TreeHugger has covered the saga of 461 Dean, the world’s tallest modular structure in Pacific Park, Brooklyn, since 2011, when I wrote that “the whole thing boggles the mind. Having worked in prefab for a number of years, I can tell you that it's complicated.” (See all our coverage here) One of the claims that boggled most was the anticipated cost savings of 15-20 percent. The savings from building in prefab come with practice- could they be achieved with a new system, a new factory, a new team?
Many of our posts drew on the work of Brooklyn journalist Norman Oder, who since 2005 has covered the overall project almost obsessively on his Atlantic Yards/Pacific Park Report blog. The following is a guest post by Norman Oder, looking particularly at the purported savings from the project- and the way the media still cover it. The views that he expresses so frankly and strongly are his; It is a taste of the book he is writing about the Pacific Park project, which I am certainly looking forward to.
"Applying this R & D to high-rise construction and other markets creates a business opportunity that provides an unparalleled competitive advantage," it told potential partners in a January 2012 Opportunity Brief.
More than four years later, the results are sobering: delays, losses, and a bitter, unresolved set of lawsuits with former partner Skanska USA, which had been contracted to construct the tower and manage the jointly owned module-assembly factory. (Forest City blames Skanska's execution; Skanska blames Forest City's R&D.)
The Claim of 20 Percent Savings
Astoundingly, some recent press accounts claim that the use of modular techniques purportedly allowed Forest City to save 20 percent on the 32-story, 363-unit tower, which contains half market-rate units and half "affordable" ones. Consider:
- CNN: "Ultimately, the method saved Forest City Ratner Companies about 20% on construction costs
- Business Insider: "That allowed them to save 20% on construction costs, Forest City's VP of residential development, Adam Greene, tells Business Insider."
- Architectural Digest: "it also cost developers 20 percent less than building a traditional skyscraper."
- Bisnow: [Forest City's Greene was] "pointing out that building units in a weatherproof factory gave his company a 20% discount and predicting the method would soon be the norm."
It reflects both unskeptical, lazy journalism and a developer brazen even by the standards of real-estate puffery.
Benchmarks of Failure
Not only did 461 Dean (aka B2) take more than twice as long as expected, thus indicating higher costs, it bears numerous negative benchmarks.
As parent Forest City Enterprises (now Forest City Realty Trust) told the Securities and Exchange Commission (see p. 60ff.) in February 2015, they had to buy out the building's 75 percent co-investor, the Arizona State Retirement System. They also recorded a huge $146 million impairment, or write-down in value.
Later in 2015, they had to pay off their tax-exempt bond decades early and fund the rest of construction from equity. The tower, Forest City told the SEC, was once expected to cost $155 million; last year, it estimated $195.6 million.
No wonder outside investor Scopia Capital, in a slide presentation filed with the SEC last August (see below), called 461 Dean (aka B2) a "value-destructive transaction."
Similarly, another investor, Land & Buildings, in a press release last month, counted B2 among the company's most significant losses.
Given that it took Forest City more than twice as long to build the tower as originally promised, you'd think they'd practice some humility. Instead, they've portrayed the delay as a mere bump in the road, even though they exited the modular business.
More Accuracy, But Also Spin
Not everyone got it so wrong. Wired recently called the experiment "mixed" and concluded that "The building took more time and money to build than originally promised." National Real Estate Investor pronounced, "Those savings didn’t materialize at Dean Street."
Still, the latter article quoted former Forest City executive Roger Krulak, whose new company Full Stack Modular bought the business (for an unspecified sum): "I would say to [sic] cost of modular construction is 80 percent of the cost of conventional construction.”
Note Krulak's use of conditional language. Indeed, that's how proponents were talking four years ago, before the real-world test that proved it wrong. In fact, that figure was more aspiration than guarantee.
Early On, 20 Percent Was a Hope
In November 2011, New York magazine reported that "the whole process is designed to shave as much as 20 percent off construction costs." Fast Company quoted Forest City Ratner CEO MaryAnne Gilmartin as "figur[ing] the modular experiment could potentially shave 10% to 20% off her costs."
John Dolan, a project executive with Skanska, told Curbed, "We estimated that it's around 12 to 15 percent in savings."
Gothamist reported that the builders expected to get faster and thus save more: "As of now, the units are around 10 percent cheaper than their conventional counterparts, though [Forest City's] Melissa Burch... anticipates savings of up to 30 percent as the technique is perfected."
Shifting Goals on Timing
Skanska's Dolan, by the way, said they expected to be finished in 18 months, compared with 24 months for conventional construction. But that benchmark was fuzzy.
Hubris and Hype Versus Reality
The modular story reflects hubris, hype, and an often unskeptical press. Forest City placed enthusiastic articles in Gizmodo, Gothamist. Fast Company, and Forbes, even as construction was troubled. Krulak, ironically enough, got a Breakthrough Award from Popular Mechanics just as the problems mounted.
The structure suffered leaks, mold emerged, and the first four floors "were largely gutted," a state monitor reported, in documents I acquired through a Freedom of Information Law request.
The builders got so skittish that, on two floors, drywall ceilings and walls were omitted from modules, to be installed later—thus undermining the modular concept. When Forest City announced the start of B2’s 11th floor, after a delay, it announced the need to realign 10th floor modules both horizontally and vertically.
Builders couldn't install wood flooring as planned at the factory, given warranty challenges, so they did so on-site. They did deliver modules with facades attached, unlike some other modular builders, but that didn't work well, so some facades required repair.
Some Lessons Learned
Now we know more about what went wrong. As Krulak told Wired, the diversity of modules--32 types!--meant "we probably went a little overboard on the complexity."
Also, 461 Dean may just have been too tall. Krulak told Fast.Co Design, "Our sweet spot for modular in an urban environment is in the 10-to-18-story, 80,000-to-120,000-square-foot buildings." He told Multi-Housing News his new company would focus on hotels, student housing and multifamily rentals.
Similarly, architect Christopher Sharples of SHoP told Fast.Co Design, in the publication's summary, "After 15 stories, extra brace framing is needed to support the building. Stay under that height, and the building's complexity is significantly less."
Such humility is a far cry from "cracking the code." It should be heeded in any account of 461 Dean's purported savings, as well.
Brooklyn journalist Norman Oder writes the watchdog blog Atlantic Yards/Pacific Park Report, and is working on a book about the project.