At a meeting with other TransitMatters people, I had to explain various distinctions in what is called in American parlance regional rail or commuter rail. A few months ago I wrote about the distinction between S-Bahn and RegionalBahn, but made it clear that this distinction was about two different things: S-Bahns are shorter-distance and more urban than RegionalBahns, but they’re also more about service in a contiguous built-up area whereas RegionalBahns have the characteristics of interregional service. In this post I’d like to explore the different travel markets for regional rail not as a single spectrum between urban and long-range service, but rather as two distinct factors, one about urbanity or distance and one about whether the line connects independent centers (“interregional”) or a monocentric urban blob (“intraregional”).
This distinction represents a two-dimensional spectrum, but for simplicity, let’s start with a 2*2 table, so ubiquitous from the world of consulting:
|Connection \ Range||Short||Long|
|Intraregional||Urban rail, S-Bahn||Big-city suburban rail|
|Interregional||Polycentric regional rail||RegionalBahn|
The notions of mono- and polycentricity are relative. Downtown Providence, Newark, and San Jose all have around 60,000 jobs in 5 km^2. But Caltrain and the Providence Line are both firmly in the RegionalBahn category, the other end being Downtown San Francisco or Boston, 70-80 km away with 300,000-400,000 jobs in 5-6 km^2. Newark, in an essentially contiguous urban area with New York, 16 km from Midtown and its 1.2 million jobs in 6 km^2, is relatively weaker and does not fit into the interregional category; a New York-Newark line is an S-Bahn.
On the 2*2 table, the appellations “big-city” and “polycentric” are necessary. This is because longer-range rail lines are likelier to get out of the city and its immediate suburbs and connect to independent urban centers. Exceptions mostly concern the size of the primary urban cluster. If it is large, like New York, it can cast a shadow for tens of kilometers in each direction: commuter volumes are high from deep into Long Island, as far up the Northeast Corridor as Westport, as far up the Hudson as northern Westchester, and so on. In Paris, I wouldn’t be comfortable describing any of the RER and Transilien lines as RegionalBahn. In London, the closest independent cities of reasonable size are Cambridge, Brighton, Oxford, and Portsmouth, the first two about 80 km away and the last two about 100.
Tokyo, about as big as New York and London combined, casts an even longer shadow. In my post on S-Bahns and RegionalBahns I called some of its outer regional rail branches RegionalBahn, giving the examples like the Chuo Line past Tachikawa. But even that line is not really interregional in any meaningful way. It stays within the Tokyo prefecture as far as Takao, 53 km from Tokyo Station, and commuter service continues until Otsuki at kp 88, but everything along the line is bedroom communities for Tokyo or outright rural. The branching and short-turns at Tachikawa mean that the Chuo Line through Tachikawa is a long S-Bahn, and past Tachikawa is really a suburban commuter line too long to be an S-Bahn but too monocentric and peaky to be Regionalbahn (the peak-to-base frequency ratio is about 2:1, whereas German RegionalBahn is more commonly 1:1).
At the other end, we can have regional rail that is short-range but connects two distinct centers. This occurs when relatively small cities are in proximity to each other. In a modern first-world economy, these cities would form a polycentric region, like the Rhine-Ruhr or Randstad. Smaller regions with these characteristics include the Research Triangle, where relatively equal-size Raleigh and Durham are 40 rail kilometers apart, and Nord, where Lille is 30-50 km from cities like Douai and Valenciennes. This may even occur in a region with a strong primary center, if the secondary center is strong enough, as is the case for Winterthur, 28 km from Zurich, which has Switzerland’s fourth highest rail ridership.
Size is measured in kilometers, not people. Stockholm is a medium-size city region, but Stockholm-Uppsala is firmly within RegionalBahn territory, as the two cities are 66 km apart. Randstad’s major cities are all closer to each other – Amsterdam-Rotterdam is about 60 km – and that’s a region of 8 million, not 3 million like Stockholm and the remainder of Uppland and Södermanland.
The issue of frequency
The importance of the 2*2 table is that distance and urban contiguity have opposite effects on frequency: high frequency is more important on short lines than on long lines, and matching off-peak frequency to peak frequency is more important on interregional than intraregional lines.
Jarrett Walker likes to say that frequency is freedom, but what frequency counts as freedom depends on how long passengers are expected to travel on the line. Frequency matters insofar as it affects door-to-door travel time including wait time, so it really ought to be measured as a fraction of in-vehicle travel time rather than as an absolute number. An urban bus with an average passenger trip time of 15 minutes should run every 5 minutes or not much longer; if it runs every half hour, it might as well not exist, unless it exists for timed connections to longer-range destinations. But an intercity rail line where major cities are 2 hours apart can easily run every half hour or even every hour.
The effect of regional contiguity is more subtle. The issue here is that an intraregional line is likely to be used mostly by commuters at the less dense end. The effect of distance can obscure this, but within a large urban area, a 45-minute train will be full of commuters traveling to the primary city in the morning and back to the suburbs in the afternoon or evening; the same train between two distinct cities, like Boston and Providence, will not have so many commuters. In contrast, the same 45-minute trip will get much more reverse-commute travel and slightly more non-commute travel if it connects two distinct cities, because the secondary city is likelier to have destinations that attract travelers.
In no case are the extreme peak-to-base ratios of American commuter lines justifiable. Lines with tidal commuter flows can run 2:1 peak-to-base ratios, as is common in Tokyo, but much larger ratios waste capacity. The marginal cost of service between the morning and afternoon peaks is so low until it matches peak service that having less midday than peak service at all is only justifiable in very peaky environments. The 45-minute suburbs of New York, Tokyo, and other huge cities can all live with a 2:1 ratio, but other lines should have lower ratios, and interregional lines should have a 1:1 ratio.
The implication is that just as peak-to-base ratios going as high as 2:1 are acceptable for long-range intraregional lines, short-range interregional lines must run a constant, high frequency all day. I would groan at the thought of even half-hourly frequency on a 40-km interregional line; the worst I’m comfortable with is 15-20 minutes all day. Of note, such lines are necessarily pretty fast, since by assumption they make few intermediate stops to speed up travel between the two main cities – if there are significant cities in the middle then the lines connect even shorter-range cities and should be even more frequent.
Urban, suburban, intercity
Individual lines may have the characteristics of multiple variants of regional rail. They pass through urban neighborhoods on their way to outlying areas, which may be suburbs or independent cities; they may also pass through multiple kinds of independent areas.
In practice, in big cities this leads to three tiers on the same line: urban at the inner end, suburban at the middle end, interregional at the outer end. Inversions, in which there are independent cities and then suburbs, are possible but extremely rare – I can’t think of any in Paris, London, or New York, and arguably only three in Tokyo (Chiba, Saitama, Yokohama); fundamentally, if there are suburbs of the primary city beyond your municipality, then your municipality is likely to itself be popular as a suburb of the primary city.
That regional lines have these three tiers of demand type does not mean that every single regional line does. Some lines don’t reach any significant independent city. Some don’t usefully serve close-in urban areas – for example, the Providence Line barely serves anything urban, since the stop spacing is wide in order to speed up travel to high-demand suburbs and to Providence and the closest-in urban neighborhoods have Orange Line subway service. In rare cases, the suburban tier may be skipped, because there just isn’t much tidal suburban commuter ridership; in Boston, the Newburyport Line is an example, since its inner area has unbroken working-class urban development almost all the way to Salem, and then there’s almost nothing between Salem and Newburyport.
This does not mean that suburbs are always in between urban areas and independent cities – this is just a specific feature of large metropolitan areas. In smaller ones, the middle tier between urban and long-range interregional service is occupied by short-range interregional service rather than suburban commuter rail. Skipping the suburban tier, which is rare enough in large cities that in the cities I think about most often the only example I can come up with is the Newburyport Line, is thus completely normal in smaller cities.
There are common best practices for commuter rail: electrification, level boarding, frequent clockface schedules, timed transfers, fare integration, proof of payment fare collection.
However, high frequency means different things on lines of different characteristics. An interregional line should be running consistent all-day frequency, and if it is long enough could make do with half-hourly trains with timed connections to suburban buses; an urban line should be running every few minutes as if it were a metro line. Regional rail lines with characteristics off the main diagonal of the S-Bahn to RegionalBahn spectrum have different needs – suburban lines can have high peak frequency to reduce road congestion, although they should still have useful off-peak frequency; short-range interregional lines should run every 10-20 minutes all day.
The distinctions between intraregional and interregional lines and between short- and long-range lines may also affect other aspects of planning: station spacing, connections to local surface transit, connections at the city center end, through-running, etc. Even when the best industry practices are the same in all cases, the relative importance of different aspects may change, which changes what is worth spending the most money on.
