In the last post, I criticized the idea of large-radius suburban circle, using the example of the Berlin Outer Ring, at radius 10-26 km from city center. In comments, Andrew in Ezo brought up a very good point, namely that Tokyo has a ring at that radius in the Musashino Line, and ridership there is healthy enough to fill a train every 10 minutes off-peak. Of course, the Musashino Line’s intersections with the main JR East lines, like Nishi-Kokubunji and Minami-Urawa, have the ridership of a city center station in Germany rather than that of a station 25 km out. So to discuss this further, let’s drop midsize cities like Berlin and look at an actually large city: New York. Consider the following possible circle in New York, at radius 20-25 km:
See full-size version here (warning: 55 MB).
Most of the radial extensions I’ve already discussed in previous posts – for example, here. Here these extensions go somewhat further in order to meet the ring, including at Newark Airport, on Staten Island, in Bay Ridge, at Floyd Bennett Park, in Canarsie, at Starrett City, near the Queens/Nassau County line, and in Yonkers.
The ring is 151 km, of which around 87 km would be above ground, mostly replacing highways like the Belt Parkway to reduce costs. Of note, this cannot be done adjacent to an extant highway – the fast car traffic deters nearby development, making transit-oriented development impossible. So key road links around the region have to go, which is fine, since people should be transitioning from driving to taking trains. With some additional elevated construction including through City Island, across the Long Island Sound, and in low-density parts of North Jersey where demolishing houses even at $1 million per unit is cheaper than tunneling, construction costs could be reduced further. But it’s still a $20-25 billion project at average world costs, maybe $15 billion at Nordic or Korean or Southern European or Turkish costs.
The only way to pay off the costs of such a line, not to mention to fill enough trains to support frequency that can take untimed transfers (at worst a train every 10 minutes), is to have very high ridership, on the order of 400,000-500,000 per day. This is for a line that misses Manhattan and all of the big secondary job centers, like Downtown Brooklyn and Long Island City. Is this plausible?
The answer is not an obvious no. Sufficiently aggressive TOD could plausibly create ridership. But it’s still questionable. There are really a few different forces pulling such a line in different directions:
- Using existing rights-of-way to reduce costs, hence the use of the Belt Parkway and not the denser development around Avenue U or even Flatlands.
- Serving secondary nodes like JFK, Coney Island, EWR, and Yonkers. Potentially it would be plausible to veer inward in New Jersey in order to hit Downtown Newark, at the cost of a few extra kilometers of tunnel, making the line radial from Newark’s perspective, whereas the line as depicted above is circumferential from Newark’s perspective since it goes around city center.
- The need to connect to radial subway and commuter rail lines, which means serving stations, opening plausible infill stations, and extending some lines toward the ring.
There are different ways to resolve this tension; the line I depicted is not the only one. For example, a higher-cost, higher-ridership version could veer inward in the Bronx and Queens, aiming to connect to Flushing and Jamaica and then replace the AirTrain JFK, leading to a ring of radius closer to 16 km than to 20-25.
I only bring this up to point out how many things have to work if you want such a ring to work out. Keeping costs to even semi-reasonable levels requires demolishing highways and engaging in aggressive TOD, which is only possible in an environment of total political victory over NIMBY and pro-car interests (note: these two are not the same!).
This is not the history of the Musashino Line. The Musashino Line originates in a freight bypass around the built-up area of Tokyo, which eventually turned into a circumferential passenger line. This is why it connects to the radial lines near but not at the busiest regional stations – at Nishi-Kokubunji and not Kokubunji, at Minami-Urawa and not Urawa, at Shin-Matsudo and not Matsudo or Kashiwa.
But even when the line is new, there are always compromises on right-of-way. Uncompromised right-of-ways are 100% possible, but not at 25 km radius, because the cost is too high to always go to the most important secondary centers. They happen when the radius is smaller, like Paris’s 8-10 km for M15, because then ridership can be high enough (M15 projects nearly a million riders a day). Farther away, ridership drops and costs rise because the line gets longer faster than per-km costs drop, so compromises are inevitable.
I am not proposing the ring above as a definitive crayon. I’m just mentioning it as something that highlights the difficulties of circumferential public transportation in the suburbs. Even as it is, the strongest segment of the ring is most likely the one in the city taking over the Belt Parkway, which could replace busy buses like the B15, B1, B3, B6, and B82. The suburban segments are weaker – there isn’t that much commuting across the Hudson that far north, and building up such commuting requires heavy commercial TOD in Yonkers, Mount Vernon, and New Rochelle.
In dense urban cores, it’s valuable to run circular rail lines. They connect dense near-center neighborhoods to one another without going through the more congested center, and help make transferring between parallel lines more efficient, again through avoiding central business district congestion. Some of the largest cities in the world even support multiple circles, line Lines 2 and 10 in Beijing, or the various overlapping circles of Moscow, Tokyo, and soon Paris. However, this system of radial lines through the center and circular lines around the center cannot go on forever. There is a limit to how far out one can build circles, which is much sharper than the limit of how far radial lines can go. Lower-density suburbs can have radial lines connecting them to city center or to near-center nodes of activity, but circumferential lines are likely to be weak.
For a concrete example, take Berlin. It has the Ring through fairly dense neighborhoods, supporting 5-minute frequency on the S-Bahn during most of the day. But it also has the Outer Ring, built in the 1950s through East Berlin and the Brandenburg suburbs to surround West Berlin and permit the construction of the Wall; today it runs regional trains, and one segment through East Berlin runs the S75 every 10 minutes, but there is no train making the entire orbit, just trains using short segments to position themselves to a better radial entry into the center of Berlin. It looks frustrating – there is circular infrastructure, why not use it? But there’s a solid reason not to run it as a true circle.
See map below:
A schematic of service patterns can be seen here.
