach the World?

On the 27th, Secretary of Transportation Pete Buttigieg announced the creation of a new program called Momentum, to export what he calls best practices around the world. Buttigieg said he invites global civil society to engage with USDOT, linking to the Momentum website – not so that the US can learn from the rest of the world, but so that the rest of the world can learn from the United States on matters of transportation and climate change.

It’s remarkable that the areas covered by Momentum are consistently ones on which the only thing to learn from the United States is what not to do. There are seven target areas: transport infrastructure projects, climate change mitigation, transport safety, regional corridors, logistics supply chains, emerging tech (e.g. smart cities), good regulatory practices.

That the US has the world’s worst urban rail construction costs is just the beginning. Climate change, so central to this plan, is another example of American failure; Wikipedia’s list has the US near the top in CO2 emissions per capita, and the US is lagging in not just decarbonizing transport, which the entire developed world is failing at, but also in installing renewable energy (or nuclear power). Transport safety is almost always better in rich countries than in poor ones, but in 2018 the US had the highest per capita car accident death rate in the developed world, and rates rose during corona (in Germany, they fell). The supply chain issues in the US are often localized to the one country – the baby formula shortage is worse than in Europe. Good regulatory practices are to be learned from countries with strong apolitical civil service apparatuses, and not from the US, with its grabbing-hand regulators and government by lawsuit.

There is approximately one thing the US has to teach other countries, but it’s nowhere within USDOT’s portfolio: people who are familiar with the history of infrastructure construction in the early 20th century, when it was labor-intensive because everything was labor-intensive then by today’s standards, should teach these methods to countries with similar GDP per capita to Gilded Age and Progressive Era America, like India or Nigeria, so that they can use their advantage in low-cost labor and avoid importing expensive machinery or use techniques that only make sense with modern first-world wages.

However, exporting that history requires taking the exact opposite approach of Momentum. Momentum tells other places “you can be like the US!”. The historical approach tells them “your GDP per capita is $5,000, get over your cultural cringe and your tendency toward isomorphic mimicry and think how to get from $5,000 to $20,000.” As it is, any country that participates in the Momentum program is likely to be importing bad practices, including a politicized civil service, anti-housing NIMBYism, slow government that is supposed to protect civil rights and environmental standards but doesn’t, and a can’t-do attitude.

I don’t know where the idea for such a stupid scheme came. I know USDOT was interested in dialog with other countries to learn best practices, but I don’t know how far up that idea went. Not knowing Washingtonian well, I can’t tell from Buttigieg’s language whether his junket trip to Germany impressed him with how public transport here is run. But somewhere in that game of telephone, the notion that the US should learn from other countries turned into that the US should teach other countries, and that’s just wrong.

I’ve heard the Momentum program analogized to having Saudi Arabia export its human rights practices. And this analogy, unfortunately, goes further than intended. It’s not just that Saudi Arabia is a notorious human rights abuser and the US is (among people with comparative knowledge) notorious for the poor state of its transport infrastructure. It’s that Saudi Arabia does in fact export its human rights practices – dictators all over the world are impressed by Mohammed bin Salman and wish they had the ability to murder international journalists with a US green card. Countries with wealth or cultural cachet have soft power like this.

And unfortunately, this is not just hypothetical when it comes to infrastructure. A lot of public transit construction badness originates in the United Kingdom, where the privatization of the state in the late 20th century exerts considerable soft power anywhere that interacts with the London elite. The peripheral Anglosphere learned those practices and has subsequently seen its construction costs explode: Canada, Singapore, and Hong Kong all built subways at reasonable costs until, depending on the country, 15-25 years ago, and in Canada the explosion can be traced to the adoption of bad practices like design-build contracts and poor oversight of consultants. The Nordic countries and France are British-curious as well – the bibliography of the Stockholm cost report, to appear very soon, is replete with papers discussing how Sweden should privatize infrastructure construction and maintenance on the British model, written by the Swedish civil service or by academics who are contracted to do research for it, none questioning whether such privatization is wise.

The US has fortunately not been able to export its own variety of dysfunction so far, which differs in some key ways from the British dysfunction that consultants so often recommend. This is because Americans have been insular in both directions so far; after the failure of programs in the 1960s to create heaven on Earth (and defeat communism) within the span of one presidential administration, the US reduced its global presence, and now it’s much more likely that a poor country seeking infrastructure advice will buy Japanese or Chinese dysfunction (and almost never the positive things in those countries’ infrastructure – Chinese investments in African railways build palatial stations outside city centers, but not actual high-speed rail).

Unfortunately, Momentum seems set up to export this dysfunction after decades of neglect. And even more unfortunately, American dysfunction is worse than British dysfunction and much worse than Japanese or Chinese dysfunction. Japan builds subways domestically for maybe $400 million per km – more in Central Tokyo, less in very suburban areas; Japanese-financed projects elsewhere in Asia, such as the Jakarta, Ho Chi Minh, and Dhaka Metros, are largely elevated, but correcting for that, they’re more expensive, and the mostly-but-not-wholly-underground Dhaka MRT manages to get up to $600 million per km even without such correction. But Los Angeles, San Jose, and Seattle are all worse than this, and New York is worse than all three with its $2 billion/km projects. Far from acknowledging that these are all failures, the Biden administration named San Jose’s Nuria Fernandez as the head of the Federal Transit Administration, and in her capacity as FTA head, Fernandez gave a keynote talk at Eno’s symposium on construction costs that displayed total indifference to the problem and consisted of a litany of excuses.

