Shut Down the Brooklyn-Queens Expressway

New York’s high construction costs are not just a problem for public transit. Roads exhibit the exact same problem. Case in point: replacing 2.5 km of the deteriorating Brooklyn-Queens Expressway (BQE) in Brooklyn Heights is slated to cost $3-4 billion, take 6-8 years, and require temporarily closing the pedestrian promenade supported on top of the highway. This is not even a tunnel – some local NIMBYs have proposed one in order to reduce the impact of construction, but the cost would then be even higher. No: the projected cost, around $1.5 billion per kilometer, is for an above-ground highway replacement.

The section in question is between the Brooklyn-Battery Tunnel and the Brooklyn Bridge; the Promenade is the northern half of this section.

Is it worth it?


There exist infrastructure projects that are worth it even at elevated cost. Second Avenue Subway Phase 1 cost $4.6 billion where it should have cost $700 million, but the expected ridership was very high, 200,000 per day, and so far ridership is on track to meet projections: the three new stations had a total of 138,000 boardings and alightings between them in 2017, and the revamped 63rd Street station went up by another 8,000. The BQE replacement is not such a project. Current traffic on the highway is stated as 153,000 vehicles per day, so on a per-vehicle basis it’s similar to Second Avenue Subway’s per-rider projection, around $23,000. But cars are not transit and cities need to understand that.

The construction of a subway creates noise and traffic disruption, but once the subway is up, all of that is done. Even elevated trains cause limited problems if built properly from materials that minimize noise – the trains on the viaducts on the Paris Metro are less noisy than the cars on the street below. There are operating costs involved with subways, but fixed costs are so dominant that even in New York a busy line like Second Avenue Subway should be at worst revenue-neutral net of costs; for reference, in Vancouver the projection for the Broadway subway extension’s operating costs is well below the revenue from the projected extra ridership.

Cars are not like that. They are noisy and polluting, and greenwashing them with a handful of expensive electric cars won’t change that. There are benefits to automobility, but the health hazards cancel out a lot of that. The Stern Review estimates the cost of unmitigated climate change at 20% of global GDP (e.g. PDF-p. 38), which in current terms approaches $500 per metric ton of CO2. The US has almost the same emissions intensity per dollar of production as the rest of the world; the negative impact of cars coming from climate change alone is comparable to the total private cost of transportation in the US, including buying the car, maintenance, fuel, etc. Now add car accidents, noise, and local air pollution.

In a region where cars are an absolute lifeline, there’s a case for building connections. The costs are low since grading a road for medium speed with level crossings is not expensive. In cities, the situation is different. Drivers will grumble if the BQE is removed. They will not lose access to critical services.

Is anyone proposing removing the BQE?

Yes, there are some proposals to that effect, but they’re so far only made haltingly. Council Speaker and 2021 mayoral frontrunner Corey Johnson’s report on municipal control of the subway includes the following line: “Before spending $4 billion to reconstruct a 1.5 mile stretch of highway, the City should study alternatives to the reconstruction of this Robert Moses-era six lane road, including the removal of the BQE in its entirety.” The halting part here is that to study does not mean to enact; Johnson himself opposes repurposing car lanes for bus service in his own district.

City Comptroller Scott Stringer, who has relied on a lot of the information I have brought up in this space in his reports, proposes to keep the BQE but only allow access to trucks. Bloomberg’s transportation commissioner Janette Sadik-Khan agrees with the idea and even pitches it as a brave alternative to the car. In other words, per the comptroller and former commissioner, billions of dollars are to be spent on the reconstruction of a somewhat narrower structure for 14,000 trucks per day. Stringer’s report even says that the comparable urban freeways that have been removed did not allow trucks in, which is incorrect for the Embarcadero Freeway in San Francisco and for the Voie Georges Pompidou in Paris (look for “camions” here). In reality, if closing the BQE means adding just 14,000 vehicles to surface streets, then it’s an almost unmitigated success of road dieting, since it means much less pollution and noise.

The Regional Plan Association proposes its usual quarter-measures as well, sold under the guise of “reimagining.” It does not mention closure at all – it proposes rebuilding the structure with four lanes, down from the current six, and even dares to cite the closure in Paris as precedent. Everything in its analysis points out to the benefits of full closure and yet the RPA feels too institutional to propose that. Presumably if the RPA had opined on lynchings in the midcentury American South it would have proposed a plan to cut total lynchings by 25% and if it had opined on Fourth Republic-era colonialism in Algeria it would have proposed to cut the incidence of torture by a third while referencing the positive precedent of British decolonization in India.

What should replace the BQE?