The line’s origin as a bypass means it doesn’t serve any of the nodes near its radius, like Potsdam (too built-up), Spandau (in West Berlin), or Märkisches Viertel (also in West Berlin). The only node it does pass through is the soon-to-close Schönefeld airport, which only became important well into the Cold War; moreover, a branch parallel to the line to the southeast serves the soon-to-open Berlin-Brandenburg Airport, with plans to run many different kinds of regional services entering Berlin from both the Stadtbahn and the North-South Main Line. So a circular service would, by itself, just connect various outlying areas like Marzahn, Hennigsdorf, and Falkensee to the airport. By itself, this doesn’t support very high frequency.
Now, what the line could do is work as a network together with radial lines, connecting to them to facilitate travel not passing through the center of Berlin. However, there is not much point in transfers unless they are either high-frequency or timed. High-frequency transfers are out – the radial lines that penetrate the Outer Circle run 2-3 trains per hour. This forces the transfers to be timed.
Timed connections on lines that intersect crosswise rather than parallel with cross-platform transfers are completely possible. The trains can’t be too long, but that’s fine, a 4-car train with stair and elevator connections could have 2-3 minute transfer windows and still exchange passengers in all directions. It’s worth establishing at sufficiently important stations where a cross-platform transfer is not possible; as a four-way transfer, it’s not even that much more involved than a cross-platform transfer with timed wrong-direction transfers like Wittenbergplatz between U2 and U1/3. However, this is for one station.
All of this goes out the window when a circle intersects 12 different radial lines. Such a scheme can only work if all of the transfers are timed, or at least a large majority of them. Otherwise, people might as well take the train through the center and connect at Berlin Hauptbahnhof, or even stay on the same train if it runs through like RE 1 or RE 3.
In theory, you can time a short succession of transfers on the same line. All it really takes is to make sure that the circular line takes a half-integer multiple of the takt interval between every pair of transfer points, allowing both-direction transfers everywhere. On a few stretches of the line, it’s even plausible, with a 20-minute takt – the line would be fast because it’s so far out and has to few stops, so 7-10 km in 8 minutes (10 minus 2 for the transfer window) is not outside the realm of possibility.
Except that some segments between transfer points are still bad, like between the two just west of Spandau, or on both sides of the crossing with S5 and RE 1 in Lichtenberg. And even if they weren’t, this runs into the problem that trains are not infinitely punctual. Having 12 knots between a circular line and radials around Berlin, or even just 10 if weak ones are dropped, means that suburban Berlin would have more knots every 20 minutes than Switzerland has today every half hour (8), and not too many fewer than Switzerland is planned to have every half hour in the 2030s. The required schedule discipline is intense, especially in a big city defined by crowded rush hour trains.
This has implications elsewhere. Paris has its Grande Ceinture, which is tempting for a regional rail ring, but the frequency at which it can support a full RER line is not high; instead, the region is breaking the line into segments, to be turned over into tram-trains, with some segments diverging from the mainline to serve nodes near but not on the line.
In general, what this means is that if you’re not connecting to a major city center, there’s only so much service you can run. If you’re within the densely built-up area, as the Ring is or as the various orbitals Paris has (M2/M6, T3) or plans (M15), then it’s fine – untimed transfers are fine when trains come every 5 minutes, and overlapping one-seat rides like Prenzlauer Berg-Neukölln and Ostkreuz-Tempelhof and so on can help fill the train as well. But once frequency drops below about a train every 10 minutes, untimed transfers no longer work, which means that services that rely on connections only work if the connections are at a handful of key points, not at 12 different radii around the city.
Outside a city core with very high frequency of transit, say 8 minutes or better, bus and train services must be timetabled to meet each other with short connections as far as possible. Normally, this is done through setting up nodes at major suburban centers where trains and buses can all interchange. For example, see this post from six months ago about the TransitMatters proposal for trains between Boston and Worcester: on the hour every half hour, trains in both directions serve Framingham, which is the center for a small suburban bus system, and the buses should likewise run every half hour and meet with the trains in both directions.
This is a dendritic system, in which there is a clear hierarchy not just of buses and trains, but also of bus stops and train stations. Under the above system, every part of the Framingham area is connected by bus to the Framingham train station, and Framingham is then connected to the rest of Eastern New England via Downtown Boston. This is the easiest way to set up timed rail-bus connections: each individual rail line is planned around takt and symmetry such that the most important nodes can have easy timed bus connections, and then the buses are planned around the distinguished nodes.
However, there’s another way of doing this: a bus can connect two distinct nodes, on two different lines. The map I drew for a New England high- and low-speed rail has an orbital railroad doing this, connecting Providence, Worcester, and Fitchburg. Providence, as the second largest city center in New England, supplies such rail connections, including also a line going east toward Fall River and New Bedford, not depicted on the map as it requires extensive new construction in Downtown Providence, East Providence, and points east. But more commonly, a connection between two smaller nodes than Providence would be by bus.
The orbital bus is not easy to plan. It has to have timed connections at both ends, which imposes operational constraints on two distinct regional rail lines. To constrain planning even further, the bus itself has to work with its own takt – if it runs every half hour, it had better take an integer multiple of 15 minutes minus a short turnaround time to connect the two nodes.
It is also not common for two suburban stations on two distinct lines to lie on the same arterial road, at the correct distance from each other. For example, South Attleboro and Valley Falls are at a decent distance, if on the short side, but the route between them is circuitous and it would be far easier to try to set up a reverse-direction timed transfer at Central Falls for an all-rail route. The ideal distance for a 15-minute route is around 5-6 km; bus speeds in suburbia are fairly high when the buses run in straight lines, and if the density is so high that 5-6 km is too long for 15 minutes, then there’s probably enough density for much higher frequency than every half hour.