I hope that nobody should make the mistake of participating in the Momentum program. USDOT should take it down and replace its pretense of teaching the world with the humility of learning from it. The Bipartisan Infrastructure Law touted by Buttigieg in the video spends tens of billions on urban mass transit and tens more on intercity rail. Done right – that is, done not in the American way – it can create amazing things for American transportation and set up a success that will leave Americans wanting more and then going ahead and building more. But the US needs to lower its head and learn from places that build urban rail for $150 million/km instead of stepping on a soapbox and towering over everyone else.

Quick Note: Bureaucratic Legalism in the United States

After I wrote about the absurdly high construction cost of wheelchair accessibility in New York and the equally absurd timeline resulting from said cost, I got some criticism from people I respect, who say, in so many words, that without government by lawsuit, there’s no America. Here, for example, is Alex Block extolling the notion of accessibility as a right, and talking about how consent decrees can compel change.

But in reality, accessibility is never a right. Accessibility is a feature. The law can mandate a right to certain standards, but the practice of accessibility law in the United States is a constant negotiation. The Americans with Disabilities Act mandated full accessibility everywhere – but even the original text included a balancing test based on the cost of compliance. In practice, legacy public transportation providers negotiated extensive grandfather clauses, and in New York the result was an agreement to make 100 key stations accessible by 2020. Right now, the negotiation has been extended to making 95% of the system accessible by 2055.

And this is why adversarial legalism must be viewed as a dead-end. The courts are not expert on matters of engineering or planning, and in recognizing their technical incompetence they remain extremely deferential to the state on matters of fact. If the MTA says “we can’t,” the courts are not going to order the system shut down until it is compliant, nor do they have the ability to personally penalize can’t-do or won’t-do managers. They can impose consent decrees but the people implementing those decrees can remain adversarial and be as difficult as possible when it comes to coordinating plans; the entire system assumes the state cannot function, and delivers as intended.

So as adversarial legalism is thrown into the ashbin of history as it should be, what can replace it? The answer is, bureaucratic legalism. This already has plenty of precedents in the United States:

  • Drug approval is a bureaucratic process – the courts were only peripherally involved in the process of corona vaccination policy.
  • The US Army Corps of Engineers can make determinations regarding environmental matters, for example the wetlands that the deactivated railway to be restored for South Coast Rail passes through, with professional opinions about mitigations required through Hockomock and Pine Swamp.
  • Protection of National Parks is a bureaucratic process: the National Park Service can impose regulations on the construction of infrastructure, and during the planning for the Washington Metro it demanded that the Red Line cross Rock Creek in tunnel rather than above ground to limit visual impact.
  • While the ADA is said to be self-enforcing via the courts, in practice many of the accessibility standards in the US, such as ramp slope, maximum gap between train and platform, elevator size, maximum path of travel, and paratransit availability are legislative or regulatory.

Right now, there’s an attempt by the FTA to improve public transit access to people with limited English proficiency. This, too, can be done the right way, that is bureaucratically, or the wrong way, that is through lawsuits. Last year, we wrote a response to an RFI about planning for equity highlighting some practices that would improve legibility to users who speak English poorly. Mandating certain forms of clarity in language to be more legible to immigrants who speak poor English and recommending others on a case-by-case basis does not involve lawsuits. It has no reason to – people who don’t speak the language don’t have access to the courts except through intermediaries, and if intermediaries are needed, then they might as well be a regulator with ethnographic experience.

What’s more, the process of government by lawsuits doesn’t just fail to create value (unless 95% accessibility by 2055 counts as value). It also removes value. The constant worry about what if the agency gets sued leads to kludgy solutions that work for nobody and often create new expected and unexpected problems.

Public Transportation in the Southeastern Margin of Brooklyn

Geographic Long Island’s north and south shores consist of series of coves, creeks, peninsulas, and barrier islands. Brooklyn and Queens, lying on the same island, are the same, and owing to the density of New York, those peninsulas are fully urbanized. In Southeastern Brooklyn, moreover, those peninsulas are residential and commercial rather than industrial, with extensive mid-20th century development. Going northeast along the water, those are the neighborhoods of Manhattan Beach, Gerritsen Beach, Mill Basin, Bergen Beach, Canarsie, Starrett City, and Spring Creek. The connections between them are weak, with no bridges over the creeks, and this affects their urbanism. What kind of public transportation solution is appropriate?

The current situation

The neighborhoods in the southeastern margin of Brooklyn and the southern margin of Queens (like Howard Beach) are disconnected from one another by creeks and bays; transportation arteries, all of which are currently streets rather than subway lines, go north and northwest toward city center. At the outermost margin, those neighborhoods are connected by car along the Shore Parkway, but there is no access by any other mode of transportation, and retrofitting such access would be difficult as the land use near the parkway is parkland and some auto-oriented malls with little to no opportunity for sprawl repair. The outermost street that connects these neighborhoods to one another is Flatlands, hosting the B6 and B82 buses, and if a connection onward to Howard Beach is desired, then one must go one major street farther from the water to Linden, hosting the B15.

For the purposes of this post, the study area will be in Brooklyn, bounded by Linden, the Triboro/IBX corridor, and Utica:

This is on net a bedroom community. In 2019, it had 85,427 employed residents and 39,382 jobs. Very few people both live and work in this area – only 4,005. This is an even smaller proportion than is typical in the city, where 8% of employed city residents work in the same community board they live in – the study zone is slightly smaller than Brooklyn Community Board 18, but CB 18 writ large also has a lower than average share of in-board workers.

