I got a bunch of accolades and a bunch of flaming replies over a tweetstorm imagining a bigger, better New York. Some people complained about my claim that subway trains in Brooklyn are underfull; I urge everyone to read my analysis of data from 2016 – it’s still relevant today, as the only big change is that Second Avenue Subway has reduced Upper East Side crowding. The point of this post is to demonstrate where zoning should definitely focus on adding more apartments, to fill trains that are not yet full.
Instead of using the current subway map, let us start with a deinterlined map:
The reason for using this map is that it’s cleaner than the real map, since there is no track-sharing between routes of different colors, and not much route-sharing (one color local, one express). Getting from here to this map is cheap but not free, as it requires certain junction rebuilds, especially on the 2/3. I ask that my commenters resist the temptation to argue over the details of this map, since the point about zoned capacity does not really depend on questions like whether the E runs local in Queens and the F runs express or the reverse.
Where there is capacity
In 2016, three directions on the subway were truly at capacity, surpassing 4 standees per square meter: the 2/3 and 4/5 coming into Midtown from Uptown, and the L. The analysis looks at crowding on trains entering the Manhattan core, so it lumps lines from Queens based on which tunnel they enter from, which underestimates crowding on the E, since it shares tracks with the under-capacity M. Counted properly, the express Queens Boulevard trains should be viewed as near or at capacity as well, the F having 3.33 standees per square meter and the E having somewhat more.
Additional lines with capacity crunches, with about 3 standees per square meter or more, include the A/D coming in from Uptown, the 6, and the Astoria Line (then the N/Q, now the N/W). The 1 and 7 trains have capacity crunches as well in outlying areas: the 7 is overcrowded until it hits the transfer points to the E/F and N/W but has plenty of space in Long Island City, and the 1 is fairly crowded north of the junction with the express trains and then unloads passengers onto the overcrowded 2/3. These areas should not be deemed to have much spare capacity until such time as operations on the subway improve, permitting higher frequency and eventually more lines.
In contrast, the remaining lines have space, often plenty of space. Everything in Brooklyn except the L and to some extent the J/M/Z is underfull: these trains have high frequency as determined by crowding guidelines at the Uptown or Queens end, but in Brooklyn there are fewer people today so the ridership is weaker. The local lines on the Upper West Side both have plenty of space on the trains as well as space on the tracks for more trains if need be. The 7 downstream of Queensboro Plaza has plenty of space, and the local Queens Boulevard trains downstream of Jackson Heights have nowhere for passengers to transfer to an overcrowded express service.
Since I’m relying on data from 2016, there’s no accounting for Second Avenue Subway. Even then, the 4/5 was only the third most overcrowded trunk line entering the Manhattan core, and it’s likely that there’s additional capacity coming from the new line. There’s certainly space on the tracks for more trains on Second Avenue, and one of the goals of deinterlining specifically is to make it feasible to run more service on this line, which currently only runs a train every 6-8 minutes at rush hour.
The map of where New York could add housing
The map excludes parts of Lower and Midtown Manhattan where the highest and best use is commercial rather than residential. But the boundaries there are deliberately crude: Downtown Brooklyn, NYU, and the Meatpacking District are drawn, to avoid excessive fragmentation of the drawn area, while Chelsea and Hell’s Kitchen are excluded as too close to Midtown.
The map also does not look at considerations other than capacity. Some of the highlit areas on the Upper East and West Sides and Lower East Side are already built to very high density, at least on the avenues and major streets; these areas should be the template of how the rest of the city should look. At the other end, East New York has too weak demand for massive construction, especially if everything to its west is upzoned.
However, large swaths of desirable, close-in areas with relatively short buildings are highlit. Rich inner Brooklyn neighborhoods like Park Slope and South Brooklyn are currently built to missing middle density, with a floor area ratio of about 1.5 away from corner lots. A more appropriate floor area ratio in these neighborhoods is 12, corresponding to tapering buildings in the 20-30 story range, as on the avenues on the Upper East and West Sides. Park Slope is half an hour from Midtown by subway, and less than that from Lower Manhattan. The population of these neighborhoods is perhaps 150,000, and should be more than a million given their proximity to job centers.
Subway deserts and future additions
The map is designed to work with more or less the same service as today, maybe with slightly more frequency on lines that could handle it easily (that is, Second Avenue Subway). But what about future service? The L train is overcrowded, but only runs 19 trains per hour at the peak due to electrical limitations, and could go up to 26 with better electrical capacity, or for that matter lighter trains drawing less power during acceleration. Further extensions of Second Avenue Subway could more effectively relieve pressure off the 4/5, to the point of creating more capacity in the Bronx, which remains well below peak population. Commuter rail modernization opens up large swaths of Queens. Decades in the making extensions on Nostrand and Utica fill in the transit desert in southeast Brooklyn, currently served by buses that nominally come every 2 minutes and in practice comes in platoons of 4 every 8 minutes.
As with the map above, a hypothetical map of development sites assuming reasonable subway expansion includes areas that would be unlikely to actually see new development. Williamsburg and Greenpoint may turn into forests of towers given the opportunity, but in neighborhoods like Sheepshead Bay and East Flatbush developers might well stick to the occasional 6-to-10-story mid-rise building that would not look out of place in Paris. In Eastern Queens, the desired density is probably spiky, with clusters of tall buildings around LIRR stations surrounded by single-family houses and missing middle, much like the structure of density in Toronto and Vancouver.
This post is a cautionary note for everyone who proposes, advocates for, or plans public transportation: please avoid numerology. What I mean by numerology is, it’s easy to target round numbers for trip time, ridership, capacity, or cost, but this may not be based on good design principles. Round numbers are memorable, which makes them attractive for marketing, but quite often the roundness percolates from public communications to system design, and then it tends to lead to bad results: excessive amounts of money spent on meeting a particular trip time, useful scope cut from a project to stay under a too tight budget, and general overpromising.
I’m tagging this incompetence because it is always bad, but even people who are generally good may unwittingly engage in numerology. I’m pretty confident I’ve done this in previous posts by accident. So I’m exhorting myself and good transit advocates and not just the usual politicians and power brokers.
10x and tech
The worst numerology that I’ve seen in technology is not specifically in transportation, but in the software industry of the American West Coast, which is obsessed with the concept of 10x, that is 10 times as good as normal. The most common variation of this is the 10x engineer, that is the programmer who gets 10 times the productivity of the average programmer, but (by implication) does not demand 10 times the average salary, or even 1.5 times the average salary.