The upshot is that connections between two nodes are valuable, especially for people in the middle who then get easy service to two different rail lines, but uncommon. Brockton supplies a few, going west to Stoughton and east to Whitman and Abington. But the route to Stoughton is at 8.5 km a bit too long for 15 minutes – perhaps turning it into a 30-minute route, either with slightly longer connections or with a detour to Westgate (which the buses already take today), would be the most efficient. The routes to Whitman and Abington are 7 km long, which is feasible at the low density in between, but then timetabling the trains to set up knots at both Brockton and Abington/Whitman is not easy; Brockton is an easy node, but then since the Plymouth and Middleborough Lines are branches of the same system, their schedules are intertwined, and if Abington and Whitman are served 15 minutes away from Brockton then schedule constraints elsewhere lengthen turnaround times and require one additional trainset than if they are not nodes and buses can’t have timed connections at both ends.
Planners then have to keep looking for such orbital bus opportunities. There aren’t many, and there are many near-misses, but when they exist, they’re useful at creating an everywhere-to-everywhere network. It is even valuable to plan the trains accordingly provided other constraints are not violated, such as the above issue of the turnaround times on the Old Colony Lines.
I recently found myself involved in a discussion about Boston regional rail that involved a proposal to do more thorough regional rail-subway integration. Normally, S-Bahn systems mix some aspects of longer-range regional rail and some aspects of urban metro systems. They provide metro-like service in the urban core – for example, Berliners use the the three trunk lines of the S-Bahn as if they were U-Bahn lines. But, unlike proper metros, they branch in the suburbs and tend to have lower frequency and lower quality of infrastructure. However, there is a limit to this integration, coming from timetabling.
The characteristics of metro-like S-Bahn
When I call some S-Bahns, or some S-Bahn trunks, “metro-like,” what I mean is how users perceive them, and not how planners do. A metro line is one that users get on without concern for the timetable. It may run on a clockface schedule, for example on a 5-minute takt in Berlin, but passengers don’t try to time themselves to get on a specific train, and if the train is 1-2 minutes behind schedule then nobody really minds. This user behavior usually comes from high frequency. However, in New York, despite extensive branching and 10-minute frequencies, I classify the subway as fully metro-like because the trains are not dispatched as a scheduled railroad and even if they were, passengers don’t ever think in terms of “my Queens-bound N train arrives at :06 every 10 minutes.”
S-Bahn lines have trunks like this, but also branches that work like regional rail. The regional rail pattern in the sense of RegionalBahn is one in which passengers definitely look at timetables and try to make them, and connecting public transit lines are planned to make timed transfers. On lines branded as RegionalBahn service comes every half hour or every hour, and usually S-Bahn tails are every 15-30 minutes (occasionally 10), but the printed schedule is paramount either way; when I rode the RER B to IHES in the last three months of 2016, I memorized the 15-minute takt and timed myself to it.
The key aspect of S-Bahns is combining these two patterns. But this leads to a key observation: they have to interline a number of different service patterns, which requires planning infrastructure and service to permit both. They can’t run on pure headway management in the core, because the branches must be scheduled. But they have to use a timetabling system that permits high core frequency nonetheless.
Finally, observe that I am not discussing the type of equipment used. A subway train that extends far into the suburbs may qualify as regional rail – the Metropolitan line in London qualifies as an example on account of its highly branched service pattern in Metro-land. In the other direction, a train built to mainline standards that runs consistent service pattern with little to no branching at a range typical of metros is not, for the purpose of this issue, regional rail – examples include the Yamanote and Keihin-Tohoku Lines in Tokyo, which run identical trains to those that run deeper into suburbia but have literally no (Yamanote) or almost no (Keihin-Tohoku) variation in service patterns.
The limit of interlining
A large degree of interlining tends to reduce timetable reliability. Trains have to make junctions at specific times. This is compounded by a number of different factors:
1. Trunk throughput
The busier the trunk is, the harder it is to keep everything consistent. If you run 15 trains per half-hour, that’s 15 opportunities for a 2-minute delay to mess the order in which trains arrive, which has implications further down. If you run 4 trains per half-hour, that’s 4 opportunities, and a 2-minute delay is easily recoverable anyway.
2. Trunk length
Longer and more complex trunks introduce their own problems. If many passengers treat trains as interchangeable and don’t care what order they arrive in, then this may not be good for timekeeping – a slight delay on a branch may lead to grossly uneven headways on the trunk, which compound on busy metro lines for similar reasons as on buses. Berlin’s Stadtbahn has 14 stations from Ostkreuz to Westkreuz counting both, and this may make the branches with their 20-minute frequencies a little too difficult to fit together – evidently, peak throughput is 18 trains per hour, hardly the cutting edge. The RER A has 7 trunk stations from Vincennes to La Défense inclusive, and around 27 peak trains per hour.
3. Branch infrastructure quality
In the limit, the branches have to have excellent infrastructure quality, to be resilient to 1-2 minute delays. Timed meets on a mostly single-trunk line, routine on 15-minute branches like some lines in suburban Zurich and Tokyo, become dicey on lines that feed very busy trunks. Tokyo does this on the Yokosuka Line, which is far from the busiest (it peaks around 20 trains per hour) and Zurich on the right bank of Lake Zurich, which feeds into an S-Bahn trunk with 4 stations inclusive from Stadelhofen to Oerlikon. The busiest S-Bahn lines tend to have all-doubled outer ends.
4. One vs. two ends
If the line is single-ended, then inbound trains can just run metro-style in city center without regard for the printed schedule, use the terminal for schedule recovery, and then go outbound on schedule. Non-through-running lines are by definition single-ended, and this includes what I believe is Tokyo’s busiest regional rail line, the Chuo Rapid Line. But even some through-running lines are de facto single-ended if demand is highly asymmetric, like the Stadtbahn, which has far more demand from the east than from the west, so that one branch even turns at Westkreuz. Double-ended lines do not have this opportunity for recovery, so it’s more important to stay on schedule, especially if the end is not just busy but also has extensive branching itself.
I was recently asked about the issue of incrementalism in infrastructure, with specific reference to Strong Towns and its position against big projects (e.g. here). It’s useful to discuss this right now in context of calls for a big infrastructure-based federal jobs program in the United States. The fundamental question to answer is, what is the point of incremental projects?