In contrast with the limited extent of in-zone work travel, nearly all employed zone residents, 76,534, work in the city as opposed to its suburbs (and 31,685 of the zone’s 39,382 jobs are held by city residents). Where they work looks like where city workers work in general, since the transportation system other than the Shore Parkway is so radial:

Within the zone, the southwestern areas, that is Mill Basin and Bergen Beach, are vaguely near Utica Avenue, hosting the B46 and hopefully in the future a subway line, first as an extension of the 4 train and later as an independent trunk line.

To the northeast, Canarsie, Starrett City, and Spring Creek are all far from the subway, and connect to it by dedicated buses to an outer subway station – see more details on the borough’s bus map. Canarsie is connected to the L subway station named after it by the B42, a short but high-productivity bus route, and to the 3 and 4 trains at Utica by the B17, also a high-productivity route. Starrett City does not have such strong dedicated buses: it is the outer terminus of the circumferential B82 (which is very strong), but its dedicated radial route, the B83 to Broadway Junction, is meandering and has slightly below-average ridership for its length. Spring Creek is the worst: it is a commercial rather than residential area, anchored by the Gateway Center mall, but the mall is served by buses entering it from the south and not the north, including the B83, the B84 to New Lots on the 3 (a half-hourly bus with practically no ridership), the rather weak B13 to Crescent Street and Ridgewood, and the Q8 to Jamaica.

The implications for bus design

The paucity of east-west throughfares in this area deeply impacts how bus redesign in Brooklyn ought to be done, and this proved important when Eric and I wrote our bus redesign proposal.

First, there are so few crossings between Brooklyn and Queens that the routes crossing between the two boroughs are constrained and can be handled separately. This means that it’s plausible to design separate bus networks for Brooklyn and Queens. In 2018 it was unclear whether they’d be designed separately or together; the MTA has since done them separately, which is the correct decision. The difficulty of crossings argues in favor of separation, and so does the difference in density pattern between the two boroughs: Brooklyn has fairly isotropic density thanks to high-density construction in Coney Island, which argues in favor of high uniform frequency borough-wide, whereas Queens grades to lower density toward the east, which argues in favor of more and less frequent routes depending on neighborhood details.

Second, the situation in Starrett City is unacceptable. This is an extremely poor, transit-dependent neighborhood, and right now its bus connections to the rest of the world are lacking. The B82 is a strong bus route but many rush hour buses only run from the L train west; at Starrett City, the frequency is a local bus every 10-12 minutes and another SBS bus every 10-12 minutes, never overlying to produce high base frequency. The B83 meanders and has low ridership accordingly; it should be combined with the B20 to produce a straight bus route going direct on Pennsylvania Avenue between Starrett City and Broadway Junction, offering neighborhood residents a more convenient connection to the subway.

Third, the situation in Spring Creek is unacceptable as well. Gateway Center is a recent development, dating only to 2002, long after the last major revision of Brooklyn buses. The bus network grew haphazardly to serve it, and does so from the wrong direction, forcing riders into a circuitous route. Only residents of Starrett City have any direct route to the mall, but whereas Starrett City has 5,724 employed residents (south of Flatlands), and Spring Creek has 4,980 workers, only 26 people commute from Starrett City to Spring Creek. It’s far more important to connect Spring Creek with the rest of the city, which means buses entering it from the north, not the south. Our bus redesign proposal does that with two routes: a B6/B82 extension making this and not Starrett City the eastern anchor, and a completely redone B13 going directly north from the mall to New Lots and thence hitting Euclid Avenue on the A/C and Crescent Street on the J/Z.

What about rail expansion?

New York should be looking at subway expansion, and not just Second Avenue Subway. Is subway expansion a good solution for the travel needs of this study zone?

For our purposes, we should start with the map of the existing subway system; the colors indicate deinterlining, but otherwise the system is exactly as it is today, save for a one-stop extension of the Eastern Parkway Line from New Lots to the existing railyard.

Starrett City does not lie on or near any obvious subway expansion; any rail there has to be a tram. But Canarsie is where any L extension would go – in fact, the Canarsie Line used to go there until it was curtailed to its current terminus in 1917, as the trains ran at-grade and grade-separating them in order to run third rail was considered impractically expensive. Likewise, extending the Eastern Parkway Line through the yard to Gateway Center is a natural expansion, running on Elton Street.

Both potential extensions should be considered on a cost per rider basis. In both cases, a big question is whether they can be built elevated – neither Rockaway Parkway nor Elton is an especially wide street most of the way, about 24 or 27 meters wide with 20-meter narrows. The Gateway extension would be around 1.3 km and the Canarsie one 1.8 km to Seaview Avenue or 2.3 km to the waterfront. These should cost around $250 million and $500 million respectively underground, and somewhat less elevated – I’m tempted to say elevated extensions are half as expensive, but this far out of city center, the underground premium should be lower, especially if cut-and-cover construction is viable, which it should be; let’s call it two-thirds as expensive above-ground.

Is there enough ridership to justify such expansion?

Let’s start with Canarsie, which has 28,515 employed residents between Flatlands and the water. Those workers mostly don’t work along the L, which manages to miss all of the city’s main job centers, but the L does have good connections to lines connecting to Downtown Brooklyn (A/C), Lower Manhattan (A/C again), and Midtown (4/5/6, N/Q/R/W, F/M, A/C/E). Moreover, the density within the neighborhood is uniform, and so many of the 28,515 are not really near where the subway would go – Rockaway/Flatlands, Rockaway/Avenue L, Rockaway/Seaview, and perhaps Belt Parkway for the waterfront. Within 500 meters of Rockaway/L and Rockaway/Seaview there are only 9,602 employed residents, but then it can be expected that nearly all would use the subway.

