Uday Schultz has a thorough post about New York’s subway service deterioration over the last decade, explaining it in terms of ever more generous maintenance slowdowns. He brings up track closures for renewal as a typical European practice, citing examples like Munich’s two annual weekends of S-Bahn outage and Paris’s summertime line closures. But there’s a key aspect he neglects: over here, the combination of regional rail and subway tunnels means that different trunk lines can substitute for one another. This makes long-term closures massively less painful and expensive.
S-Bahn and subway redundancy
S-Bahn or RER systems are not built to be redundant with the metro. Quite to the contrary, the aim is to provide service the metro doesn’t, whether it’s to different areas (typically farther out in the suburbs) or, in the case of the RER A in Paris, express overlay next to the local subway. The RER and Métro work as a combined urban rail network in Paris, as do the S- and U-Bahns in German cities that have both, or the Metro and Cercanías in Madrid and Barcelona.
And yet, in large urban rail systems, there’s always redundancy, more than planners think or intend. The cleanest example of this is that in Paris, the RER A is an express version of Métro Line 1: all RER A stops in the city have transfers to M1 with the exception of Auber, which isn’t too far away and has ample if annoying north-south transfers to the Champs-Elysées stations on M1. As a result, summertime closures on the RER A when I lived in the city were tolerable, because I could just take M1 and tolerate moderate slowdowns.
This is the case even in systems designed around never shutting down, like Tokyo. Japan, as Uday notes, doesn’t do unexpected closures – the Yamanote Line went decades with only the usual nighttime maintenance windows. But the Yamanote Line is highly redundant: it’s a four-track line, and it is paralleled at short distance by the Fukutoshin Line. A large city will invariably generate very thick travel markets, and those will have multiple lines, like the east-west axis of M1 and the RER A, the two north-south axes of M12 and M13 and of M4 and the RER B, the east-west spine from Berlin Hauptbahnhof east, the Ikebukuro-Shibuya corridor, or the mass of lines passing through Central Tokyo going northeast-southwest.
The issue of replacement service
In the United States, standard practice is that every time a subway line is shut for maintenance, there are replacement buses. The buses are expensive to run: they are slow and low-capacity, and often work off the overtime economy of unionized labor; their operating costs count as part of the capital costs of construction projects. Uday moreover points out that doing long-term closures in New York on the model of so many large European cities would stress the capacity of buses in terms of fleet and drivers, raising costs further.
This is where parallel rail lines come in. In some cases, these can be other subway lines: from north of Grand Central to Harlem-125th, the local 6 and express 4/5 tracks are on different levels, so the express tracks can be shut down overnight for free, and then during maintenance surges the local tracks can be shut and passengers told to ride express trains or Second Avenue Subway. On the West Side, the 1/2/3 and the A/B/C/D are close enough to substitute for each other.
But in Queens and parts of the Bronx, leveraging commuter rail is valuable. The E/F and the LIRR are close enough to substitute for each other; the Port Washington Branch can, to some extent, substitute for the 7; the Metro-North trunk plus east-west buses would beat any interrupted north-south subway and would even beat the subway in normal service to Grand Central.
Running better commuter rail
The use of commuter rail as a subway substitute, so common in this part of the world, requires New York to run service along the same paradigm that this part of the world does. Over here, the purpose of commuter rail is to run urban rail service without needing to build greenfield tunnels in the suburbs. The fares are the same, and the frequency within the city is high all day every day. It runs like the subway, grading into lower-density service the farther one goes; it exists to extend the city and its infrastructure outward into the suburbs.
This way, a coordinated urban rail system works the best. Where lines do not overlap, passengers can take whichever is closest. Where they do, as is so common in city center, disruption on one trunk is less painful because passengers can take the other. The system does not need an external infusion of special service via transportation-of-last-resort shuttle buses, and costs are easier to keep under control.
Andrew Cuomo resigned, effective two weeks from now, after it became clear that if he didn’t the state legislature would remove him. As much of the leadership of public transportation in the state is his political appointees, like Sarah Feinberg, the incoming state governor, Lieutenant-Governor Kathy Hochul, will need to appoint new heads in their stead. From my position of knowing more about European public transit governance than the New York political system does, I’d like to make some recommendations.
Hire from outside the US
New York’s construction costs are uniquely high, and its operating costs are on the high side as well; in construction and to a large extent also in operations, it’s a general American problem. Managers come to believe that certain things are impossible that in fact happen all the time in other countries, occasionally even in other US cities. As an example, we’ve constantly heard fire safety as an excuse for overbuilt subway stations – but Turkey piggybacks on the American fire safety codes and to a large extent so does Spain and both have made it work with smaller station footprints. Much of the problem is amenable to bringing in an outsider.
The outsider has to be a true outsider – outside the country, not just the agency. An American manager from outside transportation would come in with biases of how one performs management, which play to the groupthink of the existing senior management. Beware of managers who try to perform American pragmatism by saying they don’t care about “Paris or such,” as did the Washington Metro general manager. Consultants are also out – far too many are retirees of those agencies, reproducing the groupthink without any of the recent understanding by junior planners of what is going wrong.
Get a Byford, not Byford himself
Andy Byford is, by an overwhelming consensus in New York, a successful subway manager. Coming in from Toronto, where he was viewed as a success as well, he reformed operations in New York to reduce labor-management hostility, improve the agency’s accessibility program, and reduce the extent of slow orders. Those slow orders were put in there by overly cautious management, such as Ronnie Hakim, who came in via the legal department rather than operations, and viewed speed as a liability risk. Byford began a process called Save Safe Seconds to speed up the trains, which helped turn ridership around after small declines in ridership in the mid-2010s.
The ideal leader should be a Byford. It cannot be Byford himself: after Cuomo pushed him out for being too successful and getting too much credit, Byford returned to his native Britain, where Mayor Sadiq Khan appointed him head of Transport for London. Consulting with Byford on who to hire would be an excellent idea, but Byford has his dream job and is very unlikely to come back to New York.