Thanks to Elon Musk, the same concept of 10x has jumped into the transportation discourse – Musk promises a 10x reduction in construction costs for tunneling. It goes without saying he cannot deliver, but the telling thing here is the origin of the number. It does not come from some deep analysis finding that California’s tunneling costs are about 10 times as high as those of some target best practice, or even as high as those of a new method. (In fact, California is around 7 times as expensive to build in as Madrid or Seoul, the world’s cheapest cities to build in, so 10 is at the limit of plausibility.) Rather, the number came first: innovation in American tech is supposed to come in orders of magnitude, not continuous improvements, so the target was 10x, just as SpaceX’s target for space launch cost reduction is 10x even though so far the reality is maybe 1.5x or 2x.
The primary problem here is overpromising. Factor-of-10 improvements are almost nonexistent. The one example I am comfortable with in transportation is the tunneling costs in New York specifically, and even that is a problem that only emerged with the latest project, Second Avenue Subway Phase 2; Phase 1 and the 7 extension are off by a factor of 6 or 7 off the rest-of-world average (and about 15-20 off the very cheapest in the world), and East Side Access is a problem of overbuilding more than anything so I can’t even give it a specific factor. Many other things in New York are too expensive, but generally by a factor ranging from 1.5 to 3. Cutting operating costs in half, cutting rolling stock procurement costs by a third, and so on are both laudable goals, but 10x rhetoric skips them entirely. Thus comes the secondary problem with 10x-oriented numerology: just as it rounds up factor-of-7 improvements and overpromises a factor of 10, it completely ignores factor-of-2 improvements as they simply cannot plausibly be stretched to an order of magnitude.
It is common in marketing to promise round numbers for schedules: 2-hour trip times, 3-hour trip times, etc. This sometimes percolates into the planning world behind the scenes, leading to planning around discrete trip times in integer numbers of hours.
In France it’s a commonplace that high-speed rail is only competitive with air travel if the trains take 3 hours or less. The reality is very different on two levels: first, mode share is a continuous function of trip time, so the difference between (say) 2:55 and 3:05 cannot be very big. And second, in 2009, rail had a 54% mode share of all Paris-Toulon trips, on which the TGV takes 4:08-4:20, compared with 12% for air; the TGV held its own as far east as Cannes (34%), 5:26 away, and Nice (30%), 5:57 away. The 3-hour rule is alluring and may be true in one specific social class, namely airline and railway managers, but the numerology here makes it easy to stick to it even if the breakeven point keeps creeping up to 3:30, 4:00, 4:30, 5:00.
A more benign example of numerology is the 30-30-30 plan in Connecticut. Governor Lamont has proposed far-reaching investments to speed up trains to take half an hour on each of three segments: New York-Stamford, Stamford-New Haven, New Haven-Hartford. This is more or less feasible: a reasonable level of investment would reduce New York-New Haven to about 1:03 on express trains, with Stamford near the exact midpoint. However, the target trip times remain numerological: there is no obvious reason why 1:00 is so much better than 1:10. So far 30-30-30 has run into resistance from incompetent traditional railroaders, but it’s easy to imagine a future in which the governor approves the plan over their objections, and then has to decide how much money to spend on the final few minutes’ worth of speedup to meet the stated goals.
In contrast with numerology based on round numbers, there is a much more solid planning paradigm based on trip times a few minutes short of a round number. In that case, the trip time is a round number including turnaround time, which makes it easy to run trains on a clockface schedule. Differences like 1:05 vs. 0:55 are not important enough to bother passengers about, but differences in frequency between hourly and every 1:10 are critical – passengers can remember 9:05, 10:05, 11:05, 12:05 much better than they can 9:05, 10:15, 11:25, 12:35. Therefore, the integrated timed transfer plan of Switzerland and the Netherlands aims at trip times that are not very memorable, but that together with connection or turnaround time enable memorable schedules.
In addition to the tech industry’s 10x concept, more traditional cost estimations can suffer from numerology as well. Here it is important to distinguish relative from absolute costs. Relative costs are relative to an already-decided budget; in that case, it is useful to force agencies to stay within their promised costs, to discourage lowballing costs in the future (“strategic misrepresentation” in Bent Flyvbjerg’s language). Absolute costs are about numbers that sound big or small, and in that case, there is no good reason to force costs to hew to a specific number.
In the case of absolute costs, politicians may fit the program to the cost in either direction. Reportedly, the size of the stimulus bill passed by the Obama administration at the beginning of 2019 was designed to be in the hundreds of billions and avoid the dreaded trillion number, even though some of the administration’s advisors argued for $1.2-1.8 trillion. In transportation, I do not know of specific examples, but there is so much political pressure among various people who think they’re fiscally conservative that there’s bound to be pressure to go underneath a round number, in other words a political equivalent of pricing a product at $99 instead of $100.
In the other direction, visionaries may think they’re being bold by making up a high number, usually a catch round figure like $1 trillion for US-wide infrastructure. The numerology here operates on a different level from the relatively small band of just under a limit vs. just over a limit: here the main problem is that the cost figure is arbitrary, and then the list of projects to be funded is chosen to match it. If there aren’t enough good projects, agencies will either bloat the budgets of projects by lading them with semi-related spending, for example bundling a light rail line with street reconstruction and tree planting, or go forward with weak proposals that would otherwise not be funded.
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.
Four years ago I brought up the concept of the small, dense country to argue in favor of full electrification in Israel, Belgium, and the Netherlands. Right now I am going to dredge up this concept again, in the context of intercity trains. In a geographically small country, the value of very high speed is low, since trains do not have stretches of hundreds of kilometers over which 300 km/h has a big advantage over 200 km/h; if this country is dense, then furthermore there are likely to be significant cities are regular intervals, and stopping at them would eliminate whatever advantage high-speed rail had left.
Nonetheless, unlike with electrification, with high-speed rail there is a significant difference between Israel and the Low Countries. Israel does not have economic ties with its neighbors, even ones with which it does have diplomatic relationships, that are strong enough to justify international high-speed rail. Belgium and the Netherlands do – the high-speed rail they do have is already internationally-oriented – and their problem is that they have not quite completed their systems, leading to low average speeds.
The situation in Israel
Israel is a country of 20,000 square kilometers, with about 9 million people. Both figures exclude the entirety of the Territories, which are not served by intercity trains anyway, and have such geography that not even the most ardent annexationists propose to build any.
The country is long and narrow, and the maximum north-south distance is almost 500 km, but the cities at the ends are very small, and the population density in the South is exceptionally low. Eilat, at the southern tip of the country, is a city of 52,000, and is 170 km from the nearest Israeli city, Dimona. A low-speed line for freight may be appropriate for this geography, offering an alternative to the Suez Canal, but there is no real point in investing in high passenger rail speed. For purposes of fast intercity trains, the southern end of Israel is Beer Sheva, less than 100 km from Tel Aviv.
