What’s the Infrastructure’s Highest Value?
A piece of land and infrastructure may have multiple uses. Land might be needed for urban development or for a highway. A two-track structure might be needed for freight or passenger service. A right-of-way might be needed for multiple kinds of rail, or a road, or a power line easement, or a park. In all cases, the correct policy choice is to allocate the land to the use that has the highest social value, and this use depends on the situation at hand. It should not be allocated to whatever one fancies.
Concretely, let us consider the following cases:
1. The High Line. Occasionally, railfans grumble about the linear park, and say it should’ve had passenger rail service instead; read the comments on Ben Kabak’s post on linear parks, or New York City subway forums. But in reality, the High Line is very useful as a park in a busy neighborhood that doesn’t have other parks. In contrast, it’s nearly worthless as a transit line: it’s parallel to a north-south subway that’s operating well below capacity, it would be nightmarishly difficult to connect to any existing line, and the only east-west service it could possibly be useful for is connecting to 14th Street, not the most important job destination in the city.
2. The Northeast Corridor in Rhode Island, south of Providence. The expansion of MBTA commuter rail southward into sprawling exurbs is a major failure of regional transportation policy. Providence is not all that congested by the standards of the larger Northeastern cities; auto-oriented commuter rail toward it is doomed to fail, and near-downtown parking is cheap and plentiful. (The commute market from Warwick and Wickford Junction to Boston is trivial.) In contrast, the line is perfect for intercity service, since it has relatively gentle curves outside city limits, and is straight south of East Greenwich. The South County project not only costs $200,000 per weekday rider, but also makes poor use of high-speed track. Since the line is more important as high-speed rail than as a commuter line, Amtrak should be more aggressive about demanding that commuter projects create their own capacity.
3. The Northeast Corridor in Maryland, north of Baltimore. For the same reasons as the MBTA extension’s eventual failure, MARC underperforms north of Baltimore. Although the line has extensive three- and four-track segments, the bridges are two-tracked, and high-speed rail should again be given priority, including canceling commuter rail if necessary. Ironically, because of more extensive four-tracking, the need for bypasses around Wilmington and perhaps North East, and the at-grade track layout, Perryville is quite easy to connect to Philadelphia by commuter rail without interfering with intercity rail.
4. Caltrain to San Jose, the MBTA to Providence, MARC to Baltimore. In contrast with the situation in points #2-3, those three lines are all useful commuter lines; they are all similar in that they connect two distinct cities that share suburbs, with a rump extension that exists purely for show (into Gilroy, Perryville, and soon to be Wickford Junction). Any and all high-speed rail use of these corridors should permit a reasonable frequency of commuter trains, with timed overtakes when possible and full four-tracking otherwise. On Caltrain, in particular, interference with commuter rail is one reason why the chosen Pacheco Pass alignment is inferior to the Altamont alignment.
5. The Lower Montauk Line. Despite perennial railfan desires (and an empty Bloomberg campaign promise, since scrubbed from his campaign website) to restore passenger service, there’s not much point in regional rail that stub-ends in Long Island City. To give an idea how much demand there is, the LIRR currently runs 5 trains per day per direction into Long Island City. Thus, the line is more useful for freight trains than for passenger trains. This will change if, and only if, there is a way to connect the line to Manhattan through the existing LIRR tunnels, or perhaps new tunnels, but then the cost is going to be orders of magnitude higher than just restoring service.
6. Urban freeways, e.g. the BQE. American freeways were built at a time when, even more so than today, land was allocated based on political power rather than any sort of social consensus or market pricing concept. While Japanese cities have to make do with 4-lane freeways due to high land costs and strong property rights protections, American cities demolished entire neighborhoods to make room for freeways with wide exclusion zones around them. The land occupied by some would be more useful for additional neighborhood housing growth than it is for a freeway. For example, the BQE hogs prime real estate in Williamsburg, right next to the under-capacity Marcy Avenue subway station, and to a lesser extent in the rest of Brooklyn and Queens, and this land could be used for high-density development instead.
Pedestrian Observations 
Do you think there’s any feasible/legal way to convert Montauk to a subway line? Let’s say within the next 50 years and in conjunction w/ a massive rezoning/cleanup of industrial Maspeth.
Yeah, they could just abandon freight service and cordon off two tracks at Jamaica for Montauk service. There’s no real regulatory hurdle. The problem is that there’s really no point in trains that only go to LIC. There’s no way around doing some heavy lifting with concrete there, of which linking to the East River tunnels is just one option.
There are also a number of grade crossings that would have to be eliminated, mainly in Maspeth. Or the roads could just be cut.
Couldn’t Lower Montauk trains reach Penn Station by looping around Sunnyside Yard? It looks like the tracks needed to do so (the chord at 25th St and the loop at 43rd St) already exist. Would cost several minutes in each direction but would still be a lot faster than current commutes involving busses.