Since an individual line can serve multiple markets on its way from city center to a faraway outlying terminal, it may be useful to set up a timetable that works for all of these markets and their differing needs. For example, urban lines need higher frequency than suburban and interregional ones, so a regional line with significant urban service should either branch or run short-turn trains to beef up short-range frequency. If there is a suburban area in the middle with demand for high peak frequency but also a secondary city at the outer end, it may be useful to give the entire line high all-day frequency, overserving the line off-peak just because the cost of service is low.
Ultimately, regional rail is about using mainline rail to fulfill multiple functions; understanding how these functions works is critical for good public transportation.
More than a year ago, I compared Los Angeles with a number of other large American cities. I brought up issues of public transportation ridership, city center job concentration, and income differences, as in the Los Angeles region people who commute by public transit average barely half the earnings of people who drive alone. One of the things noted in that post is that in the secondary transit cities of the United States – Chicago, San Francisco, Washington, Boston – people who commute by transit outearn people who do not. I didn’t delve deeply into that issue in that post, but in this post I will, because it showcases a serious problem in all four cities. New York lacks this pattern as of 2017 – solo drivers outearn transit commuters, though by a small and declining margin, so by 2020 it may join the secondary cities.
The reason this is a problem is that in none of these cities is public transportation so good as to be a luxury good. Rather, the issue is that public transportation is mostly an option for people traveling to city center, where incomes are higher. Crosstown public transportation options are weak – there is rarely direct rapid transit, and transfer trips are inconvenient. There may also be a peak vs. off-peak artifact, but I have no data confirming that richer Americans are likely to commute at rush hour, when transit frequencies are higher and congestion is worse.
Income by mode of transportation to work
From the 2017 American Community Survey, we can grab data about median earnings for workers by their main mode of travel to work:
|Metro area||Workers||PT mode share||PT income||Solo driver income||Median income|
Other modes exist too, most notably carpooling, which has lower median incomes than both solo driving and public transport in all of the cities in the table. Also of note, public transportation user income is more polarized – even though the median is comparable to and usually even higher than the overall median, the poverty rate for transit commuters is higher than the general rate everywhere except in San Francisco, where the poverty rates are within the margin of error.
Car ownership increases with income. In Singapore, the highest-transit use city for which I have this data, the overall mode share is 58.7%, which splits as low 60s for roughly the bottom half of the income distribution and then less in higher categories, bottoming at 43% in the highest income category, covering the top 15%. It’s really weird that in American cities with public transportation we see the opposite pattern – transit usage is higher in higher income brackets.
The explanation has to be about where people work. OnTheMap doesn’t have great income data, but we can still compare the proportion of workers in the highest income category, which is $3,333/month. I’ve used different definitions of city center in different blog posts: the one about Los Angeles used a restricted one, just a few blocks by a few blocks, covering a single-digit percent of the region, whereas more recently I’ve made 100 km^2 blobs, covering one third of workers in some cities, to maintain comparability with Paris. For this post’s purposes, I’m going to use a definition around the center of a radial transit network (as in the LA post), as well as a looser definition corresponding to something like city limits; in Washington and New York the restricted definitions are somewhat looser to take into account the spread of the subway network just outside city center, but in Chicago and San Francisco the LA post’s definition is apt.
|New York||9,408,498||52.1%||City proper||4,367,781||55.4%||South of 60th||2,098,740||65.7%|
|Chicago||4,604,044||47.9%||City proper||1,373,969||53.2%||LA post||401,169||71.7%|
|Washington||2,830,896||55%||DC, Arl.||714,075||63%||Mass., 395, water||270,299||72.2%|
|Philadelphia||2,853,154||49.5%||City proper||684,869||50.9%||Center City||240,665||61.9%|
|Boston||2,682,278||56.3%||Boston, Cam.||787,287||66%||Arl., Stuart, water||228,300||72.1%|
|San Francisco||2,400,290||59.2%||City proper||723,907||65.5%||LA post||231,042||76.8%|
|Seattle||1,919,635||57.8%||City proper||585,480||64.1%||Jackson, I-5, Denny||180,482||71.2%|
In all cities, the proportion of workers earning $40,000 a year or more is higher in the city than in the rest of the region, and higher yet in the CBD. Moreover, this effect is weakest in Philadelphia, which may explain why there, unlike in the other secondary transit cities, drivers still significantly outearn transit commuters.
Crosstown public transportation
In all the cities studied in this post, public transportation carries a high share of trips into city center, especially at rush hour. This props up its usage numbers among the middle class, especially the upper middle class – professional jobs cluster in city center.
The problem is that not everyone works in city center. Midtown and Downtown Manhattan are 22% of metro New York employment going by OnTheMap’s LEHD numbers, and even that is a pretty hefty area. In smaller cities, there are necessarily fewer rapid transit lines and a smaller zone of intersection in which service is good from all directions. Improving transit service to destinations outside city center, and thus for working- and lower middle-class jobs, requires more than just disjointed center-to-bedroom-communities rail lines.
One way to have vigorous crosstown public transportation is with buses. However, buses are slow, almost by definition slower than cars. Chicago has a pretty good bus grid, but it still has the pattern of transit commuters outearning solo drivers. And that’s in the city proper – in the suburbs it’s not really possible to have a bus grid, because distances are too great and street networks are usually too broken.
Instead, a better solution has to involve diagonal trips on rapid transit, with a transfer in or near city center, and trips that stay outside city center. A good recipe includes all of the following:
- Easy downtown and near-downtown transfers, with no missed connections and a minimum of walking. San Francisco deserves especial demerits for forcing people to transfer between Muni and BART via the street, crossing two sets of faregates.
- High frequency on commuter rail in both directions, with timed bus shuttle connections from stations to office parks too far to walk. In some cases, such buses can do double duty ferrying suburban commuters to those stations for trains to city center.
- Complete fare integration, with free transfers and mode-neutral fares, to avoid forcing low-income commuters onto slow buses while richer ones get faster trains.
- Through-running when feasible, since a worker in one neighborhood may end up finding a job at a suburban job site on another line, even the opposite side of the city, e.g. between Brooklyn or Queens and Newark.
Income differences and universal design
The principles for good crosstown service are largely class-neutral. They have to be: the differences between where rich and poor people work in a deindustrialized country are real but not enormous, enough to be noticeable but not enough to play to populist clichés of two Americas. Nonetheless, better public transportation service to non-CBD destinations is especially useful for the working class, because the working class is less likely to work in the CBD than the middle class.
The relevance of class here is twofold. First, every demographic pattern in transportation mode choice has a reason, and provides hints as to how different people travel. This is the case regardless of whether the socially more dominant group commutes by public transport more (the rich, the educated) or less (the native-born, men, whites in Western countries). It remains the case even when there’s no obvious social dominance hierarchy between the groups we compare, for examples professionals versus small business owners.
And second, the people who manage public transportation agencies are drawn from one social class. They are middle-class managers working in city center at traditional peak hours. They may not be aware of how other people commute, regardless of whether those other people are retail workers working two part-time jobs in two different neighborhoods or tech workers who work 12-8. They provide the service that people who are like them can use, and neglect other use cases.
There’s a thread on Twitter by Stephen Smith bringing up Zurich’s S-Bahn as an alternative to extensive metro tunneling. It reminded me of something I’d been meaning to write about for a long time, about how S-Bahn tunnels, in Zurich and elsewhere, include not just the bare minimum for through-running but also strategic tunneling elsewhere to reach various destinations not on the mainline. Zurich’s S-Bahn includes about 19 km of tunnel built since the 1960s, which is similar per capita to the amount of tunneling built for the Washington Metro.
Such tunneling is important to ensure a regional rail network reaches destinations off the mainlines. Even cities with metro systems need to understand this as long as they have some mainline rail serving suburban destinations. For example, in the Center of Israel, Tel Aviv is getting a subway-surface light rail network, but outside the urban core rail transport will remain dominated by Israel Railways service; as Israel Railways avoids many city centers, such as Netanya, short strategic tunnels are critical.
Tunnels in Zurich
The core of the Zurich S-Bahn is three city center tunnels: the 2 km Käferberg Tunnel from Oerlikon to Hardbrücke, the 7 km combination of the Hirschengraben Tunnel and the Zürichberg Tunnel from Hauptbahnhof to the Right Bank of Lake Zurich and points northeast, and the 5 km Weinberg Tunnel from Hauptbahnhof to Oerlikon and points north. The Käferberg Tunnel is from the 1960s, the Hirschengraben and Zürichberg Tunnel opened in 1989-1990 as the core of the Zurich S-Bahn, and the Weinberg Tunnel opened in 2014 as a second S-Bahn route to add more capacity.
These 14 km of tunnel look like any standard picture of regional rail tunneling. However, Zurich has in addition built a 5 km tunnel for a loop to the airport. Without this tunnel, no regional or intercity rail service to the airport would have been possible, as the airport was at a distance from the mainline; only trams could have served the airport then.