The issue is that the legitimate reason to prefer less ambitious projects is money. If a new subway tunnel costs $5 billion, but you only have the ability to secure $1.5 billion, then you should build what you can for $1.5 billion, which may be a tram rather than a subway, or surface improvements to regional rail instead of a new regional rail tunnel, etc.
A secondary legitimate reason is that even if there is more money, sometimes you get better results out of building something less flashy. This is the electronics-before-concrete approach – in a developed country it’s almost always cheaper to invest in signaling, electrification, and platform upgrades than to build new tunnels. This can look incremental if it’s part of a broader program: for example, if there’s already investment in electrification in the region then extending wires is incremental, so that completing electrification on the commuter rail lines in New York, reopening closed suburban branches in Philadelphia with new wires, and even completing electrification in a mostly-wired country like Belgium and the Netherlands would count.
But the example of electrification in a mostly already electrified place showcases the differences between cost-effectiveness and incrementalism. The same investment – electrification – has a certain cost-effectiveness depending on how much train traffic there is. There’s a second-order effect in that the first line to be electrified incurs the extra cost of two train fleets and the last line has a negative cost in no longer needing two fleets, but this isn’t relevant to first order. Nonetheless, electrifying a system where electrification is already familiar is considered incremental, to the point that there were extensions of electrification in suburban New York in the 1980s and there remain semi-active projects to build more, whereas electrifying one that is currently entirely diesel, like Boston, is locally considered like a once-in-a-generation project.
And that is the real problem. American cities are hardly hotbeds of giant flashy construction. They barely are in highways – big highway construction plans are still done but in suburbs and not anywhere where public transit is even remotely relevant. And transit construction plans are always watered down with a lot of reconstruction and maintenance money; most of the money in the Los Angeles sales tax measures that are sold to the urbanist public as transit measures is not about rail construction, which is why with money programmed through 2060 the region is going to only have one full subway line; an extension of the Red Line on South Vermont is scheduled to open in 2067, partly because construction costs are high but mostly because there are maintenance projects ahead in line.
So in reality, there are two real reasons why incrementalism is so popular in the United States when it comes to transportation, neither of which is legitimate. Both are types of incompetence, but they focus on different aspects of it.
The first reason is incompetence through timidity. Building something new, e.g. rail electrification in Boston or in California, requires picking up new knowledge. The political appointees in charge of transit agencies and the sort of people who state legislators listen to do not care to learn new things, especially when the knowledge base for these things is outside their usual social networks. Can Massachusetts as a state electrify its rail network? Yes. Can it do so cheaply? Also yes. But can the governor’s political appointees do so? Absolutely not, they are incurious and even political people who are not beholden to the governor make excuses for why Massachusetts can’t do what Israel and Norway and New Zealand and Austria and Germany do.
In that sense, incrementalism does not mean prudence. It means doing what has been done before, because the political people are familiar with it. It may not work, but it empowers people who already have political clout rather than sidelining them in favor of politically independent technocrats from foreign countries who might be too successful.
The second reason is incompetence through lack of accountability. This is specific to an approach that a lot of American urbanists have backed, wrongly: fix-it-first, or in its more formal name state of good repair (SOGR). The urbanist emphasis on SOGR has three causes: first, in the 1980s New York had a critical maintenance backlog and neglected expansion in order to fix it, which led to positive outcomes in the 1990s and 2000s; second, in highways, fix-it-first is a good way to argue against future expansion while hiding one’s anti-car ideology behind a veneer of technical prudence; and third, Strong Towns’ specific use case is very small towns with serious issues of infrastructure maintenance costs and not enough residential or commercial demand to pay for them, which it then generalizes to places where there’s more market demand for growth.
In reality, the situation of 1980s’ New York was atypical. Subsequently, the SOGR program turned into a giant money pit, because here was an opportunity to spend enormous sums of capital construction money without ever being accountable to the public in the form of visible expansion. Ask for a new rail line and people will ask why it’s not open – California got egg on its collective face for not being able to build high-speed rail. Ask for SOGR and you’ll be able to brush away criticism by talking about hidden benefits to reliability. Many passengers may notice that trains are getting slower and less reliable but it’s easier in that case to intimidate the public with officious rhetoric that sounds moderate and reasonable.
Incrementalism is fundamentally a method of improving a legitimate institution. The EU needs incremental reform; China needs a democratic revolution. By the same token, in infrastructure, incrementalism should be pushed when, and only when, the status quo with tweaks is superior to the alternatives. (Note that this is not the same as electronics-before-concrete – what Switzerland did with its rail investment in the 1990s was very far-reaching, and had tangible benefits expressed in trip times, timed connections, and train frequency, unlike various American bus redesigns.) Strong Towns does not believe that there’s anything good about the American urban status quo, and yet it, and many urbanists, are so intimidated by things that happened in the 1950s, 60s, and early 70s that they keep pushing status quo and wondering why there is no public transportation outside about eight cities.
The Modernizing Rail (Un)Conference happened last Sunday. We’re still gathering all the materials, but here are video uploads, including the keynote by Michael Schabas.
We will also have slides as given by presenters who used them. But for now, here are the slides used by the keynote. You may notice that the recording does not begin on the first slide; we missed Schabas’s introduction and some remarks on his background, detailing his 40 years of experience designing public transit systems in a number of countries, mainly Britain and Canada but also elsewhere in the developed world.
My session on construction costs was slide-free (and was not recorded), since I mostly just showed people around our under-construction cost dataset and answered a lot of questions. Some of those questions were annoying, by which I mean they questioned my thinking or brought up a point I haven’t considered before. I am not talking too much about it partly because I was mostly (mostly) repeating things I’ve said here, and the full database should be out later this summer, with all the mistakes I’ve made in currency conversion rates and in not updating for cost overruns fixed.