In the Galilee the situation is not quite as stark. The main barrier to intercity rail development is not low population density – on the contrary, the Galilee averages around 400 people per km^2, not counting the Golan Heights. Rather, the physical and urban geographies are formidable barriers: the mountainous topography forces all railroads that want to average reasonable speed to tunnel, and the cities are not aligned on linear corridors, nor are there very large agglomerations except Nazareth, which is about 100 km north of Tel Aviv. A low-speed rail network would be valuable, tunneling only under mountainous cities like Nazareth and Safed, but even 200 km/h in this region is a stretch, let alone 300. Thus, just as the southern limit of any fast intercity rail planning in Israel should be Beer Sheva, the northern limits should be Haifa and Nazareth.
The box formed by Haifa, Nazareth, Jerusalem, Tel Aviv, and Beer Sheva, less than 200 km on its long side, is not appropriate geography for high-speed rail. It is, however, perfect for medium-speed rail, topping at 160 or 200 km/h. The Tel Aviv-Jerusalem high-speed line, built because the legacy line is so curvy that it is substantially slower than a bus, only runs at 160 km/h for this reason – the distance along the railway between the two cities is 57 km and there’s an intermediate airport stop, so the incremental benefit of running faster is small. The Tel Aviv-Haifa line, built in stages in the 1930s and 50s, runs in the Coastal Plain and is largely straight, capable of 160 km/h or even faster. The Tel Aviv-Beer Sheva line is slower, but it too can be upgraded. In all of these cases, the target average speed is about 120 km/h or perhaps a little faster. A high-speed train would do better, but reducing trip times from 40 minutes to 30 just isn’t worth the expense of a new line.
Nazareth is the odd one out among the major cities, lacking a rail connection. This is for both geographical and sociopolitical reasons: it is on a hill, and it is Arab. Reaching Nazareth from the south is eminently possible, on a line branching from the Coastal Railway in the vicinity of Pardes Hanna, continuing northeast along Route 65 through Kafr Qara and Umm al-Fahm, and entering the city via Afula. Modern EMUs can climb the grades around Umm al-Fahm with little trouble, and only about 4 km of tunnel are required to reach Nazareth, including a mined underground station for the city. Continuing onward requires perhaps 8 km of tunnel.
However, so far Israel Railways has been reticent to enter city centers on tunnels or els. Instead, it serves cities on the periphery of their built-up areas or in freeway medians. It would require little tunneling to enter the center of Netanya or Rishon LeTsiyon, and none to enter that of Ashdod or Ashkelon. This is the result of incompetence, as well as some NIMBYism in the case of Rishon. Nonetheless, such short tunnels are the right choice for regional and intercity rail in those cities as well as in Nazareth, which poor as it is remains the center of Israel’s fourth largest urban agglomeration.
What if there is peace?
In Belgium and the Netherlands, there is 300 km/h high-speed rail, justified by international connections to France and Germany. What if Israel reaches a peace agreement with the Palestinians that thaws its relationships with the rest of the Arab world, justifying international connections to present-day enemy states like Syria and Lebanon as well as to cold friends like Jordan and Egypt?
The answer is that the Levant writ large, too, is a relatively small, dense area. The Palestinian Territories have even higher population density than Israel, as does Lebanon. Jordan and Syria, on the desert side of the mountains, are less dense, but if one drops their low-density areas just as one would drop Israel south of Beer Sheva, the box within which to build intercity trains is not particularly large either.
Amman is 72 km from Jerusalem; it’s an attractive target for a continuation of the Tel Aviv-Jerusalem railway at 160-200 km/h, the main difficulty being the grades down to and up from the Jordan Valley. Beirut and Damascus are both about 240 km from Tel Aviv on the most likely rail routes, via the coast up to Beirut and via Nazareth and Safed up to Damascus. The only connection at a truly compelling distance for 300 km/h rail is to Aleppo, which is not large enough and is unlikely to generate enough ridership across the language and political barrier to be worth it.
Egypt presents a more attractive case. Cairo is enormous, and there is a whole lot of nothing between it and the Gaza Strip, a perfect situation for high-speed rail. However, this is firmly in “we’ll cross that bridge when we get to it” territory, as none of the required construction really affects present-day Israeli intercity rail planning. It’s not like the Levantine Arab capitals, all of which lie along extensions of important domestic Israeli routes.
Integrated timed transfers
The Netherlands and Switzerland both have national rail networks based on the idea of an integrated timed transfer, in which trains from many destinations are designed to reach major nodes all at the same time, so that people can connect easily. In Switzerland, trains arrive at every major city just before :00 and :30 every hour and depart just after, and rail infrastructure construction is designed to enable trains to connect cities in integer multiples of half hours. For example, since trains connected Zurich and Basel with Bern in more than an hour, SBB built a 200 km/h line from Olten to Bern, shortening the trip time to just less than an hour to facilitate connections. Every half hour this line carries a burst of four trains in seven minutes in each direction, to ensure trains from many different destinations can connect at Bern at the right time.
I have argued against this approach in the context of Germany, proposing high–speed rail instead specifically on the grounds that Germany is a large country with many pairs of large cities 500 km apart. In the context of the Netherlands, the integrated timed transfer approach is far superior, which is why it is adopting this approach and refining it in ways that go beyond Switzerland’s decentralized planning. Belgium, too, had better adapt the Swiss and Dutch planning approach. What about Israel?
In Israel, timed transfers are essential to any intercity rail build-out. However, a fully integrated approach is more difficult, for three geographical and historical reasons. First, most intercity traffic flows through one two-track mainline, the Coastal Railway. Using advanced rail signaling to permit many trains to enter Tel Aviv at once is fine, but it would not be the everywhere-to-everywhere system of more polycentric countries like Switzerland.
Second, Israeli metro areas are really a mixture of the mostly-monocentric contiguous sprawl of France and the Anglosphere and the polycentric regions of distinct cities of the Netherlands and the German-speaking world. Jerusalem’s agglomeration is entirely Anglo-French in this typology, without significant independent cores, and Tel Aviv and Haifa both have substantial Anglo-French cores ringed by far less important secondary centers. The significant secondary centers around Tel Aviv and Haifa are edge cities within the built-up area that may be near a rail line, like Herzliya Pituah and the Kiryon, but are never independent town centers like the various Randstad and Rhine-Ruhr cities.