In addition to these 19 km, there is some talk of building an additional tunnel of 7-10 km on the Zurich-Winterthur Line, called the Brüttener Tunnel, to speed up service between these two cities.
Tunnels on other regional rail systems
In Paris, the RER consists not just of legacy rail track and city center tunnels, but also outlying tunnels reaching new destinations. The RER B connection to Charles de Gaulle Airport is new construction, opening in 1976 as a commuter line just before the RER opened and incorporated it as a branch. It’s a mix of above- and underground construction, totaling 5.5 km of tunnel. Two more key RER lines, at both ends of the RER A, are new: the branch to Cergy, which opened between 1979 and 1994 and has 3 km of tunnel, and the branch to Marne-la-Vallée, which opened in stages starting on the same day as the RER A’s central tunnel and continuing until reaching its terminus in 1992.
All three new RER branches are busy. They have to be – if there weren’t so much demand for them, it would have been financially infeasible to build them and those areas would have had to make do with a bus connection to the existing mainlines. The Marne-la-Vallée branch carries about two thirds of the eastern branch ridership of the RER A, making it most likely the busiest single rail branch in Europe.
In London, the regional rail network is less modern than in Paris, Zurich, and other cities with extensive development of new tunnels. Nonetheless, the Crossrail plans do include a short outlying tunnel reaching Heathrow Airport. Moreover, one of the two eastern branches of the mainline has the characteristics of an outlying tunnel, namely the branch to Canary Wharf. Canary Wharf is only 5 km from the City of London and the tunnel connecting to it is contiguous with the central tunnel, but the branch is not really about improving connections to onward suburbs. Where La Défense was always on the way to western suburbs on the RER, Canary Wharf is only on the way to Abbey Wood. There are proposals among area railfans to extend this branch much farther to the east, but no official plans that I know of. In the currently planned paradigm for Crossrail, Canary Wharf is purely a destination.
In Munich, there is a new line toward the airport, with some tunneling on airport grounds as well as at two intermediate suburban stations. There is also a short above-ground spur connecting the airport to the western side of the S-Bahn, giving it two different routes to city center. Finally, there is a short tunnel slightly to the west of the main trunk tunnel to better connect S7 to the mainline.
Why are airports so prominent on this list?
The concept of using strategic tunnels to build new spurs and loops to connect mainlines to new destinations has nothing to do with airports. And yet, so many of these spurs connect to airports: Charles de Gaulle, Heathrow, Zurich, Munich. There are many more such examples, on regional or intercity lines: Schiphol, Arlanda, Ben-Gurion, soon-to-be Berlin-Brandenburg, Barajas. Why is that?
The answer is that the purpose of a spur or loop is to connect to a destination off the mainline. European cities for the most part developed around the railway or metro line. Virtually every important destination in London is on a legacy railway because during the city’s 19th and early 20th century growth period, the railway was the only way to get to Central London. Airports are consistent exceptions because they’re so land-intensive that it’s hard to site them near existing railways.
Where non-airport destinations somehow had to be developed away from the mainline, they’re attractive targets for spurs as well. Canary Wharf sits on the site of a disused dock, which generated some freight rail traffic but little demand for passenger rail. Cergy is one of several new towns built around Paris to act as suburban growth nodes, together with Marne-la-Vallée and Évry (served on a loop of the RER D).
In smaller cities than Paris and London, suburban growth often came together with a metro line. In Stockholm, the Metro was planned together with public housing projects, so many of the Million Program projects are right next to stations, facilitating high public transportation usage. There’s usually no need to build many new regional rail spurs, because such sites are close enough to the center for metro service to be quick enough.
The situation of regional rail in Israel
In Israel, urban development has ignored the railway almost entirely. The colonial network was weak and barely served the state’s travel needs. Investment was minimal, as the state’s political goals were population dispersal and Judaization of peripheral areas rather than efficient transportation. Towns were built around the road network, connected to one another by bus since people were too poor to afford cars.
Rail revival began in the early 1990s with the opening of the Ayalon Railway, providing through-service between points north and south of Tel Aviv. In the generation since, ridership has grown prodigiously, albeit from low initial levels, and the state has built new lines, with an ongoing project to electrify most of the passenger network. However, since the cities came first and the trains second, the new lines do not enter city centers, but rather serve them peripherally near the highway, often surrounded by parking.
Thus, Netanya’s train station is located to the east of the city’s built-up area, on the wrong side of the Route 2 freeway. Ashdod’s train station is on the periphery at a highway interchange, well to the east of city center. Ashkelon’s station is on the eastern margin. The under-construction line through Kfar Saba and Ra’anana passes just south of the built-up area.
In all of these cases, doing it right would require, or would have required, just short, strategic elevated or underground lines:
- Netanya is at the northern end of the Tel Aviv commuter rail network, and so it can easily be served by a spur. The existing station can be retained as a junction for intercity rail service, but building a commuter rail spur would not compromise frequency. Such a spur would require no more than 2 km of tunnel.
- In Ashdod and Ashkelon, there are north-south arterials that are so wide, 50-60 meters, that they could host cut-and-cover subways, effectively moving the line to the west to serve those cities better. In Ashdod there is a decision between going under B’nai Brith, which offers a more convenient through-route, and Herzl, which is more central but requires some boring at the southern end of the city.
- In Kfar Saba and Ra’anana, about 8 km of tunnel under Weizmann and Ahuza are needed, and could potentially be done cut-and-cover as well, but these streets are 30 meters rather than 50 meters wide. Such a route would replace the under-construction combination of a freeway and railway.
- In Rishon LeZion, a 6km route, not all underground, is needed to connect Rishonim with Moshe Dayan via city center and the College of Management rather than via the under construction freeway route avoiding these destinations.
Unfortunately, so far the state’s investment plans keep skirting city centers. It serves them with a cars-and-trains paradigm, which assumes the rail passenger is driving or riding a bus to the train station, never mind that in that case it’s more convenient to drive all the way to one’s destination. This suppresses ridership; not for nothing, the busiest station outside metropolitan centers is Rehovot, with 2.1 million annual entries, and not Ashdod, which is second with 1.9 million. Ashdod is a city of 220,000 and Rehovot one of 140,000, but Rehovot’s station is far more walkable. Were Ashdod not poor, few people would use the station at all – they’d all just drive.
In England and Wales, 15.9% of workers get to work on public transport, and in France, 14.9% do. In Canada, the figure is close: 12.4%, and this is without a London or Paris to run up the score in. Vancouver is a metro region of 2.5 million people and 1.2 million workers, comparable in size to the metropolitan counties in England and to the metro area of Lyon; at 20.4%, it has a higher public transport modal share than all of them, though it is barely higher than Lyon with its 19.9% share. Calgary, Ottawa, Edmonton, and Winnipeg are likewise collectively respectable by the standards of similar-size French regions, such as the departments of Bouches-du-Rhône (Marseille), Alpes-Maritimes (Nice), Gironde (Bordeaux), Haute-Garonne (Toulouse), and Bas-Rhin (Strasbourg).
As a result, Jarrett Walker likes telling American cities and transit agencies to stop envying Europe and start envying Canada instead. Canada is nearby, speaks the same language, and has similar street layout, all of which contribute to its familiarity to Americans. If Europe has the exotic mystique of the foreign, let alone East Asia, Canada is familiar enough to Americans that the noticeable differences are a cultural uncanny valley.
And yet, I am of two minds on this. The most consistent transit revival in Canada has been in Vancouver, whose modal share went from 14.3% in 1996 to 20.4% in 2016 – and the 2016 census was taken before the Evergreen extension of the Millennium Line opened. TransLink has certainly been doing a lot of good things to get to this point. And yet, there’s a serious risk to Canadian public transport in the future: construction costs have exploded, going from Continental European 15 years ago to American today.
The five legs of good transit
I was asked earlier today what a good political agenda for public transportation would be. I gave four answers, like the four legs of a chair, and later realized that I missed a fifth point.
- Fuel taxes and other traffic suppression measures (such as Singapore and Israel’s car taxes). Petrol costs about €1.40/liter in Germany and France; diesel is cheaper but being phased out because of its outsize impact on pollution.
- Investment in new urban and intercity lines, such as the Madrid Metro expansion program since the 1990s or Grand Paris Express. This is measured in kilometers and not euros, so lower construction costs generally translate to more investment, hence Madrid’s huge metro network.
- Interagency cooperation within metropolitan regions and on intercity rail lines where appropriate. This includes fare integration, schedule integration, and timetable-infrastructure integration.
- Urban upzoning, including both residential densification in urban neighborhoods and commercialization in and around city center.
- Street space reallocation from cars toward pedestrians, bikes, and buses.
