MGM raised about $20B selling its casinos — and signed the clause that broke Red Lobster

Fintok has already published the version of this story that ends badly. Red Lobster was sold in 2014 largely with the proceeds of selling its own restaurants, on leases bundled into cross-defaulted pools, and filed for bankruptcy a decade later owing $190.5 million a year in rent its own court filing called above market.

MGM did the same thing on purpose, at roughly ten times the scale, and it worked. The interesting part is why — and it is not that MGM’s lease was gentler. The clause that made Red Lobster’s leases inescapable is in MGM’s, in almost the same words.

The same clause, in almost the same words

Read the Master Lease MGM signed in April 2016 and the mechanism is stated without euphemism. It is, in its own text, “one indivisible lease”, and:“An Event of Default with respect to any portion of the Leased Property is an Event of Default as to all of the Leased Property.”²

That is the Red Lobster trap exactly. A tenant cannot hand back the properties that are losing money and keep the ones that are not, because failing on one is failing on all of them. It is the clause that makes a portfolio of leases behave like a single obligation, and it is why closing a weak location does not relieve the pressure — the landlord has arranged matters so the weak locations are carried by the strong ones or not at all.

Ten properties went in at the start, against $10 billion of gross book value, at $550 million of rent a year, escalating two per cent annually for the first stretch of the term.^1,3 Then more properties were added and the rent went up with each one: $650 million after Borgata, $756.7 million after National Harbor, $946.1 million by April 2019.^3,4

What was sold, and what stayed

The buildings left. The business that needs them did not.

Mandalay Bay from street level. Its real estate was contributed to MGM Growth Properties in 2016 and moved into a Blackstone joint venture in 2020; MGM has operated it throughout.Photo: S. Clyde, public domain, via Wikimedia Commons

What it raised, and what nobody has added up

The test came in March 2020, and it passed

On 17 March 2020 every MGM property in the United States closed. Revenue went to nearly nothing — the filing’s own words are that the properties were“effectively generating no revenue”.¹⁰ The rent did not move. This is precisely the scenario that ends a cross-defaulted triple-net tenant.

MGM paid all of it. From the landlord’s side, in August:“we continue to receive our rental payments in full and on time”, and MGP was “not engaged in any conversations with MGM to modify the economic terms of our leases”.¹¹ No deferral, no abatement, no renegotiation.

The reason is the whole lesson, and it is not that the rent was small. MGM had access to capital: it drew on its revolver, raised debt, and in April 2020 amended its credit facility for covenant relief, pledging its operating partnership units and agreeing to keep at least $600 million of liquidity available.¹⁰ A company that can still borrow can pre-fund a year of fixed obligations inside a fortnight. A restaurant chain owned by a private equity firm, carrying acquisition debt on top of its rent, cannot. The instrument was identical. The buffer was not.

Ratings agencies were not reassured, and they said so in terms that name the mechanism. Fitch cut MGM in March 2020 citing “decreased financial flexibility following the recent sale-leaseback transactions”, which it said reduced the company’s ability to monetise assets and increased its rent obligations to unaffiliated parties. A year later S&P lowered MGM, MGM China and MGP together on lease-adjusted leverage. The rent was paid and the structure was still treated as a real constraint — contained, not absent.

Where the money actually went

The reel says the proceeds lowered MGM’s cost of capital, paid down debt, stabilised margins and funded international expansion. The record supports one of those cleanly, one partly, and one not at all.

Debt paydown is real and documented: $3.1 billion retired in the fourth quarter of 2019 out of $8.2 billion of net proceeds, itemised by the chief financial officer.⁹ That is the “fortress balance sheet” the Bellagio announcement promised.⁸

Buybacks are the larger and later answer, and the sequence matters. MGM said in January 2020 that a substantial portion of the MGM Grand proceeds would go to repurchases and dividends — and then the pandemic arrived, 2020 buybacks collapsed, and the two $700 million unit redemptions that followed were assigned to the revolver and to general corporate purposes. By November 2023 the company was reporting $6.2 billion of buybacks since the start of 2021.¹⁶ The share count is down substantially. That is where the money went, two years later than the story usually implies.

International expansion has no sourced link at all. MGM Osaka’s own disclosed funding is a yen credit facility — borrowing, not proceeds. Anyone connecting the Bellagio to Japan is drawing a line the filings do not.

The record has a stain

In August 2020 MGM made 18,000 permanent layoffs, out of a United States workforce of roughly 68,000 before the pandemic, after an initial furlough of about 63,000. That was the same month its landlord was telling investors the rent had been paid in full and on time.¹¹ No source draws a causal line between those two facts, and this brief will not draw one either — but they are the same summer, and a reader is entitled to see them next to each other.

There is also a question about whose idea this was. In March 2015 the activist investor Land & Buildings publicly pushed MGM to convert to a REIT. The campaign failed: the full incumbent board was re-elected that May and the activist dropped its board bid. MGM announced substantially the same real-estate logic five months later, on its own terms.¹² Whether that was a delayed capitulation, independent arrival at the same conclusion, or something between is not settled by anything in the record.

What transfers

Read beside Red Lobster, this study says something more useful than either study says alone. The instrument does not determine the outcome. The same structure — triple-net, long-dated, escalating, cross-defaulted — produced a bankruptcy in one case and a defensible strategic repositioning in the other.

What differed was everything around it. MGM sold assets it could still fill at premium rates in a market with genuine scarcity, kept access to capital markets that let it pre-fund a year of rent inside a fortnight, and used the proceeds on debt and equity rather than to fund the purchase price of the company itself. Red Lobster’s buyer used the proceeds to buy Red Lobster, which meant the rent was there from day one with nothing on the other side of it.

So the question to ask of any sale-leaseback is not whether the terms are good. It is what the proceeds are for, and what the tenant will be holding when revenue stops for a quarter. Rent is indifferent to that answer; it arrives either way. And note the one transaction here that created no obligation at all — MGM sold Circus Circus outright and walked away. That is the alternative, and it is always available. Choosing not to take it is the decision.

The last thing to say is what is missing. MGM has never published what it raised in total, and no independent party has assessed whether the separation created value or merely moved it from one pocket to another. Nine years and roughly twenty billion dollars in, the answer to “did this work” rests on a claim made by the chairman of the entity being sold, on the day it was sold.¹²