And third, Israel completely lacks the large railway terminals of Western countries that built their mainlines in the 19th century. Integrated pulses require one station track per branch coming out of the station, since the point of such timetables is to have trains from all branches arrive at the station at once. Within Germany there is criticism of the Stuttgart 21 project on the grounds that the new underground Stuttgart station will only have eight tracks, whereas there are about 14 planned branches coming out of the city.
So does this mean timed transfers are a bad idea? Absolutely not. Israel Railways must plan around timed transfers at junction stations like Lod, the closest thing the Tel Aviv region has to a German-style secondary core, as well as at future branch points. Entering secondary city centers like Netanya and Ashdod would involve tunnels and els, but more significantly to the national network, these would all be branches, and adding more branches to the mainline would require planning better transfers at the branch points and in the center.
Moreover, Israel still has significant intercity bus service, and most likely always will. Timed connections between buses and trains at outlying terminals like Ashdod are a must, and nationwide coordination of bus schedules to enable such connections is a must as well.
Intercity rail for a small, dense country
The situation in Israel – as in Belgium and the Netherlands – favors a different kind of rail development from that of larger countries like France and Japan. Short distances between major urban areas, frequent stops for intermediate cities, and cities that are not really located along easy lines call for the following design principles:
- The maximum speed should be 160-200 km/h – lines should not be designed for higher speed if that requires more tunneling or bypassing existing mainlines, unless there is a compelling international link.
- All trains should be electric, and run electric multiple units (EMUs) rather than locomotives, making use of EMUs’ fast acceleration to serve many stops.
- Significant cities that do not have rail links or have circuitous links should get new lines, using short tunnels or viaducts if necessary to reach their centers.
- Transfers at junction stations should be timed, as should transfers between buses and trains in cities with significant travel volumes to areas not served by the railway.
- The state should coordinate timetables and fares at the national level and engage in nationwide integrated planning, since a change in one city can propagate on the schedule 100-200 km away.
In Israel, public transportation planners understand some of these points but not others. Rail planning is based on medium rather than high speed; there are some calls for a high-speed train to Eilat, but so far what I’ve seen is at least partly about freight rather than passengers. The state is electrifying most (though not all) of its rail network – but it’s buying electric locomotives as well as EMUs. New rail lines go in freeway medians and on tangents to built-up areas, as if they were 300 km/h lines, rather than low-speed regional lines for which if people have to drive 5 km they may as well drive the remaining 50 to their destination. Schedule coordination is a mess, especially when buses are involved.
Going forward, Israel should aim to have what the Netherlands has, and even more, since the Netherlands has not fully electrified its network, unlike Switzerland. Israel should aim for very high traffic density, connecting the major cities at a top speed of 160-200 km/h and average speed of about 120 km/h, with easy transfers to slightly slower regional lines and to buses. Its cities may not be Tokyo or Paris, but they’re large enough to generate heavy intercity traffic by public transportation, provided the rail network is there.
There’s an interesting discussion on Twitter, courtesy of Adam Batlan, about federal subsidies for capital funding versus operations. It’s become a popular reform proposal among American public transport advocates, who are frustrated with the status quo of federal funding for capital but not for operations. Unfortunately, the proposed change to the status quo – federal funding of operations and maintenance – is even worse than the status quo. The hazards of outside funding sources for operations are considerable and unavoidable, whereas those of outside funding for capital expansion can be mitigated by defining expansion appropriately, to the exclusion of ongoing maintenance.
Why federal funding should only go to expansion
Public transportation has ongoing operating expenses, and capital funding. Ongoing expenses can only change gradually – rail service in particular is dominated by fixed costs, like maintenance, and service changes have little effect on operating costs. This argues in favor of steady funding for operations.
Can federal funding be this steady? The answer is no. The federal government is where politics is. People with serious differences in opinion over issues including the overall size of federal spending, spending priorities, and how sensitive spending should be to economic conditions contest elections, and if one side has a majority, that side will get its legislative way. Nor is this some artifact of two-party majoritarianism. In consensus democracies the salience of a majority is if anything higher – there are big differences in policy, including transportation policy, between the various parties of Switzerland or the Netherlands, as the parties have to deliver results to attract voters rather than relying on polarization and partisan identity.
This kind of politics is very good when it comes to debating one-time capital projects. A center-right government committed to austerity with little attention to climate change, for example Germany since 2005, will not spend much money on rail expansion, and railroads will formulate their plans accordingly. The key here is that planning around maintaining current operations without expansion is not difficult, whereas planning around sudden cuts in operating funding is.
The issue of ongoing capital expenses
Current US policy is for the federal government to fund capital expenses, but not necessarily expansion. Normal replacement of equipment and long-term maintenance both receive federal funding. This is bad policy, because the way agencies respond to changes in funding levels is to defer maintenance when the federal government is stingy and then cry poverty when the federal government is generous.
The most extreme case of this is the state of good repair (SOGR) scam. The origins of SOGR are honest: New York City Transit deferred maintenance for decades, until the system collapsed in the 1970s, leading to a shift in priorities away from expansion and toward SOGR in the 1980s and 90s. There were tangible improvements in the last era, raising the mean distance between failures on the subway from about 10,000 km in 1980 to 250,000 in the 2000s. But this process led to a trend in which agencies would deliberately defer maintenance, knowing they could ask for SOGR funding letting them spend money without having anything to show for it.
By the 21st century, New York’s SOGR program turned into such a scam. The MTA capital plans keep having line items for achieving SOGR, but there are no improvements, nor does the backlog appear to shrink. If anything, throughout the 2010s service deteriorated due to slowdowns, until Andy Byford began the Saving Precious Seconds campaign. The same scam appears elsewhere, too: Amtrak deferred maintenance in the 2000s under political pressure to look profitable for privatization, a Bush administration priority, and when Obama was elected and announced the stimulus, Bush-installed CEO Joe Boardman began to talk about SOGR on the Northeast Corridor as a way of hogging billions of dollars without having to show increases in speed.
The forward solution to this problem is to credibly commit not to fund maintenance, ever. The fix-it-first maxim is for local governments only. The maxim for outside funding should be that any request for funding for maintenance or replacement is a tacit admission the agency cannot govern itself and requires an outside takeover as well.
The issue of frequency
The problem the thread linked to at the beginning of this post sets to solve is that some cities get money to build a light rail line but then only run it every 20 minutes. This, however, is a problem of incompetence rather than one inherent to the incentives.
A long-term revenue-maximizing agency, confronted with an urban rail line that runs every 20 minutes, will increase its frequency to at worst every 10 minutes, secure in the knowledge that the long run elasticity of ridership with respect to frequency in this range is high enough that it will make more money this way. This remains true even for a dishonest agency, which has no trouble maximizing long-term revenue by deferring maintenance and then asking for SOGR money when funding is available.