We can rate how Canada (by which I really mean Vancouver) does on this rubric:
- The fuel tax in Canada is much lower than in Europe, contributing to high driving rates. In Toronto, gasoline currently costs $1.19/liter, which is about €0.85/l. But Vancouver fuel taxes are higher, raising the price to about $1.53/l, around €1.06/l.
- Canadian construction costs are so high that investment in new lines is limited. Vancouver has been procrastinating building the Broadway subway to UBC until costs rose to the point that the budget is only enough to build the line halfway there.
- Vancouver and Toronto both have good bus-rapid transit integration, but there is no integration with commuter rail; Montreal even severed a key commuter line to build a private driverless rapid transit line. In Vancouver, bus and SkyTrain fares have decoupled due to political fallout from the botched smartcard implementation.
- Vancouver is arguably the YIMBYest Western city, building around 10 housing units per 1,000 people every year in the last few years. Toronto’s housing construction rate is lower but still respectable by European standards, let alone American ones.
- There are bike lanes but not on the major streets. If there are bus lanes, I didn’t see any of them when I lived in Vancouver, and I traveled a lot in the city as well as the suburbs.
Vancouver’s transit past and future
Looking at the above legs of what makes for good public transport, there is only one thing about Canada that truly shines: urban redevelopment. Toronto, a metro area of 6 million people, has two subway mainlines, and Montreal, with 4 million people, has 2.5. Vancouver has 1.5 lines – its three SkyTrain mainlines are one-tailed. By the same calculation, Berlin has 6.5 U- and 3 S-Bahn mainlines, and Madrid has 2 Cercanías lines and 7 metro lines. Moreover, high construction costs and political resistance from various GO Transit interests make it difficult for Canadian cities to add more rapid transit.
To the extent Vancouver has a sizable SkyTrain network, it’s that it was able to build elevated and cut-and-cover lines in the past. This is no longer possible for future expansion, except possibly toward Langley. The merchant lawsuits over the Canada Line’s construction impacts have ensured that the Broadway subway will be bored. Furthermore, the region’s politics make it impossible to just build Broadway all the way to the end: Surrey has insisted on some construction within its municipal area, so the region has had to pair half the Broadway subway with a SkyTrain extension to the Langley sprawl.
Put in other words, the growth in Vancouver transit ridership is not so much about building more of a network, but about adding housing and jobs around the network that has been around since the 1980s. The ridership on the Millennium and Canada Lines is growing but remains far below that on the Expo Line. There is potential for further increase in ridership as the neighborhoods along the Canada Line have finally been rezoned, but even that will hit a limit pretty quickly – the Canada Line was built with low capacity, and the Millennium Line doesn’t enter Downtown and will only serve near-Downtown job centers.
Potemkin bus networks
When Jarrett tells American cities to envy Canada, he generally talks about the urban bus networks. Toronto and Vancouver have strong bus grids, with buses coming at worst every 8 minutes during the daytime off-peak. Both cities have grids of major streets, as is normal for so many North American cities, and copying the apparent features of these grids is attractive to American transit managers.
And yet, trying to just set up a bus grid in your average American city yields Potemkin buses. Vancouver and Toronto have bus grids that rely on connections to rapid transit lines. In both cities, transit usage is disproportionately about commutes either to or from a city core defined by a 5 kilometer radius from city hall. Moreover, the growth in public transport commuting in both cities since 1996 has been almost exclusively about such commutes, and not about everywhere-to-everywhere commutes from outside this radius. Within this radius, public transportation is dominated by rail, not buses.
The buses in Toronto and Vancouver have several key roles to play. First, as noted above, they connect to rapid transit nodes or to SeaBus in North Vancouver. Second, they connect to job centers that exist because of rapid transit, for example Metrotown at the eastern end of Vancouver’s 49. And third, there is the sui generis case of UBC. All of these roles create strong ridership, supporting high enough frequency that people make untimed transfers.
But even then, there are problems common to all North American buses. The stop spacing is too tight – 200 meters rather than 400-500, with frequency-splitting rapid buses on a handful of very strong routes like 4th Avenue and Broadway. There is no all-door boarding except on a handful of specially-branded B-line buses. There are no bus lanes.
One American city has similar characteristics to Toronto and Vancouver when it comes to buses: Chicago. Elsewhere, just copying the bus grid of Vancouver will yield nothing, because ultimately nobody is going to connect between two mixed-traffic buses that run every 15 minutes, untimed, if they can afford any better. In Chicago, the situation is different, but what the city most needs is integration between Metra and CTA services, which requires looking at European rather than Canadian models.
Is Canada hopeless?
I don’t know. The meteoric rise in Canadian subway construction costs in the last 15 years has ensured expansion will soon grind to a halt. Much of this rise comes from reforms that the Anglosphere has convinced itself improve outcomes, like design-build and reliance on outside consultants; in that sense, the US hasn’t been copying Canada, but instead Canada has been copying the US and getting American results.
That said, two positive aspects are notable. The first is very high housing and commercial growth in the most desirable cities, if not in their most exclusive neighborhoods. Vancouver probably has another 10-20 years before its developable housing reserves near existing SkyTrain run out and it is forced to figure out how to affordably expand the network. Nowhere in Europe is housing growth as fast as in Metro Vancouver; among the cities for which I have data, only Stockholm comes close, growing at 7-8 net units per 1,000 people annually.
Moreover, with Downtown Vancouver increasingly built out, Vancouver seems to be successfully expanding the CBD outward: Central Broadway already has many jobs and will most likely have further commercial growth as the Millennium Line is extended there. Thus, employers that don’t fit into the Downtown Vancouver peninsula should find a home close enough for SkyTrain, rather than hopping to suburban office parks as in the US. Right now, the central blob of 100 km^2 – a metric I use purely because of limitations on French and Canadian data granularity – has a little more than 30% of area jobs in Vancouver, comparable to Paris, Lyon, New York, Boston, and San Francisco, and ahead of other American cities.
The second aspect is that Canadians are collectively a somewhat more internationally curious nation than Americans. They are more American than European, but the experience of living in a different country from the United States makes it easier for them to absorb foreign knowledge. The reaction to my and Jonathan English’s August article about Canadian costs has been sympathetic, with serious people with some power in Toronto contacting Jonathan to figure out how Canada can improve. The reaction I have received within the United States runs the gamut – some agencies are genuinely helpful and realize that they’ll be better off if we can come up with a recipe for reducing costs, others prefer to obstruct and stonewall.
My perception of Canadian politics is that even right-populists like Doug Ford are more serious about this than most American electeds. In that sense, Ford is much like Boris Johnson, who could move to Massachusetts to be viceroy and far improve governance in both Britain and Massachusetts. My suspicion is that this is linked to Canada’s relatively transit-oriented past and present: broad swaths of the Ontarian middle class ride trains, as is the case in Outer London and the suburbs of Paris. A large bloc of present-day swing voters who use public transport is a good political guarantee of positive attention to public transport in the future. American cities don’t have that – there are no competitive partisan elections anywhere with some semblance of public transportation.
These two points of hope are solid but still run against powerful currents. Toronto really is botching the RER project because of insider obstruction and timidity, and without a strong RER project there is no way to extend public transportation to the suburbs. Vancouver is incapable of concentrating resources where they do the most good. And all Canadian cities have seen an explosion in costs. Canadians increasingly understand the cost problem, but it remains to be seen whether they can fix it.
When you ride a subway train, and the train decelerates to its station, you feel your body pulled forward, and your muscles tense to adjust, but then when the train reaches a sudden stop, you are suddenly flung backward, since you are no longer decelerating, but your muscles take time to relax and stop fighting a braking that no longer exist. This effect is called jerk, and is defined to be change in acceleration, just as acceleration is change in speed and speed is change in position. Controlling jerk is crucial for a smooth railway ride. Unfortunately, American mainline rail is not good at this, leading to noticeable jolts by passengers even though speed limits on curves and acceleration rates are very conservative.
This is particularly important for speeding up mainline trains around New York and other legacy cities in the US, like Boston. Speeding up the slowest segments is more important than speeding up the fastest ones; my schedules for New York-New Haven trains, cutting trip times from 2:09 to 1:24, save 4 minutes between Grand Central and 59th Street just through avoiding slowdowns in the interlocking. The interlocking is slow because the switches have very conservative speed limits relative to curve radius (that is, lateral acceleration), which in turn is because they are not designed with good lateral jerk control. The good news is that replacing the necessary infrastructure is not so onerous, provided the railroads know what they need to do and avoid running heavy diesel locomotives on delicate infrastructure.
Spirals and jerk
In practice, the worst jerk is usually not forward or backward, except in the last fraction of a second at the end of acceleration. This is because it takes about a second for train motors to rev up, which controls jerk during acceleration. Rather, the worst is sideways, because it is possible to design curves that transition abruptly from straight track, on which there is no lateral acceleration, to curved track, on which there is, in the form of centrifugal force centripetal force.
