Yes: FTX Was Really Bankrupt. And the Standard Excuses Are as Invalid as They Typically Are

The standard excuses made by flacks for bankrupt financiers have long fallen into three categories: (1) we were solvent but for the discount risk imposed on the value of our assets, (2) our gamble...

The standard excuses made by flacks for bankrupt financiers have long been three: (1) we were solvent but for the discount risk imposed on the value of our assets, (2) our gamble for resurrection would have succeeded had our bets not been cancelled while the wheel was still spinning, (3) bad actors who are really to blame forced a crisis in which we had no time to find counterparties who would understand what the situation really was. Flacks gotta flack, and may well be paid handsomely in cash or social-network points for doing so. But do not take them too seriously…

---

Cryptogrifters…

It is really important to push back against the cryptogrifting ecosystem now, because it would be a bad thing if it were allowed to continue to grow, or even shrink slower than it is currently shrinking. And so I find myself, today, once again assuming the role of Sisyphus here.

And I really do consider myself—or have persuaded myself that I consider myself—happy doing this.

---

My initial reaction was: WAIT!! WHAT?!

Occasionally, I run across something from people who are very smart and play it straight that leads me to startle and think: “they really hope nobody is going to read and remember this, do they?” 

Today’s example is from Ian Ayres of Yale and Jack Donohue: a column about the collapse and bankruptcy of the crypto exchange FTX titled “FTX Was Never Really Bankrupt” <https://www.project-syndicate.org/commentary/ftx-never-really-bankrupt-can-make-creditors-and-customers-whole-by-ian-ayres-and-john-donohue-2024-01>. And then they go on to say that “the public would view [FTX principal Sam] Bankman-Fried very differently if they realized that FTX had sufficient assets to make whole its customers and other creditors all along”. 

Presumably—they do not spell it out—they believe that if FTX had been a bank rather than an exchange, then Sam Bankman-Fried would not have committed a crime. Sam Bankman-Fried wound up:

1. taking $8 billion of FTX customers’ fund that was supposed to not be touched— beheld by FTX—and 

2. lending it out—on no security—to himself in his guise as principal of Alameda 

3. so that he, although underwater, could then gamble-for-resurrection, 

4. and then losing it.

Their argument I think—again, they do not spell it out—is that that would have been self-dealing and imprudent financial management, but not criminal, had FTX been a bank and not an exchange.

And—again they do not spell it out—the fact that FTX’s customers will eventually get some (they claim all) of their money back should, they think, lead “people… [to] think much more highly” of Sam Bankman-Fried. 

Why? Because in the end there will, they claim, be no victims. And while Sam Bankman-Fried committed crimes in his role as FTX, those actions would not have been crimes in some counterfactual world in which FTX had been a bank.

The real villains here, in Ayres’s and Donohue’s estimation, are bankruptcy trustee John J. Ray III and the other post-bankruptcy declaration “lawyers and financial advisers… [who] billed in excess of $400 million by the end of 2023… Sullivan & Cromwell alone has billed at least $180 million…” and whom Ayres and Donohue claim are looting the bankruptcy estate via a “sluggish asset distribution”.

I genuinely do not get this: People think Sam Bankman-Fried was a chaos monkey who, when his back was against the wall, stole $8 billion of customer funds he had no legal authority to touch and lent it to himself in an attempt to gamble for financial resurrection. 

That is generally not something a “brilliant businessman”—as Ayres and Donohue characterize Sam Bankman-Fried—does. 

That is something a chaos monkey and cryptogrifter does.

And I would be very, very surprised—I would have to greatly revalue my opinions of Ayres and Donohue and of the trust I place in their past work—if they do not know this. Which raises the question of how to read Ayres and Donohue’s statement that they are “friends and colleagues of Bankman-Fried’s parents”: Are we to read this piece as an act of friendship and pass on, or are we supposed to take it as Ayres’s and Donohue’s considered and firmly held analytical opinion?

The thing that disturbs me the most is Ayres’s and Donohue’s claim that it is relevant for the FTX case that today, after the crypto recovery and the start of the AI-boom, the assets now in the FTX bankruptcy estate are sufficient to pay out all claims. 