This fact regarding frequency is doubly true if the trains already run frequently at rush hour and only drop to 20-minute frequency off-peak. Fleet costs are determined by the peak, and large peak-to-base service ratios require expensive split shifts for crews. Therefore, a bump in off-peak frequency, especially from such a low base as 20 minutes, will increase ridership for very little increase in operating cost.
The thread does not mention the issue of connecting bus service much – I got yelled at for proposing half-hourly local buses timed with commuter trains – but there, too, the rule of only subsidizing expansion rather than maintenance or operation leads to good enough incentives. In Seattle, light rail expansion has led to bus service changes designed to feed the trains, increasing bus ridership even as rail service replaces the most crowded corridors.
The bus cuts of (for example) San Mateo County in response to rail expansion should then be put in the same basket of pure incompetence with the light rail line that runs every 20 minutes off-peak. The incentives line up in one direction, but due to such factors as unfamiliarity with best practices and managers who do not ride the trains they run, management goes in the other direction.
The forward solution here is to stick to funding by expected ridership. If the service plan involves low frequency, this should show up in the ridership screen and penalize the project in question, while urban rail lines that run every 5 minutes get funded.
Most urban rail networks in the world use color to distinguish lines, either alone or in combination with line names or numbers. Moreover, most of these networks have different train fleets for different metro lines – for examples, the trains on the Northern line are used only on the Northern line, and the trains on Paris Metro Line 1 are used only on Line 1. The interiors of these trains have static line maps dedicated to the lines they serve. Occasionally, the trains are also painted in their thematic colors, as in Boston. So, why not extend this and not only paint trains in their thematic colors, but also have different art on each trainset, using the thematic color?
A blue line, like the Piccadilly line or the RER B, would use drawings that incorporate the color blue in some essential way. For example, one trainset could depict an endless ocean, one could depict the sky, one could depict glass-clad skyscrapers that appear blue, and so on.
The key here is to make each trainset visually distinct and recognizable. Part of the reason is pure art: it introduces more interesting variability to a mundane activity, serving the same purpose as street sculptures. This exists in Japan to some extent, with public mascots and Hello Kitty trainsets, but this could generalize to every trainset. In a large city, this would require finding several dozen different paint schemes per color, ideally each by a different artist using a variety of styles.
But there’s another reason for this scheme: it makes it easier for passengers to remember which train they were on if they lost something or wish to report a crime. Right now, trains are tracked by model number, which passengers have no reason to remember after getting off the train. In contrast, a heraldic system is easier for passengers to retain, especially if the art covers both the exterior and the interior of the vehicle.
For the latter reason, it’s fine to be repetitive and paint every car in a trainset with the same scheme: passengers can roughly remember if they were near the front or back of the train, so if they lost something on the train, they can give enough information to reduce the search space to maybe two cars. Trainsets on modern urban rail systems are almost always permanently coupled, often in open gangways – even New York permanently couples cars into half-trains and joins two sets at a time to form a train, making it feasible to associate paint schemes with entire sets rather than individual cars.
The choice of art should rely on local history, geography, mythology, and culture whenever possible. For example, in the Eastern United States, one red trainset could depict brilliant fall foliage, but in Europe, trees do not turn red in the autumn so the reference would not be easily understood. In Japan, trees turn red in the spring and not the fall, so a red trainset could be painted with the cherry blossom. While Paris does not associate red with the color of leaves in any season, it was historically a center for impressionist art, so one blue trainset could have an impressionistic painting of foliage depicting it in blue.
Iconic food may be another intensely local element to paint in some cities. Everyone in New York knows what a bagel, a New York-style pizza, and a hero sandwich are, and New Yorkers of all ethnic and social groups eat them. At the deli, the professor and the security guard may well order the same pastrami hero. The same is true of döner and currywurst in Berlin, and bento boxes and yakitori in Tokyo.
Mythology and history add more recognizable symbols that are specific to the region or country. London and Paris may each find famous battles to commemorate, just as London names one of its intercity train stations after Waterloo and Paris names one of its after Austerlitz. An American city, especially Washington, may depict Union troops in the Civil War or the raising of the flag at Iwo Jima. Every major city can find an episode of its labor history to paint on one of its trainsets, in red of course. Mythology can add recognizable elements, such as fire-breathing dragons in red, Poseidon in blue, and pots of gold in yellow. Those elements would naturally look differently in a non-Western city like Tel Aviv or Singapore, but the principle is the same.
Diverse cities especially benefit from being able to depict their various cultural backgrounds, making different trainsets more visually distinct. Paris can paint some of its green and black trains with Arabic calligraphy, New York and Chicago can depict black Union troops with blue uniforms, Washington can depict the March on Washington with a blue sky or green lawn background, London can depict the Windrush and lotus art and Muslim South Asian architecture. These cities are all predominantly Western, but have large and growing minorities from non-Western backgrounds or from backgrounds with different takes on Western cultural production (such as black and Hispanic Americans), and should reflect the majority culture as well as the minorities, treating the transit network as a microcosm of the entire population.
Commercial culture and advertising
The plan should be to keep each design for a long time, potentially the entire life of the trainset, or at least through a midlife refurbishment. History, mythology, and geography all provide themes that are sufficiently long-run to remain relevant over the long life of a train.
In some cases, commercial properties can both be expected to exist for a long time and have well-known thematic colors. Examples include Star Wars and the iconic light saber colors, the best-known Pokemon, Hello Kitty, many superheroes, and the Smurfs. Transit agencies could enter long-term advertising contracts with Disney, Nintendo, and other long-lived corporations producing popular culture, and paint their properties on trainsets.
Advertising on the subway has a long history, and can coexist with painting the train if the regular ads are contained to the usual posters. It’s already spilling into painting an entire train: the Hello Kitty train is one example, but negative examples exist as well, when New York wraps an entire subway trainset in an ad for a television show that will be forgotten in a few years.
This kind of long-term advertising, in contrast, reinforces the recognizability of individual trainsets as no two trainsets should ever be painted with the same property (though trains of different colors may be painted with different Pokemon, or one with Jedi and one with Sith, etc.). Moreover, the paint scheme should be stable over 20 years – temporary modifications to help advertise a new film, video game, TV series, or book in the franchise should cost extra, and potentially be treated as regular ad posters.
However, there should be a limit to commercialization: the majority of subway paint schemes should not be based on global brands, but on local factors. Pokemon is everywhere, but the cherry blossom, recognizable skylines, picturesque mountains, and historical battles are specific to a country or region.