Look outside the Anglosphere
High operating costs are a New York problem, and to some extent a US problem. Canada and the UK do just fine there. However, construction costs, while uniquely bad in New York, are also elevated everywhere that speaks English. The same pool of consultants travel across, spreading bad ideas from the US and UK to countries with cultural cringe toward them like Canada, Australia, and Singapore.
The MTA has a $50 billion 5-year capital plan. Paris could only dream of such money – Grand Paris Express is of similar size with the ongoing cost overruns but is a 15-year project. The ideal head of the MTA should come from a place with low or at worst medium construction costs, to supervise such a capital plan and coordinate between NYCT and the commuter rail operators.
Such a manager is not going to be a native English speaker, but that’s fine – quite a lot of the Continental European elite is fluent in English, though unfortunately this is not as true in Japan, South Korea, or Taiwan. If it is possible to entice a Spanish manager like Silvia Roldán Fernández of Madrid Metro to come in, then this is ideal, given the number of Spanish-speaking New Yorkers; Madrid of course also has legendarily low construction costs, even today. Gerardo Lertxundi Albéniz of Barcelona is a solid option. Italian managers are an option as well given the growing networks in Italy, not just building new lines but also making old stations accessible: Stefano Cetti of Milan’s public works arm MM, Gioia Ghezzi of the operating company ATM, Giovanni Mottura of Rome’s ATAC, etc. Germans like Munich’s Bernd Rosenbusch or Ingo Wortmann or Berlin’s Eva Kreienkamp have experience with juggling conflicting local and state demands and with more labor militancy than people outside Germany associate Germany with. Laurent Probst may well be a good choice with his experience coordinating an even larger transit network than New York’s – assuming that he wouldn’t view New York as a demotion; the same is true of RATP’s head, the generalist Catherine Guillouard.
This is not meant to be a shortlist – these are just the heads of the transit organs of most of the larger Continental Western European systems. Japanese, Korean, and Taiwanese heads should be considered too, if they speak English and if they don’t view working in the US, in a city smaller than Tokyo or Seoul, as a demotion.
Let the civil service work
American civil service is broken – or, more precisely, was never allowed to become an administrative state, thanks to postwar anti-state paranoia. Professionals learn to be timid and wait for the word of a political appointee to do anything unusual. Cuomo did not create this situation – he merely abused it for his own personal gain, making sure the political appointees were not generic liberal Democrats but his own personal loyalists.
The future cannot be a return to the status quo that Cuomo exploited. The civil service has to be allowed to work. The role of elected politicians is to set budgets, say yes or no to megaproject proposals, give very broad directions (“hire more women,” “run like a business,” etc.), and appoint czars in extreme situations when things are at an impasse. Byford acted as if he could work independently, and Cuomo punished him for it. It’s necessary for New York to signal in advance that the Cuomo era is gone and the next Byford will be allowed to work and rewarded for success. This means, hiring someone who expects that the civil service should work, giving them political cover to engage in far-reaching reforms as required, and rewarding success with greater budgets and promotions.
The bipartisan infrastructure framework (BIF) just passed the Senate by a large margin, with money for both roads and public transportation. Unlike the 2009 Obama stimulus, the BIF has plenty of money for high-speed rail – not just $8 billion as in the 2009 bill, but a total of $66 billion to be spent on mainline rail. The Northeast Corridor program gets $24 billion out of this $66 billion in a dedicated program and another $6 billion out of another program within this bucket dedicated to Amtrak. This is $30 billion, which should be more than enough for high-speed rail on the Northeast Corridor. Together with other buckets for other parts of the US, it can even build some non-Northeastern lines, for example serving Chicago or Los Angeles.
I say should because the current plans are to waste the money. But better things are possible, so at the Transit Costs Project, we’re planning to embark on a project to write a report on how to do this better. The construction cost report will be done in early 2022, but we can overlap to some extent. A one-year program, to debut in early 2023, will include a Northeast Corridor proposal; a two-year one will also include tie-ins and starter lines elsewhere, such as Chicago-Cleveland/Detroit or Los Angeles-San Diego.
But for this, we need funding. We’re a good deal of the way there, I think around two-thirds for the two-year option – and this isn’t quite enough for the one-year option, some of the money needs to be matched. This is not the same as my Patreon in either scale (the difference is more than an order of magnitude) or scope (my Patreon funds the blog and vlog, which are way more general); if you know grants for such projects, please let us know, we can send a fuller proposal.
What’s the project’s scope?
Lots and lots of analysis, for one, like what we’re doing for subways. Intriguingly, high-cost countries for high-speed rail tend to also have high subway costs and vice versa, and this remains true even as it is easier to explain high-speed rail costs in terms of unnecessary scope and leakage. But this is not the dominant part of the project – rather, we are going to be synthetic and make a proposal. We’re not committing to an investment figure; my guess is that in 2021 dollars it should be around $15 billion to cut Northeast Corridor trip times to about 1:45 on each of New York-Boston and New York-Washington, but some variation is possible in either direction.
If there’s $30 billion for the Northeast Corridor, and high-speed rail is doable for half that, then the other half should be spent on tie-ins, for example improving regional rail in all four major metropolitan areas. Naturally, this should only include useful spending for rail operations and connections, but the Northeast doesn’t lack for those; New York can spend $17 billion on new tunnels and that’s at the per-km cost of Citybanan, one of the cheaper city center regional rail projects in our database.
We are happy to announce that on Sunday the 29th of August we will hold this year’s Modernizing Rail conference, on the heels of the success last year.