The BQE should be removed all the way from the Brooklyn-Battery Tunnel to the Williamsburg Bridge. Its curves in Downtown Brooklyn with the loops to the Brooklyn and Manhattan Bridges consume valuable real estate, and farther east they divide neighborhoods. The new Navy Yard developments are disconnected from the rest of Brooklyn because of the BQE.

Going east through Fort Greene, the BQE is flanked on both sides by Park Avenue. Buildings face the street, though many of the lots are empty or low-value. Thus, the surface streets have to stay. Selling what is now Park Avenue as parcels for residential and commercial development and mapping a street on the BQE’s 30-meter footprint is probably not viable. Instead, most of the footprint of the expressway should be parceled into lots and sold, converting Park Avenue into a one-way pair with streets about 12-15 meters in width each. East-west buses will continue running on Flushing and Myrtle, and north-south buses should probably not make stops at Park.

In contrast, going south through South Brooklyn, buildings do not face the abutting surface street, Hicks. They present blank walls, as if it was midblock. This is a prime opportunity to narrow the street as if the highway has never been there, creating an avenue perhaps 20 or 30 meters in width. The wider figure is more appropriate if there are plans for bus lanes and bike lanes; otherwise, if buses stay on Columbia, 20 is better.

In South Williamsburg, the road is nearly block-wide. The neighborhood is pro-development due to high birthrates among the Haredi population. Thus the footprint of the freeway must be used for private housing development. The area next to the Marcy Avenue subway station on the J/M/Z is especially desirable for the non-Haredi population, due to the proximity to Manhattan jobs. The city should retain an avenue-width roadway for Williamsburg Bridge access from the south, but beyond that it should restore the blocks of the neighborhood as they were before the BQE was built.

Heal, don’t placemake

If there’s a common thread to the various proposals by local politicians and shadow agencies (that is, the RPA), it’s an attempt at placemaking, defined to be any project that they can point to and say “I built that!”. A BQE rebuilt slightly narrower, or restricted to trucks, achieves that goal, with some greenwashing for what remains a waste of billions of dollars for motorist convenience.

But the same can be said of a park, as in one architect’s proposal for the BQE. I can see a case for this in Brooklyn Heights, where the Promenade is an important neighborhood destination, but elsewhere, the case is extraordinarily weak. In South Brooklyn, the most important benefit of removing the BQE is easier pedestrian access to the waterfront; recreation space should go there. Fort Greene and the Navy Yard are both rich in parks; BQE removal makes the large parks on both sides of the motorway easier to access. And Williamsburg is hungry for private development, whether near the subway for Manhattan workers or elsewhere for Haredi families.

Thirty years from now, nobody is going to walk on the remade street grid of South Williamsburg or the narrowed Hicks Street and wonder which politician set this up. But people may well notice the lower rents – and they may well notice them within a few years of the deconstruction of the road and the sale of the land for housing development. Ultimately city residents do notice if things are improving or deteriorating. It’s on the city to nudge infrastructure development in the direction of less pollution and fewer boondoggles.


  1. Henry

    What would your proposal for freight be? The Cross Harbor Tunnel is basically pushed by a single person, you probably don’t want to waste valuable Manhattan rail capacity on freight, and rail on the Tappan Zee was abandoned. Not to mention that the BQE itself serves as a link to industrial areas.

    • Alon Levy

      1. Belt Parkway.
      2. Surface streets, it’s not that many trucks.
      3. Maybe don’t pollute the city to hold on to waterfront manufacturing on the wrong side of the harbor just so that some politician can play to a national mythology of industrial jobs.

      • AstoriaBlowin

        You have to raise the overpasses on the Belt to accommodate trucks, not a major problem in theory, but with NYC incompetence and corruption that would probably cost north of a billion and take 3-4 years. Bring in that crew from the Netherlands who replaced the highway tunnel in three days and we’d be ready to shut down the BQE by summer time!

      • Henry

        The Belt is incredibly roundabout, particularly for distribution of goods to Long Island via the LIE. (It’s also not currently capable of handling trucks at all, and incredibly congested even today, but I’ve no idea how expensive fixing either of those problems is. Given Jamaica Bay’s status as a wildlife preserve it might not be cheaper at all.)

        Not everything in Maspeth is industrial production. A lot of it is warehousing and the freight yards and such for distribution of goods in an island with 7.5M people on it.