Just as cities often have art exhibits at subway stations, and just as they sometimes paint the trains on each color line with the color it’s named after, subway and regional rail networks can paint trains individually in thematic colors. In the largest cities, like New York and London, this could well involve more than a thousand distinct paint schemes; this is fine – those cities have enough artists and enough inspiration for a thousand trainsets.
Overall, the combination of some commercial properties with various aspects of history, geography, tourism, food, and mythology, curated from the majority group as well as from various ethnic and religious minorities, is exactly the mosaic that makes the city’s culture. One of the two prime reasons to do this is as a tool to help passengers remember what train they were on. But the other one is art, which simultaneously is aesthetic and sends a message: on the train we are all New Yorkers, or Londoners, or Parisians, or Berliners.
After feedback regarding the post I wrote last month about high-speed rail in Germany, here is an updated proposal:
Blue indicates lines that already exist or are under-construction, the latter category including Stuttgart-Ulm and Karlsruhe-Basel. Red indicates lines that are not; some are officially proposed, like Frankfurt-Mannheim and the Hanover-Hamburg-Bremen Y, others are not but should be.
Würzburg and capacity
The primary difference with the older map is that there’s more service to Würzburg, connecting it to Nuremberg, Frankfurt, and Stuttgart, in addition to the already existing line north toward Hamburg.
The reason for the added connections is not so much that they are by themselves great. Würzburg is not a large city. The through-services have some value, but the Stuttgart-Würzburg line saves travelers from Stuttgart or Zurich to Hamburg or Berlin half an hour, which is nice but not a big game-changer. The Frankfurt-Nuremberg connection is likewise of noticeable but not amazing value: Munich-Frankfurt and Munich-Cologne are shortened by about 15 minutes, and Nuremberg itself gets direct service to Frankfurt and points northwest but is only a medium-size city.
Rather, the most important reason for these connections is capacity. Today, the Frankfurt-Mannheim railway is the busiest in Germany; a high-speed line between the two cities is proposed for capacity more than for speed. However, under a more expansive high-speed rail program, this line would soon reach capacity as well. The demand for trains connecting Frankfurt to Basel, Zurich, and Munich in two hours is likely to be high, at least a train every half hour to each. Moreover, all of these cities would be connected with Cologne in three hours, and Stuttgart would be three hours from Berlin and three and a half from Hamburg. Raw demand may turn the Frankfurt-Mannheim trunk into the busiest high-speed rail trunk in the world off-peak, even ahead of the Tokaido Shinkansen and its six off-peak trains per hour in each direction. Moreover, this trunk would exhibit complex branching, in particular entering Frankfurt from either direction for through-service to either Cologne or Berlin and Hamburg.
The Würzburg connections change this situation. Trains from Stuttgart to Hamburg and Berlin do not need to pass through Mannheim and Frankfurt, and trains from Munich to Frankfurt do not need to pass through Stuttgart and Mannheim.
Paris-Marseille fills about two trains per hour most of the day, Paris-Lyon counting both Part-Dieu and the airport fills around 1.5 trains per hour off-peak and 4 per hour at the peak. The TGV averages higher seat occupancy than the ICE, about 70% vs. 50%, because it varies service by time of day and has practically no seat turnover. It also runs trains with more seats, about 1,100 on a TGV Duplex vs. 900 on a single-level Velaro. This means that for the same ridership, Germany needs to run about two-thirds more frequency than France, which for the most part means matching the frequency France runs at the peak all day.
The largest metro region in Germany is the Rhine-Ruhr, with around 10 million people, not many fewer than Paris. It is polycentric, which normally works against a region – passengers are more likely to be traveling to a destination far from the central train station – but in this case works in favor of it, since the east-west network branches and makes stops at all major cities in the region. The second largest region is Berlin, with around 5 million people, twice as many as Lyon and three times as many as Marseille. Comparing this with Paris-Lyon and Paris-Marseille, an all-day frequency of six trains every hour is reasonable, two connecting Berlin to each of Cologne, Wuppertal-Dusseldorf, and the Ruhr proper from Dortmund to Duisburg.
In general, it’s best to think of this system as a series of city pairs each connected every half hour. The following list looks reasonable:
Not counting international tie-ins like Dresden-Prague, Munich-Vienna, or Aachen-Brussels, these lines total around 9,000 km with repetition, so the total service provision over 15 daily hours of full service is to be 540,000 train-km, maybe somewhat less if the weaker lines (especially Berlin-Leipzig) are served with single 200-meter trainsets rather than double trainsets. Filling seats at today’s rate, say with an average trip length of 350 km, requires ridership to be on the order of 250 million a year, which is about twice what it is today, and around two-thirds that of the Shinkansen. Germany has two-thirds Japan’s population, and the proposed network nearly doubles the average speed on a number of key city pairs, so at least on the level of a sanity check, this ridership level looks reasonable.
The half-hourly connections should be timed so that passengers have easy transfers on city pairs that do not have direct trains. For example, there are no direct Berlin-Karlsruhe-Basel or Hamburg-Stuttgart-Zurich trains, so the Berlin-Zurich and Hamburg-Basel trains should have a timed transfer at Fulda. A wrong-way timed connection between one of the Zurich-Stuttgart lines and the Munich-Stuttgart line toward Strasbourg should speed up Zurich-Munich travel, replacing the current slog through Austria.
Frankfurt, the center of the universe
Frankfurt is the most served station in this scheme, making it the key bottleneck: it has six connections in each direction, for a total of 12 trains per hour in each direction through the central tunnel. Berlin, in contrast, is the terminus on eight out of nine connections, so it only gets 10 trains per hour through the North-South Main Line (not counting Gesundbrunnen stub-ends), which has four tracks at any case.
The implication is that the Frankfurt tunnel should be used exclusively by high-speed trains, and regional trains should terminate on the surface. There may be capacity for a few regional connections in the tunnel, but unless they are extremely punctual, one delay would propagate to the entire country. An ICE network running largely on dedicated tracks would not have this problems – delays would be uncommon to begin with. In Berlin, the same is true in two tracks of the North-South Main Line; some regional trains can mix in the other two tracks, as well as on the express tracks of the Stadtbahn.
West of Frankfurt, eight trains per hour travel up the existing high-speed tracks to Cologne. This may be excessive, but six is not excessive given the sizes of the cities so connected. Passengers from all over central and southern Germany would have regular train access to Frankfurt itself as well as to the airport and some of the major cities of the Rhine-Ruhr. This is likely to be one of the two biggest long-distance bottlenecks, alongside Frankfurt-Mannheim, which is to get six trains per hour, two entering Frankfurt from the west to continue to Hamburg and four from the east to continue to Cologne.