To reduce jerk, the transition from straight track to a circular arc is done gradually. There are a number of usable transition curve (see Romain Bosquet’s thesis, PDF-p. 36), but the most common by far is called the clothoid, which has the property of having constant change in curvature per unit of arc length – that is, constant jerk. Different countries have different standards for how long the clothoid should be, that is what the maximum lateral jerk is. Per Martin Lindahl’s thesis, the limit in Sweden is 55 mm/s (PDF-p. 30) and that in Germany is 69.44 mm/s (PDF-p. 38), both measured in units of cant deficiency; in SI units, this is 0.367 m/s^3 and 0.463 m/s^3 respectively. In France, the regular limit is 50 mm/s (Bosquet’s thesis, PDF-p. 35), that is 0.333 m/s^2, but it is specifically waived in turnouts.
Track switches are somehow accepted as sites of very high jerk. A presentation about various technical limits in France notes on p. 106 that in switches (“appareils de voie” or “aiguilles” or “aiguillages,” depending on source, just like “switch” vs. “turnout” in English), the jerk can be increased to 100 and even 125 mm/s. On p. 107 it even asserts that in exceptional circumstances, abrupt change in cant deficiency of up to 50 mm on main track and 100 on the diverging direction on a switch is allowed; see also PDF-pp. 13-15 of a pan-European presentation. Abrupt changes are not good for passengers, but will not derail a train.
Turnout design in the advanced world
Second derivative control, that is acceleration and cant deficiency, can be done using calculus and trigonometry tools. Third derivative control, that is clothoids and jerk, requires numerical calculations, but fortunately they are approximated well by pretending the clothoid is half straight line, half circular arc, with the length determined by the maximum jerk. Working from first principles, it’s possible to figure out that at typical turnout needs – e.g. move a train from one track to a parallel track 4 meters away – the clothoid is far longer than the curve itself, and at 50 mm/s jerk and 150 mm cant deficiency it’s not even possible to hit a curve radius of 250 meters.
Turnouts are inherently compromises. The question is just where to compromise. Here, for example, is a French turnout design, in two forms: 0.11 and 0.085. The numbers denoting the tangent of the angle at the frog, and the radius is proportional to the inverse square of the number, thus the speed is proportional to the inverse of the number. The sharper turnout, the 0.11, has a radius of 281 meters, a maximum speed of 50 km/h, and a total length of 26 meters from point to frog (“lead” in US usage), of which the clothoid curve (“point”) takes up 11, to limit jerk to 125 mm/s at a cant deficiency of 100 mm. The 0.085 turnout has a radius of 485 meters, a maximum speed of 65 km/h, a lead of about 38 meters, and a point of about 14.5 meters.
In Germany, turnouts have somewhat independent numbers and radii – some have shorter leads than others. The numbers are the inverse of those of France, so what France calls 0.11, Germany calls 1:9, but at the end of the day, the curve radius is the important part, with a cant deficiency of 100 mm. A higher cant deficiency may be desirable, but lengthening the point requires almost as much space as just increasing the curve radius, so might as well stick with the more comfortable limits.
Turnout design in the United States
American turnouts look similar to French or German ones, at first glance. I’ve seen a number of different designs; here’s one by CSX, on PDF-pp. 22 (#8) and 24 (#10), the numbers being very roughly comparable to German ones and inverses of French ones. CSX’s #10 has a curve radius of 779.39′, or 238 meters, and a lead of 24 meters, both numbers slightly tighter than the French 0.11. The radius is proportional to the square of the number, and so speed is proportional to the number.
However, the cant deficiency is just 50 mm. The point is not always curved; Amtrak’s low-number switches are not, so the change in cant deficiency is abrupt. Judging by what I experience every time I take a train between New York and New Haven, Metro-North’s switches have abrupt change in cant deficiency even on the mainline. The recommended standards by AREMA involve a curved point, but the point is still much shorter than in France (19.5′, or just under 6 meters, on a #12), so a 125 mm/s jerk only gets one up to about 62 mm cant deficiency.
The reason for this is that European turnouts are curved through the frog, whereas American ones are always straight at the frog. Extremely heavy American freight trains do not interact well with curved frogs and long points.
One might ask, why bother with such turnout design on rail segments that never see a heavy freight locomotive or 130-ton freight car? And on segments that do see the odd freight locomotive, like the approaches to Grand Central and Penn Station with the rare dual-mode locomotive, why not kick out anything that doesn’t interact well with advanced track design? Making a handful of passengers transfer would save around 4 minutes of trip time on the last mile into Grand Central alone for everyone else, not to mention time savings farther up the line.
Public transportation companies may have the ability to raise fares arbitrarily based on market demands, for examples British buses outside London and American freight railroads. Or they may be subject to regulations capping the fare, for example Japanese railroads. Mixed systems exist as well, such as British rail fares. In Britain, the privatized, mostly deregulated approach is so commonly accepted that a Conservative recently called Labour dangerous socialists for proposing municipalizing bus systems, as in such socialist states as the US, Japan, Germany, etc. In reality, in the case of rail specifically (and perhaps buses as well), there’s a theoretical case with some empirical backing for why reasonable fare caps as in Japan can lead to more investment and more capacity, whereas wholly unregulated fares lead to hoarding and capacity cuts to create shortages.
I’m stealing the economic model for this post from Paul Krugman, who used it to explain the California blackouts of 2000-1. The demand curve is inelastic: the demand is 1,000 units at $20/unit, decreasing to 900 units at $1,000/unit, at which point the curve goes flat. The supply curve is a constant $20/unit, but the market is oligopolistic (say, there are very high barriers to entry because building your own power plant is hard), and there are 5 producers, each with 200 units. If the price is regulated at $20/unit, each producer will supply 200 units. If the price is unregulated, then each producer alone gets an incentive to hold back production, since 100*1000 > 200*20, and then production will be curtailed to 900 units.
The model is simplified in a number of ways: real supply curves slope up; the part about demand going flat at 900 units is unrealistic and exists purely to avoid dealing with optimizing where at 800-something units each producer has an incentive to go back to producing more; capacity constraints involve escalating production costs rather than a God-given restriction on the number of suppliers and their capacity. But with all these caveats, it fits markets that have the following characteristics:
- There are steep barriers to entry, for example if large amounts of capital are required to enter (to build a power plant, set up a rail operating company, etc.).
- Demand is highly inelastic.
- Adding new capacity is expensive.
The issue of capacity
In rail, we can start plugging real numbers for both demand elasticity and the cost of new capacity.
In the above model the price elasticity is -0.0244 in the 900-1,000 units range, which is ridiculously inelastic, on purpose so as to highlight how the model works. TCRP Report 95 says the elasticity in a number of large cities studied is about -0.18, and a VTPI review in a mixture of cities and circumstances (peak vs. off-peak, bus vs. rail, etc.) asserts a short-term average of about -0.3. Unregulated fares will lead to supply reductions if the elasticity times the number of producers is more than -1 (or less than 1 if you flip signs); if no producer has <18% of the market, there will be supply restrictions under unregulated fares, just as a monopolist will hold back supply and raise fares if demand is inelastic.
The cost of new capacity of course depends on the line and the characteristics of competition between different railroads. It’s higher in Japan, where separate railroads run their own lines and trains, than in Britain, where different companies franchise to run trains on the same tracks. But even in Britain, getting a franchise requires a commitment to running service for many years. The significance of this is that the long-run public transport ridership elasticity with respect to fare is more elastic (VTPI recommends a range of -0.6 to -0.9), with a few estimates even going below -1.
For the purposes of this section, we do not distinguish capital from operating costs. Thus, the cost of new capacity is not given in units of capital costs, but in units of operating costs: if increasing service by 1% raises operating expenses by 2% counting the extra investment required, then we say the supply elasticity is 2. Note that supply curves slope up so the elasticity is always positive, but the elasticity can be below 1, for example if economies of scale are more important than the need to invest in new capacity.
Set the following variables: u is quantity of service, r is total revenue (thus, fare is r/u), c is total costs. The railroad is assumed profitable, so r > c. We are interested in the change in profit based on quantity of service, i.e.
The important thing to note is that price controls keep dr/du higher in an oligopoly (but not in a competitive environment, like housing – a single landlord can’t meaningfully create a housing shortage). With price controls, we get
whereas without price controls, with elasticity , we get
And likewise, with supply elasticity , we get
Note, moreover, that price controls as construed in Japan let operating companies recover profits, letting them raise prices if they invest in more capacity, so that dr/du is actually higher than r/u.
