Airport New York

Lavishing Money on Access to Stewart Airport

Stewart Airport Transit AccessNew York State has many transit needs, but this shouldn’t be one of them.

For years, New York City’s airports have suffered from massive congestion, with the airspace over the city continually trapped in traffic. Delays at Kennedy, LaGuardia, or Newark Liberty airports have the tendency to back up the entire American flight system. As a result, in 2007, the Port Authority of New York and New Jersey, which operates the other New York area airports, agreed to take over operations of Stewart International, an under-utilized facility located in Orange County dozens of miles north of Manhattan. Stewart would serve as a new base for air service expansion.

The problem is that the airport is difficult to get to, not only because it’s far from New York’s 8 million inhabitants, but also because it offers no direct public transit links. In 2007, the Metropolitan Transportation Authority began a study of possible connections to the airport, pinpointing ferry, commuter rail, light rail, and bus rapid transit as potential options. Two years later, the MTA is closing in on a final decision about how to make the link.

To hear MTA talk about this project, you’d think that it’s also about improving service to Upstate New York, providing residents of Orange, Putnam, Ulster, Dutchess, and Rockland Counties a new airport from which to take flights. An additional benefit of improving links to Stewart would be improved transit for any commutes in the region. MTA argues that this area is the fastest-growing in New York State and therefore that it needs more transit.

But here’s the truth: of the 84 million trips taken on Metro-North commuter rail lines last year, only 2 million were made on the west side of the river, where the airport lies. Only 500 daily passengers ride the ferry between Newburgh on the west side of the river to Beacon on the east, where there’s access to Metro-North’s quicker Hudson Line. In other words, this area may be growing, but it’s still a very small travel market. Improved transit links in the region are going to affect a very small number of people. This project is really about connecting passengers in New York City to the airport.

Current options, according to the MTA, include constructing a dedicated rail link between the Stewart Mills Station on the Metro-North Port Jervis Line (west side) and improving bus service through new express service from Manhattan and from the surrounding counties. A ferry to New York City has been eliminated because it would take too long and be too expensive; a rail link to the Metro-North Hudson Line across the river has been removed from consideration because of cost and environmental concerns. The project, according to an article back in 2007, could cost one billion dollars.

Ben over at Second Avenue Sagas has written in both 2007 and 2008 on the poor thinking behind this project. The airport is quite far away from the city’s population centers and will therefore have difficulty attracting crowds from the city; the airport’s current offerings of flights to just five destinations — Philadelphia, Atlanta, Orlando, Ft. Lauderdale, and Detroit — indicate that a serious increase in demand there from locals is unlikely over the next few years. Few commuters are going to be willing ride the 90 minute plus train between Penn Station and the airport, so why is this link a priority? It certainly doesn’t seem likely to cut down on air congestion.

Let’s imagine that the $1 billion existed to build this project, unlikely enough considering the MTA’s dismal fiscal situation. Wouldn’t it make more sense, from the perspective of improving transit, to spend it on desperately needed projects such as the Second Avenue Subway? People in Orange County — population 350,000 — may want more transit, but so do the roughly 350,000 people who live in East Harlem and the Upper East Side, and the latter group, to say the least, is far more likely to use public transportation than the former. Certainly, cheap express buses should be considered, but a rail link seems completely unnecessary.

But what about relieving congestion at New York’s airports? As I already pointed out, I’m not sure how many people downstate are going to be willing to make the commute up to Stewart, so an investment there may be pointless. Here’s another option, however: invest the $1 billion in New York’s high-speed rail program, reducing the dependence of New Yorkers on planes to get to Albany, Syracuse, Rochester, Buffalo, and even Montréal and Toronto. Those destinations collectively account for 202 daily departures from New York area airports (109 to New York State destinations, 93 to Canadian destinations). Eliminating half of those flights would open up the equivalent of a full hour of rush-hour capacity at JFK Airport. Talk about congestion relief!

Image above: Stewart Airport Access map, from Metropolitan Transportation Authority

29 replies on “Lavishing Money on Access to Stewart Airport”

Don’t forget this Freakonomics post that argues that the cheapest way to alleviate congestion at Newark and Kennedy is to shut down La Guardia. Add high-speed rail to the mix and we’d never know that La Guardia is gone. As it is now, with the AirTrain it’s easier to get to JFK from most of Western Queens.

Nice post, but I don’t think the mass-transit numbers for Rockland and Orange counties that you quote are really relevant. Plenty of people live in those areas; they just don’t take the train, bus, or ferry.

Stewart airport has excellent road access from I-84 and I-87. The problem that I see is that Newark and JFK also have excellent highway access from northern NJ and Westchester, respectively. Why drive to a new airport when you can drive to your old favorites? If you want to make Stewart more attractive, I suggest raising tolls on the Garden State Parkway, the NJ Turnpike, the Throgs Neck bridge, and the Whitestone bridge. For extra oomph you could take out the tolls on the Thruway south of Stewart, too.