Frankfurt’s position is not surprising given its geography. It’s near the center of western Germany’s north-south spine, right between the Rhine-Ruhr and the major cities of southern Germany and Switzerland. To its west lies Paris, two and a half hours away once a high-speed line to the French border opens. Berlin may be the larger city center, but it is located in Germany’s eastern margin, the capital of one historic state rooted in the east; Frankfurt is in a region that has always been denser and more economically developed, and high-speed rail is likely to strengthen its role as its distance from Paris and northern Switzerland is especially convenient by fast trains.
An environmental activist who saw the map asked why it was so thin in northwest Germany, mentioning a continuation of the line from Bremen to Oldenburg and even west to Groningen and Amsterdam as a possibility, as it has proven demand for intercity bus service. This connection may be prudent, I am not sure. My skepticism comes from the fact that northwest Germany does not have very big cities other than Hanover and Bremen, and medium-size cities like Oldenburg, Osnabrück, and Münster do not lie on convenient linear corridors.
Nonetheless, Oldenburg itself could be usefully served by a continuation of Berlin-Bremen or Hamburg-Bremen trains on legacy track. The same is true of a number of lines not indicated on the map, for example Hamburg-Kiel, or potentially some connections from Berlin and Hamburg to cities in Mecklenburg-Vorpommern branching off of the Berlin-Hamburg line. Moreover, among the four lines running on Frankfurt-Cologne, the one that does not run through to either Duisburg or the Netherlands could turn west to serve Aachen and maybe even continue to Brussels. Connections beyond Brussels are undesirable as Paris gets a faster direct link to Frankfurt, and London is a morass of delays due to border controls and Eurostar boarding slowness.
At the other end of the country, tie-ins to proposed tunnels across the Alps may be desirable. The problem is that these tunnels still leave the tracks with tens of kilometers of slow approaches that are not fixable without extensive tunneling. The air line distance between Zurich and Milan is 216 kilometers. The idea that a train could ever connect the two cities in an hour is complete fantasy, and even two hours is a stretch; Switzerland’s plans for the Ceneri and Zimmerberg base tunnels go down to about three hours. Farther east, the Brenner Base Tunnel’s northern portal is deceptively about a hundred kilometers by air from Munich, but half of that distance is across the Karwendel Alps and fast trains would require an entirely new route of complexity approaching that of the under-construction base tunnel.
Whither the Deutschlandtakt?
The Deutschlandtakt plan was meticulously developed over the years with the input of technical rail activists aiming to imitate Europe’s two best intercity rail networks, those of Switzerland and the Netherlands. Detailed maps of service in each region as well as nationwide for intercity trains are available, aiming to have timed connections between medium-speed trains wherever possible. But it is not the right way forward for a large country. With so many city pairs that high-speed trains could connect in two to four hours, Germany can and should build a network allowing trains to run largely on dedicated tracks, interlining so that most lines would see four to six trains per hour in each direction to ensure high utilization and return on investment.
At high service levels, trying to design lines to be utilized in bursts every half hour is not feasible or desirable. It’s more useful to space trains on intermediate connections like Berlin-Hanover to overlie to provide walk-up frequency, as high frequency is useful on short trips and encourages higher ridership. Moreover, key links like a tunnel through Frankfurt can’t really be used in bursts, as activists are pointing out in connection with Stuttgart 21. This is fine: Switzerland’s design methodology works well for a small country whose largest city would be Germany’s 16th largest, and Germany ought to see what France and Japan do that works and not just what Switzerland and the Netherlands do.
Is this feasible?
This high-speed plan does require high investment levels. But this is not outlandish. After fourteen years of stonewalling on climate change, with a flat fuel tax and more concern for closing nuclear plants than for closing coal plants, Angela Merkel has begun showing flexibility in face of massive climate change protests and announced a plan for a carbon tax.
Millennial and postmillennial Green voters lack the small-is-beautiful mentality of aging hippies. I did not see references to high-speed trains at the climate march a week ago (see selected signs on my Twitter feed), but I did see many calls for replacing cars with trains, and few small-is-beautiful signs, just one NIMBY sign against tall building and one anti-nuclear sign held by someone who looked 35-40 and someone who looked 60-70. Felix Thoma pointed out to me that as the Greens’ voter base is increasingly weighted in favor of educated millennials who travel often between cities, the next generation of the German center-left is likely to be warm to a national and international high-speed rail program.
The barrier, as always, is money. But Germany is not the United States. Costs here are higher than they should be, but they’re rarely outrageous – even Stuttgart 21 costs mostly in line with what one would expect such extensive regional rail tunnels to amount to. The core domestic network I’m proposing, that is excluding lines within Germany that are only useful for international connections like Stuttgart-Singen toward Zurich, adds 1,900 km of new high-speed rail, of which maybe 100 km is in tunnel. An investment of 60 billion euros would do it with some error margin.
A green future for Germany requires a network like the one I’m proposing. A green future can’t be one exclusively based around slow travel and return to the living standards of the early 20th century. It must, whenever possible, provide carbon-neutral alternatives to the usual habits that define modern prosperity. Trans-Atlantic travel may be too hard, but domestic travel within Germany is not, and neither is travel to adjacent countries: high-speed trains are an essential tool to permit people to travel conveniently between the major and medium-size cities of the country.
The Metropolitan Transportation Authority has just released its capital plan for 2020-4. The cost is very high and the benefits substantial but limited, and I urge people to look over criticism by Henry Grabar at Slate about elevators and Ben Kabak’s overview at Second Avenue Sagas. Here I am going to focus on one worrying element: the cost of the trains themselves, on both the subway and commuter rail.
I started comparing subway construction costs nearly ten years ago. Here’s an early post on Second Avenue Sagas, hoisting something I wrote in comments. Over here I started writing about this in 2011. Early on, I was asked about the costs of the trains themselves rather than the tunnels, and said that no, there’s no New York premium there. At the time the most recent rolling stock order for the subway was the R160, for which the base order cost was $1.25 billion for 620 cars (source, PDF-p. 34), or about $110,000 per meter of length. Commuter rail was similar, about $2 million per 25-meter-long M7 in the early 2000s and $760 million for 300 M8s of the same length in the mid-2000s. London’s then-current order, the S Stock, cost £1.5 billion for 191 trains and 1,395 cars, around $90,000 per meter of length for narrower trains; Paris’s MP 05, a driverless rubber-tired train, cost €474 million for 49 trainsets, around $140,000 per meter.