The B42 an B17 provide a lower limit to the potential ridership of a subway extension. The subway would literally replace the B42 and its roughly 4,000 weekday riders; nearly all of the 10,000 riders of the B17 would likely switch as well. What’s more, those buses were seeing decreases in ridership even before corona due to traffic and higher wages inducing people to switch away from buses – and in 2011, despite high unemployment, those two routes combined to 18,000 weekday riders.

If that’s the market, then $500 million/18,000 weekday riders is great and should be built.

Let’s look at Gateway now. Spring Creek has 4,980 workers, but first of all, only 3,513 live in the city. Their incomes are very low – of the 3,513, only 1,030, or 29%, earned as much as $40,000/year in 2019 – which makes even circuitous mass transit more competitive with cars. There’s a notable concentration of Spring Creek workers among people living vaguely near the 3/4 trains in Brooklyn, which may be explained by the bus connections; fortunately, there’s also a concentration among people living near the proposed IBX route in both Brooklyn and Queens.

The area is the opposite of a bedroom community, unlike the other areas within the study zone – only 1,114 employed people live in it. Going one block north of Flatlands boosts this to 1,923, but a block north of Flatlands it’s plausible to walk to a station at Linden at the existing railyard. 51% of the 1,114 and 54% of the 1,923 earn at least $40,000 a year. Beyond that, it’s hard to see where neighborhood residents work – nearly 40% work in the public sector and OnTheMap’s limitations are such that many of those are deemed to be working at Brooklyn Borough Hall regardless of their actual commute destination.

There’s non-work travel to such a big shopping center, but there are grounds to discount it. It’s grown around the Shore Parkway, and it’s likely that every shopper in the area who can afford a car drives in; in Germany, with generally good off-peak frequency and colocation of retail at train stations, the modal split for public transit is lower for shopping trips than for commutes to work or school. Such trips can boost a Gateway Center subway extension but they’re likely secondary, at least in the medium run.

The work travel to the mall is thankfully on the margin of good enough to justify a subway at $50,000/daily trip, itself a marginal cost. Much depends on IBX, which would help deliver passengers to nearby subway nodes, permitting such radial extensions to get more ridership.

Adversarial Legalism and Accessibility

New York State just announced that per the result of a legal settlement, it is committing to make 95% of the subway accessible… by 2055. Every decade, 80-90 stations will be made accessible, out of 472. Area advocates for disability rights are elated; in addition to those cited in the press release or in the New York Times article covering the news, Effective Transit Alliance colleague Jessica Murray speaks of it as a great win and notes that, “The courts are the only true enforcement mechanism of the Americans with Disabilities Act.” But to me, it’s an example not of the success of the use of the courts for civil rights purposes, in what is called adversarial legalism, but rather its failure. The timeline is a travesty and the system of setting the government against itself with the courts as the ultimate arbiter must be viewed as a dead-end and replaced with stronger administration.

The starting point for what is wrong is that 2055 is, frankly, a disgrace. By the standards of most other old urban metro systems, it is a generation behind. In Berlin, where the U-Bahn opened in 1902, two years before the New York City Subway did, there has been media criticism of BVG for missing its 2022 deadline for full accessibility; 80% of the system is accessible, and BVG says that it will reach 100% in 2024. Madrid is slower, planning only for 82% by 2028, with full accessibility possible in the 2030s. Barcelona is 93% accessible and is in the process of retrofitting its remaining stations. Milan has onerous restrictions such that only one wheelchair user may board each train, but the majority of stations have elevators, and 76% have elevators or stairlifts. In Tokyo, Toei is entirely accessible, and so is nearly the entirety of Tokyo Metro. Even London is 40% accessible, somewhat ahead of New York. Only Paris stands as a less accessible major world metro system.

The primary reason for this is costs. The current program to make 81 stations accessible by 2025 is $5.2 billion. This is $64 million per station, and nearly all are single-line stations requiring three elevators, one between the street and the outside of fare control and one from just inside fare control to each of two side platforms. Berlin usually only requires one elevator as it has island platforms and no fare barriers, but sometimes it needs two at stations with side platforms, and the costs look like 1.5-2 million € per elevator. Madrid the cost per elevator is slightly higher, 3.2 million €. New York, in contrast, spends $20 million, so that a single station in New York is comparable in scope to the entirety of the remainder of the Berlin U-Bahn.

And this is what adversarial legalism can’t fix. The courts can compel the MTA to install elevators, but have no way of ensuring the MTA do so efficiently. They can look at capital plans and decree that a certain proportion be spent on accessibility; seeing $50 billion five-year capital plans, they can say, okay, you need to spend 5-10% of that on subway accessibility. But if the MTA says that a station costs $64 million to retrofit and therefore there is no room in the budget to do it by 2030, the courts have to defer.

This, in turn, is a severe misjudgment of what the purpose of civil rights legislation is. Civil rights laws giving individuals and classes the right to sue the government already presuppose that the government may be racist, sexist, or ableist. This is why they confer individual and group rights to sue under Title VI (racial equality in transportation and other facilities), Title IX (gender equality in education), and the ADA. If the intention was to defer to the judgment of government agencies, no such laws would be necessary.

And yet, the nature of adversarial legalism is that on factual details, courts are forced to defer to government agencies. If the MTA says it costs $64 million to retrofit a station, the courts do not have the power to dismiss managers and hire people who can do it for $10 million. If the MTA says it has friction with utilities, the courts cannot compel the utilities to stop being secretive and share the map of underground infrastructure in the city or to stop being obstructive and start cooperating with the MTA’s contractors when they need to do street work to root an elevator. Judges are competent in legal analysis and incompetent in planning or engineering, and this is the result.

