Express Airport Connectors are a Scourge of Public Transit

Earlier this month, Andrew Cuomo unveiled a proposal to spend $10 billion on improvements to JFK Airport, including new terminals, highway expansion, and public transit access. I encourage readers to look at the plan: the section on highways proposes $1.5-2 billion in investment including adding lanes to the Van Wyck Expressway and to on-ramps, and has the cheek to say that this will reduce fuel consumption and greenhouse gas emissions. This while the section on mass transit gives it short shrift, only proposing superficial improvements to the AirTrain; in the unlikely the case that this is built, highway mode share will grow and transit mode share will fall. Put in plainer terms, the environmental case for the plan includes fraud.

However, this is not really the topic of this post. That Andrew Cuomo lies to the voters and doesn’t care about good transportation is by now a dog-bites-man story. Instead, I want to focus a little on a throwaway line in the plan, and more on the Regional Plan Association’s reaction. The throwaway line is that almost every major world airport has a one-seat train ride to city center, and by implication, so should JFK.

As an organization dedicated to environment-friendly public transit, the RPA should have made it very clear it opposes the plan due to its low overall transportation value and its favoring of highways over transit. Instead, the RPA immediately launched a brief detailing possible new airport connectors between JFK and Manhattan. The RPA has a lot of good technical people, and its list of the pros and cons of each option is solid. It correctly notes that using the LIRR and Rockaway Beach Branch would compete for traffic with LIRR trains serving Long Island, although it doesn’t mention associated problems like low frequency. The brief is based on prior RPA proposals, but the timing, just after Cuomo came out with his announcement, suggests an endorsement. There are several intertwined problems here:

There is no no-build option

A good study for public transit should not only consider different alignments and service patterns, but also question whether the project is necessary. The US requires environmental impact statements to include a no build option; European countries require a cost-benefit analysis, and will not fund projects with a benefit/cost ratio under 1.2, because of cost escalation risk.

The RPA study does not question whether a one-seat ride from JFK to Manhattan is necessary or useful. It assumes that it is. Everything else about the study follows from that parameter. Thus, it considers entirely express plans, such as the LIRR option, alongside local options. Everything is subsumed into the question of connecting JFK to Manhattan.

One of the alignments proposed is via the LIRR Atlantic Branch and Second Avenue Subway, which the RPA has long believed should be connected. The brief says that it would be slow because it would have to make many local stops; I’ll add that it would serve Midtown, where nearly all the hotels are, via a circuitous alignment. But with all these stops on the way, shouldn’t this be considered as primarily a new trunk line connecting Eastern Brooklyn with Second Avenue? The question of whether the eastern terminus should be Jamaica or JFK must be subsumed to a study of this specific line, which at any rate is unlikely to offer faster service to JFK than the existing AirTrain-to-E option. After all, the most optimistic ridership projection for a JFK connector is maybe 40,000 users per day, whereas the projection for the full Second Avenue Subway is 500,000. I don’t think a Second Avenue-Atlantic Branch connection is warranted, but if it is, the question of whether to serve JFK at the end is secondary.

Express airport connectors are a fetish

I lived in Stockholm for two years, where I went to the airport exclusively using the Arlanda Express, a premium express link running nonstop between the airport and city center. I imagine many visitors to Stockholm use it, are satisfied, and want to replicate it in their own cities.

Unfortunately, such replications miss something important: any air-rail link must go to the areas that people are likely to want to connect to. For locals who wish to travel to the airport, this means good connections to the local transit network, since they are likely to come from many neighborhoods. Not even a small city like Stockholm worries about providing rich areas like Vasastan and Roslag with a one-seat ride. For visitors, this means a one-seat ride to where the hotels are.

Stockholm is a largely monocentric city, with one city center where everything is. (It has an edge city in Kista, with more skyscrapers than Central Stockholm, but Kista can’t be reasonably connected to the airport). The situation in other cities is more complicated. And yet, express air links prioritize serving a big train station even if it’s poorly connected to the transit network and far from the hotels. Let us consider London and Paris.

In London, the five-star hotels cluster around the West End. Only two are at Paddington Station, and only a few more are an easy walking distance from it. This is where the Heathrow Express and the slower Heathrow mainline trains go. No wonder the Heathrow Express’s mode share, as of 2004, is 9%, whereas other Heathrow connections, mainly the Piccadilly line, total 27% (source, PDF-p. 28). The Piccadilly line beautifully passes through the parts of the West End with the largest concentration of hotels, and last time I was in London, I chose it as my Heathrow connection. Nonetheless, the government chose to build the Heathrow Express.

In Paris, the five-star hotels cluster in the west of the city as well, in the 8th arrondissement. The current airport connection is via the RER B, which offers express service in the off-peak when there’s capacity, but not in the peak, when there isn’t. Even so, it is a local commuter rail service, with good connections to the city transit system, and a two-seat ride to the 8th. Because of slow perceived speeds, the state is planning to build an express connector, originally planned to open in 2015 but since delayed to 2023. The express connector will dump passengers at Gare de l’Est, with no hotels within walking distance, no access to Metro lines serving the hotel clusters (Metro 7 does so peripherally, M4 and M5 not at all), and a long walk to the RER for passengers wishing to connect to longer-range destinations such as parts of the Left Bank.

I bring this up to show that the idea of the express air-rail link is a fetish rather than a transportation project, and by analogy, so is the one-seat ride. There is value in faster service and in minimizing the number of transfers, but express airport connectors attempt both even at the cost of building a line that doesn’t go where people want to go.

Ultimately, Cuomo doesn’t care about good transit

Cuomo has many concerns. The chief one is most likely winning the 2020 presidential primary. He has been running for president since the moment he was elected, and many of his policies – gay marriage, the feuds with Bill de Blasio, the desperate attempt to build shiny infrastructure with his name on it – are best viewed through that lens. To the extent that he is not running for president, he has attempted to cement absolute power within the state. He backed a palace coup in the State Senate that secured a Republican(-ish) majority even though the Democrats won most seats; a Democratic majority would be led by a different faction of the party, one more beholden to Democratic interest groups, and might send Cuomo bills that he would lose political capital if he either signed or vetoed them.

This is why I keep giving him as an example of an autocrat in various posts; here is the major takedown, but see also here. Autocrats are always bad for the areas that they govern, which as two separate implications. The first is that their choice of spending priorities is compromised by the need to expand their own power and glory: even if you believe that New York needs $1.5-2 billion in new highway spending, is the Van Wyck really the best place for it?

The second and worse implication is that it is hard for outside groups to convince autocrats to do better. Autocrats don’t have to listen; if they did, they would be democratic leaders. Cuomo happens to be an anti-transit autocrat, and this means that pro-transit groups in New York need to view him as an obstacle and work to weaken him, rather than to ask him to please consider their plans for an air-rail link.

The difficulty is that, precisely because local- and state-level democracy in the US is so weak, it is difficult for issue-oriented groups to go out and oppose the governor. Planners in Democratic cities are hesitant to attack budget-cutting Republican governors like Charlie Baker and Larry Hogan; attacking Democratic governors like Cuomo is a nonstarter. Nonetheless, the RPA needs to understand that it needs to oppose governments hostile to public transit rather than ask them to improve. When Cuomo proposes a bad transportation project, say “no” and move on to more important things; don’t try to work with him, because nothing good can come of that.


  1. Steve

    The problem with Heathrow express isn’t the route, changing at paddington is simple, its the cost especially for the on the day fare. Having once lived in Hounslow I wouldn’t wish the Piccadilly line on anyone, its the slow train to no-where, I would never use it when arriving at Heathrow.

    • Alon Levy

      I will say that my familiarity with Heathrow options is purely as a tourist. But in Paris the RER B is quite fast. Moreover, it is not difficult to enable express trains at rush hour, since the common RER B trunk line from Gare du Nord to Aulnay-sous-Bois is four-tracked. Two of the tracks are dedicated to Transilien K and longer-range trains, but Transilien K ridership is a rounding error and is so is intercity ridership to Laon, and both can just share express tracks with the RER B. What’s more, the CDG Express plan is justified on grounds of capacity… but the big capacity crunch is between Gare du Nord and Chatelet-Les Halles, and that’s not getting fixed by an express train that goes to Gare de l’Est.

      • michael.r.james

        Alon Levy wrote:

        What’s more, the CDG Express plan is justified on grounds of capacity… but the big capacity crunch is between Gare du Nord and Chatelet-Les Halles, and that’s not getting fixed by an express train that goes to Gare de l’Est.

        I suppose it must remove some, maybe most, of the 8m who currently take the RER-B from the airport. As you pointed out in the article, it is a bit of a slog (through pedestrian underground tunnels) from Gare de l’Est to the RER-B station at Gare du Nord so most people will opt to take M4 or M7 which are right at Gare de l’Est, especially as both give you options to avoid Chatelet. (Both M4 and M5 stop at both Gare du Nord and Gare de l’Est but I’m pretty sure you can walk it too–ie. by tunnel; these are so close together I recall you can see the next station along the line from the platforms.) Remember that M4 is the only line that intersects all other metro lines (so max of one correspondance) and actually M7 is only one fewer IIRC. Hmm, in fact maybe this is another factor in why they have done this termination at Gare de l’Est.

        • Alon Levy

          The north-south lines all intersect the east-west lines, yes… but then air travelers would need a three-seat ride: CDG Express-M4-M3, for example. It eats away all the time saving coming from the express service. It’s nothing like the Arlanda Express, which is expensive as fuck but still so much faster than any other option, even with some backtracking connecting from the Red Line. Of course, I’m in an especially convenient location for the RER B – I live in Nation, so it’s a two-seat ride for me with a cross-platform transfer at Les Halles – but the RER B/A transfer is decent even wrong-way, so service to Etoile is pretty good too.

  2. michael.r.james

    No wonder the Heathrow Express’s mode share, as of 2004, is 9%, whereas other Heathrow connections, mainly the Piccadilly line, total 27% (source, PDF-p. 28).

    While it is true that the Piccadilly line is more convenient for more people, one very real reason for lower patronage of the Express line is sheer cost: US$33 one way. (Typical with British transit there are a dozen conditional fares.) And also British this line was almost certainly built on a “user pays” financing model, but equally naturally the high fares to avoid government subsidy means a lot fewer people choose to use it than otherwise. It’s sole advantage is that it is “express”, ie. fast, while the Piccadilly line is painfully slow and at plenty of times of the day is crowded with suburban commuters, and it has no provision for luggage so can be a real bore. There are 20 stations to reach Piccadilly which is about the centre, and as I recall it takes about 75 minutes. I hate buses but the bus to Victoria Station can be a better option.

    Nonetheless, the government chose to build the Heathrow Express.

    The amazing thing is (well, it’s British transport so not so surprising) that the tracks were there all along! given that (and it was a uninterrupted ROW freight line) I don’t know why HE would be so expensive to build but again … British. The real question is why it wasn’t used from half a century ago.

    Incidentally, your focus on where hotels are is a bit odd. First, you are really focussing on the top hotels and dare I say people who can afford those (especially in London) are not the type to take public transit. Second, though the Piccadilly line intersects about 10 other Underground lines, if you have to make one correspondance then it doesn’t make a whole lot of difference to end at Paddo. (Except I’ll bet in that fine British nit-picking tradition that $33 ticket does not include a free trip on the LU, say the way the RER-B ticket from Paris-CDG includes the Metro so you can change seamlessly. Indeed, given it is a separate company I’d bet the Heathrow Express can’t even sell you the LU ticket so you’ll have to scrabble at Paddo to buy it.)

    Re Paris, I agree with your points about the future Express line: it will make travel from the airport less not more convenient; and will be about as expensive (€24) as the Heathrow Express for the same reasons (they figure they can make the airport travellers pay thru the nose for the cost of building this special line, though there will be little convenience–even in time of travel it will only be a nominal saving to Gare de l’Est, let alone beyond to your ultimate destination.) This will probably drive more people to using taxis and Uber which will make the roads more congested … and so it goes. It is an imperfect solution to the problem of ever-rising suburban commuter traffic on RER-B which they have chosen to prioritise. Perhaps fair enough but I reckon they should have simply made airporters change at Aulnay-s-Bois. The Express terminates at Gare de l’Est because there is no tunnel space southwards without spending a fortune. Same reason it uses G l’Est and not the more convenient Gare du Nord. However if it stops at Aulnay one would have the option of changing to the regular RER-B which traverses all of Paris north-to-south and intersects ten of 14 Metro lines. OTOH people like me will be furious paying €24 just to transfer to the old €10 route! No, not ideal either way.
    BTW, this is about 8m pax per annum.
    But again, in terms of hotels, they are all over the city and again with that cluster you speak of, those people are not going to be using public transit to get there. In fact there are plenty of cheaper hotels around the Gare du Nord/l’Est.

    A shame to hear about Cuomo. Aren’t there already mutterings that he might consider a prez run in 2020?

    • Alon Levy

      Cuomo was mooted as a serious candidate for 2016 before Hillary announced and cleared the field. Before it was clear that she’d run, I thought the main candidates were Cuomo, O’Malley, and Elizabeth Warren, and once Hillary did run, I was glad that because of home state issues Cuomo wouldn’t be her VP.

      I bring up the most expensive hotels because, well, if you’re spending the extra money on a premium airport train, you’re probably not staying at a youth hostel. Of course now that AirBnB is taking over Paris, there are listings all over… but that just makes a train that terminates at Gare de l’Est even less convenient, relatively speaking. Metro 4, 5, and 7 don’t have great coverage of the city. (Speaking of coverage: M1 and M14 don’t have great coverage either, and as Paris’s most important intercity transportation gateway, Gare de Lyon needs more, which means M5 should be rebuilt to serve the station.)

      The reason I do think one-seat rides have extra value to airport travelers is twofold. First, changing trains can be terrifying if you don’t know the city well. Chatelet-Les Halles is the bane of every commuter who has to change there between the Metro and the RER, but it was especially confusing the first time I changed trains there, in 2010. And second, luggage. Maybe London is different, but here, the transfers between surface and underground stations aren’t great.

      The capacity argument re Gare de l’Est is really unconvincing. After all, there are no additional slots south of Gare du Nord until the tunnel gets quadrupled, and so far the region is unwilling to spend 700 million euros on it (more, if more platforms at Les Halles are needed, which they are not). Ile-de-France just generally has weird transportation priorities: some good stuff like M14 and M15, but then wastes of money like M17 and M18, compared with no investment into core RER capacity beyond the RER E extension, and less investment than I’d like into Metro extensions into inner suburbs.

      • michael.r.james

        Alon Levy wrote:

        Metro 4, 5, and 7 don’t have great coverage of the city.