After my breakout, I was uncertain between which of two sessions to attend – one on HSR-legacy rail compatibility by María Álvarez, and one on equity issues in rail planning, by Grecia White and Ben She. I ended up going to the latter, which featured interesting discussions of inclusion of low-income people and minorities, both as riders (that is, serving people who are not middle-class whites better on regional rail) and as workers (that is, diversifying planning and engineering departments).
It went well in that there was no monopolization of discussion by people who have more a comment than a question, or any open racism or sexism; but it was somewhat frustrating in that while there was a lot of productive discussion of racial equality in rail planning, there was very little of gender equality even though we did intend to talk about both; Grecia was specifically interested in discussing these, for example women’s perceptions of public safety. This is in line with conference demographics – the organizing team and the breakout presenters were each one-third people of color, in line with US demographics; but the organizing team had 2/18 active women and the presenters 3/15. TransitMatters is similar in that regard – racial diversity is comparable to that of the Boston region, and the proportion of regulars who are queer is enormous, but there are very few women.
Finally, I hosted a session on how to set up a transport association, a.k.a. Verkehrsverbund. Christof Spieler did the most talking, and German attendees explained a lot about the difference between a transport association and agency amalgamation. But for the most part that session felt like an ersatz conclusion to the entire conference; it technically lasted an hour, but once the hour had lapsed, people from other sessions came to the room and the conversation continued naturally, talking a bit about different transit planning issues in Germany and a bit about applicability to rail reform in the Northeastern US.
It’s been a while since I last wrote this series, where I covered the American, Soviet, and British traditions of building urban rail. I’d like to return by focusing attention on the French tradition, which has been influential not just within France itself but also to some extent former French colonies, especially Quebec.
An issue I hope to return to soon is the extent to which France has not truly decolonized; former French colonies in Africa, especially the Maghreb, rely on French technical expertise for construction, and often outsource their monetary policy (as with the CFA franc, but Morocco too has a peg to a dollar and euro mix). This matters, because this means the French way of building urban transit is influential in former French colonies in Africa, whereas the British tradition’s impact on India, Nigeria, and so on is limited.
The history of Paris
Like Britain, the USSR, and the US, France has a dominant financial center that its smaller cities aim to imitate. This imitation has been much more extensive than in the US and UK – to the extent that secondary French cities diverge in design principles from the capital, they do things that were fashionable in Paris at the time they built out their rail networks rather than things that were fashionable in Paris when Paris built the Métro. Thus, it is especially valuable to look at the history of urban rail in Paris.
The Paris Métro opened in 1900, as the world’s fifth metro system. Already then, it had a critical feature that the previous four (London, Budapest, Chicago, Glasgow) lacked: it was a centrally planned multi-line system. The city planned a coordinated system of what would become Lines 1-6, in the shape of a # in a circle: Lines 1 and 3 would run east-west, Lines 4 and 5 would run north-south, and Line 2, eventually split into Lines 2 and 6, would run the trace of the wall that delineated the city’s pre-1860 boundary.
The Métro was a municipal effort run by the municipal CMP, designed around the city’s needs, which included not just good transportation but also separation from the working-class suburbs. Whereas the London Underground was mostly technologically compatible with the mainline system, the Métro was deliberately designed not to be, to protect the urban middle class from transport integration with the suburban poor. This led to the following features:
- The trains are extremely narrow, 2.4-2.44 meters wide, compared with about 2.9 m on the mainline; the deep Tube trains in London, held to have the narrowest loading gauge on a standard-gauge railway, are 2.68 m wide.
- The interstation distance is very short, 562 meters on average. Paris is compact and dense and the short interstations are only a real problem in the suburbs.
- The trains run on the right, like French road traffic, whereas French trains run on the left.
- No legacy lines were incorporated into the system, unlike in New York and London, and thus the shape of the network looks much more like how one would design a metro network from scratch and less like how old West London branches or Brooklyn excursion lines looked.
Like New York and Berlin and unlike London, Paris built the Métro cut-and-cover. The lines built before the 1990s all closely follow streets except when they cross the river – and in the 1900s the Line 4 river crossing was the hardest part of the system to build, opening in 1908 whereas the rest of the network had opened by 1906. This was done entirely by hand, forcing the lines to curve where the streets did, which led to two notable warts. First, while most of the system had a design standard of 60 meter curve radii, Line 1 goes down to 40 at Bastille. And second, Line 5, which crosses the Seine on a bridge, cannot serve Gare de Lyon; the engineers could not get it to curve that way while still running through to Gare d’Austerlitz and the Left Bank, so instead the transfer point between Lines 1 and 5 is Bastille, and more recently the RER A and Line 14 both cross Line 5 without a transfer as they run express from Gare de Lyon to Châtelet.
That said, the missed connection between Lines 5 and 14 is the only one in the system, though two more are under construction on Line 14 extensions. Only one among the major metro systems of the world runs entirely without missed connections, the Mexico City Metro, which has unusually low line density in the core and unusually many tangential lines.
The suburbs and the RER
The Métro’s deliberate exclusion of the suburbs made sense from the point of view of a middle-class Parisian in 1900 who was mortally afraid of the working class. But by the 1930s, it was leading to serious design constraints. Further Métro extensions both densified the network and extended it outward, and in the 1930s, lines began to extend past city limits, to such suburbs as Lilas, Issy, Neuilly, and Montreuil. The short interstations made longer extensions infeasible, and some solution involving regional rail was needed.
In 1938, CMP bought and electrified the Ligne de Sceaux, which alone among the Paris commuter lines had reached close to city center, terminating at Jardin du Luxembourg rather than at the farther away rail stations, which are located at or just inside the M2/M6 ring. Then after the war, as suburbanization intensified and commuter traffic at Gare Saint-Lazare grew increasingly congested, CMP’s successor RATP collaborated with SNCF on connecting regional rail branches to form an express system, that is the RER; the Ligne de Sceaux became the southern half of the RER B, while a similar branch going east paired with one of the Saint-Lazare lines to form the RER A. Through-service opened in 1977, roughly at the same time as the German S-Bahn through-tunnels, but the system grew much larger as Paris was and remains far larger than any German city.