Worse, the adversarial process encourages obstructive behavior. The response to any request from the public or the media soon becomes “make me”; former Capital Construction head and current MTA head Janno Lieber said “file a Freedom of Information request” to a journalist who asked what 400 questions federal regulators asked regarding congestion pricing. Nothing goes forward this way, unless accessibility in 33 years counts, and it shouldn’t.

How Washington Should Spend $10 Billion

The planned $10 billion expansion of Washington Union Station is a waste of money, but this does not mean that money appropriated for public transportation in the National Capital Region is a waste. The region has real transportation needs that should be addressed through urban rail expansion – just not through a rebuild of the intercity rail station. Those needs include local and regional travel, to be addressed through investment in both the Metro and the commuter rail networks. It is fortunate that when I probed on Twitter, there was broad if imperfect agreement among area advocates about what to do.

A $10 billion budget should be spent predominantly on new Metro Rail lines, carefully chosen to satisfy multiple goals at once: physical expansion of the reach of the system, additional core capacity, and deinterlining to improve reliability and increase the capacity of existing lines. For the purposes of the question I posed to area advocates, I set the expansion budget at $7.5 billion, good for 30 km at average global prices, leaving the rest for commuter rail improvements.

What to do about commuter rail

Washington does not have a large legacy commuter rail network, unlike New York, Chicago, Boston, or Philadelphia. It is not as old as those cities, and its conception as the southern end of an East Coast region stretching up to Boston is postwar, by which point investment in passenger rail was largely relegated to the past. Nonetheless, it does have some lines, three to the north as the MARC system and two to the south as the VRE system. They should be upgraded to better commuter rail standards.

Union Station already has the infrastructure for through-running. The junction between the through-tunnel and the terminal tracks is flat, and almost all intercity trains terminate and most will indefinitely no matter how much investment there is in high-speed rail to points south. This requires delicate scheduling, which is good up to about 18 trains per hour in each direction, either six through- and 12 terminating or the other way around. Running half-hourly all-day service on each of the lines, with some additional urban overlay in Virginia and extra service on the Penn Line to Baltimore, should not be too difficult.

Thus, the main spending items on the agenda are not new tracks, but electrification and high platforms. MARC runs diesel trains even under catenary on the Northeast Corridor, which problem requires no additional electrification to fix, but its other two lines are unelectrified, and VRE has no electrification infrastructure. Those lines total 327 route-km of required wiring, with extensive single-tracking reducing per-km cost; this should be around $600 million. But note that they all carry significant freight traffic, and additional accommodations may be necessary.

As far as platforms go, there are nearly 50 stations requiring high platforms (I think 49 but I may have miscounted). At Boston costs it should be $1 billion or a bit more, but that’s for long trains, and MARC trains are not so long, and a system based on shorter trains at higher frequency would be somewhat cheaper. Infill stations are probably unnecessary – there are Metro Rail lines along the inner sections of most of the lines providing the urban rail layer.

Metro Rail expansion

The most pressing problem WMATA’s trains have is poor reliability. Two changes in the late 2000s and 2010s made the system worse: the 2009 elimination of automatic (though not driverless) operations worsened ride quality and reducing capacity, and the 2014 opening of the Silver Line introduced too much interlining reducing both reliability and capacity. WMATA is aware of the first problem and is working to restore ATO; the Silver Line’s problems should be fixed through judicious use of deinterlining. Deinterlining by itself only requires a short extension of the Yellow Line to separate the lines, but it can be bundled with further expansion.

Consensus among area advocates is that there should be separate tunnels for the Yellow and Blue Lines and a new trunk line under Columbia Pike, which three lines total 21 km. Additional lines can consist of another trunk line going northeast from Union Station between the Brunswick and Camden Lines or an extension of the Columbia Pike line from Bailey’s Crossroads, the present outer limit of high density, to Annandale, which would require extension transit-oriented development along the line.

A full-size version can be found here; note that the lines at Union Station are moved around to get rid of the Red Line’s awkward U-shape. The northeast extension option is colored red but should be a Blue Line extension, but the Red Line taking over H Street and going to Largo.

No New Washington Union Station, Please

A new presentation dropped for Amtrak’s plans to rebuild Union Station. It is mostly pictorial, but even the pictures suggest that this is a very low-value project, one with little to no transportation value and limited development value. The price tag is now $10 billion (it was $7 billion 10 years ago; the increase is somewhat more than cumulative inflation), but even if two zeros are cut from the budget it’s not necessarily worth it.

What are the features of good train stations?

A train station is interface between passengers and trains. Everything about their construction must serve this purpose. This includes the following features:

  • Platforms that can effectively connect to the trains (Union Station has a mix of high and low platforms; all platforms used by Northeast Corridor trains must be raised).
  • Minimum distance from platform to street or to urban transit.
  • Some concessions and seats for travelers, all in an open area.
  • Ticketing machines.
  • An information booth with maps of the area and station facilities.
  • Nothing more.

In particular, lavish waiting halls not only waste of money but also often have negative transport value, as they either force passenger to walk longer between street and platform or steer them to take an option that involves a longer walk; the new Moynihan Train Hall in New York is an example of the latter failure. Berlin Hauptbahnhof, a rare example of a major urban station built recently in a rich country, has extensive shopping, but it’s all designed around fast street-platform and S-Bahn-intercity connections.

What are the features of Washington Union Station expansion?

The presentation highlights the following features:

  • A new concourse beneath the platforms.
  • A new concourse on H Street with a prominent headhouse, with bus and streetcar connections.
  • An enclosed bus facility.
  • Underground parking.
  • Future air rights development.