        No, trust me, that is not true. As I mentioned in earlier post M4 and M7 have the best coverage of the city (and both go considerably beyond Paris at both their ends). You’re probably thinking of east-west coverage and line 1 which does carry the most pax (but it covers fewer inhabitants, just more of the commercial areas and government & tourist hotspots and of course La Defense). These are the two lines I would choose to live on.

        Re your other comments, it’s partly a matter of approach. When arriving in a new city, especially a big one and a non-English-language one, I much prefer to trust using its Metro system than anything else, including taxi and especially bus including shuttles. If you can read a map then every Metro system is pretty elementary. (And unlike other modes there are always good maps in every station if not on every train; in Paris there is a Plan de Quartier in each station.) Paris (intramuros) is effectively impossible to get lost in, because if you keep walking in a straight line it will be only a few hundred metres at most before you can see the Metro sign and thus no longer be lost, plus of course an extremely easy means to get back to wherever you want. I may be biased by familiarity but I think Paris is one of the easiest to navigate (and I recall being impressed on my very first journey there especially as I was comparing it to London which does not impress me). It is true that, mostly being built a hundred years ago, street access and some correspondances are not ideal if you are less than able-bodied. Just like your wish to link M5 to Gare de Lyon, some things are just too much work to try to fix.
        Though, to take a bit of a diversion, I am disappointed those experiments with very fast travellators didn’t work out because I reckon they could be a solution to some of these problems besetting dense cities with increasing transit crushes; ie. linking longer distances for pedestrians. For example that would be better to link very high trafficked routes like Chatelet to Leftbank etc. than contemplating more hyper-expensive deep-bore tunnels to double RER-B capacity (and increase yet more the traffic thru Chatelet!).
        As to overall transport planning of greater Paris, I have a certain trust in their ability to do this (I reckon their record is pretty damned good). Remember that Gare de l’Est is getting upgraded to also host the terminus of TGV l’Est. You’ve also nicely described in previous articles how the large number of cicrumferential tramways link up the radial Metro & RER lines; we will begin to see over the next decade or so how these impact on specific line usage. Re M17 & M18 I imagine they are part of an overall urban development project–M17 in the poor north-eastern suburbs which is the fastest growth area but has fewer employment centres; M18 the same for the Universite Paris-Saclay complex and I assume a push for new urban centres and new industries.

        • Alon Levy

          Quadrupling the RER B+D tunnel isn’t that expensive! If it’s done with a hookup to the existing platforms then, per a study from 2003, it’s €700 million. If it follows a fanciful routing with new platforms plus a stop at Republique then it’s €2-4 billion, which is pretty bad, but still cheaper than M17 and M18. The problem is that M17 isn’t really good at linking Seine Saint-Denis to anything – neither to the center (as with M14 and other extensions like M12) nor to other suburban centers (as with M15 and M16). M18 is worse – it’s a circumferential link between two suburbs with nothing in between, and with already pretty good connections at Chatelet and at Saint-Michel; university travel between the Left Bank and Orsay is effectively reverse-peak, or short-of-peak, so it’s free money to STIF without a capacity crunch.

          The reason I say M4 and M7 have poor coverage is that, while they’re good for two-seat rides, they’re bad for one-seat rides. Obviously it’s better than a non-interchange, but the point is that M4, M5, and M7 all serve roughly the same direction of travel. You can transfer… but then with the airport service it’s a three-seat ride. Nobody likes these, especially not people dragging suitcases.

          I trust planners here more than I do the ones in New York and such… but premium airport connectors are a global plague. And there’s really no excuse for M18 when the suburbs served at both ends have more pressing transportation needs (namely, improving RER B capacity by quadrupling the damn tunnel, and connecting Transilien L to the RER).

          • michael.r.james

            Alon Levy wrote:

            Quadrupling the RER B+D tunnel isn’t that expensive! If it’s done with a hookup to the existing platforms then, per a study from 2003, it’s €700 million.

            I assume that is the shared tunnel section (ie. B+D) from Chatelet to Gare du Nord, and I agree it is a bargain if it was really a mere €700m, for 2.5km! It is deep-bore (probably the deepest in Paris) and one wonders how they would even get the giant boring machines into place.

            The thing is that as much as I love the density of the Metro + RER system within Paris, perhaps it is reaching an endpoint. I mean the upgrade of RER-A is at the limits (double-deckers, shortest headways, platform-aligned doors) and now carries 300m+ p.a. That’s why I return to discussion of the Metropole Grand Paris plan: to develop greater Paris (to improve its urbanity and livability) which should also take some pressure off central Paris. I think this is long-term (maybe medium-term) planning for a future Paris which is why you don’t like parts of it. M16 +17 in fact duplicate parts of RER-B route and today’s airport RER-B so it presumably will relieve RER-B (as it pushes further eastwards to serve more banlieusians). M18 is pushing thru what you say is too sparse to be worthy of such a thing but this region has long been designated for great plans–eg. the Saclay complex is probably the biggest agglomeration of research institutions and universities in Europe and they always want more (light) industry (hi-tech, biotech, agri-tech etc) here. Other than connections to RER-B (and C; and the M14 extension which should also relieve RER-B) there is to be a TGV station at Massy-Palaiseau. This is a pretty nice part of affordable suburbia (if you’re into such) and with the billions they are putting into Saclay etc I am sure it will be a growth area.

            Instead of continuing to pour billions into duplicative service over a few km of central Paris, this makes more sense. And I’m fairly sure you are one who has complained about the pampering central Paris receives compared to Seine-St-Denis etc. You should be embracing the changes.

            The reason I say M4 and M7 have poor coverage is that, while they’re good for two-seat rides, they’re bad for one-seat rides.

            Sorry to labour the point, but no, simply not true. Both are extremely useful lines and including one-seat rides. M7 was my line for years (from Villejuif medical centre 5km south of Paris) and it serves many places you want to go, from Leftbank (Jussieu, admittedly need to transfer to M10 for most of Leftbank but it is an easy correspondance), to Marais & Bastille (Sully-Morland, Pont Marie) & Ile St Louis (and thus Ile de la Cité), Chatelet, Pont Neuf/Rivoli, Palais Royal/Louvre, Opera, Grand Magasins, Gare-de-l’Est, Villette. M4 is equally useful: Denfert-Rochereau, Montparnasse and Leftbank (St-Sulpice (for Le Bon Marché); St-Germaine-des-Pres; Odeon, St-Michel-Notre-Dame, Cité); Les-Halles (& Rivoli), Gare-de-l’Est & Gare-du-Nord, Barbas-Rochechouart (Arabe quarter etc) to the Clichy marchés-aux-Puces. Believe me, as a resident (and as a visitor once you have checked off all the must-dos) you spend far time in these areas than you ever do on the Champs Elysée or the Eiffel tower. Notice how M4 connect three mainline train & TGV stations (Montparnasse, de l’Est & du Nord).
            Also let me remind you that the Paris Metro runs more frequently than many other big city systems (eg. London where my abiding memory is waiting on the platforms) so the main issue with what you call “2 seat” or “3 seat” rides, is the correspondance rather than the waiting for the train. So you soon learn certain ones to avoid such as Chatelet-les-Halles, Montparnasse and Concorde. As I said earlier, I don’t think there is another city as easy to get around.

          • Alon Levy

            The problem with transferring in Paris is never the wait times. It’s walking between the platforms. The bigger interchange stations are confusing; I still sometimes get to the wrong exit at Nation when I use anything other than the RER A or Metro 1.

            The various destinations you mention on M4 and M7 are places where tourists visit, not places where they stay. Ile de la Cite has the Notre Dame and the Conciergerie, but no hotels. It’s also entirely commercial, so no Airbnb listings. The Marais is a nice place for tourists to go to, and I sometimes go there for the Israeli shawarma joints, but I don’t think anyone actually stays there. The Opera is entirely commercial – it has the highest job density and lowest residential density in the built-up area; some hotels are nearby, but most are well to the west, closer to Etoile than to Auber. Saint-Michel has some hotels… but then you’d just take the RER B direct rather than transfer between an express link and M4; in general, the Left Bank is more accessible from the RER B than from M4, because of things like Luxembourg Gardens, not to mention M4’s brutally short stop spacing. The RER B/M10 transfer is ghastly, but M10 is the lowest-ridership line in the system, and for good reason.

            I sincerely doubt M17 will relieve anything. It’s a weirdly-shaped line, not really circumferential and certainly not radial.

            Quadrupling the RER B+D tunnel is an investment that would take place in Paris, but not really serve Paris proper. It’s like East Side Access: the money is spent in Manhattan and Long Island City, but the benefits go to the suburbs, and in the late 1990s and early 2000s its champion was George Pataki, who was supporting it to appeal to Long Islanders and not to New Yorkers. Precisely because the RER B has a capacity crunch, with more crowded trains than the RER A, adding capacity would be a major boost for suburban service. Most likely the resulting service pattern would allow incorporating more Transilien H branches into the RER, improving service to working-class suburbs in Val d’Oise.

            The main places where I think the regions underinvests in transit service to Seine-Saint Denis are toward the east and northeast, not the north. The north only really has one service gap, toward the west, and that’s being addressed by M15. But the other sectors aren’t so good. I appreciate the M11 extension and the M1 extension, but they’d still leave coverage gaps, which could be beautifully plugged by M9 and M3 extensions. M3 isn’t very fast, but it hits the CBD better than M1 does; M9 is a bit circuitous, but it has direct service to job centers in the 2nd and a transfer to the RER for points farther west. I’d also be happier if the M11 extension went farther north, serving Noisy-le-Sec and Bondy, but water under the bridge. Really, the issue is that the east sector is served well by RER, but only its more suburban parts; the inner areas, like Montreuil, aren’t so well-served; the northeast is split between lines going in through the north like the RER B and lines going in through the east like the RER E, and M16 will only partly bridge the divide.

            In contrast, I don’t think there are big service gaps within the city to be filled, once Rosa Parks opens. I can see an infill RER B/D station at Ney for the transfer to T3, and maybe M10 to BNF… but the RER C isn’t at capacity, and M10 wouldn’t really do anything the RER C doesn’t already do. I guess M10 connects to Jussieu, but the people at Jussieu don’t really need to go to BNF. There’s the M5 (or failing that, M10) connection to Gare de Lyon, but that’s mainly for suburbanites. Same is true of this crayon: city tunnels, service mainly to the suburbs (exception: the 13th, the poorest part of the Left Bank).

          • michael.r.james

            Alon Levy 2017/01/17 – 01:59

            The various destinations you mention on M4 and M7 are places where tourists visit, not places where they stay. Ile de la Cite has the Notre Dame and the Conciergerie, but no hotels. It’s also entirely commercial, so no Airbnb listings. The Marais is a nice place for tourists to go to, and I sometimes go there for the Israeli shawarma joints, but I don’t think anyone actually stays there. The Opera is entirely commercial – it has the highest job density and lowest residential density in the built-up area;

            No, only some of the destinations are devoid of places to stay. You and I wouldn’t stay in the Marais because we are too miserly but there are plenty of places to stay. (Oh, and we’re not hip enough! Let’s face it, you’re a 12th arrondissement kind of guy 🙂 And the 9th (Opera) has a residential density of almost 27,000/km2 so pretty much the average of all Paris and of course well above arrondissements 1, 2, 4 & 8th. (Hmm, perhaps you meant the 2nd which borders Opera? It still has rez of 21,000/km2.) Re the issue of access from the airport, I’ve agreed right at the beginning that I would prefer to use the old RER-B instead of the proposed new Express line that will terminate at Gare de l’Est, but if it happens then M4 & M7 are very useful lines to get to where a lot of people (and I count a lot of Parisian travellers, not just transient visitors) need to. OTOH if M17 also serves the airport then I am perhaps more likely to take that (and it will presumably be Metro tickets not rip-off express-airporter fares?). But isn’t the point of the new airport express line that it allows RER-B mainline to much better serve those banlieusians? Incidentally there doesn’t seem to be any hint of double-deckers for RER-B or aligned doors, to increase pax handling?

            The main places where I think the regions underinvests in transit service to Seine-Saint Denis are toward the east and northeast, not the north.

            But M16 & M17 are to the East and NE (respectively), yes? Along with M15 (terminates Noisy-Champs where M16 & M11 extension also terminate; I’ll bet the property prices at that future TOD are on the up.). But still I think your point appears to be that the currently denser Petit Couronne areas of the east need more servicing than perhaps the Grand Couronne areas to be covered by M16/17? You may be correct on your various points, and certainly I don’t have enough info to make a judgement. It could be local politics or any number of complicated future urban planning issues. But I hope you write articles on this as (and if) these lines get built.

          • Alon Levy

            When I say the Opera has low residential density, I analyze this at the level of quartier, not arrondissement. I also use actual data on job density, but that’s only available at the arrondissement level; the order is 2 > 9 > 8 > 1 > rest. The least residentially dense central quartier is Saint-Germain, but half of it is a park and a lot of the rest is the Louvre and Ile de la Cite (some employees but not a lot). Excluding that and other quartiers with a lot of parkland (Invalides, Odeon, Champs-Elysees) or railyards (Bercy), the quartier in the 9th that has the Opera, Chaussee-d’Antin, has the lowest residential density, with just 6,424/km^2. The quartier in the 2nd bordering the Opera, Gaillon, has the second lowest density. Then come a bunch of surrounding quartiers in the 1st, 2nd, and 8th.

            I don’t actually know what the fares are going to be on Grand Paris Express. To me it doesn’t matter too much because I have a Navigo, so to me the marginal cost of going to the airport on the RER is zero. My suspicion is that STIF will abolish zoned travel even on single-ride tickets and carnets, but I won’t be surprised if it carves out the airport as an exception, as in Vancouver. In Vancouver, the airport is in zone 2 and accepts regular zone 2 tickets, provided you buy them elsewhere (or buy zone 1 tickets and AddFare to zone 2 at the airport); if you buy tickets at the airport, you’re slapped with a surcharge.

            M16 is indeed in the east. It is precious, almost as much as M11, M14, and M15, but it still provides circumferential rather than radial service. It’s also farther out, whereas the bigger gaps in the upcoming Metro/GPX + RER network are in inner areas like Montreuil. M17 is in the north rather than northeast, and only serves two new stops on the way to the airport: Le Bourget Aeroport, and Triangle de Gonesse. Triangle de Gonesse is in farmland, and Le Bourget Aeroport is sandwiched between an autoroute and a general aviation airport. I guess they’re planning to develop Triangle de Gonesse, but it’s in really poor location: hemmed by autoroutes, right downstream of one of the Bourget runways. It’s about the same distance from the center as Houilles-Carrieres-sur-Seine, which has empty land within walking distance, and where all residential development creates reverse-peak rather than peak travel demand.

          • michael.r.james

            Alon Levy 2017/01/17 – 13:35

            When I say the Opera has low residential density, I analyze this at the level of quartier, not arrondissement.