Please register using this form. And please give details on what you’d like to see, and if you’re willing to lead sessions – the schedule of the breakout sessions is still up in the air depending on popular demand. Even the number of breakouts depends on how many registrants we get, compared with the about 200 we had last year. Perhaps the news of the infrastructure bill will tilt the demand toward more political sessions regarding how to ensure what is built is good and less toward technical best practices.
Our keynote is certainly political: Rep. Seth Moulton (D-MA), who represents the northern suburbs of Boston (6th district) and for years has been pushing the North-South Rail Link. He will give brief remarks at 16:00 Eastern time, or 22:00 Central Europe Summer Time, to be followed by a Q&A; if you have a question that you’d like to hear an answer to, you can mention it in the registration form, or email the organizing committee at email@example.com. We will be taking questions throughout the conference, which will start 11:00 Eastern, so if your questions depend on what you hear at the breakouts, you’re in luck.
After my last post on poor timetabling in the New York area, I got a lot of feedback comparing New York’s zonal system with existing high-quality commuter rail networks. Some of it was in comments, but most interesting was a post by the pseudonymous socialist Emil Seidel, who compares the situation in New York with that of Munich.
I’m going to go over some best practices here – this is not intended as a highlight of poor American practices. That said, because of the application to New York, I’m going to go over Paris and Tokyo, as they’re both very large cities, in addition to cleaner German examples, including Berlin (where I live), Nuremberg (where Herbert in comments lives and where a Twitter commenter pointed out express service), and finally Emil’s example of Munich.
The upshot is that yes, commuter trains do often have express service, and it’s common for the express service to run local on an outer segment and then express closer in. However, this is not really the New York zone theory. Most importantly, high-quality local service always comes first, and everything else is an overlay. This is common to all of the examples we will look at, and is the most fundamental fact of commuter rail: S-Bahn service is urban rail on mainline tracks.
Infrastructure for local trains
Local service always comes first, ahead of any longer-range regional service. This can be readily seen in infrastructure allocation: in all examples I know of in the German-speaking world, Paris, and Tokyo, when there’s scarce infrastructure built for through-service, local trains get it ahead of longer-range regional ones.
- In Paris, the RER is defined as what runs through on newly-built tunnels, whereas Transilien service terminates at one of the historic terminals of Paris. This distinction is fundamental and precedes other distinctions, such as frequency – there are sections of Transilien H, J, and L that have higher frequency than some RER branches. And where the two systems run side-by-side, the RER is the more local one.
- In Germany, newly-built tunnels are for S-Bahn service. For example, in Munich, the S-Bahn gets to use the tunnel, while other trains terminate on the surface; this is also the case in Frankfurt, Stuttgart (until the upcoming Stuttgart 21), and Berlin (until the North-South Main Line opened).
- In Zurich, there are two through-tunnels under Hauptbahnhof. The older one is used principally by the S-Bahn; the newer one is used by the S-Bahn as well as longer-distance trains. But many long-distance trains stay on the surface.
- In Tokyo, local commuter trains get preference in JR through-running. The original set of through-tracks at Tokyo Station was used for local trains on the Yamanote and Keihin-Tohoku Line, while faster, longer-distance regional trains were demoted, and through-running ceased entirely when the Shinkansen took their space in the 1990s. Regional trains only resumed through-running when the Ueno-Tokyo Line opened in 2015. The Shinkansen’s use of space over regional train is justified because it serves large secondary cities in the Tohoku region and not just suburbs.
Timetabling for local trains
Local trains are also the most important priority for high frequency. In all of the five example cities for this post, local frequency is high, even on branches. In Tokyo and Paris, the trunks don’t really run on takts; Japan and France overall have less rigid takts than Germany but do have off-peak takt patterns, it’s just not very important to passengers when a train on the RER A or the Chuo Line comes every 4-5 minutes off-peak.
Elsewhere, there are takts. There are also takts on the branches in Paris. Typical frequencies are a train every 10, 15, or 20 minutes; they may be lower on outer branches, especially ones that are operationally half-branches, i.e. branches of branches like the two halves of S1 and S2 in Munich. All of this depends on city size; Berlin is bigger than Munich, which is bigger than Nuremberg.
- In Berlin, S-Bahn branches run every 10 or 20 minutes, but the ones running every 10 usually have short-turning variants, so the outer portions only get 20-minute service. The outer ends of 10-minute service – Spandau, Buch, Frohnau, Friedrichshagen, Teltow Stadt, Grünau – tend to be 15-18 km from the center, but one, Potsdam, is almost 30 km out.
- In Munich, S-Bahn branches likewise run every 10 or 20 minutes at rush hour, with some tails that have ugly 40-minute headways. Off-peak, the numbered branches run every 20 minutes.
- In Nuremberg, frequency is weaker, as it is a small city. But S2 has a 20-minute takt up to Schwabach, about 15 km out.
Let us now compare larger cities. Just as Berlin has higher frequency at a given radius than Munich and Nuremberg, so does Paris have even higher frequency, and Tokyo yet higher. On the RER A, branches run every 10 minutes all day; Marne-la-Vallée, home to Disneyland Paris as well as a suburban office park, sees trains every 10 minutes off-peak, 37 km outside city center. At the other end, Cergy sees a train every 10 minutes all day at similar distance, and at rush hour this rises to 5 minutes, but half the trains run on Transilien L rather than the RER.
Some of these Parisian RER trains run express. The RER B, off-peak, has a pattern with three services, each running every 15 minutes: at each end these go minor branch (Robinson or Mitry-Claye), major branch express (major stops to Massy and then local to Saint-Rémy or nonstop to CDG), major branch local (local to Massy or CDG). So yes, nonstop trains exist, in the special context of an airport, but local trains still run every 15 minutes as far as 20-30 km from city center. At rush hour, frequencies rise and there’s no more room for express trains to the north, so trains run every 6 minutes to each of CDG or Mitry, all local: local service always comes first.