        • adirondacker12800

          Assuming it has clearances. The Belt Parkway is a Parkway. Most of the day in the literal sense, that you can get out and walk faster. I think he’s forgetting about the Gowanus Expressway, the Belt ends in Sunset Park and the Gowanus connects the Verrazano Bridge with the BQE. That is all I-278 is a hint. Makes me wonder what those 14,000 trucks a day are carrying. Probably stuff like food and toilet paper. Bidets all around and have Amazon drones deliver food from distribution centers in New Jersey? I wonder how big they would have to be to deliver half kegs of beer to bars. …. All the way out to Greenport and Montauk.

  2. The Economist

    While I would love seeing the BQE disappear completely, I highly doubt that this would happen. This is the US and the car is the king in the outer boroughs. I am not familiar with all proposals on the BQE. Has anyone proposed open cut excavation for a tunnel? While it will be expensive, no doubt, it will be much cheaper than boring the tunnel and also will allow for buildings, parks and whatever else to be put on top in pretty much the way you are proposing. Open cut excavation cannot be any more disruptive or noise generating than the currently proposed aboveground replacements, but does result in elimination of the thing from the surface and opening up the area to development.

    • The Economist

      Also open cut excavation allows for the tunnel to be relatively shallow which is good for egress, ventilation and overall cost. Vibration will be worse than a deep bored tunnel, but cannot be any worse than with above surface replacement. Also shallow is good for easier protection of the thing from raising ocean levels and climate change.

      • Alon Levy

        I don’t even think open-cut excavation is feasible along that alignment? It’s feasible if you remove the highway first, but otherwise, there’s still no right-of-way for it in Brooklyn Heights or in DUMBO. This isn’t a subway, where you can make do with 10-12 meters of street width – freeways have wide footprints.

        And as for “car is king,” this is New York, a city where the majority of households are car-free.

        • Ross Bleakney

          >> freeways have wide footprints.

          Typically yes, but the only reasonable suggestion so far (other than demolition) is the one by the comptroller, which is to build a two lane highway (one lane each direction) to be used only by trucks. If this could be built using trenches, it seems like a very reasonable suggestion.

          Of course the other option is to just destroy it, and aggressively deal with the surface streets by adding congestion taxes or restrictions in general on cars. I have no idea what it would like, or if the idea has been studied. It is easy to see how bus lanes can help quite a bit, but what about truck lanes, or even the combination?

    • Henry

      One of the proposals is essentially capping a surface-level highway where the cantilever currently is. So you would replace Furman St with a highway, put a cap on it with a local street and a park, and then maintain the promenade or build some kind of terrace connection to Brooklyn Heights. So it should be less expensive than the Big Dig since it involves very little actual tunneling.

    • Alon Levy

      Most of it follows the Ringbahn, so the neighborhood-splitting issue isn’t really valid. That said, if Berlin wants to build more housing where people want to live, then the Westkreuz area could make nice TOD (although to be honest the most desirable part of Charlottenburg is closer in) if the Autobahn goes away. But it doesn’t look like the gash that the BQE is, or the the waterfront blocker that the Voie Georges Pompidou and the Embarcadero Freeway were.

      • Herbert

        The insanity about the A100 is that they’re actually still expanding it. Oh and it blocks southward extension of the U4, Berlin’s second most pointless subway line (after the accident of history that is U55)

          • Herbert

            There used to be a maintenance and storage tunnel south of the current endpoint back when the subway was the pride of the city of Schönefeld…

  3. Onux

    Comparing the RPA’s suggestion for a four-lane BQE to lynchings and torture is far out of line. Firstly, there is absolutely no moral equivalence between transportation policy debate and murder as a public spectacle or torture. Second, there is nothing evil about proposing different policies that stem from different assumptions or goals. You yourself have advocated keeping the Providence Line south to Attleboro two tracks with overtakes instead of three tracking the entire length. How would you feel if someone compared your plan to vaccinating only two-thirds of the population against smallpox in the 1950’s and leaving the rest to die?

    You may agree or disagree with the RPA’s plan, and you may or may not be right about the better plan for the BQE, but you can argue via facts, reason, and appeal. There is no need to resort to inflammatory false equivalences or default ad hominem attacks.

    Note that I have no objection to criticism of the RPA or other organizations for organizational traits (institutionalism, timidness, etc.) related to the type of policies they do or don’t support.

    • Alon Levy

      I would respond by noting that timed overtakes have higher capacity than a third track and talk about the need to integrate schedules with infrastructure. What I’m proposing is a way to get more capacity at lower cost.

      This is not the same as saying that freeways are bad and pointing out great examples of freeway removal and then recommending spending billions on rebuilding a freeway at slightly less capacity, with all the problems that construction would cause (which the RPA agrees are problems!).

      Would I compare this with torture and lynchings if someone said “driving is good, we need to maintain and expand road capacity”? No. Give me some credit. (I’d say “it was a beautiful planet.”) I do this in the context of an RPA that constantly compromises with itself on these issues instead of saying “tear down this road.”