            Right, but that’s getting extremely fine-grained: Gaillon is only 18 hectares (probably <10 ha minus streets), the smallest quartier in Paris. So, you know, that is hardly comparable to yer average quartier. Stochastic variation could produce that result in random 18 ha selections in any city. About the size of Parc Monceau and about half Parc du Buttes Chaumont, or a fraction of any of the three big cemeteries–so you know, you could obtain a density of zero! Having said that, ok, it is a place with a lot of commerce (bank headquarters). Still, I'm a tad surprised because all those streets (Av. de la Opera; de la Paix, Bvd Capucines etc) appear to be lined with Haussmannian apartment blocks? There are still a couple of large hotels (and a total of 8 in the two quartiers around Opera; and just 50m from the Ritz; only big ones are marked on my map; incidentally also within this area–just fyi so you can increase your hipster index–is the Hôtel de Pourtalès, the exclusive residence that Kim K stays in during the shows and got robbed recently.). The quartiers directly adjoining these are much higher rez density. In any case Gaillon is low but still has 7100/km2 and Chaussee-d’Antin has 6,400/km2 as you say; these are similar to the av. density of San Francisco (the second densest American city) and only exceeded by NYC and Manhattan (not looking at fine grain pic.). Just to point out that one needs to put these "high" and "low" density areas in perspective.

            Re the Metro/RER expansions I would guess that some of those stops are industrial workplaces and/or planned to be future rez. Of course I don't know that part of Paris (ie. Seine-St-Denis and east) but on one of the southern new lines I see a future station at Villejuif-IGR which is the centre of a very big hospitals & bio-med research area (where I worked; biggest Cancer Hospital and cancer research centre in Europe, or used to be), so one can see what they are planning.

          • Alon Levy

            It’s not stochastic – it’s six contiguous quartiers, straddling the 1/2/8/9 boundary. It’s also borne out by arrondissement-level job density data, and by past transportation planning. RATP originally wanted the RER A to be parallel to Metro 1 and have a station at Concorde; this was changed to Auber partly because SNCF wanted the connection to Saint-Lazare but partly because that specifically would serve the Opera CBD.

            The residential density in dense CBDs is indeed pretty high by the standards of non-dense cities. Not just in Paris: in Manhattan, the residential density of community board 1 (TriBeCa and Lower Manhattan) is 15,000/km^2, and that of community board 6 (Midtown and Madison Square) is 10,000. In all cases, the density is rising, quite rapidly for Manhattan CB 1 and slowly for CB 6 and the central Paris arrondissements. This is because housing demand is so high that at the high end, it can compete with office uses; Lower Manhattan also has somewhat of a glut, in that the city and Port Authority keep trying to develop it even though office demand is still centered at Midtown (WTC is half empty).

          • michael.r.james

            Alon Levy 2017/01/18 – 11:59

            OK, I think we have reverted to your argument (from previous articles) about the “need” to densify this part of inner Paris? On this thread you used the argument that there were not many residents nor many hotels for those visitors (travelling from CDG) to require a Metro/RER servicing this district (to be of utility in this argument). I don’t think we are all that far apart but I pointed out that the “low” residential density is only low relative to the rest of Paris, and would be considered high compared to any inner-city area of anywhere in north America outside Manhattan. Further there are actually quite a few big hotels: I don’t have my relevant map with me but last night I counted 8 (Vivienne quartier would add the Ritz and a few more).
            Obviously you have the data in front of you but I thought for clarity (though alas the table will inevitably end up not perfectly legible):

            Arr. Qtr Qtr_Name Pop. Ha. Density (perkm2)
            1st 4th Place-Vendôme 3,044 26.9 11,316
            2nd 5th Gaillon 1,345 18.8 7,154
            8th 31st La Madeleine 6,045 76.1 7,943
            9th 34th Chaussée-d’Antin 3,488 54.3 6,424
            (4 qtrs) 13,922 176.1 7,905
            2nd 6th Vivienne 2,917 24.4 11,955
            9th 33rd Saint-Georges 20,850 71.7 29,079
            (6 qtrs) 37,689 200.5 18,798

            Thus: in the first arrondissement the Place-Vendôme (#4 of 80 in Paris) has 3,044 residents in an area of 26.9 hectares for a density of 11,316/km2. (For comparison Jardin du Luxembourg is 23 hectares; Washington Square Park NYC is 4 Ha and NYC Central Park is 346 Ha.)
            Thus the four quartiers touching Opera have a total area of 176 Ha and density 7,900/km2; while the 6 quartiers that surround the quartier containing Opera comprises 200 Ha with density of 18,798/km2.

            So, we can agree that the Opera district has lower density than many other parts of Paris (av. is 28,000/km2 with highest being the 11th arrondissement of 41,000/km2; the densest quartier is the Folie-Mericourt in the 11th with 45,000/km2, though the Rochechouart quartier in the 9th is close at 44,500/km2). But even the central 4 averages 7,900/km2 while the somewhat bigger district is 18,798/km2 which is 67% of Paris’ overall average. To me this shows that even when business (banks, finance) are concentrated the Opera district still supports high residency, especially when you move a mere few hundred metres away. To you this looks like a problem requiring remediation whereas to me it looks perfectly fine for a healthy city with mixed business, residential and “entertainment” (Opera Garnier, Comedie Francaise, Theatre des Capucines, Opera Comique, Theatre Daund, Th. la Pepiniere-Opera, Th. de la Michodiere, Th. des Varietes, Th. des Bouffes Parisiens, Comedie Coumartin, Th. Edouard VII, Th. Mogador, Th. Athenee L. Jouvet, Casino de Paris–in fact seeing this concentration, these theatres also help explain the displacement of residents). And the grand magasins on the north side of Opera Garnier (Au Printemps and Galeries Lafayette and assorted lesser names along Bvd Haussmann).

            Thus, I would say there is plenty enough residential building and hotels to use any Metro services (eg. M7 via Gare de l’Est; RER-A at Auber, via Chatelet etc.). Don’t you think the Auber (St Lazare) RER was justified? I assume some of the custom of those 8 hotels (and even the Ritz which is probably where the Goldman-Sachs, IMF, World Bank etc masters-of-the-universe shack up when in Paris) are the constant flow of financial types to all those banks etc.
            IIRC, this area is one of the most intensively remodelled Haussmann districts; he ripped up two entire streets of perfectly good housing to build avenue de la Opera. And it shows: it is one of the most photographed and cited examples of his new urbanism. I know you like making your data-heavy arguments but I don’t think you are really serious about changing it?
            It can also support some actual modernism. When I lived in Paris one of my weekend walks inevitably led me thru the back streets from Rivoli-St-Honoré thru the Place du marché St Honoré on my way to Brentanos on Av. de l’Opera. (Luckily an Iranian-Frenchman saved the illustrious bookstore from the Borders/KMart collapse in 2009-10.) In my day the marché/square hosted a hideous raw-concrete 50s hi-rise car-park (with fire-brigade on ground floor); today it hosts the gleaming glass headquarters of BNP-Paribas with the public Passage des Jacobins thru the middle of it. Got to approve of that kind of redevelopment. Not many visitors see it because it is a bit hidden away (though just metres from the St Honoré fashion strip) and the pittoresque and calm streets around the place with their al fresco restaurants feature in many movies (though they unnecessarily avoid showing the ultra-modern BNP building; IIRC there is a Bourne chase scene thru here?).
            I wouldn’t change any of it.

          • Alon Levy

            It’s not about densification at all. I bring up the low residential density in these areas to argue that there must be elevated commercial density, since the buildings aren’t any shorter than in the rest of the city. The point I’m making is that these areas are predominantly commercial, which suppresses their use as Airbnb pads; it’s hard to measure exactly because Airbnb lists single units so any map will look too cluttered, but when I just checked there did seem to be a lower density of listings around the Opera. So that’s not really where tourists are heading, at least not straight from the airport. I bring this up because there really is a benefit to doing RER B to A, even with the annoying climbing of stairs or escalators and scurrying to the other side of the station; Metro 7 just doesn’t serve most demand.

            Of course Auber was justified. If the station were at Concorde, it’d make the problem of poor RER connection to M3 and M9 even worse than it is with Auber. Unfortunately, my two sources on RER A ridership by station disagree strongly on how many people use Auber – one RATP study says 24 million a year, but the RATP Metro + RER data by station says 6 million. I’m fairly certain the higher figure is correct; it matches facts on the ground about capacity, which say the most crowded segment in the morning rush is Chatelet-Les Halles -> Auber. From SNCF data, the top stations for afternoon peak boardings, broken down by line, are Gare de Lyon on RER D and Gare du Nord on RER B, but Haussmann is third, and all of the numbers count transfer passengers, even if they transfer across the platform. So someone who connects A -> D at Gare de Lyon counts as an RER D boarding at Gare de Lyon, someone who connects B D at Gare du Nord counts as an RER B or D boarding at Gare du Nord, etc. Just because of how nasty Metro-to-RER transfers are, there are bound to be way fewer transfers at Haussmann than at Gare de Lyon or Gare du Nord, which means that it’s almost certainly the single busiest station for afternoon boardings on one line.

            Hard to say where I’d prescribe building the most residential density in the city proper. The most commercial density has a clear answer, namely, Les Halles and points west through Auber, but residential density is harder. The problem is that ideally the maximum density should be in the most in-demand area; unfortunately, in Paris that is Saint-Michel, one of the few parts of the city with genuine historic conservation value. Etoile could stand to have taller buildings – I have a post in my queue (maybe here, maybe on Streetsblog) about how unwalkable that place is, especially when compared with superficially similar Nation. The building height to street width ratio there is really low, nothing like the pleasant enclosures on medium-size streets like Saint-Michel, Saint-Jacques, Rivoli, everything that comes out of Nation except Cours de Vincennes, etc.

          • michael.r.james

            Alon Levy 2017/01/19 – 06:40

            I bring this up because there really is a benefit to doing RER B to A, even with the annoying climbing of stairs or escalators and scurrying to the other side of the station; Metro 7 just doesn’t serve most demand.

            Of course–for now–because travellers would sensible stay on RER-B to get closer to their destination, such as StMichel-Notre Dame for Leftbank (or Luxembourg), or for a Metro interchange station. However we were discussing about after the planned Airport Express comes in and dumps travellers at Gare de l’Est. Then I’d be surprised if quite a lot don’t transfer to M7 since it does service lots of places (and lots of interchange stations).

            Incidentally here is a letter (about three weeks ago) to an Australian Travel mag that highlights the problems and traps:

            WAITING GAME
            As I was going to Paris for the first time, I booked a pick-up through Parishuttle, which would pick me up at my terminal and take me to my accommodation.
            On arrival I had to call them and advise them which terminal to pick me up from. When I called I was told they would be there in 30 minutes. Three more phone calls and two hours later, they finally arrived. This after a 23-hour flight. No explanation as to the delay nor any apology.

            The following week another person responded with similar complaints. The best advice, bien sür, is to just head straight to the RER station in T2. She would have saved about 90 minutes, probably €100 and an awful lot of anxiety. But I’ve noticed that fancy travel sections of newspapers etc never recommend the simplest, cheapest and most reliable method. Here are the current costs of alternatives (since the govt imposed fixed fares on the taxi industry in Jan 2016.):

            The price is 50 euros between Paris-Charles de Gaulle airport and destinations on the Right Bank, and 55 euros to the Left Bank.
            UBER estimate the same journey (for example to St Germain) would cost around 33 euros. For a shared Uberpool, the price would be about 18-25 euros per passenger. Depending on the time of day, UBER can be cheaper than a taxi, but keep in mind that UBER drivers can’t use the bus lanes like normal taxis can (which bypass traffic).
            To use a private cab transfer ( up to 4 people could travel from CDG airport into Paris for 100 euros, and up to 8 people in a private minivan for 139 euros. Good if you are travelling with friends or family.

            Just like that letter writer, these options have risks. If you share a cab or Uber with strangers, you won’t know when your destination will prioritized. I remember taking shuttles in the US in which I sat for an hour or longer while we toured suburbia/exurbia for the other sharers before finally heading to the city centre for me. Oh, and of course waiting around at the airport rank because the driver wants to cram the vehicle full. I’m sure I’ve been in vehicles which were illegally overloaded.

          • michael.r.james

            Alon Levy 2017/01/19 – 06:40

            all of the numbers count transfer passengers, even if they transfer across the platform

            I assume that pax are counted only upon entry since there is no mechanism, except on RER, to count on exit. So crossing the platform is not measurable (including RER to Metro because you can’t tell if they are using the Metro or just exiting the system). Unless you are saying all the stats are compiled by having men with click-counters and clipboards doing surveys?

            Etoile could stand to have taller buildings – I have a post in my queue (maybe here, maybe on Streetsblog) about how unwalkable that place is, especially when compared with superficially similar Nation. The building height to street width ratio there is really low, nothing like the pleasant enclosures on medium-size streets like Saint-Michel, Saint-Jacques, Rivoli, everything that comes out of Nation except Cours de Vincennes, etc.

            Yes, but I wouldn’t change it. Etoile and its grand avenues and perspectives were in place well before Haussmann and he left them intact for good reasons. Plonking a Tour Montparnasse anywhere near it would simply destroy one of the magnificent cityscapes in the world. I say that even as a Parisian I hardly ever went up there. The only reasons Parisians go there is for the wide-screen cinemas, the FNAC & Virgin Megastore and perhaps Sephora (no, not really) and perhaps the Drugstore (again, no not really, it is packed with tourists not to mention poseurs willing to overpay). Oh, and the most common reason: as reluctant tour-guide to visitors. While I agree it is not exactly the best flaneur territory, in fact the sidewalks are exceptionally wide and there is an outer narrow “street” (can’t remember what it is called) around the Etoile (ie. inside the main traffic chaos) that makes it easier enough (if, I suppose, a bit slow). The height/street-width is low naturally because the streets are so wide; and entirely intentional. In fact IIRC most of the buildings have more floors (if not higher absolute ht) at 9-10 than most of Paris as on your “the pleasant enclosures on medium-size streets like Saint-Michel, Saint-Jacques, Rivoli” (6-7-8 floors), but I wouldn’t want the overall height limits to change. I’m confident neither would Le Notre or the Baron.
            Other than the obvious in-fill (eg. in the 13th) the best bet for new housing in Paris is displacing the railway marshalling yards–as in Batignolles’ Martin Luther King development–and eventually over whatever tracks will remain. Though, speaking of Etoile, is Anne Hidalgo pushing thru with her notion of building affordable housing on all that excessive green space on Avenue Foch (and most expensive residential real estate in the city).