Tokyo has even higher local frequency. Rapid lines tend to have their own dedicated pair of tracks, there is so much traffic. For example, the Chuo Line has four tracks to Mitaka: the local tracks carry the Chuo-Sobu Line, and the express tracks carry the Chuo Rapid Line farther out. Both patterns are very frequent.
What Tokyo does have is a melange of express services with names like Special Rapid, Limited Express, or Liner. However, they are timetabled around the local services, or the regular rapid ones if there’s a rapid track pair as on Chuo, even in environments with competition between private railways for commuter traffic. The Chuo Rapid Line’s basic pattern, the vanilla rapid, runs irregularly every 3-8 minutes off-peak, with Special Rapid trains making limited stops timetabled around those, with timed overtakes at major stations. Thus frequency stays very high even as far out as Tachikawa, 37.5 km from Tokyo Station. Moreover, at rush hour, where frequency is denser, there is less, sometimes no, special express service.
Timetabling for express trains
All of our five example cities have express trains. In Berlin, Munich, and Nuremberg, they’re branded as RegionalBahn, distinct from the S-Bahn. In Paris, some RER trains run express, but mostly Transilien provides extra express service. In Tokyo, it’s all branded as part of the Kanto area commuter rail network. This is the core of Emil’s argument: express service exists in Germany, but has separate branding.
Nonetheless, there are best practices for how to do this. In Jarrett Walker’s bus-based terminology, it is better to run limited, that is make major stops, than to run express, that is have long nonstop sections from outer areas to city center. Sometimes patterns are somewhat of a hybrid, like on some New York subway lines, but the basic principle is that regional trains never skip major stations.
- In Berlin, the Stadtbahn, built in the 1880s, has four tracks, two dedicated to local S-Bahn trains and two to everything else. Intercity trains on the Stadtbahn only stop at Hauptbahnhof and Ostbahnhof, but regional trains make roughly every other S-Bahn stop. Elsewhere, some stations are never missed, like Lichtenberg and Wannsee. Note also that as in Paris, Berlin likes its airport express service, branded FEX, which skips the RegionalBahn station and S-Bahn branch point Schöneweide.
- In Munich, some RegionalBahn services express from the S-Bahn terminal, where they always stop, to Hauptbahnhof; some also make a few stops on the way. It depends on the line – Dachau and Laim are both popular RegionalBahn stops.
- In Nuremberg, I encourage people to look at the map. Express trains abound, at fairly high frequency, each named service running hourly, and they always make certain major stations like Erlangen and Fürth.
The stopping pattern can be more local once there’s no S-Bahn, but it’s not really local. For example, at both ends of Berlin’s RE 1, a half-hourly regional line between Brandenburg an der Havel and Frankfurt an der Oder with half the trains continuing west to Magdeburg and south awkwardly to Cottbus, there are stops spaced 7-10 km apart between the built-up area of Berlin-Potsdam and those of Brandenburg and Frankfurt.
In Paris and Tokyo, similarly, express trains stop at major stations. The RER B’s express pattern does run nonstop between Gare du Nord and CDG, but to the south of Paris, it makes major stops like Bourg-la-Reine rather than trying to run nonstop from Massy to Paris; moreover, the RER trains make all stops within the city core, even neighborhood stops like Cité-Universitaire or Nation. Tokyo’s Special Rapids likewise stop at major stations like Kokubunji, and don’t run nonstop from outer suburban branches to Shinjuku and Tokyo.
What this means for New York
New York does not run its commuter rail in the above way. Not even close. First, local frequency is weak. The pre-corona timetables of the New Haven and Harlem Lines have 30-40 minute gaps at rush hour at radii where Berlin still has some 10-minute service. Off-peak the schedule is more regular but still only half-hourly. Hourly S-Bahn systems exist, for example in Mannheim, but those are mocked by German railfans as not real S-Bahns but barely upgraded regional rail systems using the term S-Bahn for marketing.
And second, express trains are not designed to provide an express overlay on top of local trains with transfers where appropriate. When they’re zoned, they only make a handful of stops at rush hour and then express, often without overlapping the next zone for a transfer. This is the case even where the infrastructure is a four-track line set up for more normal express service: the Hudson Line is set up so that Ossining, Tarrytown, and Yonkers have express platforms, but its timetable largely ignores that in favor of long nonstops, with 20-minute gaps at Yonkers.
In the future, it is critical to focus on a high-quality local takt, with frequency depending on city size. In Boston, a Berlin-size city, the TransitMatters plan calls for a 15-minute takt, sometimes 10 minutes, generally as far out as 20-30 km. But New York is a larger city, and needs 5 minutes within the city and 10 well into suburbia, with a strong local schedule that express trains can go around if appropriate. S-Bahn service, by whatever name or brand it has, is always about using mainline infrastructure to operate urban rail and extend the city into the suburbs.
The ideal use of a politically-determined, external infusion of funds into public transit is for a capital expansion that is not critical. The service provided should be of great usefulness – otherwise, why fund it? – but it should fundamentally be not a safety-critical package, which should be funded locally on an ongoing basis. The best kind of project is one with a high one-time capital cost and long-term benefits, since a debt-issuing sovereign state can borrow cheaply and obtain the financial and social return on investment without much constraint.
Outside infusions, such as from a stimulus bill or an infrastructure package, are best used on expansion with short-term costs and long-term benefits. This includes visible projects that extend systems but also ones that reduce long-term operating and maintenance costs. For examples:
- High-speed rail: it’s operationally profitable anywhere I know of, and then the question is whether the ROI justifies the debt. Because a one-time cost turns into a long-term financially sustainable source of revenue, it is attractive for outside investment.
- Railstitution of a busy bus route, or burial of a busy tramway. This produces a combination of lower operating expenses and better service for passengers. The only reason not to replace every high-ridership city bus with a subway is that subways cost money to build, but once the outside infusion of money comes, it costs less to run a modern rapid transit system, or even a not so modern one, than a bus system with its brigades of drivers.