But since then, costs have rapidly risen. The gap is still far smaller than that for infrastructure, which New York builds for an order of magnitude higher cost than the rest-of-world median. But it’s no longer a rounding error. Subway rolling stock costs are rising, and commuter rail rolling stock are rising even faster. The latest subway order, the R211, costs $1.45 billion for 535 cars, or $150,000 per meter, for the base order, and $3.69 billion for 1,612 cars, or $130,000 per meter, including options. Commuter rail equipment costs, once about $100,000 per meter of train length, inched up to $2.7 million per car in 2013, or $110,000 per meter, and then rose to $150,000 per meter for the M9 order.
Construction costs: subway trains
The 2020-4 capital plan has showcased even further rolling stock cost escalation. Go to the link for the MTA capital plan again. On PDF-p. 23 there’s a breakdown of different items on the subway, and rolling stock is $6.057 billion for a total of 1,977 cars, of which 900 are 15 meters long and the rest (I believe) 18, for a total of $185,000 per linear meter.
I’ve blogged before about comparative costs of light rail and regional rail rolling stock. In Europe, both still cluster around $100,000 per linear meter for single-level, non-high-speed equipment. There is no apparent premium over early- and mid-2000s cost even without adjusting for inflation, which is not surprising, as the real prices of manufactured goods tend to fall over time. But what about metros? Here, too, we can look at first-world world comparisons.
In London, a recent Piccadilly line order is, in exchange rate terms, $190,000/meter (the trains are 103 m long) – but it includes 40 years of maintenance and spare parts. In Singapore, a recent order is S$2.1 million per car, which is about $70,000 per meter in exchange rate terms. Grand Paris Express’s first tranche of orders costs €1.3 billion for 183 trains totaling 948 cars, each (I believe) 15 meters long, around $120,000 per meter. Metro Report states Busan’s recent order as ₩55.6 billion for 48 trainsets (replacing 140-meter long trains), which is almost certainly an error; assuming the actual cost is ₩556 billion, this is $70,000/meter in exchange rate terms and $90,000/meter in PPP terms (PPP is relevant as this is an entirely domestic order).
In Berlin, the situation is the diciest, with the highest costs outside New York (not counting London’s maintenance-heavy contracts). An emergency order of 20 52-meter trains, tendered because cracks were discovered in the existing trains, cost €120 million, around $150,000 per linear meter. A longer-term contract to supply 1,500 cars (some 13 meters long, most 16.5 meters long) for €3 billion by 2035 is on hold due to litigation: Siemens had already sued over the emergency order of Stadler cars, but now Alstom made its own challenge. But even here, costs are well below the levels of New York, even before we adjust for inflation since Berlin’s future contract is in 2020-35 prices and New York’s is in in 2020-24 prices.
Construction costs: New York-area commuter rail
Commuter rail is faring even worse. On PDF-p. 27 the LIRR is listed as spending $242 million on 17 coaches and 12 locomotives, and on PDF-p. 29 Metro-North is listed as spending $853 million on 80 EMU cars and 30 locomotives.
Figuring out exact comparisons is not easy, because locomotives do cost more than multiple-units and unpowered coaches, and there is a range of locomotive costs, with uncertainty due to currency conversions, as most information I can find about European locomotives is in Eastern Europe with its weak currencies, since Western Europe mostly uses multiple-units. Railway Gazette’s pages on the world rolling stock market suggest that a European locomotive is around €5 million (e.g. the PKP Vectron order), or $6.5 million; PKP’s domestic order (including some dual-modes) is around $4.2 million per unit measured in exchange rate terms, but twice as much in PPP terms; Bombardier has a sale to an undisclosed customer for about $4.8 million. Siemens claims the Vectron costs €2.5 million per unit, although all the contracts for which I can find prices are substantially more expensive.
For what it’s worth, in the US dual-mode locomotives for New Jersey Transit cost around $9.5 million apiece, which is still evidently lower than what the LIRR and Metro-North plan on spending. 242 – 9.5*12 = 128, and 128/17 = 7.5, or $300,000 per linear meter of unpowered coach; similarly, 853 – 9.5*30 = 568, and 568/80 = 7.1, or $280,000 per linear meter of new Metro-North EMU. If we take the normal-world cost of a locomotive at $6 million and that of an EMU or coach at $2.5 million per US-length car, then the LIRR has a factor-of-2.1 cost premium and Metro-North a factor-of-2.2 premium.
The equipment is conservative
The FRA recently realigned its regulations to permit lightly-modified European mainline trains to run on American tracks. Nonetheless, no American commuter rail operator has taken advantage of the new rules – the only ones buying European equipment had plans to do so even before the revision, going through costly waiver process that increased costs. At a public meeting last month, Metro-North’s vice president of engineering did not even know FRA rules had changed. The LIRR and Metro-North are buying the same equipment, to the same standards, as they have for decades.
The subway, likewise, is conservative. It is a laggard in adopting open gangways: the R211 order is the first one to include any, but that is just two test trainsets, the rest having doors between cars like all other older New York trainsets. It is not buying any of the modular products of the global vendors, like Bombardier’s Movia platform or the Alstom Metropolis. It is buying largely the same kind of equipment it has bought since the 1990s.
Despite this conservatism, costs are very high, consistent with a factor somewhat higher than 2 on commuter rail and somewhat lower than 2 on the subway.
But perhaps the conservatism is what increases costs in the first place? Perhaps the reason costs are high is that the world market has moved on and the MTA and some other American operators have not noticed. In Chicago, Metra found itself trying to order a type of gallery car that nobody makes any longer, using parts that are no longer available. Perhaps the same kind of outmoded thinking is present at the MTA, and this is why costs have exploded in the last 10 years.
A secular increase in costs of infrastructure construction is nearly universal. No such trend can be seen in rolling stock: nominal costs in Paris are 15% lower than they were 15 years ago, and real costs are about 30% lower, whereas in New York nominal costs are 70% higher than 10 years ago and real costs about 40% higher. Paris keeps innovating – M1 and M14 have the highest frequency of any metro system in the world, a train every 85 seconds at the peak, and M1 is the first driverless line converted from earlier manual operations rather than built from scratch. In contrast, New York is stuck in the 1990s, but far from keeping a lid on costs, it has seen rolling stock cost explosion.
Update 9/24: I just saw a new commuter rail coach order in Boston. These are bilevels so some cost premium is to be expected, but $345 million for 80 unpowered coaches, or $170,000 per meter, is excessive, and TransitMatters tried hard to fight against this order, arguing in favor of EMUs on the already-electrified Providence Line.