          • michael.r.james

            Alon Levy 2017/01/19 – 06:40

            Hard to say where I’d prescribe building the most residential density in the city proper. …
            Etoile could stand to have taller buildings – I have a post in my queue (maybe here, maybe on Streetsblog) about how unwalkable that place is, especially when compared with superficially similar Nation. The building height to street width ratio there is really low, nothing like the pleasant enclosures on medium-size streets like Saint-Michel, Saint-Jacques, Rivoli, everything that comes out of Nation except Cours de Vincennes, etc.

            Amazing timing. Take a look at this if you haven’t already seen it, from today’s CityLab newsletter:
            (This is a taster/report of the primary source–Haussmanhattanblog–which I dare not click on, in dread of what it would do to my meagre mobile broadband limits; later.)

            New York’s Early Skyscrapers, Transplanted to Paris
            The Haussmanhattan blog shows what Manhattan-On-Seine might have looked like.
            FEARGUS O’SULLIVAN, Jan 19, 2017

            You could spend all day dreaming about a mashup of the French capital’s grand boulevards interspersed with Manhattan’s skyscrapers—or you could jump head-first into a beautifully vintage alternate reality created by Paris architect and photographer Luis Fernandes and shared on his blog, Haussmanhattan. Using old photos of both cities to create striking collages of urban landscapes that seem strangely real. In Haussmanhattan’s elegant visual creations, the Flatiron Building reappears on the prow of the Île de la Cité, the Woolworth Building pokes out from clouds next to the Eiffel Tower, while the neck of the Invalides’ dome is stretched into a cloud-scratchingly high cylinder towering over the city.

            Luis Fernandes read your mind by placing the Met Life building on the Champs Elysées. I also liked the Woolworth building forming a twin with the Eiffel Tower on the Champs de Mars. There’s even a mock-up with the tips of both towers poking thru low fog. But the CityLab author made a good point re how Paris would tend to look like Moscow-sur-Seine:

            The results are eerily effective thanks to the careful choice of black and white photos. It also sticks to Manhattan landmarks finished well before World War II, whose masonry facades blend more easily into Parisian surroundings. Curiously, the result of this combination of aesthetics of 1920s Paris and 1920s Manhattan turns out to be somewhat familiar to people who have visited the Eastern Bloc. The combination of broad European-style boulevards with tall towers still clad in historicist embellishments at times looks rather like Moscow or Kiev, where elaborate Stalinist towers hover over wide, tree-lined avenues. Haussmanhattan’s hybrid city might look like a delightful place to explore, but most of us will be glad that it’s striking combination of Baron Haussman and Ernest Flagg remains fictional.

            The only disappointment is no Trump Tower. But actually Fernandes’ fantasy ’20s version doesn’t wreck Paris as much as post-war redevelopment would have, as we can see from the Tour Montparnasse, Porte Maillot, the Jussieu university admin tower or the medium-hi-rise in the 13th and Front-de-Seine.

    • Matthew

      King’s Cross to Heathrow via Picadilly is about 55 minutes, not 75 minutes. So annoying (especially on a Tube train) but not unbearable. But yes, the cost of Heathrow express is truly bizarre for someone coming from outside the UK, but sadly typical of Britain and its broken transport policies. Crossrail will build yet another link to Heathrow, not sure how much better that will be, if at all.

      The real answer is: use Gatwick instead.

      Gatwick also has its somewhat quirky 30-minute every-15-minutes Express train that terminates at Victoria station. This is not so bad for many connections, and the cost is reasonably normal (for Britain), but I never use it. Instead, I just take the regular Thameslink commuter/regional train, which also takes about 55 minutes from St Pancras to Gatwick, running 3-4 times per hour, and has proper seats. According to Google the Tube trip from St Pancras to Victoria + Gatwick Express might save me 10 minutes, but I don’t really see it as being worth the trouble. Of course, the recent Southern troubles might have me thinking twice…

      Alon is, of course, completely correct about Cuomo and the terrible way things are done in NY. I would say that as far as Britain is still considered to be “in Europe” though, I’m not sure that the thinking is so much better here. Local government is thinking about spending £200mil on a project with a benefit/cost ratio less than 1, and the central government is just drooling £billions over very questionable motorway schemes, seemingly everywhere you look. Outside of London, public transport is operated with very poor practices, zero planning, high fares and a great deal of fragmentation. It’s more politics than transport. In the current state of things, councils are able to acquire permits to bulldoze people’s homes, but they cannot tell bus companies how to do anything, no matter how stupidly and senselessly those bus companies operate.

      • Alex B.

        I’ve used the Gatwick Express before, and I’d say it has two major benefits as a tourist. One is the reasonable cost (compared to the Heathrow Express) and the other is simplicity. For someone visiting London, it wasn’t that hard to navigate the Underground. But without an Underground connection to Gatwick, the simplicity of the Express train worked quite well.

      • michael.r.james

        The real answer is: use Gatwick instead.

        Not an option for almost all the international airlines I fly with (or want to fly with). Other than Emirates and BA, it is mostly LCC and short-haul to Europe or North Africa. Not a single US carrier uses it.

        But anyway, the real alternative to Heathrow is to avoid London altogether. If you really must go to the city then I suggest flying into Paris-CDG (or alternatively Brussels or Amsterdam), spending the day in the city and catching the evening Eurostar.

          • michael.r.james

            Norwegian uses Gatwick for flights to the US.

            I said “US carrier”, not carrier to the US! (Many US travellers will stick to their major carriers. Norwegian is an LCC, perhaps one of the better ones?)

    • threestationsquare

      “The amazing thing is (well, it’s British transport so not so surprising) that the tracks were there all along!” No, they weren’t. The Heathrow Express project required building brand new tunnels under the active airport runways, and underground stations at Heathrow Central and at Terminal 4. The Great Western Main Line tracks from Airport Junction (near Hayes & Harlington) to Paddington were of course already there, but the Heathrow Express project did include electrifying them. Of course, the fact that Heathrow Express was such a relatively difficult project (given its meagre usefulness) makes the decision to pursue it all the more bizarre.

      • michael.r.james

        threestationsquare wrote:

        No, they weren’t. The Heathrow Express project required building brand new tunnels under the active airport runways, and underground stations at Heathrow Central and at Terminal 4.

        Well, yes they were. The tracks from Paddington station to the Heathrow site were there. What you’re talking about is the “last mile” in which they didn’t plan for it. Typically British. While every other comparable city (Paris, NYC, Tokyo etc) were building new airports well outside their congested cities, London persisted with Heathrow and continue to suffer the poor planning (or zero planning). The cost of the 3rd runway is more than the cost of brand new 2-runway airports elsewhere. CrossRail is 40 years late too (modeled on Paris RER-A opened in 1977!).

        • Gag Ha;frunt

          We did build new airports well outside London, namely Gatwick and Stansted, but the government has allowed Heathrow to keep expanding at the same time instead of funnelling all new traffic to other airports

          As for Heathrow Express, you seem to be under the impression that there was already a line “to the Heathrow site”, built for the purpose of serving the airport, and Heathrow Express simply brought the tracks closer to the terminals. This is incorrect; the Heathrow Express line connects directly to the Great Western Main Line. Everything between Airport Junction and the airport, including the junction itself, was built new and is owned by Heathrow Airport.

          • michael.r.james

            We did build new airports well outside London, namely Gatwick and Stansted

            Yep, perhaps both those might have been a viable option; one about 50km south and one about 50km north of central London (while closing Heathrow), each would still be serving a catchment of about 15m+ people. Build dedicated fast train connections with CrossRail/RER-like line across/inder London etc. (Gatwick has always been ok to reach though you can only catch it at Victoria; but the few times I used Stansted I found it an immense bore from either London or Cambridge. I imagine a lot of people in Cambridgeshire and Norwich still traipse to Heathrow for lots of flights.). Incidentally all this could have been done cheaper than the absurd sums they spent on Heathrow’s second runway and now the third (current estimate is £18 billion!). Today, expansion anywhere including at Gatwick is opposed, takes forever and costs a fortune.
            OTOH I think history shows a single mega-airport works best for a mega-city. And just make sure transport to it is superb. For all his clownishness I think Boris was correct about a single airport at the Thames estuary. Like when Paris created CDG in the late 50s they made sure it had enough space (including flight corridors that minimize disturbance of most of the city) for whatever the future brings. Or when JFK was created at about the same time. Or Tokyo’s Narita. Or Berlin closing down all but one of its three airports to create a single big one (further out).

            Of course sensible planning has been missing for half a century in the UK, and the privatisation of all three London airports into a single monopoly entity was …. unspeakably awful. But only what we have come to expect from neo-liberalism where competition is for the populace while extraordinary monopolistic profits is ok for private companies, and even worse, privatised formerly public ones (indeed they’ll continue to receive public subsidy!). It made any logical planning difficult and even today the privatised Heathrow won’t be funding the third runway; wonderful.

            you seem to be under the impression that there was already a line “to the Heathrow site”, built for the purpose of serving the airport,

            No, that is not at all what I said.
            And ok, “the last 5 miles” not “last mile”.
            I see there is speculation that London CrossRail may use the tunnels and track & platforms, and HE may be discontinued.

          • Alon Levy

            What do you mean, the government didn’t funnel traffic to other airports? For 30 years there was a rule prohibiting most airlines from flying between the US and Heathrow.

            The easiest thing the state can do to reduce airport congestion in London is join Schengen and get rid of the passport checks on Eurostar. Between the passport checks and the security theater, train travelers between London and Continental destinations lose half an hour. Paris is still one of London’s top air travel destinations, even though in-vehicle train travel time is short enough to eliminate air service as on Paris-Brussels and Paris-Lyon.

          • michael.r.james

            Alon Levy 2017/01/17 – 12:55

            For 30 years there was a rule prohibiting most airlines from flying between the US and Heathrow.

            Part of that was an attempt to take pressure off Heathrow, but it shows the problem of trying to run more than one international airport in the same city. When they could, all the US airlines relocated to Heathrow.

            The easiest thing the state can do to reduce airport congestion in London is join Schengen and get rid of the passport checks on Eurostar. Between the passport checks and the security theater, train travelers between London and Continental destinations lose half an hour.

            Agreed. But it has been pretty efficient any time I used it, certainly compared to flying. Eurostar has about 75% of the London-Paris market and I’m not sure the rest is bus (which has become absurdly cheap). ie. I don’t know why anyone would fly. In my files is this bit of comparison:

            It estimates that the London to Paris journey on its trains takes 2hrs 47mins (including check in, security, customs and journey time) while the plane service would end up taking 3h 55mins of a passenger’s day (this includes averages for transfers, check-in, security, customs, baggage reclaim and journey time).

          • Matthew

            How long ago did you last fly Stansted? There is a Stansted Express leaving every 15 minutes from Liverpool Street. From Cambridge, there are two options: the Crosscountry train runs every hour from Birmingham to Stansted via Cambridge, and Abellio Greater Anglia runs a dedicated Stansted train from Cambridge every hour during weekdays (this will, in the future, be extended into a Norwich to Stansted train via Cambridge). Both take about 35-40 minutes to complete the journey from Cambridge station.

            Stansted is by far the most accessible London-area airport to people coming from Cambridgeshire and probably from Norwich as well. I don’t know anyone who would rather hike to Heathrow, which is either a 3 hour bus ride or a 2 hour train ride away from Cambridge.

            Gatwick is well served by commuter/regional rail trains running via Thameslink, albeit slower. You don’t have to “go to Victoria”.

          • michael.r.james

            Matthew 2017/01/20 – 06:23

            a dedicated Stansted train from Cambridge every hour during weekdays (this will, in the future, be extended into a Norwich to Stansted train via Cambridge). Both take about 35-40 minutes to complete the journey from Cambridge station.

            Every hour, huh? I rest my case.

            I don’t know anyone who would rather hike to Heathrow, which is either a 3 hour bus ride or a 2 hour train ride away from Cambridge.

            No one would do it by preference. They would do it because most international airlines don’t like to fly out of all these “international” London airports so you get a restricted choice at Stansted. Most major airlines prefer to operate out of Heathrow. That was my point.
            If I lived/worked in Cambridge I would still traipse to Heathrow because I would want to fly Cathay Pacific to HK/Australia etc. Or SIA via Singapore etc. (I only fly when I can’t rail it so am not interested in any of those LCCs to Europe that use Stansted & Gatwick.)

            Gatwick is well served by commuter/regional rail trains running via Thameslink, albeit slower. You don’t have to “go to Victoria”.

            Correct. My (implied) point was that the Gatwick train terminates at Victoria, as compared to Paris RER-B which traverses Paris and connects with almost all the Metro and RER lines. In fact in all 4 places I lived in Paris, across 4 arrondissements from north to far south, it happens I was always within walking distance of RER-B; I suppose it spoiled me for ease of getting to airports. (Indeed Paris made me very intolerant of London on almost every count you can think of.) About 40+ years later this is finally what CrossRail will do in London for Heathrow. (London CrossRail is explicitly modelled on Paris RER-A opened in 1977.)

          • threestationsquare

            > “Correct. My (implied) point was that the Gatwick train terminates at Victoria, as compared to Paris RER-B which traverses Paris”
            The tourist-trap Gatwick Express does, but Gatwick is also served by Thameslink trains that run across London just like the RER, normally stopping at London Bridge, Blackfriars, City Thameslink, Farringdon, St Pancras, Kentish Town, West Hampstead, and continuing on to Bedford. (Though due to the in-progress total rebuild of London Bridge these trains are currently skipping it and using a somewhat slower approach to London; they should be back to normal within a year I think.)

            When I lived in Cambridge my preference order was:
            1. Stansted (reached by twice-an-hour direct train); I would sometimes change in Dublin on the way to the US in part so I could fly through Stansted since it’s so much more convenient from Cambridge.
            2. Gatwick (reached by twice-an-hour express train to King’s Cross and then frequent Thameslink train from St Pancras straight across London to Gatwick).
            3. Heathrow (reached by express train to King’s Cross and then interminable cramped ride on the Piccadilly Line).
            (I never happened to have any reason to use Luton while I was there, and only used City once.)
            The premium Heathrow Express and Gatwick Express were never any use to me; even if I hadn’t been put off by the high fares, the better connections available via the through-running Piccadilly and Thameslink trains made those much more convenient. This is exactly what Alon predicts will happen in Paris; most travellers will keep riding the RER B from CDG because of its better coverage and connections, and an express train to Gare de l’Est will have difficulty competing even at equal fares much less premium ones (though some small number of snobs and confused tourists will presumably use it).

          • Matthew

            threestationsquare has already pointed out that the Thameslink trains do what the RER does and serve Gatwick frequently. So I’m not quite sure why you don’t seem to realise that.