The real world
I do not know to what extent the lack of fare regulation on many British trains contributes to capacity shortages. However, there is some evidence that the same situation is holding back investment in the United States, on Amtrak. Amtrak is a monopolist facing some fare regulations, for example congressional rules limiting the spread between the lowest and highest fares on a given train, but within its ability to set its own capacity in the medium run, it has relatively free hand, and in fact a strong incentive to maximize fares, in order to subsidize money-losing trains outside the Northeast Corridor.
Amtrak generally runs the trains it has on the Northeast Corridor, without explicitly holding back on capacity. However, this is in an environment with very low utilization rates. There are 20 Acela trainsets, but only 16 run in service at a given time, giving them the moniker “hangar queens.” There is no real interest within Amtrak at raising speed just enough to be able to run consistent service intervals, for example hourly with two trainsets coupled to form a 16-car train south of New York. Nor is there any interest in making small investments to permit such long trainsets to run – most Acela stops from New York to the south have platforms long or almost long enough for such trains, but the rest need to be lengthened, within right-of-way so that the cost is positive but low.
In the future, capacity cliffs may prove serious enough to stymie American passenger rail development. Right now the main obstacle are Amtrak itself and obstructive commuter railroads such as Metro-North, but assuming competent, profit-maximizing investment plans, it is not so expensive to invest in capacity and speed so as to permit around 4 long high-speed trains per hour north of New York (or even New Haven) and 6 south of it. But then the next few trains per hour require further bypasses, for example four-tracking most of the Providence Line. High supply elasticity – let’s say around 2 – is plausible. Then eventually a dedicated pathway to intercity trains through New York becomes necessary, raising supply elasticity even higher. In an environment with uncapped, profit-maximizing fares, a rational Amtrak management may well just keep what it has and jack up prices rather than build more capacity.
I gave a second talk this week about transportation, this time at Hartford Station, concerning the plans for Connecticut transportation. The starting point is Governor Lamont’s $21 billion plan for investment, including both expansion and repairs (read: the State of Good Repair black hole), of which $14 billion is highways, $6.2 billion is rail, and $450 million is buses. But most of the talk concerns what Connecticut should be doing, rather than the specifics of Lamont’s plan.
Here are my slides. The talk itself took around 40-45 minutes out of a nearly 2-hour meeting, so it was designed around taking many questions, and around further explanations. Something I didn’t put in the slides but explained verbally is how easy the modern track renewal process is. Nowadays, there are machines that use no infrastructure except the tracks themselves, running on the tracks at very low speed (slower than walking) and systematically replacing the rails, ties, and ballast. They can also regrade the tracks’ superelevation angle independently of the drainage angle, changing the tracks’ cant as they go. The upshot is that increasing the cant on tracks is almost cost-free, and would enable large increases in train speed on both regional and intercity trains.
Other technology that has negative cost in the future is getting higher-performance EMUs than the current equipment. The current trains are obsolete technology, built around superseded federal regulations. There’s no point in getting more of the same. They’re okay to run until end of life, but new purchases should involve electrification and modern European EMUs. Whereas infrastructure costs are rising (see here and here), technology costs are falling in real terms. The fall in train costs is not so quick as that of computer costs, but still the rolling stock factories are designed around making products for the 2020s, not the 1990s, and retooling them for older technology costs extra.
Hence my slogan from the talk: better things are possible, on a budget.
One question I was asked at the talk that I didn’t have an answer to was, why is construction in Connecticut so expensive? Plans for infill stations are budgeted extravagantly, ranging between $50 million and $100 million without any special construction difficulties. Boston builds infill stations (counting high-platform upgrades as infill since the preexisting stations have no facilities) for $20-30 million counting various hidden costs (e.g. regular MBTA employees, like project managers, count as operating and not capital costs even if they only work on capital costs); Berlin does for €10-20 million.
After the talk, Roger Senserrich explained to me (and a planner at the MBTA confirmed to me) that in Connecticut there’s no in-house design at all. Massachusetts has a mix of in-house design review, with the team stymied by uncompetitive wages making hiring and retention difficult, and outsourcing work to consultants. CDOT exclusively outsources to consultants, and has no in-house expertise to evaluate whether the contracts are fair or whether it’s being overcharged.
In the third and last installment of my series posting sample commuter rail schedules for New York (part 1, part 2), let’s look at trains in New Jersey. This is going to be a longer post, covering six different lines, namely all New Jersey Transit lines that can go to Penn Station, including one that currently does not (Raritan Valley) but could using dual-mode locomotives.
As on Metro-North and the LIRR, very large improvements can be made over current schedules, generally reducing trip times by 30-43%, without straightening a single curve. However, electrification is required, as is entirely new rolling stock, as the electric locomotives used by NJ Transit are ill-fit for a fast schedule with many stops. Moreover, all low platforms must be raised to provide level boarding and some must be lengthened to avoid overuse of selective door opening, which may require a few new grade separations on the North Jersey Coast Line. As a first-order estimate, 50-something trainsets are required, each with 8-12 cars. This is not quite free, but the cost is low single-digit billions: about $1.5 billion for trains, maybe $400 million for 160 km of electrification, and around $700 million for what I believe is 70 low- or short-platform stations.
Here is a spreadsheet detailing speed zones for all New Jersey Transit lines passing through Newark. In support of previous posts, here are other similar spreadsheets:
- New Haven Line (express schedule, add stop penalties as appropriate for locals) – the spreadsheet is about a minute too fast, missing some slowdowns in the terminal, and the version in my post (part 1) corrects for that
- Harlem Line
- Hudson Line locals and expresses
- LIRR Main Line (including Port Jefferson, not covered in my posts)
Line by line schedules
The New Jersey Transit timetables are less consistent than the east-of-Hudson ones; I attempted to look at local midday off-peak outbound trains whenever possible.
|Station||Current time||Future time|
|Newark South Street||—||0:12|
This fastest rush hour express trains do the trip in 1:12-1:13, and Amtrak’s Regionals range between 0:55 and 1:04, with trains making all nominal Amtrak stops (including rarely-served New Brunswick and Princeton Junction) taking 1:15.
North Jersey Coast Line
|Station||Current time||Future time|
|Newark South Street||—||0:12|
|Point Pleasant Beach||2:15||1:22|
In electric territory, that is up to Long Branch, my schedule cuts 38% from the trip time, but in diesel territory the impact of electrification nearly halves the trip time, cutting 48%.
Raritan Valley Line
|Station||Current time||Future time|
|Newark South Street||—||0:12|
The Raritan Valley Line does not run through to Manhattan but rather terminates at Newark Penn because of capacity constraints on the mainline, so the New York-Newark trip times are imputed from Northeast Corridor trains. So really the trip time difference is 1:34 versus 0:54, a reduction of 42% in the trip time thanks to electrification.
|Station||Current time||Future time|
|Newark Broad Street||0:19||0:11|
|Newark 1st Street||—||0:13|
This timetable is cobbled from two different train runs, as electric wires only run as far out as Dover, so trains from New York only go as far as Dover, and trains to Hackettstown serve Hoboken instead. Observe the 35% reduction in trip time in electric territory despite making a few more stops, and the 48% reduction in trip time in diesel territory.
|Station||Current time||Future time|
|Newark Broad Street||(0:19)||0:11|
|Newark 1st Street||—||0:13|
As the line is entirely electrified, the time saving is only 30%. Note that Gladstone Branch trains do not run through to Penn Station except at rush hour, so I’m imputing New York-Newark Broad trip times using the Morristown Line.
|Station||Current time||Future time|
|Newark Broad Street||(0:20)||0:11|
|Newark 1st Street||—||0:13|
|Newark Park Street||—||0:15|
|Montclair State U||0:50||0:33|
Beyond Dover, a handful of evening trains continue to Hackettstown. Interestingly, the saving from electrification is only 32% – and the train I drew the current schedule from is a Hoboken diesel train. Electric trains run from New York to Montclair State University, but are for some reason actually slightly slower today than the Hoboken diesels on the shared Newark-MSU segment. I suspect that like the LIRR, NJ Transit does not timetable electric trains to be any faster than diesels on shared segments even though their performance is better.
There are specific patterns to where my schedule outperforms the existing one by the largest margin and where it does so by the smallest margin.
Between New York and Newark, I am proposing that trains take 10-11 minutes, down from 18-20 today, cutting 45% from the trip time. This comes from several factors. The first is avoiding unnecessary slowdowns in terminal zones: Penn Station should be good for about 50 km/h, ideally even more if there are consistent enough platform assignments that the turnouts can be upgraded to be faster; Newark should not impose any speed limit whatsoever beyond that of right-of-way geometry.
The second is increasing superelevation and cant deficiency. The worst curve is the turn from Harrison to Newark; its radius is just shy of 500 meters, good for around 110 km/h at normal cant and cant deficiency (150 mm each), or even 120 km/h if the cant is raised to 200 mm in support of higher-speed intercity service. But the current speed limit is a blanket 45 mph, even on Amtrak, whose cant deficiency is fine. The Newark approach is then even slower, 35 mph, for no reason. It’s telling that on my schedule, the Secaucus-Newark speedup is even greater than the New York-Secaucus speedup, despite the Penn Station interlocking morass.
