The malls were 92.5% full when General Growth filed the biggest real estate bankruptcy in US history

This is the cleanest illustration in modern real estate of a distinction that costs people money in every cycle: a company can be solvent and still fail, because solvency is measured over the life of an asset and liquidity is measured on the morning a loan matures.

It is also the case that broke an assumption the entire commercial-mortgage-backed securities market was built on. That part matters more than the trade, and it is the part almost nobody tells.

The wall was built in 2004

The reel says General Growth did not have a mall problem, it had a debt wall. That is right, and the wall has a date on it.

In August 2004 the company agreed to buy the Rouse Company for about $12.6 billion, closing that November.^8,9 Rouse’s own chief financial officer described the financing without any softening at the time: General Growth was “borrowing $7 billion to finance the deal, pushing its debt to $23 billion”.⁸

That is the whole mechanism. Short-and-medium-dated borrowing against long-lived assets works for as long as the borrowing can be renewed, and the renewal is not a property question — it is a credit-market question. By the end of 2008 General Growth had $3.31 billion coming due in 2009 and $6.43 billion in 2010, into a market where commercial mortgage securitisation had stopped functioning.¹ An analyst told the New York Times that October, six months before the filing: “The reality is, given this credit crunch, they’re up against the wall.”¹⁰

The operating business, meanwhile, was fine — with one honest qualification the reel does not make. Occupancy at the end of 2008 was 92.5 per cent, down only slightly from 93.8 the year before, and net operating income rose 4.5 per cent to $2.59 billion. But comparable sales fell 3.8 per cent, against a 1.4 per cent rise the previous year.^1,2 Tenants were still there and still paying. They were selling less. “Nothing was wrong” overstates it; “nothing was wrong that would have mattered without the maturities” is the accurate version.

One share, five marks, eleven years

The DIP loan that was never drawn

The story as usually told is that Ackman bought the stock, provided the debtor-in-possession financing, and took a board seat to steer the reorganisation. The middle of those three is not what happened, and what did happen is more interesting.

On 15 April 2009, the day before the filing, General Growth accepted a commitment from Pershing Square for $375 million of debtor-in-possession financing. The terms were what you would expect from a lender with no competition: LIBOR with a three per cent floor plus twelve per cent, a $15 million commitment fee, and warrants for 4.9 per cent of the fully diluted equity.³

Three weeks later it was gone. On 13 May the bankruptcy court approved a replacement $400 million facility led by UBS and funded by an entirely different group of lenders.⁵ The Pershing loan was never drawn. The $15 million fee had already been paid.

That is not a criticism, and it is not nothing either. Committing to lend at the moment nobody else would is what bought a seat at a table where the outcome was decided — and it was compensated at fifteen million dollars for a loan that never funded. A study that repeats “he provided the DIP financing” has described the intention and missed the transaction.

Solvent, current, and in bankruptcy anyway

The buildings kept working the entire time.

Escalators in a multi-level mall atrium, photographed in July 2007. Wikimedia's own categorisation places this at Water Tower Place; the photographer's caption does not name the location, so it is shown here as the building type rather than the building.Photo: Jeramey Jannene, CC BY 2.0, via Wikimedia Commons

The ruling that outlived the trade

Every one of General Growth’s malls was financed inside a bankruptcy-remote special purpose entity — a subsidiary structured, with independent managers and consent requirements, for the express purpose of ensuring that a parent’s failure could not drag the property into a bankruptcy. That structure is the assumption commercial mortgage securitisation is priced on. It is what lets a lender underwrite a single mall without underwriting the company that owns it.

General Growth put roughly 166 of those entities into Chapter 11 anyway, including ones that were solvent and fully current on their loans. Their lenders moved to dismiss, and on 11 August 2009 Judge Allan Gropper denied every motion.⁶ A subsidiary’s board, he held, may weigh the interests of the parent group in deciding whether to file, and the replacement of two independent managers days before the filing — which he described as“admittedly surreptitious” — was not bad faith.⁶

The lenders had bought separateness and discovered they had bought a presumption. Bankruptcy remote turned out not to mean bankruptcy proof, and law firms advising the securitisation market said so within weeks. The pricing of every subsequent deal in that market carries some of this ruling in it, which is a longer-lived consequence than any single investor’s return.

The record has a stain

The claim that old shareholders kept their shares is true, extraordinary for a bankruptcy this size, and incomplete. Each old share did become one new share plus about 0.0983 of a Howard Hughes share.¹³ But roughly 644 million new shares were sold to the investor group at $10.00 to $10.25, against 312 million shares outstanding before the filing.^14,6 Setting those two sourced figures beside each other, legacy holders came out with something like a third of the combined company. Their shares survived. Their share of the company did not.

Simon Property Group spent months and retained four advisory firms trying to buy General Growth outright, raising its offer repeatedly and promising creditors would be paid in full. It withdrew on 7 May 2010 — beaten not on price but on a warrant package the court approved for the rival bidder. Its chairman said the board had “hastily decided in less than 24 hours to accept substantially less value”.¹¹

And the thesis that prime assets endure has aged unevenly. Brookfield bought the whole company at $23.50 a share in 2018 and committed billions to redevelopment.¹⁶ Reporting since has described a portfolio where the strongest malls hold up and a long tail does not — properties sold or handed back to lenders, and at least one where the owner defaulted deliberately on a mall that was still 94 per cent occupied because its appraised value had halved. Occupancy survived at the top. It did not save the middle.

What transfers

The reel’s own summary is the right one and worth restating: liquidity crises are not solvency crises. An asset that produces cash is not worthless because the loan against it matures in a month when nobody is lending. It is worth exactly what it always was, to whoever can hold it long enough to refinance.

Three things make that usable rather than a slogan. Find the maturity schedule before you form a view on the equity — General Growth published its 2009 and 2010 maturities in the annual report that also showed 92.5 per cent occupancy, and both numbers were on the same page. Understand that the price of surviving a maturity wall is paid in ownership: the company survived, the shares survived, and the shareholders still ended up with about a third. And read the documents rather than the structure’s name — several sophisticated lenders had bought something called bankruptcy remote and found out in August 2009 what it actually guaranteed.

The last thing is a caution about the numbers this story is famous for. The $9.26 million and the $555.8 million are in SEC filings and exact to the cent. The $60 million and the $1.6 billion — the two figures the whole legend rests on — appear in no document, are itemised nowhere, and are the investor’s own account of his own outcome. That does not make them false. It makes them a claim, and a brief that treats a claim as a filing has stopped being a brief.