- Rail automation.
- Speed-up of a rail route to higher standards and lower maintenance costs.
The importance of non-critical projects
Critical projects are not good for a stimulus bill. The reason is that they have to be done anyway, and the process of stimulus may delay them unacceptably, as a local government assumes it will get an infusion of funds and does not appropriate its own money for it. The upshot is that a rational federal funding agency should be suspicious of a local or state agency that requests money for critical projects, especially safety-critical ones.
The point here is that the stimulus process is inherently political. It does not involve technical decisions of what the optimal kind of public transportation policy should be. It instead pits infrastructure investments against other budget priorities, like the military, holding down tax rates, or health care. It’s not meant to be predictable to the transportation expert, and only barely to the political insider. It depends on political vagaries, the state of the economy, and petty personal decisions about priorities.
Thus, an agency that asks for stimulus funds for a project sends (at least) one of two messages: “we think this project is great but if it’s not built people aren’t going to literally die,” or “we are run by incompetent hacks.” In the former case, the point of a benefit-cost analysis is that neither the costs nor the benefits are existential: the project is not safety-critical nor critical to the basic existence of the system, but the budget is not existential to the budget either and if it is wasted then the government will not go bankrupt.
I propose that transportation agencies hire people whose job is to keep abreast of global developments in the field and report on best practices.
Which agencies should do it?
Ideally, all urban ones. Very small ones should piggyback on large ones, or participate in metropolitan planning to increase the scale. National agencies could aid this by having their own larger offices, but each urban or metropolitan agency should keep a best practices expert for issues relevant to the specific local context.
How big should the team be?
Normally, only one person is required. A larger team may be necessary for language coverage. In Germany, one English-speaking person could interface with every agency in Europe – even in relatively monolingual places like Spain and Italy, enough experts speak English that it’s possible to work without learning the local language. However, East Asia is largely monolingual, and interfacing with experts in Japan, South Korea, Taiwan, and China is harder in English. Moreover, reading local debates and contracts should be done in the local language even in multilingual countries like the Netherlands and Sweden.
So since language coverage is needed, larger agencies should keep teams of sufficient size. It’s not possible to have full coverage, but, again, English is decent in a pinch. A team of about 5 should be fine, especially if the language coverage is random enough that nearby agencies are likely to only partially overlap; for example, if Berlin’s team includes a Japanese speaker and Hamburg’s includes a Chinese speaker, they can learn secondarily.
No large internal hierarchy is required. Not counting language issues, one person could do this. With full language accounting, as required for agencies the size of NYCT, TfL, or RATP, the team may have a director and a few reports, but the reports should still be paid as experienced professionals and have direct access to agency managers.
What are the team’s responsibilities?
- Keep abreast of global developments through reading trade publications, following media in relevant countries so as to know whether a proposed solution is locally considered a success or not, and keeping track of how relevant agencies introduce new technology.
- Go to international conferences to form horizontal relationships with peers and acquire more detailed knowledge of new methods, and follow up to discuss specifics with them.
- Connect local decisionmakers with peers elsewhere in order to discuss how to adapt outside innovations to the local social and political context.
Who should be hired?
People who are likely to have the required knowledge. Horizontal hiring from other agencies is especially valuable, especially agencies from other cultures, where existing hiring is less likely to happen. American agencies occasionally hire Brits and Canadians, so it’s valuable to hire people with Asian, Continental European, or Latin American agency experience for this team.
Such people tend to be mobile, and if they leave to another agency, that’s fine. Often, the most valuable thing is a person who one can email and ask “in Barcelona, how do you do maintenance on the Cercanías?” (and that’s a high-level question, there are more detailed questions at lower zoom level than our work on costs). A former employee who moved on to another agency is always going to remain such a point contact, provided they left on good terms.
To the extent high-wage countries underlearn from lower-wage ones, they have an easier time hiring this way. Junior engineers in Italy earn less than 2,000€/month after taxes; Northern European and American agencies can poach them with better pay.
Robert Jackel asked me an excellent question in comments: what is a pulse? I’ve talked about timed transfers a lot in the last almost 10 years of this blog, but I never wrote a precise definition. This is a critical tool for every public transportation operation with more than one line, making sure that trains and buses connect with as short a transfer window as possible given other constraints. Moreover, pulse-oriented thinking is to plan capital investment and operations to avoid constraints that make transfers inconvenient.
When are pulses needed?
Passengers perceive the disutility of a minute spent transferring to be more than that of a minute spent on a moving vehicle. This is called the transfer penalty and is usually expressed as a factor, which varies greatly within the literature. In a post from 2011 I quoted a since-linkrotted thesis with pointers to Boston and Houston’s numbers, and in a more recent post I found some additional literature in a larger variety of places, mostly in the US but also the Netherlands. The number 2 is somewhere in the middle, so let’s go with this.
Observe that the transfer penalty measured in minutes and not in a factor is, naturally, larger when service runs less frequently. With a factor of 2, it is on average equal to the headway, which is why it is likely the number is 2 – it represents actual time in the worst case scenario. The upshot is that the value of an untimed transfer is higher the higher the frequency is.
I used the principle of untimed transfers and frequency to explain why small subway networks do not look like small bus networks – they have fewer, more frequent lines. Subway lines that run every 3-4 minutes do not need transfer timing, because the time cost of an untimed transfer is small compared to the likely overall trip time, which is typically in the 15-30 minute range. But the lower the frequency, the more important it is to time transfers. Thus, for example, Berlin times the U6/U7 transfer at Mehringdamm in the evening, when trains run every 10 minutes, but does not do so consistently in the daytime, when they run every 5.