The third is reducing schedule padding. The schedules appear extremely padded for what NJ Transit thinks is a capacity problem but is not really a problem in the midday off-peak period. Between 9 am and noon, 18 trains depart Penn Station going west, 10 on the Northeast Corridor and North Jersey Coast and 8 on the Morris and Essex Lines and the Montclair Line.
On lines without electrification, the time savings from electrification are considerable, with the exception of the Boonton Line. This is especially notable on the tails of the North Jersey Coast and Morristown Lines, both of which allow for 48% reductions in trip time, nearly doubling the average speed.
This is related to the issue of low platforms. These tails have low platforms, whereas the inner segment of the Raritan Valley Line (up to Westfield), which already has mostly high platforms, does not exhibit the same potential speed doubling. Outer segments may also not be well-maintained, leading to non-geometric speed limits. Between Long Branch and Bay Head the tracks are fairly straight, but the existing speed limits are very low, at most 60 mph with most segments limited to 40 or even 25 or less.
In contrast with the enormous slowdowns between New York and Newark and on unelectrified tails, the workhorse inner segments (including the entire Northeast Corridor Line) radiating out of Newark are only about 1.5 times as slow as they can be, rather than twice as slow. The Gladstone Branch, which runs EMUs rather than electric locomotive-hauled trains, manages to be only about 1.37 times as slow, in large part courtesy of low platforms.
Of course, 1.5 times as slow is still pretty bad. This is because no line on NJ Transit is truly modern, that is running all EMUs serving high platforms. But the electric lines manage to be less bad than the diesel lines, and the suburbs less bad than the New York-Newark segment with its excessive timetable padding and terminal zone slowdowns.
How to get there from here
NJ Transit has a problem: perhaps unaware of the new FRA regulations, it just ordered bilevel EMUs compliant with the old rather than new regulations. If it can cancel the order, it should do so, and instead procure standard European EMUs stretched to the larger clearances of the American (or Nordic) railway network.
Simultaneously, it should complete electrification of the entire Penn Station-feeding system, including the Raritan Valley Line even though right now it does not run through to New York. This includes some outer branches with low traffic, not enough to justify electrification on their own; that is fine, since the 31 km of wire between Dover and Hackettstown, 25 km between Long Branch and Bay Head, 27 between Raritan (where semi-frequent service ends) and High Bridge, and 30 between MSU and Denville permit a uniform or mostly uniform fleet with no diesel under catenary. EMUs are far more reliable than anything that runs on diesel, and if NJ Transit retires diesels and only runs EMUs on the most congested segment of the network, it will be able to get away with far less schedule padding.
In Boston, at Transit Matters we’ve likewise recommended full systemwide electrification, but with priority to lines that connect to already-electric infrastructure, that is the Stoughton branch of the Providence Line, the Fairmount Line (which is short enough to use Northeast Corridor substations), and subsequently the entire South Station-feeding system. By the same token, it is more important to electrify the outer edges of the Morristown and North Jersey Coast Lines and the entire Raritan Valley Line than to electrify the Erie lines not analyzed in this post, since the Erie lines’ infrastructure points exclusively toward Hoboken and not New York.
In addition to electrification, NJ Transit must replace all low platforms with high platforms. This should generally be doable with ramp access rather than elevators to save money, in which case a double-track station should be doable for about $10 million, if Boston and Philadelphia costs are any indication. In addition to speeding up general boarding, high platforms permit wheelchair users to board trains without the aid of an attendant or conductor.
All of this costs money – the infrastructure should cost somewhat more than $1 billion, and new rolling stock should cost about $1.5 billion at European costs, or somewhat more if there’s an American premium for canceling the in-progress contract for inferior equipment. But none of this costs a lot of money. New Jersey is ready to sink $2.75 billion of state money as part of an $11 billion Gateway tunnel that would do nothing for capacity (since it four-tracks the tunnel but not the surface segments to Newark); it should be ready to spend about the amount of money on a program that is certain to cut 25-50% off of people’s travel time and perhaps halve operating costs.
In my last post about New York regional rail schedules, I covered the New Haven and Harlem Lines of Metro-North and the Main Line and Hempstead Branch of the LIRR. I was hoping to cover more lines tonight, but due to time constraints only the Hudson Line is available.
This post should be viewed as considerably more accurate than the previous one, because I’ve obtained a Metro-North track chart with exact curve radii. I had to use measuring tools in the previous posts, and although the results were generally accurate, they were not completely so, and a few short, sharp curves cost a few more seconds than depicted. I do not believe the total slowdown between New York and either New Haven or Southeast to be worse than one minute relative to the track chart, but it is a slight slowdown, more than countermanding my tendency to round all fractional seconds up in speed zones.
One key difference with my last post is that the Hudson Line is not entirely electrified. It is only electrified south of Croton-Harmon; farther north, trains run with diesel locomotives, changing to electric mode only in Manhattan. My timetable assumes electrification. This is a project Metro-North should be pursuing anyway, since the outer Hudson Line is one of the busiest diesel lines in New York, alongside the outer Port Jefferson Branch and the Raritan Valley Line.
This lack of electrification extends to part of the express tracks south of Croton-Harmon as well. As a result, this schedule, while relying on cheap investments, is not quite the near-zero cost improvement on the express line. On the local line it is, since the trains are electrified.
As before, I am not assuming any curve is straightened, merely that track geometry trains fix the tracks to have higher superelevation (150 mm) and that trains run at 150 mm cant deficiency rather than today’s 3″. In metric units, this means acceleration in the horizontal plane is 2 m/s^2, so curves obey the formula
One big-ticket item that Metro-North should look into, in addition to completing electrification, is grade-separating the interlocking at CP 5, between the Hudson and Harlem Lines. The flat junction is extremely busy – it may plausibly have higher peak throughput than the flat junctions that plague South London’s commuter rail network – and hinders a simple 2-tracks-in, 2-tracks-out operation. This is not strictly speaking a speedup, but I would be more comfortable writing aggressive, high-frequency timetables if trains did not conflict at-grade.
Local trains run up to Croton-Harmon, making all stops.
|Station||Current time||Future M-7 time||Future Euro time|
The 9-minute interstation between Ossining and Croton-Harmon represents end-of-line schedule padding – in the southbound direction, trains are scheduled to take only 4 minutes.
Observe that the travel time difference is smaller than on the other lines presented in my previous post. Current equipment could shave 21% off the travel time, which is considerable but a far cry from the 33-40% elsewhere in the system. The reason is that the Hudson Line is maintained to higher standards, with cruise speeds of 80 mph on much of the line; I am assuming a speedup to 160 km/h, but the stop spacing along the Hudson is so short that trains can’t even hit 160 km/h while accelerating. The curves are still insufficiently superelevated – the Spuyten Duyvil curve where the fatal derailment happened has only 2.5″ of superelevation – and trains are only rated for low cant deficiency. However, the other aspects of the speedup on other lines are less conspicuous.
I also suspect that there is less schedule padding on the Hudson Line than on the other lines. Its frequency is lower, the line is four-track for most of its length, and the one significant flat junction equally affects the other two Metro-North mainlines. So the schedule may already be stable enough that padding, while considerable, is less outrageous than on the LIRR.
Express trains on the Hudson Line run a variety of stopping patterns, especially at rush hour. The line’s infrastructure is set up for intermediate express stops at Harlem, Marble Hill, Yonkers, Tarrytown, Ossining, and Croton-Harmon, but the standard off-peak pattern makes slightly fewer stops. My assumption is that all the above stations will receive express service.
|Station||Current time||Future M-7 time||Future Euro time|
This is a 35-38% reduction in travel time while making four more stops, two on the inner part of the line and two on the outer part that currently only see occasional seasonal use for hiking trails. The explanation for this is simple: the rolling stock used today is not M-7 EMUs but diesel locomotives. Rush hour trains running nonstop between Manhattan and Beacon connect Grand Central with Poughkeepsie in 1:36-1:37, a stop penalty of about 2.5 minutes, twice as high as what a European regional EMU can achieve at a top speed of 160 km/h.
Moreover, the 80-90 mph speed limit, which is dead letter on local trains for most of the way because they stop so frequently, consumes a few minutes relative to 160 km/h when trains run nonstop for long stretches. Thus, an increase in top speed is necessary in addition to an increase in curve superelevation and cant deficiency.
What about Grand Central?
My schedules consistently depict 6-minute trip times between Grand Central and Harlem, compared with current timetables that have them do it in 10-11 minutes. On most of the line, the top speed is the same – 60 mph, against 100 km/h in my timetable. The difference is entirely in the last mile out of Grand Central, where the limit today is 10 mph for no good reason.