But it is not exactly correct to view the RER as identical to a German S-Bahn, or to one of the RER’s inspirations, the Tokyo through-running system. A number of features characterize it, some shared with other urban regional rail systems, some not:
- There are multiple trunk lines through the city, which form something like a coherent network among themselves, and do not share rolling stock. The biggest warts are that the RER B and D share tracks (but no platforms) on one interstation, and that the RER C mostly stays on the Left Bank, legacy of when planning in Paris conceived of the area around Saint-Michel as a central area to be served, where in reality it is decidedly secondary to the CBD stretching from Les Halles to Champs-Elysées.
- It runs largely, though not entirely, on separate tracks from non-RER lines.
- It is locally viewed as deficient to Métro service – researchers who use the RER B to get to IHES think of it as lower-quality, lower-class service than the Métro in the city and its immediate suburbs. I suspect that this is why Grand Paris Express is designed around Métro standards rather than as intensification of RER service, while RER expansion has fallen to the wayside.
- RER-Métro integration is imperfect: the fares are integrated but there are still barriers between RER and Métro platforms, and there are many missed RER-Métro connections, whereas in Berlin the S-Bahn and U-Bahn have only one missed connection between them.
- The interstation is around 2-3 km, but it’s actually slightly longer on the new urban tunnels build for the RER A, B, D, and E than on the legacy lines in the inner suburbs; this feature also exists in a much more extreme form in the United States, but in Berlin and Tokyo it is completely absent.
Exporting Parisian ideas
Parisian metro planning influenced Montreal, Mexico City, and the smaller French cities, in chronological order. We see any of the following features in those cities:
- Rubber-tired metros. This technology was in vogue in postwar Paris, which converted Lines 1, 4, and 11 to it figuring this was just better than steel wheels, and also Line 6, figuring that an elevated line would benefit from a quieter propulsion system.
- Non-radial network design. London and the systems inspired by it, including all Eastern bloc systems, have radial design, with nearly all lines entering a relatively small city center. Paris expanded its #-in-a-circle system to a combination of a radial network and a grid, with a large number of pairs of parallel lines. Mexico City, the largest system inspired by Paris, is rich in tangential lines but has only three lines serving city center, which are by far the three busiest.
- Short interstations, though this is truer domestically than in Montreal and Mexico City.
- Driverless operations. This technology became popular in the 1980s, starting with the Lille Metro, and France has used it on new lines in Paris (M14) and elsewhere (Lyon Line D, both lines in Toulouse), also innovating in converting manual lines to automatic on Paris M1 and now M4. While the Parisian lines are full-size metro lines, the other ones are light metro running shorter vehicles, often with extensive elevated service.
- Separation between regional rail and metro service. Montreal is sufficiently North American to have given up on regional rail entirely, but Lyon and Marseille are investing in better regional rail, run separately from the local urban transit system but with some degree of integration.
- Light rail. France’s modern light rail systems do not originate in Paris – Nantes opened its system in 1985, suburban Paris only in 1992 – but Paris has a notable feature that isn’t common elsewhere in Western Europe: it is a mixed system with some Métro lines and some tram lines filling in the gaps. This mixed system is also present in Lyon, Marseille, and Toulouse, whereas Bordeaux, Strasbourg, and Nice have entirely tram-centric systems. But in no case is there any subway-surface running as in the United States or Germany: lines are either clearly trams or clearly metros, rather than mixtures, and it is the system that is mixed, not the individual line.
Has France decolonized?
Like Britain, France did not take its geopolitical disempowerment at the end of World War Two easily. Both countries have maintained superpower pretensions, decolonizing but trying to treat their former colonies as their spheres of influence as much as possible. In Britain, this relationship broke down – the ex-colonies were being too loud in the Commonwealth, leading the country to seek to join the EU instead. In France, this relationship remains in Africa, and notable not in Southeast Asia, where Vietnam is buildings its urban rail networks with Chinese and Japanese financing.
But France is not just providing financing to infrastructure projects in its former (or current?) African colonies. It has a permanent presence. In researching Arab rail infrastructure, Anan Maalouf has noted that Alstom has had a subsidiary operating in Algeria since 2002, which does not exist elsewhere in the Arab world. This way, French firms maintain close knowledge of the situation in the Maghreb, where incomes and productivity levels are much lower than in France, so that different methods are optimal from those common in rich countries.
Nonetheless, what they build remains noticeably French. For example, the Sfax tramway does not look too different from what Bordeaux or Nice has. The Tunis Métro looks rather like a French tramway system too, despite the name; of note, even though the Tunis Métro branches, and has some underground segments, those segments are not on line trunks and thus the system does not form a subway-surface or Stadtbahn network.
I haven’t gone too much into intercity rail, but it is worth mentioning that Morocco has a high-speed rail system, built with French technical assistance and running TGV equipment.
Does this work?
Yes and no.
The Paris system works. It is not perfect, and in particular the integration between the Métro and the RER could be better; at least one tram line should be a full metro line (a completed T3 ring), and suburban extensions should generally use the RER, with more investment in RER capacity within the city as well. That said, public transport usage is higher in Paris than in its closest comparison, that is London; Paris’s system is also superior in both overall usage and future prospects to that of another megacity in Europe, Moscow. Only Istanbul could potentially do better in the future, in the context of extremely low construction costs.
That said, Paris is a giant that casts a long shadow, which doesn’t always work well for secondary cities. Lyon, Marseille, Toulouse, and the other secondary French cities aren’t too different in modal split from similar-size British cities, and are behind Vancouver, a North American city with extensive postwar growth. German cities in the Lyon size class do a lot better. See for example data here and here.