All of the above are wasteful. Connections to H Street can be handled through direct egress points from the platforms to the street, and passengers can get between H Street and the main historic station via those egress points and the platforms themselves. The platforms are key circulation spaces at a train station and using them for passenger movements is normal; I can see an argument against that if the platforms are unusually narrow or crowded, as is the case in New York, but in Washington there is no such excuse.

Nor is Union Station a major node for city buses. Washington’s surface transit network serves the station, but it’s not a major bus node – only a handful of buses terminate there and they don’t run frequently – and even if it were, a surface bus loop akin to what Ostbahnhof has in Berlin would have sufficed. Thus, the bus infrastructure should be descoped, and buses should keep using the streets.

So, none of the transit connections have any value. Parking, moreover, has negative value, as it encourages access to the area by car, displacing transit trips. Union Station already has a Metro connection as well as some surface transit. Better rail operations would also improve commuter rail access for intercity rail riders. Unfortunately, the plan does not improve those operations, nor is there any plan for much needed capital investment to go alongside better mainline rail operations, such as Virginia electrification and high platforms.

What about the air rights?

They are a poor use of money. Building towers on top of active railyards is more difficult and more expensive than building them on firma. Hudson Yards projects in New York came in at around $12,000 per square meter in hard costs, twice the cost of Manhattan skyscrapers on firma except those associated with the World Trade Center, which were unusually costly.

Nor is the location just north of the historic Union Station so desirable that developers would voluntarily pay the railyard premium to be there. The commercial center of Washington is well to the west of the site, comprising Metro Center and Farragut. More office towers around Union Station would be nice for rebalancing and for generating demand for future mainline rail improvements, but the place for them is on firma around the existing station and not on top of the approach tracks.

What should be done?

The plan should be rejected in its entirety and no further funding should be committed to it. Good transit activists should demand that spending on public transportation and intercity rail go to those purposes and not toward building unnecessary train halls. Moreover, it is unlikely the managers at Amtrak who pushed for it and who still are the client for the project understand modern rail operations, nor is it likely that they will ever learn. With neither need nor use for the project, it should be canceled and the people involved in its management and supervision laid off.

How Many Tracks Do Train Stations Need?

A brief discussion on Reddit about my post criticizing Penn Station expansion plans led me to write a very long comment, which I’d like to hoist to a full post explaining how big an urban train station needs to be to serve regional and intercity rail traffic. The main principles are,

  • Good operations can substitute for station size, and it’s always cheaper to get the system to be more reliable than to build more tracks in city center.
  • Through-running reduces the required station footprint, and this is one of the reasons it is popular for urban commuter rail systems.
  • The simpler and more local the system is, the fewer tracks are needed: an urban commuter rail system running on captive tracks with no sharing tracks with other traffic and with limited branching an get away with smaller stations than an intercity rail station featuring trains from hundreds of kilometers away in any direction.

The formula for minimum headways

On subways, where usually the rush hour crunches are the worst, trains in large cities run extremely frequently, brushing up against the physical limitation of the tracks. The limit is dictated by the brick wall rule, which states that the signal system must at any point assume that the train ahead can turn into a brick wall and stop moving and the current train must be able to brake in time before it reaches it. Cars, for that matter, follow the same rule, but their emergency braking rate is much faster, so on a freeway they can follow two seconds apart. A metro train in theory could do the same with headways of 15 seconds, but in practice there are stations on the tracks and dealing with them requires a different formula.

With metro-style stations, without extra tracks, the governing formula is,

\mbox{headway } = \mbox{stopping time } + \mbox{dwell time } + \mbox{platform clearing time }

Platform clearing time is how long it takes the train to clear its own length; the idea of the formula is that per the brick wall rule, the train we’re on needs to begin braking to enter the next station only after the train ahead of ours has cleared the station.

But all of this is in theory. In practice, there are uncertainties. The uncertainties are almost never in the stopping or platform clearing time, and even the dwell time is controllable. Rather, the schedule itself is uncertain: our train can be a minute late, which for our purpose as passengers may be unimportant, but for the scheduler and dispatcher on a congested line means that all the trains behind ours have to also be delayed by a minute.

What this means that more space is required between train slots to make schedules recoverable. Moreover, the more complex the line’s operations are, the more space is needed. On a metro train running on captive tracks, if all trains are delayed by a minute, it’s really not a big deal even to the control tower; all the trains substitute for one another, so the recovery can be done at the terminal. On a mainline train running on a national network in which our segment can host trains to Budapest, Vienna, Prague, Leipzig, Munich, Zurich, Stuttgart, Frankfurt, and Paris, trains cannot substitute for one another – and, moreover, a train can be easily delayed 15 minutes and need a later slot. Empty-looking space in the track timetable is unavoidable – if the schedule can’t survive contact with the passengers, it’s not a schedule but crayon.

How to improve operations

In one word: reliability.

In two words: more reliability.

Because the main limit to rail frequency on congested track comes from the variation in the schedule, the best way to increase capacity is to reduce the variation in the schedule. This, in turn, has two aspects: reducing the likelihood of a delay, and reducing the ability of a delay to propagate.

Reducing delays

The central insight about delays is that they may occur anywhere on the line, roughly in proportion to either trip time or ridership. This means that on a branched mainline railway network, delays almost never originate at the city center train station or its approaches, not because that part of the system is uniquely reliable, but because the train might spend five minutes there out of a one-hour trip. The upshot is that to make a congested central segment more reliable, it is necessary to invest in reliability on the entire network, most of which consists of branch segments that by themselves do not have capacity crunches.

The biggest required investments for this are electrification and level boarding. Both have many benefits other than schedule reliability, and are underrated in Europe and even more underrated in the United States.