There are ways of improving rail service to Rockland and Orange Counties, but none that involves Stewart Airport. A one-seat ride to Manhattan would be one of them; electrification, crosstown service to Westchester, and revival of the West Shore Line would work as well. The Port Jervis Line is so circuitous that there is no way it could cut its travel time to what more than a few hundred people are willing to take.

Low transit ridership on the west side of the Hudson is also indicative of few transit options – I think putting some more commuter rail in that area is a fine idea (not a priority over the Second Ave. Subway, but still). Nevertheless, expanding Stewart is a markedly uncreative way to deal with the NY area’s air traffic congestion. I think people who live in the northern ‘burbs may go there, but nobody else will.

Also, imagine if they improved Northeast Corridor service on Amtrak – you remove flights to DC, Boston, Baltimore and Philly (yes, there are flights from Newark to Philly – how absurd is that?), which is probably an even greater number than upstate and Canadian cities.

The best idea would be to marry the future Tappan Zee bridge project with a new Albany-NYC electric HSR line that runs down the I-87 ROW. Not only would it connect to Stewart, but also Kingston, SUNY New Paltz, Woodbury Commons and any other west shore points that are under-served. The current west shore Metro-north service is horribly slow, so a new ROW would also improve straight commuting efficiency as well. The Tappan Zee will already have the tracks for a new Hudson line connection, and former Sen. Bruno’s high-speed task force has already floated the idea of using I-87 as a ROW. That would connect Stewart to NYC in less than an hour.

This money would be better spent to expand service at Tweed airport in New Haven and/or providing transit links to HPN on the NY/CT line. Those two airports are closer to population centers, and already have close proximity to the types of heavy transit facilities that could create a significant reduction in regional traffic congestion, particularly through the Bronx.

Transit riders as a percent of all air passengers, I bet it is less than 5 per cent at LGA, JFK, and EWR. Maybe a larger share of airport employees, but travellers, very very very few. (Not counting taxi as transit.)

So now they need to spend a billion to get transit to the boonies? Have some big developers acquired vacant land near the future stations? Otherwise it’s hard to explain.

How about Giuliani’s notion to extend the Astoria line out to LaGuardia? That could serve many more workers and some more passengers.

The only good reason to invest in Stewart is if — or when — the ice caps melt and the sea levels rise. Then all three NYC airports will be waterlogged, while Stewart will still be on high ground.

While most airport users generally don’t use transit to get there, it really depends on the city:

35% of Tokyo passengers
25% of London Heathrow and Hong Kong passengers
20% of Paris CDG passengers
15% of Washington National passengers

I can’t find New York information, but considering the city’s density and size, ridership should be high, but the inferiority of transit connections probably means it isn’t.

The Newark-Philly planes are probably for connecting flights, Amtrak being cheaper and faster otherwise.

Scott: it’ll be far cheaper to straighten and then four-track the existing ROW from New York to Albany; it’ll also serve the major intermediate population centers better.

Woody: on the contrary, airport workers usually drive. The AirTrain serves terminals, rather than hangars, runways, and other places where employees are expected to report to work. As for the AirTrain, it has a bit over 10% of the ridership at JFK, and a bit over 5% at Newark.

I don’t have the link but from what I have read, this project would be largely funded by the PANYNJ. Since operating airports (including Stewart) is part of its mandate. Funding other high speed rail projects in a single state is not (especially outside of the port area).

The MTA would receive funding for improving the PJ line which it would otherwise not be able to afford.

Alon —
Please recheck your figures. I suspect they are cooking the books. Total AirTrain passengers — including those going from, say, Terminal One to Terminal Six, might be 10% of passengers at JFK.

But boardings at Jamaica and Howard Beach, not 10% of arrivals by my eyeballs, no way.

And from my own observation and belief, from the times I have ridden the A Train to Howard Beach and JFK, the airline and airport employees outnumber the foreigners, who outnumber Americans like me by about 5:1 or 10:1. Go meet the 2 a.m. arrival from Guyana and see only airport workers on the train.

AirTrain was a tremendous waste of public money.

No, the statistics only count paid riders. Port Authority has no way of knowing how many people take the free AirTrain.

I’ve only ridden the Howard Beach AirTrain once each way. Jamaica is far more useful if you’re traveling to or from Manhattan, and there I see no employees and a lot of travelers.

Alon & Woody, my two cents is this: I see far more airport employees on the JFK airtrain than on any other intra-airport train, such as the ones in Newark, Dallas or Atlanta. Those peoplemovers (I don’t want to dignify them as “monorails,” or “trams”) are populated in my experience exclusively with travelers.

If I assume that most airport employees are assigned to a single job site, then it could follow that the JFK workers are using the airtrain to get to and from work, or to and from lunch.

Alon’s source had this explanatory comment:
“people employed at the airport who used the two rail systems to travel free of charge between terminals, parking, and car rental areas.” So that explains why we see employees. Maybe all the passengers take the Jamaica Station line.

Maybe I’ll try that next time. AirTrain to Jamaica, change to the E-train, ride to mid-town, cross to the uptown A-train — uh oh, doesn’t that mean lots of stairs? I told you I have arthritis and now consider the stairs before I enter the subways … so I dunno.