The weird features of France, like the love for rubber tires, are not that relevant overall, but do point out that France is relatively insular, and mostly adopts domestic ideas developed in Paris rather than ideas from elsewhere in Europe, let alone Asia. (Yes, I know about Japanese influence on the initial RER; however, there have been 50 years of divergence since, same as with German tram-trains and American light rail.) This has been especially problematic with regional rail. France does not have frequent takts anywhere – even Paris only has takt timetables off-peak, running a separate schedule at rush hour, whereas the German takt plan is repeated throughout the day and the peak can only have supplemental service.
The issue is that Paris does not need to think in terms of repeating schedules, because it is so big that the RER trunks run every 5 minutes off-peak. It thinks of the RER as mostly separate trunk lines with dedicated fleets, because the primary problem is train capacity through city center. In Lyon, let alone smaller cities, this is not the main issue. There do exist a handful of individual lines running an off-peak takt elsewhere in France, but integration with urban rail remains imperfect and a comparison with Vienna, Copenhagen, Zurich, Stuttgart, and Hamburg would not be favorable. It matters that, like Britain, France has such a dominant capital that it doesn’t know how to scale down to provide rail service in a metropolitan area where if the transfers aren’t perfectly timed, people won’t ride.
On Sunday the 12th of July, a few of us public transit activists are going to hold a conference online called Modernizing Rail, focusing on better service and integration in the Northeastern United States. Our keynote speaker will be Vukan Vuchic, the Serbian-American UPenn transportation professor who imported German rail modernization schemas from the 1970s, including the concept of regional rail; he will speak about the history of this in the context of SEPTA, which built much of the S-Bahn infrastructure (e.g. S-Bahn through-running tunnel) but has not done many other important things such as fare integration and coordinated planning with urban transit.
Update 2020-07-04: due to a family health emergency, Vuchic cannot make it. Therefore we will have an alternate keynote address by Michael Schabas, entitled Using Business Case Analysis to Design Better Railways.
Schabas has been finding ways to make railways deliver more and cost less for 40 years, shaping urban, intercity, and high speed rail projects in Canada, England, and the USA, and operating passenger and freight railways in England and Australia. He is the author of The Railway Metropolis – how planners, politicians and developers shaped Modern London. Since 2014 he has been advising Toronto’s Metrolinx on the $20 billion upgrading and electrification of the GO Rail system, and the $28.5 billion expansion of Toronto’s subway system. Michael is a Partner in FCP, a rail strategy boutique based in the UK advising clients on rail developments and projects around the world
The keynote will be between 11 am and noon Eastern time.
After the keynote, we will hold unconference-style sessions. For people who have not seen this style before, this means that we solicit ideas from the entire body of attendees for breakout sessions, and then by consensus, depending on the number of attendees and what they are interested in, split into rooms for further discussion of the selected topics. There will be three slots for breakouts: 1-2, 2:15-3:15, 3:30-4:30 pm, all Eastern time; the number of breakouts will depend greatly on the number of attendees, which at this point we are uncertain about. The breakouts may include pure discussions or presentations, and we also solicit expressions of interest in presenting if there’s an issue you have particular interest and expertise in.
There will be more information available on social media, but to register, please complete this form. You can create an account on Journey for this if you’d like, in which case you can save your progress and come back later, but this is not a long form and you can complete it in one go without registration.
The conference will be held on Zoom, with link emailed shortly before the event takes place.
Update 2020-07-11: here is the timetable. Email us at email@example.com for the Zoom password if you’ve registered.
Since the 1960s, Paris has gradually built itself to have a 5-line regional rail network connecting the city and its suburbs, with more than a billion riders a year. Unfortunately, investment has been slow in the last 20 years; the fifth line, the RER E, is being extended to the west, but other problems are not being fixed through more investment. Some regional rail lines remain disconnected from the system, including one of the city’s six intercity rail terminals, Gare Montparnasse. While east-west capacity is being augmented through the RER E extension, north-south traffic is jammed and yet is not slated to receive any relief, despite past studies.
Taking everything together, this is what Paris needs to do to complete the conversion of all commuter rail in Ile-de-France to RER standards:
Full-size image can be seen here; warning: 71 MB.
Dashed lines are new tunnels to be built. Most of the dashed green line is the under-construction RER E extension from Saint-Lazare to La Défense and points west. The remainder, between Les Halles and Auber/Saint-Lazare, is a new tunnel that should be built, giving away the extension to the RER D instead. With a full line extended, the RER D could take over the entire SNCF-run part of the RER A while also continuing west to Mantes-la-Jolie as is planned for the RER E extension, so the RER A can gain the Transilien L branches to the southwest with a short curve from La Défense to Puteaux.
In addition, what is now the shared RER B and D tunnel between Gare du Nord and Les Halles should be four-tracked; the stations at both ends thankfully already have separate platform tracks for the RER B and D, and in 2003 a somewhat disruptive plan to four-track the tunnel was estimated to cost €700 million. Since the RER D tracks are to continue west, the new dual track tunnel should continue south across the river and connect to Montparnasse, creating the RER F; the RER F should take over the current northern branches of the RER B to form a southwest-northeast line, while the current southern branches of the RER B should be connected to what are now the northern branch of the RER D and the branches of Transilien H.
The RER C and E should be broken and recombined using a short four-track tunnel across the river, creating northeast-southwest and northwest-southeast trunk lines. Today, the RER C misses the Paris CBD and has an awkward connection to the RER A; with this recombination, the connections would still require a lot of walking at transfer stations but they’d exist and passengers would get solid two-seat rides.
Finally, a handful of outer-urban and suburban fixes would be useful: a few infill stations, depicted with gray filling; using all four tracks on the RER F trunk line to Aulnay (currently the RER B) to make it easier to run express trains to Charles-de-Gaulle; building a short suburban tunnel through Chaville to connect the RER F and A branch; continuing T3 to form a full circle using the Petite Ceinture in lieu of the awkward RER C branch today; constructing an infill station at the RER E/F junction in Meudon.
