The Second Avenue Subway extension to East Harlem broke ground on its major construction stage on June 8, 2026, and the MTA awarded the last of its big civil packages three weeks earlier. Phase 2 is now a live heavy civil job: shafts going down, utilities being rebuilt under Second Avenue, and a tunnel boring machine scheduled to be lowered into the ground in early 2027 at 120th Street.
What makes this project unusually instructive is that its three stations are being built three different ways. That is not an accident of design. It is the cost-containment strategy, and it is the most transferable lesson on the job.
Two Correct Answers to "What Does It Cost?"
Phase 2 gets quoted at $6.97 billion and at $7.7 billion, and both figures are defensible.
The MTA's base budget is $6.968 billion — the number in agency press releases and board materials. Federal reporting from the U.S. Department of Transportation carries the project at roughly $7.7 billion. The gap is about $731 million in financing costs: interest and fees the MTA expects to pay on bonds issued to fund its share. Neither number is wrong. One is construction and program cost; the other is what the public ultimately pays including the cost of borrowing.
Contractors and estimators should care about the distinction for a practical reason. When you benchmark "cost per mile" across transit megaprojects, you are frequently comparing a capital cost on one project against a fully financed cost on another. At 1.5 miles, Phase 2 works out to roughly $4.6 billion per mile on the base budget and about $5.1 billion per mile if you fold in financing. Those are very different headlines from the same job. If you are building comparables for a bid or a board presentation, Buildermuse's construction cost data is a better starting point than a news figure with no stated basis.
The MTA also projects Phase 2 will serve about 110,000 additional daily riders and claims the lowest cost per rider of any active heavy rail project in the country — a framing that shifts the conversation from cost per mile to cost per unit of service, which is the metric the agency would rather be judged on.
Three Stations, Three Construction Methods
Phase 2 extends the Q train about 1.5 miles from the existing 96th Street terminal north to 125th Street and Lexington Avenue, with new stations at 106th Street, 116th Street and 125th Street. Tunnel depth ranges from roughly 35 to 120 feet below Second Avenue.
116th Street reuses tunnel bored in the 1970s. A tunnel segment built between 110th and 120th Streets during the 1970s Second Avenue attempt is still there. Rather than mine a new station, the MTA is outfitting that existing bore to accommodate the future 116th Street station — a decision the agency credits with saving $500 million. The catch is what "outfitting" means in a 50-year-old tunnel: hazardous material abatement before any new work, then structural tie-ins between 1970s cast-in-place concrete and modern segmental lining. Matching a new liner to a half-century-old structure with unknown as-built tolerances is finicky, low-production work, and it is the kind of scope that generates change orders if the pre-construction survey was thin.
106th Street is cut-and-cover. Contract 3 builds new station structural shells between 105th and 110th Streets using cut-and-cover, which means decking Second Avenue to keep traffic running over an open excavation, relocating and reconstructing utilities, structural tie-ins to existing tunnels, and building at-grade and below-grade structures for two station entrances. Cut-and-cover in a dense residential corridor is a surface-impact problem more than a geotechnical one: decking panel maintenance, night noise limits, utility support of unmapped services, and access for buildings that never close.
125th Street is the deep terminal. Contract 2 covers boring twin tunnels north from 120th Street toward 125th Street and Malcolm X Boulevard, plus excavation for the 125th Street terminal station and station shell work. This is the deep, mined portion of the alignment and the piece with the least schedule float.
Three methods on 1.5 miles is a lot of mobilization overhead. The MTA accepted that overhead because it beat the alternative — mining three new deep stations at Manhattan cavern prices.
The Contracts and What They Are Worth
Contract 2 — Connect Plus Partners, a joint venture of Halmar International and FCC Construcción, covers the tunneling scope. The value is reported inconsistently: the MTA and the Governor's office describe it as $1.9 billion, while ENR reported $1.97 billion. It runs 1,471 days. Scope includes boring the twin tunnels, rehabilitating the 1970s tunnel south of 120th Street, excavating seven shafts for station entrances and ancillary facilities, and building station shells at 116th and 125th Streets.
Two procurement details are worth studying. More than 30 alternative technical concepts were approved during the bid phase — a mechanism that lets bidders propose deviations from the reference design, priced, before award. And Connect Plus Partners' bid came in nearly $200 million below the next bidder. Those two facts are related. ATCs move cost out of a bid by letting the contractor bring its own means and methods to the risk allocation rather than pricing contingency against someone else's drawings.
Contract 3 — Skanska / Traylor Bros. / Walsh Construction JV, awarded June 2026 at $1.02 billion, delivered design-build. Skanska's own share is about $498 million. Work started in May 2026 with completion expected in the third quarter of 2030.
The delivery-method shift is the story. Contract 2 is a conventional heavy civil package; Contract 3 is design-build. Nationally the same primes keep appearing across delivery models — Traylor, Walsh and Skanska are simultaneously the joint venture boring under the Hudson River on the Gateway Program's Hudson Tunnel Project, where that $1.29 billion package was procured design-bid-build. The bench of firms capable of billion-dollar soft-ground and mixed-face tunneling in the United States is small enough that the same three names cover both harbors, and the same short list of primes recurs across nearly every active TBM drive in the country.
Across all Phase 2 initiatives the MTA claims roughly $1.3 billion in savings against earlier estimates, of which the $500 million legacy tunnel reuse is the single largest line.
The Machine
The TBM is a 750-ton machine with a roughly 22-foot diamond-studded cutterhead. MTA materials for the tunneling award describe the machine in the singular; other reporting has referenced a pair of machines on order for delivery during 2026. Either way, the launch is scheduled for early 2027 at 120th Street and Second Avenue, which is the site where the June 2026 groundbreaking was staged.
A 22-foot bore is modest by megaproject standards — the Hudson River tubes are 29 feet, and the Hampton Roads highway tunnel used a 46-foot machine. Diameter is not the difficulty here. The difficulty is the length of the drive relative to the setup: on a bore this short, the launch shaft, muck handling, segment delivery logistics and the removal shaft dominate the cost curve. The machine itself may only be productive for a fraction of the contract duration. Anyone pricing urban tunneling should build the estimate around shaft and logistics costs first and treat the linear advance as the easy part.
What Happens Next
The near-term milestones are shaft excavation and utility reconstruction through the rest of 2026, TBM delivery and assembly, and the machine going into the ground in early 2027. Contract 3 cut-and-cover work at 106th Street runs through Q3 2030. Revenue service is targeted for September 2032.
For contractors, the opening is the subcontract tier now going to market behind two billion-dollar primes: support of excavation, decking, utility relocation, precast segment supply, waterproofing, station finishes, traction power, signals, ventilation and fire-life-safety. Two primes of that size generate a very large second-tier procurement, and the federally funded portions surface on the public bid board as they are advertised.
The larger question Phase 2 is trying to answer — why heavy rail costs what it does in New York — will be judged on whether ATCs, design-build on the station package and the reuse of a 1970s bore actually hold the $6.968 billion number through 2032. The savings are claimed now. They are proven at closeout.
For related coverage of transit and public infrastructure delivery, see Buildermuse's public works section.