But note: while the value of an untimed transfer is higher at higher frequency, the value of a timed transfer is the same – it is zero-penalty or close to it no matter what. So really, the relative value of timing the transfer decreases as frequency increases. But at the same time, if frequency is higher, then more passengers are riding your service, which justifies more investment to try to time the transfer. The German-speaking planning tradition is the most concerned with transfer timing, and here, it is done commonly at 10 minutes, occasionally at 5 minutes, and never that I know of at higher frequency.
Easy mode: one central station
If all your buses and trains serve one transit center, then a pulse means that they all run at the same frequency, and all meet at the center at the same time. This doesn’t usually happen on urban rail networks – a multi-line urban rail system exists in a high-ridership, high-frequency context, in which the value of serving a mesh of city center lines is high, and the cost of bringing every subway tunnel to one location is high. Instead, this happens on buses and on legacy regional rail networks.
The pulse can be done at any frequency, but probably the most common is hourly. This is routine in small American towns with last-resort bus networks serving people too poor or disabled to drive. Two and a half years ago a few of us on Transit Twitter did a redesign-by-Twitter of the Sioux City bus network, which has ten bus routes running hourly, all pulsing in city center with timed connections. A similar network often underlies the night buses of a larger city that, in the daytime, has a more complete public transport network, such as Vancouver.
Even here, planners should keep two delicate points in mind. First, on buses in mixed traffic, there is an upper limit to the frequency that can be timetabled reliably. The limit depends on details of the street network – Jarrett Walker is skeptical that timetabling buses that run every 15 minutes is feasible in a typical American city, but Vancouver, with no freeways within a city and a rich arterial grid, manages to do so every 12 minutes on 4th Avenue. A half-hourly pulse is definitely possible, and even Jarrett writes those into his bus redesigns sometimes; a 20-minute pulse is probably feasible as well even in a typical American city. The current practice of hourly service is not good, and, as I point out in the Sioux City post, involves slow, meandering bus routes.
The second point is that once the takt is chosen, say half an hour, the length of each roundtrip had better be an integer multiple of the takt, including a minimal turnaround time. If a train needs 5 minutes to turn, and runs half-hourly, then good times for a one-way trip from city center are 10, 25, 40, 55 minutes; if there is no turnaround at city center, for example if there is through-running, then half as many turnarounds are needed. This means that short- and long-term planning should emphasize creating routes with good trip times. On a bus, this means straightening meanders as needed, and either extending the outer end or cutting it short. On a train, this means speedup treatments to run as fast as necessary, or, if the train has a lot of spare time, opening additional infill stops.
The issue of branching
Branches and pulses don’t mix well. The ideal way to run a system with a trunk and branches is to space the branches evenly. The Berlin S-Bahn runs every 3-4 minute on the Stadtbahn trunk and on the North-South Tunnel, mixing services that run every 10 and 20 minutes at roughly even intervals. In such an environment, timed transfers in city center are impossible. This is of course not a problem given Stadtbahn headways, but becomes serious if frequency is sparser. A one-trunk, two-branch regional rail system’s planners may be tempted to run each branch every half hour and interpolate the schedules to create a 15-minute headway on the trunk, but if there’s a half-hourly pulse, then only one branch can participate in it.
This is visible when one compares S-Bahn and RegionalBahn systems. High-frequency S-Bahn systems don’t use timed transfers in city center, because there is no need. I can get from Jannowitzbrücke to Ostkreuz without consulting a schedule, and I would get to the Ring without consulting a schedule either, so there is no need to time the crossing at Ostkreuz. There may be sporadic transfer timing for individual branches, such as between the S9 branch of the Stadtbahn, which diverts southeast without serving Ostkreuz, and the Ring, but S9 runs every 20 minutes, and this is not a pulse, only a single-direction timed connection.
In contrast, RegionalBahn systems, running at longer ranges and lower frequencies, often tend toward timed transfers throughout. The tradeoff is that they don’t overlie to create high-frequency trunks. In some cases, trains on a shared trunk may even platoon, so that all can make the same timed transfer, if high trunk frequency is not desired; this is how intercity trains are run on the Olten-Bern line, with four trains to a platoon every 30 minutes.
Medium mode: dendritic networks
A harder case than the single pulse is the dendritic network. This means that there is a central pulse point, and also secondary pulse points each acting as a local center. All cases I am aware of involve a mainline rail network, which could be S-Bahn rather than RegionalBahn, and then bus connections at suburban stations.
Already, this involves more complex planning. The reason is that the bus pulse at a suburban station must be timed with trains in both directions. Even if planners only care about connections between the suburban buses and trains toward city center, the pulse has to time with inbound trains for passengers riding from the suburban buses to the city and with outbound trains for passengers riding from the city to the buses. This, in turn, means that the trains in both directions must arrive at the station at approximately the same time. A few minutes of leeway are acceptable, since the buses turn at city center so the connection always has a few minutes of slack, but only a few minutes out of what is often a half-hourly takt.
Trains that run on a takt only meet every interval equal to half the takt. Thus, if trains run half-hourly, they can only have suburban pulses every 15 minutes of travel. This requires planners to set up suburban pulses at the correct interval, and speed up or sometimes slow down the trains if the time between suburban nodes. Here is an example I’ve worked on for a Boston-Worcester commuter train, with pulses in both Framingham and Worcester.
Hard mode: meshes
The next step beyond the dendritic network is the multi-node network whose graph is not simply connected. In such a network, every node must have a timed transfer, which imposes considerable planning constraints. Optimizing such a network is an active topic of research in operations and transportation in European academia.
Positive examples for such networks come from Switzerland. Large capital investments are unavoidable, because there’s always going to be some line that’s slower than it needs to be. The key here is that, as with dendritic networks, nodes must be located at consistent intervals, equal to multiples of half the headway, and usually the entire headway. To make multiple timed transfers, trains must usually be sped up. This is why pulse-based integrated timed transfer networks require considerable planning resources: planning for rolling stock, infrastructure, and the timetable must be integrated (“the magic triangle”) to provide maximum convenience for passengers connecting from anywhere to anywhere.