  4. Onux

    I see your point, but I respond by noting that the examples you choose are needlessly inflammatory compared to the subject at hand. There is a massive moral and ethical implication in suggesting someone wants to only reduce some lynchings vs stopping them entirely. I would be more receptive to your argument if you had said “This is like the RPA proposing Second Avenue Subway Phase 1 with stops at only 86th and 96th” or “This is like the RPA proposing off board payment everywhere but bus stops south of 14th St” (both in the the context half measures when implementing a good idea).

    Your comment regarding timed overtakes just proves my point: there are good financial and technical arguments for timed overtakes vs three tracks, just as there are arguments for (and against) various BQE options (6 lanes, 4 lanes, removal). The appropriate path is to engage the arguments, not make simplistic and specious comparisons to acts of evil (2/3 tracks = 2/3 vaccination, 25% reduction in lanes = 25% reduction in lynchings).

    • Alon Levy

      4 lanes isn’t some clever way to increase capacity at lower cost than 6 lanes. It’s less capacity for marginally less money (since the structure needs to be replaced either way if it’s not removed).

      I don’t want to analogize this to partial off-board fare collection, because the lynching analogy is frankly better for a Northern organization, in the sense that there’s widespread understanding that there’s a problem there and that it should be fixed with all deliberate speed. In the specific context of a proposed rebuild, rebuilding the same thing but narrower just screams “we’re negotiating with ourselves” in ways that (say) a partial road diet elsewhere does not. It’s this desperate attempt to appeal to everyone at once even when the middle path is truly nonsensical; I’d understand an outfit that proposed an 8-lane rebuild better than one that’s proposing a 4-lane rebuild like the RPA is.

  5. marvin gruza

    Sorry, but there is another view:
    There is a need for highways and mobility
    We want to be able to get off Long Island/Queens
    People and goods need to get to/from Queens, Brooklyn and Long Island from Staten Island and NJ
    We do not want people driving through Manhattan to get to NJ
    A city our size should have many more lanes leaving to the verazano bridge
    Perhaps given the value to drivers provided by the BQE, it should be tolled to allow it to be rebuilt in a user friendly way
    Maybe tolls should be higher for those who do not actually cross the bridge, thus discouraging local usage which masstransit could serve
    perhaps an expressway replacing or along side the the Bayridge RR line should be revisited as a way to provide capacity in a more friendly place

    We need a good transportation mix and eliminating highways where alternative do not exist is not a good idea.

    • Alon Levy

      But alternatives do exist. There’s a subway, which includes the G train roughly parallel to the BQE as well as express trains through Manhattan that don’t get stuck in city traffic.

      Every NIMBY argument against closing the highway can equally be lobbed at tolling: it would incentivize drivers to use surface streets, it would make it harder to drive, etc. Ultimately it’s a question of whether you value clean air and safe streets or motorist convenience. Tolling is just part of the way toward elimination of the road, and given the high cost of replacement, it gets into the 20%-of-the-benefit-for-80%-of-the-cost problem.

      • adirondacker12800

        Where are the trucks going to go? The G train doesn’t go to Hunts Point Markets. Or distribution centers in New Jersey and Pennsylvania.

        • Alon Levy

          Surface streets, or nowhere (they’re 9% of expressway traffic). And Hunts Point is way better served by the Bronx expressway network anyway.

          • adirondacker12800

            You do understand how distribution centers work don’t you? Truckloads or railcars of stuff come in from long distances, lots of people shuffle stuff around and then mixed loads of stuff go back out to local destinations. The Bronx expressways don’t go to supermarkets in Brooklyn. I suppose Amazon can train the drone to hover outside your bathroom window with a six pack of toilet paper. And carry a bag of garbage back out.

          • Josh

            The Manhattan and Williamsburg are the only direct entrances to lower Manhattan for full size trucks, including everything from the port.

        • Richard

          You know you don’t have to tolerate the same incoherent infant ruining the comments section of every posting.

      • marvin gruza

        The G train gets one from brooklyn to queens but does not allow someone from Queens to access the rest of the united states – yes subways are great for intra-city transportation. Highways are needed for much the non-intra-city transportation. Thus we need both.

      • marvin gruza

        “Ultimately it’s a question of whether you value clean air and safe streets or motorist convenience”
        No! it is a question of how to strike a balance. A large part of the creation of the US constitution and the federal government was about interstate commerce and the ability to get from one city/region to another. What you call motorist convenience is really a lack of appreciation the need to be able to travel beyond NY.