            I will go further and note that Cambridge is not Paris. Cambridge has about 120,000 people, a small city. Growing, perhaps, but far to go. We are fortunate enough to have 2 trains per hour serving Stansted airport (plus a third if you count the connection through Bishop’s Stortford). As a small city on the fringe of of an enormous urban region, we’re fairly well served.

            In fact, as the Thameslink upgrade finishes in 2018, we will also enjoy additional trips to London with one-seat through-service to Gatwick Airport and Brighton.

            Now it seems that some of your attitude is due to a disdain for “low cost carriers”. Well, I suppose we are not all so fortunate to be able to drop hundreds of pounds or euros on a trip by rail when a few dozen can be spent on an equivalent flight. I was certainly happy to take Eurostar to Paris last year but I paid a premium to do so. There are times when the Eurostar fare is not so bad, and I believe that most people would happily pay a bit more for its convenience, but most of the time the flights are ridiculously cheaper. When I moved to Europe I had been under the impression that I would get to use the train system to travel the continent, but instead I found that I largely cannot afford to do so, and so wound up on the LCCs instead.

            Having said that, up til now my experience with trans-Atlantic flights has been that European low cost carrier Norwegian is far superior to the United States-based carriers. So I will happily go to Gatwick to get on a modern 787 for a lower price rather than go to Heathrow to get on a cramped ‘time machine to the 1990s’ (my most recent experience flying from Heathrow) for a higher price. Perhaps the ‘brand name’ doesn’t fit your lifestyle. So be it.

            I probably go to airports a lot more than the typical UK resident, if only because I’m a foreigner to begin with, so it looms larger in importance to me. For most people, for everyday travel, there are many things wrong with the UK transport system in general, but you seem to be picking fights that are somewhat off-target. There’s the fares (astronomical!), the complete lack of coordination or integration, the disaster that is privatisation, difficulty of obtaining information, and apart from London (mostly) the general lack of competence at operating public transport. Among many other things!

          • michael.r.james

            Matthew 2017/01/20 – 19:42
            “Norwegian is far superior to the United States-based carriers.” Hah, anything is superior to them!
            My avoidance of LCCs is because I am not convinced, whenever I have made the comparisons, that they really offer what they claim. When I lived in Europe I quickly adopted VA (an early kind of better version of LCC) for transatlantic flights (though if I was sponsored by my US hosts I was often obliged to fly with an American carrier; this is NIH policy). I haven’t live in Europe for a long time now so maybe I would re-examine things, such as Norweigan though really I find the LCCs are disruptive to the industry which has been precarious for too long. Take the effect on crowding in seats in economy and the subsequent crush.
            I am also going to annoy you again by doubting your claims about train travel versus airline within Europe. I strongly suspect you are not comparing all your costs. The trains get you from city centre to city centre, while many LCCs use very distant and inconvenient airports often greater than 50km from your destination city. False economics and not just measured in dollars.
            I should also admit that I developed an aversion to flying. Sad, because I always used to love it. But it is a pretty awful experience today so a slight premium (if real after adding in all those extra costs one inevitably incurs on going to airports) for a smooth, hassle-free train ride is worth it to me. As I tell everyone, if I could I would travel by train to Europe from Australia! (In fact once I did about 80% of it via the TransSiberian.)

            threestationsquare 2017/01/20 – 12:19

            I don’t disagree with most of your post (as your earlier one too) but surely you must agree that changing planes in Dublin so as to arrive at Stansted is rather a big thing? The context of many of my comments on this thread relate to the long-running topic of the overall poor planning of London’s airports, both the persistence with Heathrow, the privatisation under Thatcher (who also blocked building HS1, thus delaying it for 12 years!), and the poor airport transport arrangements (while simultaneously ripping you off). Anyone with an open mind who flies into the rest of the world’s major city airports, especially Asia (Singapore-Changi; HKI, Shanghai, Beijing, KL, Seoul, etc) cannot deny this. (Obviously I exclude the rest of the Anglophone world, USA + Australia, who have the same political-philosophical problems; only now, well into the 21st century are LAX and JFK being brought into the modern world! Sydney is awful, another privatised monopoly.)

            For the record I lived in Oxford for a long time and though I fundamentally hate bus/coach travel, would have to begrugdingly admit that the bus service to Heathrow was pretty good. It still amazes and confounds me that those giant coaches seem to navigate the M40 and M25 and even into central London more efficiently than private cars!

  3. Ryan

    One of the alignments proposed is via the LIRR Atlantic Branch and Second Avenue Subway, which the RPA has long believed should be connected. The brief says that it would be slow because it would have to make many local stops; I’ll add that it would serve Midtown, where nearly all the hotels are, via a circuitous alignment.

    How is that alignment at all circuitous? Granted, there’s a whole mess of other problems with converting the Atlantic Branch, but I don’t agree that “circuitous routing” is a problem or even necessarily correct.

    • Alon Levy

      It detours through Downtown Brooklyn and Lower Manhattan. The main destination for air travelers is Midtown and not Lower Manhattan. Lower Manhattan is a single-use office district, with no draws for tourists except for the WTC memorial, no hotels, and no residents.

      • Tower18

        This is increasingly untrue re:Lower Manhattan (especially in terms of residents), but it’s definitely no longer true of Downtown Brooklyn. Not only are there large residential populations now and more towers going up every day, but there’s lots of new hotel inventory being built in Downtown Brooklyn, both to capture the current cachet of Brooklyn, but also to offer tourists and frugal business travelers more convenient cheaper options. Just off the top of my head, opening in the last 2-3 years in Brooklyn have been a Hilton, Sheraton, Holiday Inn, Hampton Inn, and a couple independent hotels that I think are just occupying space until a chain believes the demand has reached the right point and slaps a franchise on them.

        This is in addition to the cheaper options like Fairfield, etc. that went up along Third Avenue a few years earlier.

        • Tower18

          Let me be clear that I’m not trying to posit that Lower Manhattan and Downtown Brooklyn are destinations on the level of Midtown, but nor are they deserts anymore (especially Brooklyn).

        • Alon Levy

          There are also hotels in Long Island City. I stayed in one a few times (it’s like $110 a night, close to Midtown, and right on the E). Neither LIC nor Downtown Brooklyn is where the mass of visitors is going. There’s a reason self-respecting New Yorkers avoid Times Square and not Cadman Plaza or Queensboro Plaza.

  4. Fredrik

    I was in Hong Kong recently, and the Airport Express there has the unique feature that you can check in your luggage before getting on the train. Does that make it worth 35 USD for my family instead of 8-9 USD for MTR to Tung Chung and then S1 to the terminal? Yes, homeward bound and lots of luggage, it was. But probably only because the Tung Chung line terminates short of the airport.

    Visitors to a city often get the impression that the premium airport train is the only way to get to the airport. I guess the Piccadilly line connection to Heathrow is well known, but do they know about Thameslink from Gatwick?

    About Stockholm. Arlanda Express is reasonably connected to the transport system. The walk to the commuter trains or the subway is no longer than the walk between the commuter trains and the subway. Also, the airport is served by the Uppsala-Älvsjö commuter train line, which stops at Helenelund, a 600m walk from Kistamässan.

    Unfortunately, there is a special charge for going from/to the airport on the commuter trains. Three or more people going to Uppsala, it is the same price as a pre-booked cab. (Cabs in Sweden are deregulated. Don’t just hop into one)

    • Alon Levy

      The Arlanda Express is more frequent than the regular commuter trains to the airport, which is a big factor in why I’d use it. How much mode share is SL expecting to get on an airport train that comes every half hour?

      The checked baggage on the Airport Express in Hong Kong is a nice feature, but it seriously raises operating costs. You can’t do a lean OPTO, you need dedicated tracks because of dwell times, you can probably forget about through-running.

      • Fredrik

        Sure, checked luggage increases operating costs. But premium prices should come with premium service. The checked luggage is the reason I can imagine that if I lived in Hong Kong I might use the Airport Express.

        One other interesting thing is the intermediate stop at Tsing Yi. Is that purely a park and ride stop? There is no stop at Sunny Bay, so they are not even trying to get airport-disneyland traffic.

        The interchanges from Airport Express to other MTR lines are obviously not a priority. If you take the S1 bus to Tung Chung, and get on the Tung Chung line, you can reach three MTR lines with one change, another four with two changes. The last one requires three changes.

        I think SL would like to run more trains Uppsala-Älvsjö, but the double tack line north of Arlanda is full. I forgot that there are also usable train connections to Arlanda from other places, e.g Gävle.

        The terms of the Arlanda Erpress concession allows other train operators to use Arlanda C, but every passenger has to pay 120 SEK (14 USD) to the Arlanda Express operator.. AEX on other has full freedom to set their own prices, so they have group prices, early booking discounts and multiple ride discounts.

        I think that Arlanda makes it clear that the best way is to build the airport on a main line. In London, the new airport should be built where HS2 intersects with the Varsity line.

        • michael.r.james

          Frederik wrote:

          Sure, checked luggage increases operating costs. But premium prices should come with premium service. The checked luggage is the reason I can imagine that if I lived in Hong Kong I might use the Airport Express.

          I am not sure too many locals use it. The reason I (a visitor) use it (to go to the airport not from the airport) is because 1. I usually leave it to the latest possible time to go to the airport (it is extremely reliable and of course, already being checked in means a very quick route thru airport security/check-in; no panics about missing absurdly early check-in deadlines–you only have to get to the gates before they physically close them) and 2. you can check in your luggage up to 36h in advance of your flight so I check in with any stowed-luggage early on my last day in HK so I have the day free to do as I want, knowing later it will be very hassle-free getting to the airport etc. (I am usually on the midnight flight to Europe or the midnight flight to Australia; both arrive early a.m. in both destinations.)
          OTOH, when I arrive in HK I usually take the bus, not solely because it is dirt cheap (it is) but 1. it services the hotels along Nathan road in Kowloon (HK island hotels are beyond my wallet or miserliness; or even the “new” Kowloon Extension area where the train also stops); the bus is superbly set up for this with big luggage storage areas (and a CCTV that displays it at every position on monitors around the bus–to reduce your nerves about opportunistic thieves, and to discourage thieves–it works) and there is an automatic announcement (in English) in advance of each stop and which hotels it serves (there is a leaflet that shows this too), so it really is convenient and foolproof (in any case stops are frequent so if you go past your stop, no big deal), almost as good as a taxi; 2. the bus gives you an excellent free tour of the route into HK: I can usually get the front seat in front of a big picture window on the top level of the (London double decker) bus–you get a terrific, thrilling ride over the 34km. From this p.o.v. the bus is heaps better than the express train which goes on the lower level of the Tsing Ma bridge (so you see nothing) and on the top level of the bus you really get a panoramic view. Many cities can be pretty dreary to arrive at in the journey from their airports but this HK journey is exhilarating (it might just be me). You arrive in the always hyper-active Nathan road, and within minutes off the bus you will have checked into your (low-to-mid priced hotel) and can be out on Temple street night-market or strolling in Kowloon Park (with views of the city) or down at the waterfront etc.

          Obviously one could also use the same bus to return to the airport. It’s very reliable but all things considered the advance check in and speed of the Express train is worth it (it may be expensive compared to the bus but at about US$10 still one third of the bloody Heathrow Express; incidentally apparently you can check in at Paddington but only if you have carry-on only, no luggage to be stowed).

          The interchanges from Airport Express to other MTR lines are obviously not a priority.

          That’s because most locals don’t live anywhere near the Express terminals and the buses from HKI are both very cheap and take them directly to their destinations in Kowloon or New Territories. About 12m use the Express train every year but about 2-3x that use the buses. The train is only convenient for those who live on HK island–as it doesn’t serve the busy part of Kowloon (where Hong Kongers and lots of retail and hotels are). I normally avoid buses but, as I described above, in HK they are kinda cool.

  5. Michael Noda

    Stub-end terminals *still* Considered Harmful. Even/especially in cities that have through-running. I’ll be interested to see how your evaluation of Heathrow access changes with the initial service via Crossrail.

    In other dog-bites-man news, Philadelphia continues to waste the premier through-running commuter rail tunnel in North America, but the Airport Line service is a decent example of getting through-running passengers, because the Airport trains tend to run through to the Reading Commuter Trunk and then on to Warminster or West Trenton, where the highway routes to PHL Airport are non-competitively indirect, and the direct road route runs through the entirety of the city grid. But it works, it runs every 30 minutes, and it’s cheap. You just need to pay the $8 or $9 in cash on board, coming from the Airport… %^$&%#^@ And we learned our lesson about not having the Airport trains be a dedicated fleet, but just having the standard fleet also accommodate Airport trips, not particularly well on the Silverliner IVs, but decently on the Vs.

    I guess Narita is a special case, just because it’s way the hell far away from where anyone is going, even in the polycentric Tokyo region.

    • threestationsquare

      In what sense is Narita a special case? The premium express Keisei Skyliner service ends at the Keisei-Ueno stub terminal, but the normal-fares Keisei Access Express service usually runs through via the Asakusa subway line (and on to Haneda), and the competing premium JR Narita Express runs through Tokyo station to Yokohama or Shinjuku (and I think this may be its main market these days since it’s slower and more expensive than the Keisei Skyliner). Of course, Tokyo has a lot of stub-end terminals in general, since there just isn’t the capacity to run everything through.

    • kclo3

      After 30-odd years, the Airport Line has settled to a pitiful 7100 daily riders, while both Denver’s A-Line and UPX have leapfrogged it in a few short years, and not just for relative O/D traffic either. It’s infuriating that the Airport and SEPTA have said nothing about how or if SEPTA Key TVMs are coming, or that the Airport still has delusions of replacing the terminal stations with a useless landside people mover. Strangely there is also an absence of Airport Line signage right after leaving the secure zone or from baggage claim, which could leave out unknowing travelers altogether.
      But miraculously, the line is still time-competitive with driving via Penrose/I-76, which leads me to think more stops could be added without much penalty. Specifically, move the current Eastwick station to Island Ave where it correctly belongs, add an 89th St station to serve the hotel complex, and even maybe 70th/63rd if ridership warrants it. If frequency is ever raised to 15 min, I would also want to see an NEC Airport Transfer station at 54th St as an alternative to the HSR plan. The Airport alignment is just about the best one could ask for, and it holds a lot of potential if the stakeholders chose to recognize its potential.

  6. Martin

    I will say that express trains like Heathrow Express and JFK Express are good compromise for women (such as my wife) who are not good with subways and directions. They are fine taking express downtown to main station cabbing the rest since taking baggage on subways, stairs and then finding hotel on foot is sometimes too much. For Heathrow Express, you can get into London with an Uber doing the last mile. JFK Express model could be the same. Just another point of view since most readers of this blog (myself included) are transit savvy and wouldn’t be overwhelmed to this level.