Excluding the ongoing RER E extension, the total length of new tunnel in city center is 8 km of two-track tunnel and about 1.5 km of four-track tunnel. This would set non-Anglosphere world records in construction costs per kilometer, just as the RER A did; costs in the €400-500 million per km can be expected given the complexity of tunneling under so many older Métro lines, so the system would cost around €5 billion, perhaps reaching €6-7 billion with the extra suburban tunnels and infill stations.
The map doesn’t go to the edge of Ile-de-France, or else it would be even bigger, but the plan should be to connect every Transilien line to this system, even ones in faraway exurbs. Frequency to the exurbs need not be very high – today they get hourly service off-peak, and half-hourly service in the future should be plenty to small towns on the edge of Ile-de-France; of course, closer-in suburbs as well as major secondary centers like Meaux and Evry should get much higher frequency, and the trunks should get a train every 3-4 minutes even off-peak.
The point of this exercise is that Paris has already done the hardest parts. The RER A and B exist, and Châtelet-Les Halles was dug at enormous expense in the 1970s. Even at the high per-km costs of connections underneath the center of Paris, the tunnels Paris needs to build in the next 10-15 years are low-hanging fruits for completing the project of connecting the entire region through a unified RER network.
Remember the Ohio Hub? Back in 2009-10, Ohio was planning on running five low-speed trains per day between Cleveland and Cincinnati and branded this exercise as high-speed rail called the Ohio Hub. The Republican victory in the gubernatorial election put it out of its misery (as unfortunately happened to the far better Florida project), but the idea of little facts-on-the-ground kinds of rail investment persists among American advocates who don’t understand how rail operations work. Now that there’s serious talk of infrastructure funding in the United States as part of a stimulus package, I’d like to explain, to prevent the debacles of the late 2000s from happening again.
The central conceit is that public transportation is not cars. It’s a different, more complex system. The road network has fewer moving parts – one just builds roads based on traffic projections. Public transportation has schedules, transfers, and equipment, all of which must be planned in coordination. “This junction gets congested, let’s build a bypass” works for road advocacy, but fails for rail, because maybe speeding up the trains by a few minutes doesn’t really help get to any timed connections and is therefore of limited value to the system.
Rail works when everything is planned together. This makes little additions not too valuable: a small speedup may not be useful if connecting lines stay the same, infrastructure investment may have limited effect on trip times if the rolling stock doesn’t change, etc.
The upshot is that it’s very easy to find 80/20 problems: 80% of the money gets you 20% of the benefits. In addition to examples of lack of coordination between infrastructure, the timetable, and rolling stock, there are issues with insufficient frequency. When frequency is low relative to trip time, the long-term elasticity of ridership with respect to service is more than 1 – that is, running more service makes the trains and buses fuller, as better service encourages more ridership. Thus, service with insufficient frequency will fail, trains and buses getting too little ridership to justify additional investment, whereas if initial frequency were higher from the start then it would succeed.
The Ohio Hub was one such example: five roundtrips a day, starter service. It makes sense to someone who thinks like a manager or a general-purpose activist: start small and build from there. But to someone who thinks like a public transportation planner, it’s a disaster. Already 10 years ago, Max Wyss in comments was warning that such service would fail – the original Intercity brand in Germany succeeded by running trains every two hours, with hourly service on stronger city pairs, often with timed transfers at junctions.
Regional rail projects suffer from a similar urge to start small. Peak-only service will invariably fail – the operating costs will be too high for ridership even if almost all seats fill. This covers just about every American effort at starting up new commuter rail service.
More fundamentally, the issue is that nobody likes failure. Insufficient, poorly-optimized service creates facts on the ground, but these facts don’t lead to any effort toward better service if people perceive what has been built to be a failure. If a handful of trains per day that average 70 km/h are called high-speed rail, then it doesn’t lead passengers to want high-speed rail; it leads them to avoid the train and conclude that high-speed rail is slower than driving on the freeway.
The passengers on such service may not be a great constituency for better service, either. If the train is very slow, then the riders will be the sort of people who are okay with slow trains. Older American railfans are filled with nostalgia for traditional railroading and openly say that slower is better. Such people are not going to advocate for modern high-speed rail, nor for learning from successful Asian and European examples.
Another group of people who ride trains and often advocate against better service is peak commuters on trains serving high-income suburbs. They are used to an adversarial relationship with the state; to them, the state taxes them to give money to poorer people, and they instead prefer hyper-local forms of government providing segregated schools and policing. Representatives of such riders engage in agency turf warfare, such as when state senators from Long Island opposed Metro-North’s Penn Station Access because it would use train slots into Penn Station that the LIRR believes are its property. On social media, people sporadically yell at me when I propose fare integration, on grounds that boil down to viewing any urban riders who would be attracted to lower fares as interlopers.
There’s an ultimate proof-of-pudding issue here. Americans have to a good approximation never seen a working public transportation system. At best, they’ve seen a megacity where people use the trains even though they are dirty and expensive to run because there is no alternative and construction was done 100 years ago when costs were lower. There is no coordinated planning; Americans do not demand it because only a handful of people know what it is, who are often young and have often lived abroad for an extended period of time, both of which make one less likely to be listened to in politics.
The result is that the sort of bottom-up activism people are used to is not useful in this context. In Germany it’s different – enough people have seen what works in Austria, Switzerland, and the Netherlands and know what to call for. But in the United States, it won’t work – the knowledge base of how to build reliable, interconnected public transportation exists but is too thinly spread and is the domain of people who do not have much political prestige.
It’s critical to then get things right from the start. Do not assume future activism will fix things. Half-measures are much more likely to lead to disillusionment than to any serious efforts to improve things to turn them into full measures. If the choice is between a high chance of bad service and low chance of good service, don’t settle for bad service and make a gamble for good service; bad public transportation is a waste of money and the general public will correctly perceive it as such.