Proof-of-payment with ungated train stations is a useful technique for reducing construction costs. It simplifies the construction of stations, since there is no need for a headhouse or mezzanine – people can go directly from the street to the platform. A station without fare control requires just a single elevator, or two if side platforms are desired, and can be built shallowly using cut-and-cover. Cities across the size spectrum, perhaps only stopping short of hypercities, should take heed and use this to build urban rail more cheaply.
Is this a common cost control technique?
No. The vast majority of low-construction cost countries use faregates, which is why I was reticent to recommend proof-of-payment as a cost mitigation strategy. Spain, Italy, Korea, and Sweden are all faregated; among the world’s lowest-cost countries, I believe only Finland and Switzerland use proof-of-payment fare collection on urban rail.
However, there are exceptions. In Italy, the Brescia Metro uses proof-of-payment. This is not typical for the country or the region – Italian metros have fare control, like the vast majority of systems outside Germany and Germany-influenced countries. However, because Brescia is small, the system was forced to engage in value engineering, removing scope that would be routine in larger cities like Milan. The majority was built cut-and-cover or above-ground; the typical urban Italian metro is entirely bored. Italian metro systems prefer short stations on new lines to minimize costs and provide capacity through automated operations and extremely high frequency; Brescia takes this to an extreme and has 30-meter trains. Among these cost minimization tactics is the lack of fare control. The result of this entire package is that Brescia spent 915 million euros on a 13.7 km metro system.
Station size and station cost
So far, we believe that the cost of the station, excavation excluded, should be proportional to the floor area. This is based on something told to us in an interview about electrical system costs for the Boston Green Line Extension, which is light rail in a trench rather than a tunneled metro system, so I recommend caution before people repeat this uncritically.
Moreover, on somewhat more evidence, it appears that the cost of station excavation should be proportional to the volume excavated. Some of the evidence for this is circumstantial: media reports and government reports on the construction of such urban rail projects as Second Avenue Subway, Grand Paris Express, and the RER specify the volume of excavation as a measure of the difficulty of construction. But it’s not just circumstantial. In Paris, the depth of some of the GPX stations has led to some construction complications. Moreover, preliminary interviews in Paris suggest, albeit not definitively, that station construction costs are predominantly a matter of dig volume. Finally, the insistence on short platforms and high frequency as a cost saving technique on new-build metro systems in Italy as well as in Denmark and on the Canada Line in Vancouver is suggestive too, even if it says nothing about whether the relationship between volume and cost is linear, degressive, or superlinear.
How does one minimize station costs with POP?
Proof-of-payment means that there is no fare control between the street and the station. This means any of the following ways of constructing station access become available:
- Cut-and-cover with the platform on level -1, with direct stair and elevator access from the street. The Berlin U-Bahn is built this way, with access points in street medians where available, such as U8 on Brunnenstrasse. It’s easy to build staircases at each end of the platform to increase access, with an elevator in the middle.
- Bored tunnel with large enough bores to fit the platform within the bore. The Barcelona method for this is to use 12-meter bores, but smaller, cheaper versions exist with smaller trains, for example in Milan. It’s also possible to use double-O-tube TBMs for this, but ordinarily they are more expensive than twin bores. Access involves vertical bores down to the platform with elevators or slant bores with escalators; there is no need for intermediate levels or entry halls.
- Bored tunnel with cut-and-cover stations, with no mezzanine levels. Here, the dig volume is unchanged, and the saving from lack of fare control is only in the finishes and elevator costs, not the excavation.
It is noteworthy that the most common technique for metro construction, by far, is the last one, where the savings from POP are the smallest. The vast majority of world metros have fare control, including in low-cost countries, and this perhaps makes metro builders not notice how two separate ways of reducing costs – cut-and-cover and POP – interact especially well together. Nonetheless, this is a real saving.
What does this mean?
A technique can be uncommon in low-cost countries and yet be useful in reducing construction costs. It is useful to think of the way Madrid, Milan, Turin, Stockholm, Oslo, Helsinki, and Seoul build their urban rail systems as good, but not always perfect. A trick that these cities might not pay attention to may still be good. The caveat is that it requires a good explanation for why they have not employed it; in the case of Italy, I believe it’s simply that the non-German world views fare control as the appropriate way to run a metro system and POP as a light rail technique and therefore only good for low-volume operations. There may also be backward compatibility issues – Brescia is a new build, like POP Copenhagen, whereas Milan is building extensions on top of a gated system.
Nonetheless, the evidence from station costs, the success of POP operations in Germany even on very busy lines, and the experience of Brescia all suggest that POP is good for metro construction in general. Cities smaller than New York building new systems should use it exclusively, and cities that already have faregates should tear them down to improve passenger circulation and facilitate the construction of POP lines in the future at lower cost.
The Swiss slogan electronics before concrete, and related slogans like run trains as fast as necessary, not as fast as possible, is a reminder not to waste money. However, I worry that it can be read as an argument against spending money in general. For many years now, Cap’n Transit has complained that this slogan is used to oppose bad transit like the Gateway Tunnel and if the money is not spent on public transportation then it may be spent on other things. But in reality, the Swiss slogans, all emphasizing cost minimization, must be reconciled with the fact that Switzerland builds a lot of concrete, including extensive regional rail tunneling in Zurich and intercity rail tunneling. Electronics precedes concrete, but does not always substitute for it; it’s better to think of these planning maxims as a way to do more with a fixed amount of money, and not as a way to do the same amount of project with less money.
The extent of tunneling in Switzerland
Here is a list of tunnels built in Switzerland since the 1980s, when its modern program of integrated timetable-infrastructure-rolling stock investment began:
- Zurich S-Bahn, including the 7 km combination of the Hirschengraben and Zürichberg Tunnels for the first S-Bahn trunk starting 1990, and the 5 km Weinberg Tunnel for the second trunk starting 2014.
