Thread by @patio11

“What happens if you invest depositor’s money in a debt platform which suffers a $100M+ operational loss of which you own 50%+ in traditional finance, Patrick?”

… That doesn’t happen.

“But play along with the hypothetical.”

Alright I’ll try:

https://twitter.com/Bitfinexed/status/1466731307447300098

We’ll start with the notion of “depositor.” Speaking broadly, if you accept retail money, by regulation you are almost certainly prohibited from investing in anything which has any reasonable probability of having a $100M operational loss.

But Ops losses do happen in the world.

So you have an agreement with your contra operating the platform that gave rise to the ops loss. That contract, which was negotiated by competent professionals on both sides, is extremely explicit as to which of the two of you owns the loss.

It’s probably your contra?

But what if your contra can’t pay? Well, since Ops losses are a thing that exist in the world, and sudden failure of a contra is a risk that finance intrinsically faces, you did two things prior to signing:

One, you underwrote your contra, to assess the probable risks involved.

This includes likelihood of loss, magnitude of loss, probability of contra being able to make good on loss, expected difficulty / timeline of that, etc etc.

Based on this, you likely required your contra to post some collateral with you, insure you against the loss, etc.

“But it all blew up. Play along.”

OK, so now it is between you and your depositors. You have an agreement with them, too, and there is a provision in it for who held the risk here.

Since you call them depositors, it is almost certainly you.

But everyone knew there were risks.

Since there were risks, you were well-capitalized; you had raised funding *that you had not invested in the blown-up platform but rather kept in something very safe with easy liquidity*, likely by selling equity.

Your equity holders have just taken a hit, in line w/ size of loss

You still have a war room. You still expect to be involved in litigation. You still expect to be called in by regulators to explain what in blazes happened. You still accumulate a reserve (accounting entry) against likely future losses, a bit in excess likely of the headline #

You may even need to recapitalize, and if you do, your existing equity holders will be further diluted because this is the worst possible timing and among worst possible reasons to attempt recapitalization.

But you soldier on, because none of this was particularly novel.

"Hey Patrick why do all this work if you say that this never happens?"

Oh ops losses absolutely happen. $100 million ops losses are rare. The thing that is almost impossible to imagine is all of the following in combination:

* $100 million ops loss
* You represent 50% of that loss because you somehow didn't underwrite your contra enough to detect that degree of concentration risk.
* Your loss is also material to you. (~20 bps of "your" AUM by my brief calculation, which is an alarming number in ops.)

I don't think I have an overly rose-tinted view of the traditional financial industry. But many geeks seem to have the opposite of that, and think this sort of thing happens all the time.

The brute fact that society continues to function should be strong evidence against that.

"Patrick you keep using this word 'ops loss' and that seems jargon-y."

You are right, voice in my head.

This is a so-called operational loss, which is a catch-all category for "Things that in theory should never happen but in practice sometimes do."

It is distinct from e.g. a credit loss (you loaned money to someone, there was some risk of non-payment or not-as-agreed-payment because that is the nature of lending), a fraud loss (malfeasance by counterparty), etc.

Here the proximate cause of the ops loss, from the perspective of the firm with $50 million on the line, is "Both we and our contra failed to secure systems with security tokens that were the only thing granting access to ~$100M+ in bearer assets."

A mistake! Not fraud (by you).

Note that I'm very, very not saying "X is not a fraud", but I will say "This is not a fraud being undertaken by X. Of that one thing, they are probably innocent."

Oh, another thing traditional finance typically has going for it is that the CFO called into the war room isn't in an Israeli prison.

...

Is that a cheap shot?

...

No, actually, that feels extremely relevant.