        • Ross Bleakney

          Per the comptrollers report (linked above): Over 80% of the traffic is for intra-New York travel. So you are saying we should spend billions of dollars so that a tiny minority of drivers can get out of town more quickly?

  6. Pingback: A Better Way than a Six Lane Highway | Sewer Socialists
  7. Josh

    The Stern Report is an outlier and generally estimates the costs of unmitigated climate change at 10%, with a possibility of up to 20%. It uses a discount rate of approximately 1.5%, which is very low. You could invest the money saved from not having such a high carbon tax and invest it at much more than that. And the Stern Review estimated SCC is currently $200 not $500.
    A freeway has much lower accident rates than surface streets, and most of that is covered by insurance that drivers pay. Trucks emit more in stop and go traffic than on a free moving freeway. And the costs of the traffic can be recovered by tolls.

    • Alon Levy

      It estimates the cost at 20% over the long term including an adjustment for the fact that most of the costs are born by low-wage countries (by the usual one-dollar-one-vote method it’s 10%, yes); you can project it back to 20% today if you use a discount rate equal to long-term global economic growth.

      And a freeway is a net increase in accidents because it induces more traffic, which is why American accident deaths skyrocketed in the years the Interstates were built. The difference in fuel consumption between city and highway driving exists but is a lot less than the effect of disappearing traffic.

  8. RossB

    I read the comptroller’s report, and I think he made a strong argument. To begin with, he didn’t say that all of the destroyed freeways lacked trucks, only that many of them did. He also spent much of the time suggesting that simply removing it is a reasonable proposal, supported by plenty of people who know what they are talking about. He isn’t the least bit worried about car traffic, but made the argument that a new facility is necessary for truck travel.

    Fair enough. To me, this should be the starting point. Is it necessary for truck travel? If it is, then is this the best way to move trucks through the city? Maybe, in both cases, the answer is yes. But if there are better ways to improve truck travel — or the trucks will be just fine with no freeway or other improvements — then it makes sense to tear it down. I think it makes way more sense to assume it is gone, and then try to figure out how best to improve truck travel (or determine whether it really is an issue). Too often I have seen agencies assume that the best way to solve a problem is to build something similar to what existed before, even if building that is extremely expensive. Then you end up with something like the SR 99 tunnel in Seattle, which was a big mistake.

  9. ragnar1

    you cant possibly live here . These proposals are crazed! ALL local streets(hicks terrace, henry) will have to be sealed off after dark from the watchtower to red hook. with . restricted pass access for deliveries and with a delivery terminal constructed in the park by the courthouse. This area cant take the traffic. It is a RESIDENTIal AREA.
    so……………. it will be replaced!

  10. Adam

    Alon! Not a fair representation of the Comptroller’s plan.

    We never said “all” teardowns were car-only. (And I’ve never seen mention that the Pompidou allowed large trucks — and don’t see any camions in the link you included. Please share documentation.)

    And we don’t want a “narrower” highway. We want to only rebuild the bottom cantilever for trucks and use the substantial savings to pedestrianize the middle level, extend the linear park all the way to red hook, and invest in local transit — higher frequency service on the G, R, 7, and a dozen bus routes.

    And I know you’ve found otherwise, but the DOT maintains that it would cost billions to make the Belt safe for trucks…

    “DOT studied the feasibility of using the already congested Belt Parkway (over 140,000 vehicles per day) as an alternate truck route during BQE construction, but making the Belt safe for trucks could take up to $3 billion and 10 years to fix:
    • Bridges over the Belt, some of which carry subway lines, are too low for trucks
    • Bridges that carry the belt were not built to carry heavy vehicles, requiring major construction projects to remedy
    • Narrow lane widths and tight turns at ramps are unsafe for trucks”

    • IAN! Mitchell

      The free market is smarter than planners.

      Tear down the BQE and see what actual business owners spending their own money choose to do.

      • Henry

        Leaving New York is not considered a good outcome, even if it is a “free market” outcome.

        The free market has spent most of its time selling infrastructure for scrap (e.g., the railroads).

  11. Stephen W. Houghton

    I don’t want to be rude but cars are transit. Until you understand that you are living in fantasy land, not reality.

    • Eric

      You are correct that cars are a means of transportation (the word “transit” typically refers only to public transit, which cars are not, but that is an unimportant question of semantics).

      However, cars are a very *inefficient* form of transportation. They take up much more road space, cost much more money, and create much more pollution than other forms of transportation. So in a city like NYC population is dense and road space is limited, it is best to limit the space available to cars, in order to create more room for efficient forms of transportation.

Leave a Reply to The Economist Cancel reply

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

You are commenting using your 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.