- Geneva RER, including the CEVA trunk, which has about 8.4 km of tunnel.
- The Mattstetten-Rothrist line between Olten and Bern is 52 km long of which a total of 21 km is in tunnel.
- A few more small intercity projects within the Bahn 2000 plan include tunnels.
This is not a small program. Zurich and Geneva are not large cities, and yet they’ve build regional rail trunk tunnels – and Zurich has built two, the most of any German-speaking country, since Berlin and Hamburg only have one of their trunk lines each in tunnel, the rest running above ground. The Mattstetten-Rothrist line likewise does not run at high speed, topping at 200 km/h, because doing so would raise the cost of rolling stock acquisition without benefiting the national integrated timetable – but it was an extensive undertaking for how small Switzerland is. Per capita, Switzerland has built far more intercity rail tunnels by length than France, and may even be ahead of Germany and Italy – and that’s without taking into account the freight base tunnels.
The issue of passenger experience
It’s best to think of organization-before-electronics-before-concrete as a maxim for optimizing user experience more than anything. The system’s passengers would prefer to avoid having to loiter 20 minutes at every connection; this is why one designs timed transfers, and not any attempt to keep the budget down. The Bahn 2000 investments were made in an environment of limited money, but money is always limited – there’s plenty of austerity at the local level in the US too, it just ends up canceling or curtailing useful projects while bad ones keep going on.
In Europe, Switzerland has the highest modal split for rail measured in passenger-km, 19.3%, as of 2018; in 2019, this amounted to 2,338 km per person. The importance of rail is more than this – commuters who use trains tend to travel by train shorter than commuters who use cars drive, since they make routine errand trips on foot at short distance, so the passenger-km modal split is best viewed as an approximation of the importance of intercity rail. Europe’s #2 and #3 are Austria (12.9%) and the Netherlands (11.2%), and both countries have their own integrated intercity rail networks. One does not get to scratch 20% with a design paradigm that is solely about minimizing costs. Switzerland also has low construction costs, but Spain has even lower construction costs and it wishes it had Switzerland’s intensity of rail usage.
Optimizing organization and electronics…
A country or region whose network is a mesh of lines, like Switzerland or the Netherlands, had better adopt the integrated timed transfer concept, to ensure people can get from anywhere to anywhere without undue waiting for a connecting train and without waiting for many hours for a direct train. This includes organizational reforms in the likely case there are overlapping jurisdictions with separate bus, urban rail, and intercity rail networks. Fares should be integrated so as to be mode-neutral and offer free transfers throughout the system, and schedules should be designed to maximize connectivity.
This should include targeted investments in systems and reliability. Some of these should be systemwide, like electrification and level boarding, but sometimes this means building something at a particular delay-prone location, such as a long single-track segment or a railway junction. In all cases, it should be in the context of relentlessly optimizing operations and systems in order to minimize costs, ensure trains spend the maximum amount of time running in revenue service and the minimum amount of time sitting at a yard collecting dust, reduce the required schedule padding, etc.
…leads to concrete
Systemwide optimization invariably shows seams in the system. When Switzerland designed the Bahn 2000 network, there was extensive optimization of everything, but at the end of the day, Zurich-Bern was going to be more than an hour, which would not fit any hourly clockface schedule. Thus the Mattstetten-Rohrist line was born, not out of desire to run trains as fast as possible, but because it was necessary for the trains to run at 200 km/h most of the way between Olten and Bern to fit in an hourly takt.
The same is true of speed and capacity improvements. A faster, more reliable system attracts more passengers, and soon enough, a line designed around a train every 15 minutes fills up and requires a train every 10 minutes, 7.5 minutes, 6 minutes, 5 minutes, 4 minutes. An optimized system that minimizes the need for urban tunneling soon generates so much ridership that the tunnels it aimed to avoid become valuable additions to the network.
The Munich S-Bahn, for example, was built around this kind of optimization, inventing many of the principles of coordinated planning in the process. It had a clockface schedule early, and was (I believe) the first system in the world designed around a regionwide takt. It was built to share tracks with intercity and freight trains on outer branches rather than on purely dedicated tracks as in the older Berlin and Hamburg systems, and some of its outermost portions are on single-track. It uses very short signaling blocks to fit 30 trains per hour through the central tunnel in each direction. And now it is so popular it needs a second tunnel, which it is building at very high cost; area activists invoked the organization before electronics before concrete principle to argue against it and in favor of a cheaper solution avoiding city center, but at the end of the day, Munich already optimized organization and electronics, and now is the time for concrete, and even if costs are higher than they should be by a factor of 2-3, the line is worth it.
Electronics before concrete, not instead of concrete
Switzerland is not going to build a French-style national high-speed rail network anytime soon. It has no reason to – at the distances typical of such a small country, the benefits of running at 300 km/h are not large. But this does not mean its rail network only uses legacy lines – on the contrary, it actively builds bypasses and new tunnels. Right now there are plans for an S-Bahn tunnel in Basel, and for an express tunnel from Zurich to Winterthur that was removed from Bahn 2000. The same is true of other European countries that are at or near the frontier of passenger rail technology. Even the Deutschlandtakt plan, compromised as it is by fiscal austerity, by high construction costs, by a pro-car transport minister, and by NIMBYs, includes a fair amount of new high-speed rail, including for example a mostly fast path from Berlin to Frankfurt.
When you plan your rail network well, you encourage more people to use it. When you optimize the schedules, fare integration, transfer experience, and equipment, you end up producing a system that will, in nearly every case, attract considerable numbers of riders. Concrete is the next step: build those S-Bahn tunnels, those express bypasses, those grade separations, those high-speed lines. Work on organization first, and when that is good enough, build electronics, and once you have both, build concrete to make maximum use of what you have.