    • Alon Levy

      Oh, I’ve definitely encountered tourists who don’t like to transfer. I was giving directions at the terminal AirTrain station to someone who insisted on taking the A to Columbus Circle instead of the E-to-B/D because they didn’t want to deal with transfers. On the other hand…

      – The mode share of these express trains is pretty low.
      – Taking a cab last-mile in a place like London or Paris or New York is pretty hectic. These stations are all big and confusing.
      – Taking a cab last-mile only makes the problem with high fares even worse.
      – Express airport service already suffers from low frequency, since the size of the market is capped by the number of air travelers. Segmenting it to only be useful for tourists who aren’t transit-savvy only makes it worse.

  7. Fredrik

    I found this on

    By Train

    From London Heathrow
    Take the Heathrow Express to London Paddington
    Take the eastbound Circle Line underground train to Baker Street
    Alight from the train at Baker Street and take the Southbound Jubilee Line underground train to Green Park
    Alight from the train at Green Park
    Turn left after the ticket barriers and exit the station
    You will see the hotel on your right
    Turn right and walk along Piccadilly, under the hotel’s arcade
    When exiting the arcade turn right and the hotel entrance is on the right hand side

    According to, this takes 41 minutes, including a 8 minute walk. Taking the Piccadilly Line directly to Green Park takes 48 minutes, with no walking.

    • michael.r.james

      Does anyone seriously imagine any Ritz patron taking public transport like this? Especially the “three seat” Heathrow Express route. Incidentally that “41 minutes” couldn’t be trusted because while the HE might be highly reliable (though its performance index of 94% allows it to be within 5 minutes of advertised arrival!) the two transfers on the LU are not consistent or reliable. The Piccadilly route is probably more consistent but of course then those Ritz hoi-polloi would have to mix it up close and personal, along with their Louis-Vuitton luggage, with regular London commuters and workers. (To be brutally honest, even I don’t want to do that …). Though there are three different stations one can board the Picaddilly line at Heathrow so that has to add about 5m depending on which terminal you arrive at.


    Heathrow Express is for Premium passengers. It’s high cost is because it was built without subsidy at a time when there was loads of spare capacity on the Great Western Mainline.

    Many passengers are happy to travel on a fast air conditioned train with plenty of room for themselves and their luggage. The express nature of the line means no one has to worry about getting off at the wrong stop, or someone is going to steal their luggage.

    Compared to with a metro or local train service jammed full of commuters and no space for luggage and dozens of stops. Your anxiety keeps building as try catch the name of the stop through the press of bodies and their death glares for your luggage.

    Express services are about diverting cab users onto transit, even when there is a perfectly good metro line etc.

    Users of the Heathrow Express love it. Go on to Airline user forums on the number of times someone in the States pipes up and wishes their city had a Heathrow Express is astounding at times. It is very popular with those who travel at the front of the plane. They also prefer Heathrow to other London airports, which is why every airline wants to fly to Heathrow as that is where all the profits are made. I expect it’s why airport links are proposed for every city. A plane is one thing you can’t be late for, a good transit service cuts through all that anxiety and an express service reduces the anxiety further, Plus an express service can be expensive enough to keep you away from the great unwashed bus users. Who cares if you don’t get dropped outside your hotel, it’s now just a short cab ride away, with no slog on the motorway in between!

    It is no surprise that Heathrow is trying to fight the closure of Heathrow Express once crossrail is up an running, despite the sense it would make. Currently it’s a stand off with Heathrow owning the access tunnels and stations and the Government about to reacquire the train paths on the Mainline, once Heathrow leases expires in 2023.

    • michael.r.james

      MR M CURRIE 2017/01/19 – 17:23

      Heathrow Express is for Premium passengers. It’s high cost is because it was built without subsidy …. It is very popular with those who travel at the front of the plane.

      First, clearly you are British (and voted for Brexit) because only they would not blink or think it odd that in this day and age that such a huge bit of infrastructure should be intended solely for an elite. What next? Put high tolls on an express road lane on the motorways so that the nomenclatura can travel unimpeded? Two legs better than four. All your negative descriptions of using the Piccadilly Line have existed for at least 40 years (since my first use of it) and constitute the prime reason why the Express should have been developed as a solution for everyone, not just the rich (or those on expense accounts; or maybe in two years you’ll have to show your British passport to use it?).
      And BTW, it would have been both much more effective and vastly cheaper, if they had used the existing tracks and stations etc of the Piccadilly Line at Heathrow for the Express line; that way they could have saved on tunnels and track etc. This is the end of the Piccadilly line so it would have actually improved that line by making it shorter and less crowded, ie. better serving its suburban commuters. Essentially this is what the Paris plan with their future Airport Express is, ie. to free up RER-B to better serve its suburban commuters; of course the RER-B was/is vastly superior for the airport travellers than the slow, crowded Piccadilly Line is for its travellers.
      Oh, and finally, the Express is “so popular” that nowhere does Heathrow (its owner, well of the bit inside the airport perimeter) reveal how many pax it carries. Of course for you, it’s a question of “feel the quality not the width”.

      Second, as to “without subsidy”, sure. Just like the third runway (£18bn and counting) will be funded by the private owners of Heathrow? (Not.) Just how many billions have been sucked out of the BAA monopoly in its two decades, straight into private pockets even as they can’t/won’t fund its expansion. There’s a series of extremely good reasons why such airports shouldn’t be privatised, and in general around the world aren’t; public/national interest and the huge capital costs of continuous improvement are just two. There is not a privatised entity that has not continued to receive government subsidy by one means or another. The railways being a prime example.

      Third, re-reading your post, now I’m not sure it isn’t parody. I think I could, with just a few tweaks, turn it into a version of that Cleese/RonnieBarker/Ronnie Corbett skit on class. Cleese would take the Express (when the Roller was indisposed), Barker would take the Piccadilly Line and Corbett would take the bus … (well, truth is Corbett would be flying on a LCC out of Gatwick …).

      • Gag Ha;frunt

        And BTW, it would have been both much more effective and vastly cheaper, if they had used the existing tracks and stations etc of the Piccadilly Line at Heathrow for the Express line; that way they could have saved on tunnels and track etc.

        The Heathrow extension was built to “tube” loading gauge, so using it for mainline trains would be impossible.

        • michael.r.james

          The Heathrow extension was built to “tube” loading gauge, so using it for mainline trains would be impossible.

          I was talking about when it was all being “planned”. They could have built the Express using the same gauge, don’t you think? Come on, you know there is some sense in my suggestion: it would have avoided what we all dislike about taking that long Piccadilly Line ride.
          But then if they were actually into anything vaguely resembling “planning”, Heathrow would have been relocated to somewhere sensible half a century ago as all these other world cities have done, and superior transport options planned and built in the most efficient manner possible.

          To M. Currie: what I was uncertain to think of as parody is at least clear now: you really are a class-bound uptight rosbif. I’ll bet you lament the removal of First Class tube carriages, so those premium passengers such as yourself could avoid “the press of bodies and their death glares for your luggage.”

      • MR M CURRIE

        God you are filled with loathing and condescension. I did not vote for Brexit what a crass assumption.

        Heathrow express was built for those passengers who would not use the Tube, it diverts millions of passengers from Taxi’s to transit. BTW a tube line that is often rammed in peak periods

        In this day and age the Express is hogging 4 express paths per hour into Paddington and it not equitable considering the number of Express commuters those same paths could carry from Reading etc. But the line was designed in the early 90’s when traffic went up and down with economy and no one expected traffic levels to reach the levels they have today or that London would grow the way it has. It’s a creature of it’s time and we have to live with it till the lease expires and a new fight begins over terms of access or whether any express paths are now available to heathrow, or we pay Heathrow for their rail infrastructure and have Network Rail take it over (can’t see that happening either). The Express link carries 16,000 passegers a day, just under 6 million a year. (it’s amazing I used to Google to find out these so called secret passenger figures).

        All I was doing was pointing out how the airport sees the line and how those premium users of the line see it. 70% of it’s passengers business class passengers who drive all the profits of the airlines and the airport. These rich and powerful people are same people who end up pushing for similar lines for their home cities.

        I am confused about your idea of reusing the Piccadilly line tunnel as vastly cheaper. If you mean cutting off the existing line and connecting it the main railway line, then there some very real problems with that. as noted by the previous poster the tube line runs to a much smaller gauge than the main railway line and would not suitable to run with 125 mph express trains into Paddington, to maintain links to the Hounslow, you’d still need an interchange with the trunkated Piccadilly line.

        I fail to see how eliminating a direct service on the Piccadilly to introduce a express service that can only go 60 mph to Paddington would be that good an idea.

        it’s almost as if you have no idea what you are talking about other throwing abuse at others.

  9. spencepatrickj

    Could you please analyze the impacts of closing and consolidating airports (LGA, for instance?) See what can happen by consolidating flights, improving regional transportation thus eliminating regional flights, analyzing the necessary replacement capacity, and seeing how much could be funded by selling off old land to private developers.

    • Alon Levy

      That works only if you can expand runway space on the other airports. Interference between LGA and JFK airspaces reduces both airports’ capacity somewhat, but not by nearly as much as the extra capacity coming from LGA’s gates and runways.

      I wrote years ago about the possibility of creating more air capacity in New York by running larger planes and using HSR to eliminate flights to nearby cities like Boston and Washington, which flights are dominated by tiny planes. See here and here.

      If LGA closes, probably the biggest benefit in terms of development is being able to build a higher-rise CBD in Flushing – provided the reduced airport access doesn’t depress demand there. The LGA area isn’t really high-demand, since it has poor access to the subway, and isn’t really on any top potential corridor like Northern or Astoria Boulevard. The Grand Central Parkway is also pretty useful for through-trips to Flushing, so closing it to turn it into a rail corridor is politically difficult.

  10. Eric

    Re Paris, if the hotel district is where you say – Rather than have the airport express terminate at Gare de l’Est, wouldn’t it be better to continue on the RER E tracks to Haussmann Saint-Lazare? (And to Port Maillot and La Defense once the RER E is extended – construction on this is apparently supposed to start soon)

    • Alon Levy

      It would be somewhat better. But Haussmann is really at the eastern edge of the mass of hotels, and Porte Maillot is well to the west. The best RER station for hotel access is still Etoile.

      • michael.r.james

        Re future Airport Express to Gare de l’Est then on RER-E tracks to Auber (correspondances to Gare St Lazare and Chausee d’Antin on Haussmann): presumably for the same reason they are removing the airport traffic from RER-B line, specifically the heavily contrained tunnel bit from Gare du Nord southbound.

        Re hotels, Adirondacker repeated the reality: people heading to those literally “ritzy” hotels in the 8th don’t use public transit. There is in fact a hotel cluster around Opera and of course in the Leftbank–some of these may well take the trains.

  11. Adirondacker12800

    In Paris, the five-star hotels cn Paris, the five-star hotels…..

    People who stay in five-star hotels don’t use the train. Maybe if they only have a carry-on and flight that leaves during rush hour. Maybe. Origin passengers almost never begin their trip at a five star hotel. Or any hotel.

    The throwaway line is that almost every major world airport has a one-seat train ride to city center, and by implication, so should JFK.

    JFK has multiple one seat rides to the AirTrain. The A, E, J, Z subway lines and LIRR that isn’t the Port Washington Branch. The AirTrain isn’t going to go away. There is never going to an AirTrain station and an LIRR station at each terminal. Most passengers will be changing to AirTrain anyway. Changing at Jamaica or Howard Beach is good enough.

    a new trunk line connecting Eastern Brooklyn with Second Avenue?

    Wall Street is the country’s third or fourth biggest employment center depending on how one defines a CBD. Connect the Second Avenue express line to the West End Line in Brooklyn ( the current D train ) Metro North to Staten Island and NJTransit to the LIRR… Or Second Avenue local to the West End Line and Second Ave express to Eastern Brooklyn or

    • Alon Levy

      JFK has multiple one seat rides to the AirTrain. The A, E, J, Z subway lines and LIRR that isn’t the Port Washington Branch. The AirTrain isn’t going to go away. There is never going to an AirTrain station and an LIRR station at each terminal. Most passengers will be changing to AirTrain anyway. Changing at Jamaica or Howard Beach is good enough.

      Yep. I’ll add that the hotels in New York cluster farther north in Midtown than Penn Station, and East Side Access is going to make things a lot better. The single biggest high-end hotel in the city, the Waldorf, is a couple blocks north of Grand Central. Even in the afternoon off-peak, it’s faster to take the subway or the LIRR from Manhattan to JFK than to slog it in a taxi, even though the E isn’t exactly high-speed rail. (Actually, the fastest route from the East Side right now is E to Forest Hills, then taxi to JFK; I did it once involuntarily when the train was emergency-rerouted to the F line, and ended up getting to the airport early. But I lucked out on finding a taxi instantly.)

      • Adirondacker12800

        …..If you are staying at the Plaza you take the R to Queens Plaza and change to the E. Because you only have a carry on and it’s rush hour.

        • Alon Levy

          For the record, I’ve taken transit with multiple checked bags. It wasn’t voluntary (it was raining, so no cabs), but I managed to make the move-out trip from my apartment in Stockholm to Arlanda on foot + subway + Arlanda Express with a carry-on and two checked bags.

  12. SouthJersey1

    In 2000, I slogged 75 minutes across London from Heathrow to Woodford (Essex) via transfer at Holborn. Yes, it was time consuming and certainly Holborn is not the Oculus, but it was cheap and it dropped me exactly 0.2 miles from the B&B. Therefore it worked.

    Heathrow Express would have done nothing of the kind, requiring two less user-friendly transfers at Paddington and Liverpool Street, plus costing a great deal more and requiring more time. I also investigated the National Rail service via bus link to/from Feltham operating on the North London Line, but sparse scheduling and lack of concise bus link information eventually led me to eliminate this option.

    CrossRail should significantly enhance the attractiveness and utilization of Heathrow Express. It would certainly be the preferred choice to reach the locations where I would most likely stay in the future.

    I also used the limited stop Stansted express (originating at Liverpool Street) between Tottenham Hale and Stansted Airport in both directions, and was favorably impressed. I found Stansted to be (by far) the most attractive London area airport. I wish there were direct flights between Stansted and the USA.

    Focusing on the original post topic, I fail to see the value of spending vast sums to provide direct JFK/Manhattan service given the existing infrastructure. I also don’t understand the allegedly pressing need to extend PATH service to EWR given the existing NJ Transit and Amtrak services and lack of capacity on PATH. New York City has a vast backlog of projects that have accumulated from the necessary diversion of limited funding toward addressing state of good repair issues in the past decades. Many of those projects are more worthy than the creation of a JFK/Manhattan express.

    • Eric

      Note that there are two services.
      Heathrow Express – £21.50, no intermediate stops
      Heathrow Connect – £10.10, same route, 6 intermediate stops.