The constrained environment of Grand Central does not leave room for high-speed switches. Nonetheless, the existing switches, called #8 switches, have a curve radius of about 140 meters, which is good enough for 40 km/h with no superelevation and a cant deficiency of 150 mm. American switches are generally rated for twice their number in miles per hour, assuming no superelevation and a 2″ cant deficiency; but higher cant deficiency is possible, and is really important as the difference between 25 and 40 km/h for a few hundred meters is considerable.
Moreover, 40 km/h is only the governing speed for a very short distance, about half a kilometer. Farther out, trains can always take the straight direction on turnouts, with one exception, turnout number 309B on the southbound local track (track 4), which is a triangular switch, i.e. one without a straight direction. Fixing the switch to have a straight direction from track 4 to track J, the westernmost approach track to the lower level of the station, should be a priority, plausibly saving 3 minutes for all trains using this track.
With trains taking the straight direction wherever possible, the central express tracks in the Park Avenue Tunnel (tracks 1 and 2) should exclusively feed the upper level, and the outer local tracks should exclusively feed the lower level; this way, there would not be any conflict. The station was originally designed for local trains to use the lower level and express trains to use the upper level, so this is nothing new, just a more rigid way of running service than today. Each of the two levels has ladder tracks permitting access to about 10 platform tracks, which is more enough for a train every 2 minutes; for reference, the 4 platform tracks of Haussmann-Saint Lazare on the RER E turn 16 trains per hour at the peak today, and were constructed with the ability to turn 18.
The upshot is that very little station reconstruction is needed at this stage. Some reconstruction is required for through-running, as it would require all approach tracks to go to the lower level, but even that would be much cheaper than the through-running tunnels. But with terminating service, only one switch needs to be changed. This is not expensive; the limiting resource is imagination to do better than today’s slow service.
A few years ago, when I started writing timetables for proposed regional rail lines, I realized how much faster they were than current schedules. This goes beyond the usual issues in Boston with electrification, which can cut the Boston-Providence trip from the current 1:10 or so to around 45 minutes. In New York the trains are already electrified, but trip times are slow, due to a combination of weak rolling stock, low platforms in New Jersey, poor maintenance in Connecticut, and obscene schedule padding in Long Island. This post collects a few before-and-after comparisons of how fast regional rail in New York could be.
Due to time constraints, not all lines are included in this post; by popular demand I can complete this and make it a two-part post. In this post I am going to focus on the New Haven and Harlem Lines and the LIRR’s Ronkonkoma and Hempstead Branches.
The LIRR and Metro-North both have reasonable if conservative equipment. Thus, it is valuable to look at the trip times that current equipment could achieve, that is the M-8s on the New Haven Line and the M-7s on the other lines. Future equipment should be higher-performance, and in particular both railroads should procure modular platforms based on proven European regional rail designs, rather than stick with overweight, overpriced equipment as in the upcoming capital plan. Thus the following tables include trip times with both current equipment and a notional regional electric multiple unit (EMU) with the specs of a Talent 2, FLIRT, Coradia Continental, DBAG Class 425, or similar train.
As a note of caution, these trip times are not achievable at zero cost, only at low cost. No curve needs to be straightened, but some curves need to be superelevated, and in some areas, particularly Connecticut, additional track work is required. All of this is quite cheap based on European maintenance regimes, though perhaps not based on American ones, but it is not literally a day one timetable – figure a few months’ worth of work systemwide. Schedules would also need to be simpler, with fewer creative express patterns, to facilitate low schedule padding, 7% as in Switzerland rather than the LIRR’s current 30% pad.
Much of this work comes from this post about the LIRR and this one about the New Haven Line, but here I’m covering the Harlem
and Hudson Line s as well, and using more recent computations for acceleration.
New Haven Line
Locals to Stamford:
|Station||Current time||Future M-8 time||Future Euro time|
|Mount Vernon East||0:27||0:18||0:16|
Some of the numbers are interpolated, but the end-to-end times as well as those to New Rochelle, Port Chester, and Riverside are exact. No curve is straightened, but all non-geometric speed limits, including those on the Cos Cob Bridge, are removed; the Cos Cob Bridge is not straight enough for high-speed rail, but a regional train could squeeze 150 km/h out of it, or 160 if it is replaced.
Expresses to New Haven are faster, as detailed in my older post on the subject:
|Station||Current time||Future M-8 time||Future Euro time|
Numbers differ from my older post by a minute to allow for slightly slower approaches to the Grand Central stub-end, at 50 km/h rather than 100 km/h as with any future through-running. This is still several minutes faster than the current 10 mph speed limit out to a mile out of the station. It doesn’t matter too much; at the end of the day, this is a difference of 1:18 vs. 2:09, with one extra station. I repeat: better track maintenance, less conservative terminal approach speeds, higher superelevation on curves, modern schedule padding, and (on the margin) higher-performance equipment could reduce trip times from 2:09 to 1:18, a cut of 40% in trip time, without straightening a single curve.
The Harlem Line today runs local and express trains, but this involves a long stretch from north of Mount Vernon West to North White Plains with three and two rather than four tracks; trains just don’t run frequently enough today that it’s a problem, but in the future they will need to. Therefore, my timetable below is all-local. Nonetheless, trip times to White Plains on the local train are comparable to those of today’s express trains.
|Station||Current time (local)||Current time (express)||Future M-7 time||Future Euro time|
|Mount Vernon West||0:32||—||0:24||0:23|
|North White Plains||1:01||0:41||0:44||0:40|
Observe that the current schedule has very long trip times before the end station – 8 minutes from White Plains to North White Plains on the local, 11 from Brewster to Southeast on the express. Southbound, both segments are timetabled to take only 4 minutes each. This is additional padding used to artificially inflate on-time performance, in lieu of the better practice of spacing out the pad throughout the schedule, at 1 minute per 15 minutes.
LIRR Main Line
The LIRR has a highly-branched system, and I’m only going to portray the Main Line to Ronkonkoma among the long express lines. This is because in the long term, the South Side lines shouldn’t be going to Penn Station but to Downtown Brooklyn and Lower Manhattan. The Port Jefferson Branch could benefit from a side-by-side comparison of trip times, but that is partly a matter of electrifying the outer part of the line, a project that is perennially on the LIRR’s wishlist.
|Station||Current time||Future M-7 time||Future Euro time|
|New Hyde Park||—||0:20||0:19|
The fastest Main Line train of the day runs between Penn Station and Ronkonkoma stopping only at Hicksville, Brentwood, and Central Islip, not even stopping at Jamaica; it does the trip in 1:08, a few minutes worse than the M7 could with less schedule padding and small speedups at terminal zones (Penn Station throat slowdowns add 1-2 minutes, it’s not the mile-long slog of Grand Central).
Finally, for local service supplementing the rapid Main Line, we can look at the Hempstead Branch, which under my regional rail maps should keep serving Penn Station along today’s alignment, continuing north along the Empire Connection to the Hudson Line. Today, only a handful of peak trains run between Penn Station and Hempstead – off-peak, Hempstead diverts to Atlantic Terminal. Here are side-by-side schedules, using the fastest peak train as a comparison:
|Station||Current time||Future M-7 time||Future Euro time|
|Country Life Press||0:47||0:34||0:32|
Across the four lines examined – New Haven, Harlem, Main, Hempstead – trains could run about 50-66% faster, i.e. taking 33-40% less time. This is despite the fact that the rolling stock today is already EMUs: the vast majority of the speedup does not come from upgrading to higher-end trains, but rather from running faster on curves as all EMUs can, avoiding unnecessary slowdowns in station throats, and reducing schedule padding through more regular timetables.
The speedup is so great that the Harlem Line could achieve the same trip times of present-day nonstop trains on locals making 14 more stops between Manhattan and North White Plains, a distance of 38 km, and the LIRR could achieve substantially faster trip times than today’s nonstops on semi-rapid trains. In fact, the LIRR could even make additional local stops on the Main Line like Forest Hills and Hollis and roughly match the fastest peak trains, but expected traffic volumes are such that it’s best to leave the locals to the Hempstead Branch and put the Main Line on the express tracks.
Good transit activists in and around New York should insist that the managers prioritize such speedups. If locals can match today’s express trip times, there is no need to run creative express stopping patterns that force trains into complex patterns of overtakes. Just run frequent local service, using the maxim that a line deserves express service if and only if it has four tracks, as the New Haven Line and shared Main Line-Hempstead Branch segment do. With the slowest speed zones sped up, curve speeds raised to the capabilities of modern EMUs (including the conservative M-7s and M-8s), and schedule padding shrunk to where it should be, the suburbs could be so much closer to Manhattan at rush hour as well as off-peak, stimulating tighter metropolitan connections.