I take the LIRR from Jamaica, not the E. It costs a bit extra, but takes 15 minutes less and is suitcase-friendlier. There are elevators at Penn Station; I think there are escalators at Jamaica, but not to all tracks and not always in the direction I need.

Guys don’t forget that NJ is west of the Hudson. One cannot mention a Port Authority project and then isolate the downstate west of Hudson MTA MNRR ops. MetroNorth’s west of Hudson service is integrated into NJ Transit operations which are very heavily used. We all know that MN’s Port Jervis line becomes NJT’s Main Line once it crosses the state border.

The PA is not building Stewart for Downstate NY residents alone. It’s a bi-state agency – both Governors have to approve the capital plans. NJ’s needs are an important part of the mix.

There is massive congestion at three NY area airports. EWR and LGA are notoriously affected by bad weather. All three airports are hemmed in by natural features or high density development. We need greater air transport capacity (yes saving is great too) Yet there is no where to grow or repair the mistakes of the past made at our legacy airports.

An FAA conceived scheme to impose congestion pricing on take offs and landings (dropped) might have made Stewart a very attractive option for leisure and business travelers looking for more reasonable options. Something similar could be in the offing that would make people take a second look at Stewart.

The PA is improving NJT’s access to Manhattan through its funding of ARC. Part of that effort will include one seat rides to Penn Station. Its not outrageous to conceive of a Stewart Airport Express train making stops at Secaucus Junction, Ridgewood and Suffern before heading to the airport.

I’m dubious that only a few hundred would find it useful. I think electrification of the west of Hudson lines is something that should be examined by the PA, NJT and MNRR.

Even in New Jersey, the Main/Bergen County Line and the Pascack Valley Line see paltry traffic levels. They’re slow, regardless of the fact that they feed into Hoboken and not Penn Station; the route of the Port Jervis Line in New York is especially circuitous.


I don’t know what you mean by paltry. Sure, the ridership is not anything like the NEC but during rush hour, the 5-6 car trains are standing room only.

I don’t know what you mean by slow. Aside from a few curves, they are 79 mph running. An express from Ramsey-Route 17 will get to Hoboken in 37 min, which is faster than a car w/o traffic.

Investment by the PANYNJ will only make this better and will help address issues that the would not otherwise be able to tackle (for example, I have a copy of the MIS which specifically mentions straightening out the curve between Suffern and Sloatsberg which is a huge expense requiring a viaduct. It also provides for double tracking to Salisbury Mills-Corwall).

The PJ Line is only really circuitous west of where a Stewart Airport Link would be.

Chris: the New York Times has ridership statistics for each station (link). The Bergen County Line, Main Line, and Pascack Valley Line have a total of 13,500 weekday riders among them, slightly more than 5-6% the systemwide total. The Metro-North west of the Hudson gets another 3,000 weekday riders, a little more than 1% the MNRR total.

I’m not sure about speeds in New Jersey, but in New York, the Port Jervis Line is particularly onerous, with its two-hour Port Jervis-to-Hoboken runtime.

two-hour Port Jervis-to-Hoboken runtime.

Respectable for a 95 mile trip through the mountains. . .it’s farther from New York, by rail, than Philadelphia.

When I flew out of EWR to Rome last month, I noticed that the departure boards showed a plethora of flights to Albany, Buffalo, Montreal, and several other cities that, with a respectable rail system, should be more easily reachable without flying. It seems to me that the billion dollars that the Stewart link would take, correctly placed, could cut out quite a bit of congestion and the need for Stewart in the first place. Further, I can’t envision an airline risking meager capital on such a remote airport, especially since it has seen such little use for decades. I’d rather see that money go to improve rail travel through New York State, bring 150-mph travel to the Northeast Corridor through NJ, ease bottlenecks through Connecticut, or even a third track on the North Jersey Coast Line through my town.

Amtrak already has a decent chunk of air/rail traffic from New York to Albany, I think 30-ish percent. Buffalo and Montreal are a different matter, but neither would be the first or even second stage of a Northeast HSR initiative. Montreal would probably be the last stage, due to little economic integration between it and the Northeast and no intermediate destination between it and Albany.

You’re right–Buffalo and Montreal are pretty outlandish at this point. My point was that there are projects where those billion(s) will have far more impact. There is no reason why trains can’t run at 150mph on constant-tension catenary through NJ on the NEC, and for that matter, there is plenty of room for improvement on the Empire Corridor.

Anyone who has flown in to Heathrow or Gatwick will know that an airport with a fast / direct connection can be miles from the city. Hong Kong’s recently built airport is another example. Newark isn’t exactly close to Manhattan, or particularly well connected, especially during rush hour. And JFK is worse than Heathrow in terms of passenger experience, and that is saying something. What other options are there?

Once Heathrow is connected to London via Crossrail, then, I think, passenger experience will be superior to JFK. However, unless your destination is walking distance from Paddington, passenger experience there is not really that much better. The breathtakingly expensive Heathrow Express is not worth £18 since it typically only saves about 15 minutes when factoring in the inevitable transfer to the tube at Paddington. The JFK Airtrain/E train combo is comparable in time, cost and convenience to taking the tube.

Leave a Reply