      £10.10 is not THAT expensive, and the one time I took the Piccadilly, all the curves and stops made me a bit nauseous. So if I had money I’d seriously consider the Heathrow Connect, final destination dependent of course.

    • Adirondacker12800

      They want to add two care to the Newark-World Trade Center trains. They don’t have enough room to store them. If they extend the current storage tracks they are almost in the airport. They can convert the storage tracks to running tracks and build a yard in the airport.

      • Alon Levy

        Somehow this “extend the storage tracks” scheme is budgeted at $1.7 billion, which is really high for a subway, let alone an el.

        • Adirondacker12800

          Which is a separate issue from increasing capacity on the Newark-World Trade Center trains by twenty five percent. It’s going to be bit more pricey than slapping an el over a wide boulevard, I count 26 lanes of highway and ten tracks of railroad to thread through.
          In a more perfect world they would be connecting the LIRR to NJTransit somewhere near Wall Street and letting the suburbanites change trains in the suburbs instead of Newark and Manhattan.

  13. Erick

    Premium airport services are often the result of groupthink and the desire for vanity projects. Chambers of commerce and other trade groups like the idea that they can get select premium service away from the hot polloi. Politicians like the idea of a “premium transit service” that won’t cost a penny(yeah, right!) to the taxpayer.

    Also typically the service is operated by a private sector operator (or maintenance while the local public transit operator provides customer service). This then is about rent seeking rather than about providing transit to the population. Thinking about it, this is probably what drives most services.

    Toronto is going through this with UPX, of course politicians there are in full blown denial but eventually this will have to be adapted at great cost to bring it back in line with the regional network. Without heavy subsidies the private sector won’t touch it.

    Good transit to airports is important for visitors, however employees at the airport never seem to factor into the plan. It’s all about that business visitor. And here is where there’s a disconnect between businesses and government.

    Business visitors may indeed bring lots of revenues to some businesses, but from a government perspective you have to be careful with the budget you have. The contribution to GDP and tax revenue from such visitors may not always justify the massive public investments into what is then a private service. Heathrow Express happened with lots of public subsidies (most of which were indirect as to avoid attracting attention).

    But politicians get to spend lots of time with business leaders and not much with airport employees, hence we get this scourge.

    A look at origin-destination inquiries would provide a good base to determine what type of transit services you need to provide to the airport, but there might not be a lot of profits in there.

    In the case of JFK, even if the service is provided by Port Authority or other public entity, the construction of that service will require very expensive contracts to the private sector.

    • michael.r.james

      Erick 2017/01/23 – 09:42

      Good transit to airports is important for visitors, however employees at the airport never seem to factor into the plan. It’s all about that business visitor. And here is where there’s a disconnect between businesses and government.

      “Heathrow Airport workers who hold a LHR ID card are entitled to a 75% discount on a weekly, monthly and annual season ticket; the discount applies to the FULL fare tickets only.”
      True that this is for travel on Heathrow Connect, the stopping service version of Heathrow Express the premium service. Couldn’t have the plebians and workers disturbing M. Currie. Numbers using the service aren’t given but there are 76,000 workers at Heathrow …

      In Paris, all workers would automatically get greatly subsidized access to RER-B via their entitlements to Carte Orange (or Navigo carte). I don’t know if there is any special arrangement for CDG workers beyond this.

      You are correct that when private for-profit companies get involved the main point (moving these large numbers of people and their luggage with maximum efficiency to their destinations while avoiding road congestion and its associated collateral costs) is often lost or deliberately sacrificed to short-term profit. Car-parking charges are a major profit centre for most airports. This instantly becomes a huge public cost burden as the roads around airports seem to need constant upgrading and it flows thru to the wider road system (such as in London’s M25 where Heathrow is a major node and cause of congestion). A few billion may be spent on rail transport but many many billions get spent on building, then maintaining, these roads. I once missed a flight at Heathrow because I was on a bus (from Oxford; rail journey is much more involved) got caught up in the congestion hell on the section between M25-at-Heathrow and the M40 junction; IIRC part of this was due to roadworks to add another lane to the M25 (which seems under perpetual repair or extension).

      • MR M CURRIE

        Sigh, again with the insults. I work at Heathrow and no doubt you earn many times what i do.

        Currently Heathrow has 38% public transport mode share which is one of the best in Europe. The airports in the World with higher shares tend to be single terminal complexes with more concentrated destinations and origins of passengers. (improving access to world airports, Coogan, 2000).

        The airport operates a strict limit on car park spaces for both employees (it can be a long wait for a car park pass) and travellers, has green travel plan audits. It subsidises discount plans on surrounding bus, coach and trains services for employees. Plus all local buses are in a free travel zone on the surrounding public roads. The airport has also subsidised extra local bus services.

        These and many other policies have been part of planning restrictions on parking at the airport.

        Parking is seen as problem at the airport.
        For example in it’s 2015 annual accounts it’s revenue was £2.7 billion, revenue from parking amounted to £107 million.

        • Alon Levy

          How common is it for Heathrow employees to take transit to work? At JFK it’s not very common. I read a study maybe ten years ago, which I’ve repeatedly tried and failed to hunt down, about New York’s secondary employment centers and public transit; JFK had the lowest transit share, maybe 15 or 20%, and was distinguished from the other centers (Downtown Brooklyn, etc.) in that even workers living right next to the airport usually drive. I don’t remember if the study goes into the higher fares on the AirTrain, out of reach for many employees. But it does mention layout: the AirTrain serves the terminals, so it’s useful for passengers and potentially for employees who work at the terminals, like concession retail workers; it doesn’t serve other places on airport grounds, like hangars and runways for maintenance workers and baggage handlers. At Heathrow, obviously, the trains serve only the terminals, too – is there a big division in mode choice between different kinds of workers?

          • threestationsquare

            The JFK AirTrain has $25 ten-trip tickets and $40 monthly passes, so the fares for employees aren’t /that/ bad. The Bx15, Q3 and Q10 buses between the airport and surrounding areas make a few other stops on the airport grounds, and the Q6 bus serves the airport postal facility but not the terminals.

        • michael.r.james

          MR M CURRIE 2017/01/24 – 04:19

          Sigh, again with the insults. I work at Heathrow and no doubt you earn many times what i do.

          You should really think of them as jokes, admittedly at your expense, rather than insults. And I’m afraid I will never let go of it, when you make such an absurdist post that feeds this anti-monarchist, anti-class Australian (who has witnessed it all in the UK). And alas, I doubt I ever earned that much more than you despite the ten years of poverty at university then another ten years of low-wages in absurdly long-hours of a post-doc en route to being a research scientist (usually forced to live in some of the most expensive cities in the world).

          For your other points, fair enough. Like Alon, it would be interesting to know the mode share by the workers. I read that there are 400 companies other than Heathrow Airport Holdings on site. At one point the retail operation was the largest single such thing in the UK (in value of turnover I guess, which is a bit of a cheat). And I assume, like at JFK, there are free shuttle buses from the public transit points to all these other sites scattered over the sprawled-out airport domain.

          Incidentally Heathrow would have always had the need to discourage car access and parking on-site because the (London) roads are always so inadequate. The “last mile” is often totally backed up. But on the other hand, for such a gigantic complex apparatus it is remarkable they function as well as they do.

          Final point: you can’t imagine that when CrossRail replaces Heathrow Express, that it will run “premium” trains or even premium carriages? No, it would be silly, and offensive. And counter-productive. Of course I suppose they will try to charge a premium price. (In the UK one gets accustomed to paying premium price for second-rate service or worse.)

          • MR M CURRIE

            Airport commission report into Heathrow shows that 47% of employees arrive by public transport 36% by bus the rest by tube/rail.

            There are 12 bus routes and 4 commuter buses to the airport (that terminate at an airport bus terminal. There are other routes that pass by the airport, but do not go in. Some Bus routes are 24 hour, others start at 3 am for the first shift. But most of these are only 2 or 3 buses an hour till 5am, when they start ramping up to daytime frequencies of at least 10 minute headways though the busiest is a 6 minute headway.

            The airport is limited by law to 45,000 parking spaces (including staff). With the new planning application for the third runway it claims it can cope with the same number of parking spaces. Hence the push for a new western rail tunnel (going through planning stages now) and studies into a southern rail tunnel as well, that avoids the pitfalls of the old airtrack scheme.

            The airport now says that there should be more short stay car parking to meet demand, as currently prices are so high that many people get people to drop them off rather than drive themselves, creating extra journeys on local road network.

            The airport is desperate to avoid the air pollution cap. By proposing to change the airport vehicle fleet to all electric, introducing road pricing and building a new airport city between the airport and the M25 where all the bulldozed offices and hotels will be relocated from the A4 into a planned development which is linked into the new terminal and terminal 5 by a new automated people mover system, so no shuttle buses and vehicles driving between buildings.

  14. Eric

    Re JFK. What would prevent building a short track connection from Airtrain to LIRR just west of Jamaica, and running direct trains to Grand Central (once ESA finishes)? The trains would have to be manned, and the Airtrain signalling system would have to be changed. Any other issues?

    • Alon Levy

      The AirTrain has a high ruling grade – I read 5.5% somewhere, but can’t find the reference.

      The other problem: AirTrain stations are short. They’d have to be lengthened to accommodate longer trains. If the trains are short, then they just hog precious slots in the ESA tunnel that should be going to full-length trains serving the suburbs.

      • Adirondacker12800

        There’s is life outside the Waldorf-Astoria. Send the train to Grand Central, all the people who want to go to the Penn Station for the West Side are screwed. The people who want the E.J. or Z still get off in Jamaica. Or the buses. Or are going out to the Island. I suppose every other one could go to Penn Station but that cuts frequency. . . AirTrain is good enough.

  15. Bruce Hain

    It is amazing to me. If you see this line from the A-train in Queens, what an impressive thing it… was. An essetially-completely-modern high-capacity line with min. 2 tracks – now completely run to seed, including the power house at Atlantic Avenue, clearly a thing designed with intended permanance in mind. They didn’t cut corners – they meant it to last. (meant with ref. to the entire branch line) Enter the engineers and corporate management of the latterly PRR et al and MTA in the past 60 years – and you get what is essentially an act of public vandalism. I mean even if they were too stupid to understand the thinking of the people who brought us the N. River and E. River tubes and Hell Gate Bridge they were at least capable of operating based on deference to the intentions of the planners of the most advanced piece of civil works anywhere so far at the time, bought at cost of life and limb. And if you think it’s going to improve now that Prendergast is gone, I don’t hold out much hope for that – at least he was a fast brain, if not operating fully in the public interest, and he had knowledge only acquired thru long experience. The disaray of the MTA, NJ Transit and PA are mind-blowing – making clear they are not up to planning daily operations, never mind BUILDING anything – whatsoever.

    Of course this line would have been the obvious, most-direct, least expensive (largely-already-built) choice when planning a line to the airport, and Giuliani was continually flummoxed (should rightly say blocked, but can’t resist the idea of having him flummoxed) by the incomprehensible insistence of these Ivy League flunkeys that using it would be impracticable. (I am trying to think of an un-crude description for anyone who would dare make such a statement. Quatsch?) It is the elephant in the room, in this case, among several. Could it be that his later promotion of the idiotic Cross Harbor line was the result of some cynical wish for revenge?

    Also puzzling is why they would insist on a grade of about 275′ per mile for a modern rail line in flat terrain. (first I heard of this) But the necessity for short cars due to the curves in the current layout, and that, can be easily dealt with – so long as there’s no precise RULE about having short cars accessing Penn Sta. The tight curves existing right past the Belt Pkwy. should be eliminated concommittantly, before it becomes easier to lay an argument of inpracticability against it. And we should be thinking about going to linear induction anyway, all over. There is credibly documented existing efficient and workable licenced technology for this and the advantages in wheel wear alone would make a quick general change-over remunerative, never mind getting rid of the expense and nuissance of the goddamn wires.

    Only problem: The line is best-and-most-obviously suited for airport access, but another line has already been built at huge expense to replace it before it was even developed as such, although the other line is inferior. Suggestions? Enter Conv. Ctr. displaced from Sunnyside – a slippery slope maybe – though by the airport isn’t a bad idea.

    The line is perfect for airport access, and provides for smooth connection to the Atlantic Avenue line for service to lower Manhattan – another one of those things, along with its extension to NJ and thru the Bergen Arches, that should be seen by now and by everyone as an agreed upon “given” – tho it is not. Re. same – It is incredibly puzzling that MTA has apparently gotten the foremost railroad architect of his generation to build a station in the wrong place: not 500′ from the Fulton Transportation Center – where possibilities for station tracks are limited if not non-existent. Whereas there exists fortuitously space on West Street opposite and S. of the World Financial Center for that, in spades. For more on local and regional transit planning blunders, in particular my new post-with-practical-solutions showing elevations and measurements – “NEC Future? or Backwardation” – see: RAIL-NYC-ACCESS.COM

    • Alon Levy

      Okay, first things first: modern EMUs can climb 3.5-4% grades with little trouble. Using this capability is critical for cost cutting; here you propose an I-95 HSR alignment in Connecticut that sticks to 2% grades, which requires a lot of tunneling, where 3.5% would only require mild cut-and-fill. The LGV Sud-Est used the ability of fast trains to climb grades to build HSR between Paris and Lyon without any tunnels, even though the legacy line, limited to 0.8%, has several tunnels (link). Modern EMUs could climb 5% if need be, but standards are still 3.5-4%. The absolute limit for a regional train is based on whether the train can start from a standstill, going uphill, in wet conditions, with one of the motors not working; that’s still more than 4%.

      It’s not how lines used to be built. In China and Japan it’s still not how lines are built, and this drives up construction costs, since it forces China to run its HSR lines on viaducts almost the entire way. But the power and reliability of a 21st-century EMU are far better than those of a steam locomotive or even electric locomotive from the early 20th century.

      The Rockaway Cutoff is indeed a pretty direct way to get from Manhattan almost to JFK. The problem is that almost is really not good enough in trains. All of the following problems combine to make it a poor corridor:

      1. It doesn’t actually go to JFK, only to Howard Beach. Cannibalizing the AirTrain has its own problems, like the grades and curves being designed for a captive linear induction train (although neither is really fatal for regional EMUs). Plus, it’s important to have high frequency for terminal-to-terminal service, and that’s only ever happening with captive driverless trains.

      2. The Manhattan terminus isn’t great. Penn Station is too far south. Grand Central would be better, but there’s a good argument for dedicating the East Side Access tunnel to LIRR trains coming in from Ronkonkoma and Huntington (see explanation here).

      3. The subway’s development going back to the els hasn’t really been configured to connect to the Rockaway Beach Branch. Thus, there is no station intersecting it on the J/Z, and never will be because it’s too close to Woodhaven. And there is no station on the A. So the line would not be great for local service.