Electrification is the subject of a TransitMatters report from last year. As far as reliability is concerned, the LIRR and Metro-North’s diesel locomotives average about 20 times the mechanical failure rate of electric multiple units (source, PDF-pp. 36 and 151). It is bad enough that Germany is keeping some outer regional rail branches in the exurbs of Berlin and Munich unwired; that New York has not fully electrified is unconscionable.

Level boarding is comparable in its importance. It not only reduces dwell time, but also reduces variability in dwell time. With about a meter of vertical gap between platform and train floor, Mansfield has four-minute rush hour dwell times; this is the busiest suburban Boston commuter rail station at rush hour, but it’s still just about 2,000 weekday boardings, whereas RER and S-Bahn stations with 10 time the traffic hold to a 30-second standard. This also interacts positively with accessibility: it permits passengers in wheelchairs to board unaided, which both improves accessibility and ensures that a wheelchair user doesn’t delay the entire train by a minute. It is fortunate that the LIRR and (with one peripheral exception) Metro-North are entirely high-platform, and unfortunate that New Jersey Transit is not.

Reducing delay propagation

Even with reliable mechanical and civil engineering, delays are inevitable. The real innovations in Switzerland giving it Europe’s most reliable and highest-use railway network are not about preventing delays from happening (it is fully electrified but a laggard on level boarding). They’re about ensuring delays do not propagate across the network. This is especially notable as the network relies on timed connections and overtakes, both of which require schedule discipline. Achieving such discipline requires the following operations and capital treatments:

  • Uniform timetable padding of about 7%, applied throughout the line roughly on a one minute in 15 basis.
  • Clear, non-discriminatory rules about train priority, including a rule that a train that’s more than 30 minutes loses all priority and may not delay other trains at junctions or on shared tracks.
  • A rigid clockface schedule or Takt, where the problem sections (overtakes, meets, etc.) are predictable and can receive investment. With the Takt system, even urban commuter lines can be left partly single-track, as long as the timetable is such that trains in opposite directions meet away from the bottleneck.
  • Data-oriented planning that focuses on tracing the sources of major delays and feeding the information to capital planning so that problem sections can, again, receive capital investment.
  • Especial concern for railway junctions, which are to be grade-separated or consistently scheduled around. In sensitive cases where traffic is heavy and grade separation is too expensive, Switzerland builds pocket tracks at-grade, so that a late train can wait for a slot without delaying cross-traffic.

So, how big do train stations need to be?

A multi-station urban commuter rail trunk can get away with metro-style operations, with a single station track per approach track. However, the limiting factor to capacity will be station dwell times. In cases with an unusually busy city center station, or on a highly-interlinked regional or intercity network, this may force compromises on capacity.

In contrast, with good operations, a train station with through-running should never need more than two station tracks per approach track. Moreover, the two station tracks that each approach track splits into should serve the same platform, so that if there is an unplanned rescheduling of the train, passengers should be able to use the usual platform at least. Berlin Hauptbahnhof’s deep tracks are organized this way, and so is the under-construction Stuttgart 21.

Why two? First, because it is the maximum number that can serve the same platform; if they serve different platforms, it may require lengthening dwell times during unscheduled diversions to deal with passenger confusion. And second, because every additional platform track permits, in theory, an increase in the dwell time equal to the minimum headway. The minimum headway in practice is going to be about 120 seconds; at rush hour Paris pushes 32 trains per hour on the shared RER B and D trunk, which is not quite mainline but is extensively branched, but the reliability is legendarily poor. With a two-minute headway, the two-platform track system permits a straightforward 2.5-minute dwell time, which is more than any regional railway needs; the Zurich S-Bahn has 60-second dwells at Hauptbahnhof, and the Paris RER’s single-level trains keep to about 60 seconds at rush hour in city center as well.

All of this is more complicated at a terminal. In theory the required number of tracks is the minimum turn time divided by the headway, but in practice the turn time has a variance. Tokyo has been able to push station footprint to a minimum, with two tracks at Tokyo Station on the Chuo Line (with 28 peak trains per hour) and, before the through-line opened, four tracks on the Tokaido Main Line (with 24). But elsewhere the results are less optimistic; Paris is limited to 16-18 trains per hour at the four-track RER E terminal at Saint-Lazare.

At Paris’s levels of efficiency, which are well below global best practices, an unexpanded Penn Station without through-running would still need two permanent tracks for Amtrak, leaving 19 tracks for commuter traffic. With the Gateway tunnel built, there would be four two-track approaches, two from each direction. The approaches that share tracks with Amtrak (North River Tunnels, southern pair of East River Tunnels) would get four tracks each, enough to terminate around 18 trains per hour at rush hour, and the approaches that don’t would get five, enough for maybe 20 or 22. The worst bottleneck in the system, the New Jersey approach, would be improved from today’s 21 trains per hour to 38-40.

A Penn Station with through-running does not have the 38-40 trains per hour limit. Rather, the approach tracks would become the primary bottleneck, and it would take an expansion to eight approach tracks on each side for the station itself to be at all a limit.

Watch Our Webinar on Construction Costs Tomorrow

The Italy case, done by Marco Chitti, is up on the website. I encourage people to read the entire report on how Italy has set things up in the last 20-30 years so as to have one of the lowest-cost urban rail infrastructure programs in the world. The Turkey case, by Elif Ensari, will be up shortly.

This is leading to a webinar, to be done tomorrow at 16:00 my time, 10:00 New York time, in which Marco and Elif will present their cases to the general public. I encourage people to register; you’ll be able to ask us questions and we’ll answer in chat or on video. But if you can’t make it, it will be recorded.