They are thus saying that it is relevant that the gamble-for-resurrection succeeded.

And I do not see how that is relevant at all to the established facts that Sam Bankman Fried the cryptogrifter is a thief. 

Gambles-for-resurrection, you see, usually do succeed. That is the point. You are financially underwater—you cannot meet the demands of your creditors tomorrow. But you can put all your chips on the wheel one last time, and make a bet that will win you enough to cover your obligations most of the time but could also go horribly south.

But that gambles-for-resurrection do usually succeed if they are allowed to run does not mean that they are good things to do, or that the counterparties and the government should let them run, or that underwater principals are guiltless in making them if they turn out for the good.

That would be a good place to stop, but there is one interesting analytical point in all this mess. That is Ayres and Donohue’s claim that “FTX was never really Bankrupt”, that “FTX was solvent… at every point before Bankman-Fried relinquished control”, that “the crypto exchange had sufficient assets to make creditors whole all along”, that Sam Bankman-Fried was a “brilliant businessman” rather than a smooth-talking practitioner of the gamblers’ ruin strategy.

But if FTX was never bankrupt, always solvent, always had assets of sufficient worth to cover all of its liabilities, and Sam Bankman-Fried was such a brilliant businessman, then why did it ever declare bankruptcy? Why was he unable to pay his debts? How could there be a “run on the exchange in November 2022” that FTX—remember, it has a custodial obligation to keep its customers’ funds so that it can repay them all immediately on demand—could not survive?

The answer is that your balance sheet always has four aspects:

1. The expected value of your assets—what, on average, they will eventually realize, discounted back to the present according to the slope of the intertemporal price system that is the interest rate.

2. The expected value of your liabilities—what, on average, you will eventually have to pay, discounted back to the present according to the slope of the intertemporal price system that is the interest rate.

3. The discount of your assets from their expected value because of risk—that while your assets may turn out in the end to be worth more than their current-expected value, they may also be worth less. Thus you cannot sell your assets today for their expected value, but rather for a lower amount to compensate your counterparties in your asset sale for bearing these risks—chief among which is that you know more about the assets than they do.

4. The discount of your liabilities from their expected value because of risk—that while your liabilities may turn out in the end to be worth less than their current-expected value, they may also be worth more. Thus you cannot pay people to take on your liabilities today for their expected value, but rather your must offer them a higher amount to compensate your counterparties in your liabilities lay-off for bearing these risks—chief among which is that you know more about the liabilities than they do.

All four of these aspects of your balance sheet are equally real. And pretending that (3) and (4) do not exist is to commit a grave analytical error.

Highly-leveraged financiers who go bankrupt can often claim that (1) the expected value of assets was in excess of (2) the expected value of liabilities. But that butters no parsnips when—as happened to Sam Bankman-Fried—(3) and (4) are large enough in everyone’s estimation to overwhelm the wedge between the expected values of assets and liabilities.

Highly-leveraged financiers who go bankrupt can often claim ex ante that their gamble-for-resurrection would have succeed, and can often point ex post to success of the gamble-for-resurrection in some counterfactual world in which their bets were allowed to ride. But that, also, butters no parsnips when you cannot find even one counterparty that is willing to value (3) the risk discount on your assets plus (4) the risk premium on your liabilities as small enough in magnitude for it to be worth their while to inject any of their funds into your business.

If Sam Bankman-Fried and FTX had not been really, truly, actually bankrupt, they would not have declared bankruptcy.

Full stop.

---

Reference:

• Ayres, Ian, & John Donohue. 2024. “FTX Was Never Really Bankrupt.” Project Syndicate, January 26. <https://www.project-syndicate.org/commentary/ftx-never-really-bankrupt-can-make-creditors-and-customers-whole-by-ian-ayres-and-john-donohue-2024-01>

• Faux, Zeke. 2023. Number Go Up: Inside Crypto’s Wild Rise and Staggering Fall_. New York: Penguin Random House. <https://www.penguinrandomhouse.com/books/711959/number-go-up-by-zeke-faux/>.