Could be, sure. Though, honestly, looking at satellite images, it looks pretty cheap to just build a viaduct and connect to the westbound southern tunnel. The eastbound southern tunnel’s portal is too close, but some underground connection may be feasible – those tunnels have an underground flying junction, and once ESA opens, the southern tunnels can even be closed temporarily with no ill effect except a very minor slowdown of Amtrak in the Penn Station throat.
Is the BQE elevated or sunken? How is it (elevated/sunken)? I think that it’s possible, with a loosening of regulations and government agencies actively seeking to gain money through ground or air-rights leasing, to urbanize the area around the freeway in spite, as it were, of the freeway. Subsidized micro-infrastructure platforms, enclosure of spaces under viaducts, hollowing of embankments to provide ground-lease opportunities, classic exercise of air rights, air rights as an urban design tool, all these things will likely impact how we deal with urban freeways, and how these roads will be able to meet budget going into the future.
The Lower Montauk or the Rockaway Beach line or whatever you want to call it doesn’t have to send trains to Hunterspoint Ave or LIC, they can go to Penn Station. You can still see the ROW and the curves the trains would take as they merged and diverged from the mainline in Rego Park. I don’t know if anyone would use it. The subway is cheaper and takes you directly to where you want to go. Maybe a trolley line from Queens Blvd down to Rockaway Blvd, but even then you’d be better off using Woodhaven Blvd. … Make it into a park and let the deed remain with the MTA and it’s successors. Who knows what someone may want to build in 100 years.
http://en.wikipedia.org/wiki/Rockaway_Beach_Branch
I’m going to have a different post on this, but these are two different lines. Rockaway Beach is the branch that figures prominently in railfan fantasy maps. Lower Montauk is the line that goes east-west from Jamaica to LIC farther south of the Main Line.
And yes, part of what I’m going to say is that Woodhaven is a much more useful corridor than the Rockaway Cutoff.
If Wikipedia is to be believed it had passenger service up until 1998. Nobody used it. Probably because it’s faster to hop on the bus that comes by frequently and go to a subway station.
It had one train a day, I think, and all the stations except Richmond Hill had low platforms, so when the LIRR got its bilevels, which are not compatible with low platforms, they just didn’t bother upgrading the stations on that line, for just one train a day. The one train a day still runs on that line, allegedly because it’s required by the LIRR’s charter.
Most of the fantasy proposals I’ve seen actually hook a restored Rockaway Beach line into the Queens Boulevard subway, as opposed to the LIRR Main Line. It leads to interesting questions about potential service patterns, to say the least.
What do you think about the proposed uses for the abandoned Essex Street trolley terminal?
To be honest, I haven’t thought about it much. But my instinct is to say that it can be given over to a different use, like Paris’s Gare d’Orsay, since the subway made the trolley terminal redundant.
As far as speculative uses, I’d like to see restored light rail over the Williamsburg Bridge, running on the surface on Utica/Nostrand/Grand in Brooklyn, and into the disused pair of Nassau St tracks from Essex to Chambers in Manhattan (where there’s potential for a future incline back up to the Brooklyn Bridge to serve other Brooklyn points). This definitely goes in the category of railfan pipedreams though.
The real problem with the Essex St proposals is that it’s difficult to imagine anybody being attracted to a “park” or shopping area in a dank (and likely noisy) hole in the ground. Most New Yorkers already spend more time underground each day than they’d really prefer; even the cleanest and most recently-renovated subway stations are not the sort of place one would want to spend leisure time (railfans notwithstanding). But there’s no particular need for transit advocates to point out this issue; it’s clear enough to all involved that I think it’s very unlikely the Essex St terminal will ever see non-transit use.
It is almost as if you wish for a more market-oriented structure in our transportation policy. How else could these optimizations happen? They won’t happen in any political sphere, where allocations of resources happen according the whims, posturing, and machinations of politicians. I guess in a theoretical world we could work this all out with some good analysis and healthy cooperation, but in the real world, we have agency turf battles, favored interest groups, and really bad analyses to inform our decisions.
In Switzerland, they work it out with good analysis and healthy cooperation.
Markets are no panacea, because they always get taken over by whoever (a) has the most information, or (b) has “market power” (single supplier, etc.), or (c) has control of the legal system which defines the rules for the markets, or (d) if there is no such system, whoever is crookedest wins….
This is essentiallly a political and social problem. How can we make our social consensus more like Switzerland when it comes to infrastructure? Good question, and I really don’t know the answer.
Market pricing of land has a lot of good uses – for one, it’s much faster than anything else. The main problem of high housing prices in places like New York is that the permit process lags the rise in demand, and construction lags even more. Thus, although on a multi-decade average basis housing is affordable, the slowness of regulation means that at certain times housing will not be affordable.
Switzerland has a lot more problems with affordability than it seems. For example, the Wall Street Journal (yes, I know – still) has an article about the Canton of Zug, which slashed taxes to attract investment, and became so successful at it that non-millionaires have all been priced out. In general, Swiss consensus democracy is very friendly to very rich people, and the level of social democracy when it comes to e.g. health care is a lot lower than in other European countries.