Intercity Rail Frequency and the Perils of Market Segmentation

SNCF loves market segmentation. Run by airline execs, the company loves to create different trains for different classes of people. Not only do individual trains have opaque pricing run on the basis of yield management, in which similar seats on the same train at the same time of day and day of week may have different fares, but also there are separately-branded trains for separate fare classes, the higher-fare InOui and the lower-fare OuiGo. On international trains, SNCF takes it to the limit and thus Eurostar and Thalys charge premium fares (both about twice as high as domestic TGVs per passenger-km) and don’t through-ticket with domestic TGVs. This has gotten so bad that in Belgium, some advocates have proposed a lower-priced service on the legacy Paris-Brussels line, which would have to be subsidized owing to the high cost of low-speed intercity rail service.

But why is market segmentation on rail so bad? The answer has to do with frequency and cost structures that differ from those of airlines. Both ensure that the deadweight loss from market segmentation exceed any gains that could be made from extracting consumer surplus.

The issue of frequency

A segmented market like that of domestic TGVs reduces frequency on each segment. To maintain segmentation, SNCF has to make the segments as difficult to substitute for each other as possible. OuiGo serves Marne-la-Vallée instead of Gare de Lyon and forcing passengers onto a 20-minute RER connection, or even longer if they’re arriving in Paris and the wave of 1,000 TGV riders creates long lines at the ticketing machines; on other LGVs it serves the traditional Parisian station and thus the segments are more substitutable.

The situation of Eurostar and Thalys reduces frequency as well: the high fares discourage ridership and send much of it to intercity buses or suppress travel. Fewer riders, or fewer riders per segment as in the case of domestic TGVs, lead to fewer trains. What’s the impact of this on ridership?

The literature on high-speed rail ridership elasticities has some frequency estimates. In Couto’s thesis (PDF-p. 225), it is stated that passenger rail ridership has an elasticity of 0.53 with respective to overall service provision. There are also multiple papers estimating the elasticity with respect to travel time: in Cascetta-Coppola the elasticity ranges from -1.6 to -2.2, in Börjesson it is -1.12, and in a Civity report it is stated based on other work that it is -0.8 to -2. The lowest values in Börjesson are associated with the premium-fare AVE, while the range for the original TGV, priced at the same level as the slower trains it replaced, is -1.3 to -1.6. The upshot is that halving frequency through market segmentation reduces ridership by a factor of 2^0.53 = 1.44, which is far more than the benefit yield management is claimed to have, which is a 4% increase in revenue per SNCF’s American proposals from 2009.

Why are trains different?

Planes and buses happily use yield management. High-speed trains do not, except for those run by SNCF or RENFE – and ridership in France isn’t really higher than in fixed-fare Northern Europe or East Asia while ridership in Spain is much lower. Why the difference?

The reason has to do with the ratio of waiting time to trip time. Thalys connects Paris and Brussels in 1.5 hours, every half hour at rush hour and every 2 hours midday. At rush hour, frequency is sort of noticeable; off-peak, it dominates travel time. This is nothing like planes – even short-distance trips involve hours of access, waiting, and egress time, and therefore trips are not usually spontaneous, and day trips are rare except for business travelers.

Buses, finally, are so small that a market like New York-Philadelphia supports multiple competitors each running frequently, and passenger behavior is such that different companies are substitutable, so that the effective frequency is multiple buses per hour.

Cost structure and bad incentives

It’s typical to price high-speed rail higher than legacy rail, even when otherwise there is no yield management. This is bad practice. The operating costs of high-speed rail are lower than those of slow trains. The crew is paid per hour; electricity costs are in theory higher at higher speed but in practice greenfield high-speed lines are constant 300 km/h cruises whereas legacy lines have many acceleration and deceleration cycles; high-speed trainsets cost much more than conventional ones (by a factor of about 2 in Europe) but also depreciate by the hour and not by the km and therefore are somewhat cheaper per seat-km.

This is comparable to the bad practice, common in the United States and in developing and newly-industrialized countries, of pricing urban rail higher than a bus. The metro is nicer for consumers than a bus, but it also has far lower operating costs and therefore a wise transit agency will avoid incentivizing passengers to take buses and instead use integrated fares. The same is true for slow and fast trains: the solution proposed by the Belgian advocates is to incentivize passengers to take a high-cost, low-price train over a low-cost, high-price one, and therefore is no solution at all.

Moreover, the cost structure of trains is different from that of planes. Planes don’t pay much for fixed infrastructure; in effect, every plane trip costs money, and then the challenge is to fill all the seats. High-speed railways instead pay a lot for infrastructure, while their above-the-rails costs are a few cents per passenger-km (€0.06/seat-km on the TGV, including trainset costs and a lot of labor inefficiency). Their challenge is how to fill the tracks with trains, not how to fill the trains with passengers. This is why the fixed clockface frequency common in Germany, Switzerland, Austria, and the Netherlands is so powerful: the off-peak trains are less full, but that’s fine, as the marginal operating cost of an off-peak train is low.

Just lower the fares

Bear in mind that frequency is not exogenous – it is set based on demand. This means that anything that affects ridership has its impact magnified by the frequency-ridership spiral. An exogenous shock, such as improvement in trip time or fare reduction, is magnified through the spiral, by a factor of 1/(1-0.53) = 2.13. In other words, every elasticity estimated in isolation must be multiplied by a factor of about 2.

And once this is understood, suddenly the optima for service look very different from what Thalys has settled on. The optimum is to charge fares to pay infrastructure costs but not much more – especially if you’re SNCF and the railway workers’ union will extract all further profit through strikes, as it did 10 years ago. And this means making sure that except at very busy times, known in advance, Paris-Brussels tickets should be 30€, not 50-100€.