      4. The line diverges from the LIRR Main Line too far west, so trains can serve either Jamaica or the airport but not both. This limits frequency too much. JFK itself isn’t enough to support high train frequency; 4 tph is reasonable, but off-peak service to Jamaica is at a minimum 5 tph and usually 7 tph. Moreover, peak ridership for passengers heading to JFK happens to coincide with the afternoon peak and its shoulders, since that’s when the transatlantic flights leave, and this means that on the LIRR to Jamaica the frequency is high.

      The contrast here is with LaGuardia service via the N/W. Such service would not compete for train slots with anything, because the same trains would be serving Astoria and LGA. The Manhattan stations come very close to the big hotels. Frequency is high because it’s already high on the N/W, buoyed by local urban service demand.

      • John

        Do China/Japan really build HSR on viaducts because of grade concerns? My impression has always been that it’s because of lower tolerance for the risk of right-of-way intrusions. Taiwan HSR (which was built in the Japanese style) has a stretch of nearly 150 km over completely flat plain that still used viaducts. I’m less familiar with Japan but my impression is that the Tokaido Shinkansen goes through fairly flat terrain as well.

        • Alon Levy

          Nothing in Japan goes through flat terrain. The Tokaido Shinkansen is full of tunnels. The reason for the shallow grades is that parts of the Tokaido Shinkansen, such as the Atami-Mishima tunnel, go back to WW2. The war-era project was not HSR, just medium-speed standard-gauge rail, and in that era, high-speed cutoffs were built with gentle grades, because most of them were still running under steam. This became dogma in Shinkansen construction standards, whereas in Europe, SNCF realized in the 1970s that the LGVs could be built with steeper grades to take advantage of high speed and decent-for-the-era initial acceleration.

  16. Bruce Hain

    I don’t think standard EMU’s can take those curves. Not that it’s like the line will be developed in this (instinctive, intuitive I would say) way, since the circuitous Van Wyck Expy route is already in place – which is unfortunate. (That the presence of the existing Howard Beach line is SO intuitive poses a some questions.) Particulars of tying Howard Beach Line into the airport people mover – assuming a complete realignment isn’t required with complete airport rebuild (is this reality?) – are far from insurmountable, and could result in a good straight alignment without huge expense, and same was true in 1995. The high cap. intentions of the original planners would stand us in good stead re. local service – There’s really no insurmountable difficulty with a few local stops, including high cap. for Aquaduct/Casino and (god knows) a big convention center if that’s what the authorities determine. But they should really take up rehabiliting the property in keeping with their obligations as MTA, as a matter of course. Hell they could have turned that power house into a fancy restaurant or something by now. As for grades in CT: part of that is to avoid divisive conflicts, which as a side effect results in the gentle grading scheme. And there is some question if any of the elevations are reliable, but am convinced it manages a viable basic outline, though it’s pie-in-the-sky as they come. Hud. Tun. Proj. team was quite specific about their standards for grades on passenger lines when dealing with me earlier re. the NY side of the Hudson which has problematic contours at the point of all the mooted tunnels. – S. of the existing one. (This despite their cofferdam thing in the ARC papers with the 4% grade.) That they intend to undertake some horrific ground stabilization project with divers there now is indication they’ve jumped the track – but the Box was built without benefit of environmental documentation for the tunnel – and Amtrak is to be congratulated for having the forethought to come up with the Box idea. It never woulda happened if left to NJ Transit et al. The Box needs to be ground down to get an appropriate grade config. There’s plenty of room laterally for wide-section tunnel. I think the connection to a West Side Line as outlined in my “Hud. Tun. Proj. vs, the Correct Alternative” plan is the best in terms of economically getting added cap. with service S. to the Financial Dist. and six add’l stations in NYC – plus freight. tho you may find the connecting curve a bit shitty. (It hasn’t been carefully workd out by me yet – don’t repeat that – but will be soon.) That while allowing partial tunnel closure at Penn in avoidance of a staged “tunnel emergency” – since don’t believe connecting the new tunnel at Penn is even possible without. Thanks for the response, this a rare thing in my experience!

    • Alon Levy

      The reason the Hudson tunnels require relatively gentle grades is that not all trains through them are EMUs. I think commuter rail should go all-EMU, but in far better-run systems it’s still a long process of conversion; Zurich still has some loco-hauled S-Bahn trains and trains without level boarding. When your initial acceleration is 0.4 m/s^2 and you’re arriving at the bottom of the hill at 90 km/h, you’re just not doing 4% grades. (Even the M7s, which are capable of 0.9 m/s^2, are derated to 0.45 m/s^2.) An HSR cutoff would not be used by anything except HSR, so legacy loco-hauled trains shouldn’t be in the design parameters.

      • Bruce Hain

        mm. And they’ve been talking about replacing the old ones with unpowered cars. Since the new unpowered ones are so popular. Guess the popular features, two levels, are not so easily transmogrified, though hear this was the original plan. As I remember this would be the 1st order ever produced so they’re leary. And the cars are expensive. A case could be made for gentle grades based on carbon production, also nightmarishness of tunnels with steep grades – for passengers. Not suggesting flatness for accommodating an inferior technology which rightfully should be replaced though. Probably ease of wide-bore will keep coming down making for less speed/tunnel encumbrance, property issues.

  17. Bruce Hain

    Yes, the LIRR Mainline problems you discuss in that article I wasn’t aware of. (One hour of one-direction-only!?) Though am always suspicious there isn’t some viable scheduling solution given the four tracks min. Sunnyside – Jamaica. Will read your article carefully.

    • Alon Levy

      The problem with LIRR scheduling is the two-track narrows between Floral Park and Hicksville. Peak traffic is 20-something tph (23, maybe?), at which point the choice is between running only local trains, and running one-way. Competent railroads run only locals; the LIRR runs one-way, because fuck reverse-peak riders.

  18. Pingback: Should there even be a One Seat Ride to the Airport? | vanshnookenraggen
  19. Ben Kurtz

    Alon’s comments on the Heathrow situation are misleading.

    As of 2015, Heathrow Express/Connect services carried about 10% of Heathrow ground traffic. The Piccadilly Line carried around 18% and buses carried 13%, for an all-in mass transit share of about 41%. Source:

    So: 10% vs. 18% — far less shabby than the 9% vs. 27% comparison that Alon intimates in his post, even with far more expensive fares (around £5 for Tube vs. £10 for Heathrow Connect and £20 for Heathrow Express).

    Moreover, the new-build parts of Heathrow Express/Connect (the bit connecting the GWR mainline to the airport terminals, including the Stockley Flyover, the runway tunnels and the airport stations) were designed and built starting in the early 1990s, not long after the 1989 Crossrail study was published. While terminating at Paddington makes the present service less useful than something that ran straight across the center of town, it will get a huge boost when the Crossrail tie-in is finally complete, allowing relatively quick and comfortable one-seat rides to Heathrow from Canary Wharf, the City and the West End — not just Paddington — and connections to a huge number of other transit lines at each of those points.

    The case for Heathrow Express/Connect, when it was built, was respectable but not overwhelming: The Piccadilly line was already crowded and was always slow, there was long-run need for additional tunnel capacity to Heathrow, there was already reasonable demand using a Paddington terminus, and (I think hugely important) it was built to tie into a major expected piece of city center infrastructure. The crying shame here was not that Heathrow Express/Connect was built, but that it took an additional ~20 years to get Crossrail built behind it.

    In the New York context, you are right to point out that the RPA proposals for express and/or one seat service between JFK and Manhattan are rubbish. A more interesting question is whether there is a case for replacing that idiotic people-mover thing at Newark with a tunnel to bring those NJ Transit and Amtrak trains that stop at Newark Airport station right to the terminal. And, of course, how to tie that into improved through-running across the region — more tunnel capacity to Manhattan allowing more express and local services, 6 tph stopping at EWR on a clockface schedule, one seat rides from the NJ side to the LIRR side, etc. A proper EWR airport station (i.e. in the terminals) would be a smaller project than a JFK one-seat heavy rail service, with a lot more potential upside — especially when tied in to other regional improvements, much like Heathrow Express/Connect was.

    There is one additional point which the British experience has taught me: Fast, reliable one-seat airport rides have real-world premium value. An air traveler — much more than the average commuter — will generally want a place to put his bags, minimal transfers (again, because of having to carry and stow bags), and reliable service (because it’s often far more inconvenient to miss a plane than it is to explain to your boss or your date why you were ten minutes late because of “wet leaves on the track”). That premium is certainly not infinite, but ought to be acknowledged when comparing costs/benefits of airport projects vs. commuter projects. It’s something the commentators here mention regularly, but which Alon barely (if ever) acknowledges.

    • Alon Levy

      A few things.

      1. The one-seat ride to the City provided by Crossrail indeed makes the Heathrow connection much more valuable. My specific criticism of the Heathrow Express is that it terminates at Paddington, which is relevant to the plans for a CDG Express terminating at Gare de l’Est. In contrast, in Stockholm, I’d get to the airport on the Arlanda Express and not on the local commuter trains, since the Arlanda Express was faster and more frequent. The Arlanda Express also benefits from serving Stockholm Central, in a monocentric city where the bulk of commercial and hotel development is next to the train station. In New York, the mass of hotel development is nowhere near Penn Station and closer to the N/Q/R/W than to Grand Central, favoring a vanilla subway extension to LGA.

      2. It’s difficult to realign the Northeast Corridor for mainline trains to enter any of the current Newark terminals. If an airport station is desired, it’s easier to move the terminals to be closer to the existing station. A branch that terminates at Newark is also a possibility, but intercity trains have no reason to stop there (it competes for precious train slots with NY-DC and NY-Philadelphia trains) and even commuter trains are dicey (nearly all commuter ridership on the NEC in New Jersey is from points south, like Metropark or Princeton Junction).

      3. 10% is not just the Heathrow Express, but also Heathrow Connect.

      4. With Crossrail, it will not be possible to have an express service to Heathrow. There’s too much expected traffic in the Crossrail tunnel for any stop-skipping, just as the RER trains run all local in the central tunnels.

      • Adirondacker12800

        There are origin passengers in New York. They do silly things like live on Long Island and want to use the LIRR from the Island. Or all the people in Queens who live along the the Queens Blvd line and can just use the subway to get to Jamaica. Then there are people in Brooklyn who live along the J and can get to Jamaica or the A and Howard Beach. Airtrain has utility within the airport and going to Jamaica and Howard Beach makes sense. Choice-riders that stay in hotels in Manhattan take the train because traffic is awful. People who take the train to save money aren’t staying in hotel within walking distance of Penn Station or Grand Central. …. Or they ferreted out a motel along the Queens Blivd line. The super express from Manhattan isn’t going to stop at all the terminals and many people who use it will still have to get on Airtrain. … I’m staying with friends in Tribeca. Do I get on the A train, go to Penn Station, get on the superexpress that doesn’t go to my terminal and use Airtrain to get to it? How much better is that than getting on Airtrain in Jamaica where I share frequency with the commuter trains? …. or do I just get on the A to Howard Beach and change to Airtrain there? It may take longer but it’s only one transfer…. Meh…. Or with friends in Morningside Heights. Meh.

        It costs billions to build airport terminals. Moving them across the highway to the train tracks would cost billions and billions. Airtrain has utility in the airport… it’s not worth it to save a few minutes for rail passengers – moving the terminals. Or building 8 stations in JFK.

        Far far in the future it makes sense to have the Kodama stop at Newark Airport.. if you are in Wilmington or Lancaster or New Haven it’s an option. It’s the same train people use to get from those places to those places or New York. It’s not if you are Boston or DC. You’d use the big airport close by.

      • Ben Kurtz

        1. Obviously, terminating at Paddington is far less useful than operating through to Canary Wharf with stops at major destinations in between. My point is by the time the British took the decision to build a train that would terminate at Paddington, they knew there was another project in the pipeline that would integrate seamlessly with Heathrow Express/Connect. The only flaw was the twenty-year wait for Crossrail.

        2. Noodling over ideas to improve mainline connections at Newark is interesting, but clearly requires a separate blog post somewhere — can’t hash all that out in the comment threads.

        3. I use the expression “Heathrow Express/Connect” consistently to make just this point; c’mon, man!

        4. I also acknowledge this point in the London context. Which is why I emphasize one seat ride and reasonable speed, not “no stops at all cost.” The point which militates against running expresses in addition to locals is mainline capacity constraint when additional Crossrail traffic comes online. But the elevated value of a reliable, comfortable, one-seat ride in the airport context remains, which should be factored into a decision regarding whether or not to build something like the Heathrow tunnels.

        • Adirondacker12800

          Newark Airport stop is relatively new, it opened in 2001. It’s on a very very straight piece of track. At the time there was a proposal for a trolley/tram line for local service between Newark and Elizabeth so space was reserved. And they put in six tracks and two island platforms. Trains that don’t stop in the very center, express trains on the center side of the island and locals on the outer side. There’s no reason to change any of that until the NEC intercity services are so busy it makes sense to bypass Philadelphia and Baltimore…. A few minutes on the people mover is good enough. Especially since the people mover has other utility and there are three terminals.

  20. Adirondacker12800

    Newark ( and JFK and La Guradia ) will still need a people mover. Getting people to the train station is one of their uses. They keep vehicles out of the terminals. The people movers carry people to the parking lots, car rentals, hotel shuttles and the kiss-n-ride/cell phone parking. Without it there would be flocks of small buses competing with the high capacity buses, taxis and private cars. . . and they carry the inter-terminal traffic.
    Newark ( and LaGuardia and JFK ) is built on a salt water marsh, While possible, it’s probably best to avoid tunnels and underground stations.

  21. Bruce Hain

    The LaGuardia train is planned with stops directly alongside the terminals – unfortunately it doesn’t go anywhere, and given the physical obstructions, this is really not possible any time soon. (see my not entirely conv. rail solution at: In the case of Newark, a conventional rail loop should have been provided for when they went to the three terminals. What possessed them to do otherwise? (or maybe they did provide it and that’s the approx. route the PM follows) It seems that now this kind of thinking is the customary MO and an excuse for the transit planning cognoscenti to continue in the same vein, to wit: “Should there even be a One Seat Ride to the Airport?” Short of Star Trek, rail is the quickest way to get to the airport, and in most instances I know requires a direct route with limited stops to achieve that – to compete. To say that providing for this is not one of the assumed “givens” of regional rail planning is an absurdity, and provisions for it around here have been in almost every instance, woefully inadequate (a phrase the chief planner of the NERL used to use – not that the First Operable[sic.] Segment isn’t same) so we have passed into the realm of (at least) absurdity.

  22. Pingback: I Almost Worked for Hyperloop One | Pedestrian Observations
  23. Pingback: Tres pifias ferroviarias catalanas » Politikon

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.