I haven’t seen any evidence that FTX promised to never invest customer deposits. Does anyone have a link? My understanding is that FTX offered customers the opportunity to make leveraged trades, i.e., to bet more than the money they had in their accounts. This suggests to me that FTX was not just an exchange but a lender, which is a very different sort of financial beast (with a different risk profile). I also understand that there was a significant interest rate on the customer accounts -- 6% -- which adds weight to that conclusion. You can’t get a return on investment without risk.
I’d also love to see links to the terms of service, as I’ve seen many state that Sam violated the terms of service but little evidence to that effect. I’ve seen one document that indicated the property of customers was their own. That’s not legally dispositive of the issue, as FTX never claimed to have an ownership interest in the customer deposits. The question is what it was allowed to do with customer deposits and, for example, whether it was required to keep segregated accounts. Far larger and more sophisticated entities, like MF Global, have failed with account segregation, so it would not surprise me if FTX failed as well.
The basic problem is this. Even if you WANT to prevent yourself from using customer deposits, if you are a lender to various account holders (including ordinary retail customers) and you are not properly segregating the funds, then when a bunch of people borrow money from you, beyond what they invested into their account, they will inevitably tap into the money of other customers. Corporate controls, account segregation, etc will prevent this. But early stage companies very often fail with this sort of thing because it’s hard, and often very labor intensive. I can see FTX not imposing a limit on Alameda’s right to leverage because: (a) they were overconfident in the ability of Alameda to manage risk; (b) they never thought that Alameda’s leverage would dip into the amount other customers needed; and/or (c) they assumed incorrectly that some form of legal segregation of accounts would prevent the technical removal of customer accounts from happening.
I don’t know what the right answer is to any of these questions. What I do know, having looked at the inside of collapsing financial institutions, is that it is far far far more complicated than most people think. It is one of the reasons why there was no criminal prosecution in the MF Global case, despite what seemed to me far clearer evidence of wrongful conduct.
Re: prediction markets, they are interesting, and I hadn’t seen them before. Slightly moves my needle towards actual fraud, maybe back to 75-25. But I’m not sure this is a scenario where we should expect prediction markets, at least in the short term, to be particularly reliable, due to herding effects. See here.
I would also add, as a long time criminal defendant myself and someone who is very much against mass incarceration, that I may have a bias here, as I do not want to see anyone sent to prison. Even my worst enemy. Jails and prisons are cruel places, and there is relatively little evidence (if any) that they serve any rehabilitative purpose. That may be biasing my assessment of Sam because I do not WANT to see him (or anyone else, really) in prison.
I haven’t seen any evidence that FTX promised to never invest customer deposits. Does anyone have a link? My understanding is that FTX offered customers the opportunity to make leveraged trades, i.e., to bet more than the money they had in their accounts. This suggests to me that FTX was not just an exchange but a lender, which is a very different sort of financial beast (with a different risk profile). I also understand that there was a significant interest rate on the customer accounts -- 6% -- which adds weight to that conclusion. You can’t get a return on investment without risk.
The issue is that as a user of FTX, you were supposed to be able to choose whether your money was being lent out or not—e.g. there was a “lend/stop lending” button in the interface. It seems totally reasonable to me that FTX loses your money if you lend it. But my current impression is that the amount lent to, and lost by, Alameda was much more than the amount that users agreed to have lent out. Agree that segregation of funds, if implemented properly, would solve the problem here.
I haven’t seen any evidence that FTX promised to never invest customer deposits.
Here is a screenshot of the (now deleted) Tweet where SBF claims FTX does not invest customer deposits:
Here is the relevant section of the FTX ToS:
(A) Title to your Digital Assets shall at all times remain with you and shall not transfer to FTX Trading. As the owner of Digital Assets in your Account, you shall bear all risk of loss of such Digital Assets. FTX Trading shall have no liability for fluctuations in the fiat currency value of Digital Assets held in your Account.
(B) None of the Digital Assets in your Account are the property of, or shall or may be loaned to, FTX Trading; FTX Trading does not represent or treat Digital Assets in User’s Accounts as belonging to FTX Trading.
I think what Sam is saying here is true, if customer deposits are being used as leverage and not as investments.
It’s arguably a bit misleading, if Alameda was a customer with a seemingly unlimited line of credit. Bu t what Sam is saying is still true.
I respond to the TOS points above. But to me—and I could be wrong on this, as I have not been in securities litigation for many years—the terms of service aren’t really dispositive.
Thanks for these replies.
I haven’t seen any evidence that FTX promised to never invest customer deposits. Does anyone have a link? My understanding is that FTX offered customers the opportunity to make leveraged trades, i.e., to bet more than the money they had in their accounts. This suggests to me that FTX was not just an exchange but a lender, which is a very different sort of financial beast (with a different risk profile). I also understand that there was a significant interest rate on the customer accounts -- 6% -- which adds weight to that conclusion. You can’t get a return on investment without risk.
I’d also love to see links to the terms of service, as I’ve seen many state that Sam violated the terms of service but little evidence to that effect. I’ve seen one document that indicated the property of customers was their own. That’s not legally dispositive of the issue, as FTX never claimed to have an ownership interest in the customer deposits. The question is what it was allowed to do with customer deposits and, for example, whether it was required to keep segregated accounts. Far larger and more sophisticated entities, like MF Global, have failed with account segregation, so it would not surprise me if FTX failed as well.
The basic problem is this. Even if you WANT to prevent yourself from using customer deposits, if you are a lender to various account holders (including ordinary retail customers) and you are not properly segregating the funds, then when a bunch of people borrow money from you, beyond what they invested into their account, they will inevitably tap into the money of other customers. Corporate controls, account segregation, etc will prevent this. But early stage companies very often fail with this sort of thing because it’s hard, and often very labor intensive. I can see FTX not imposing a limit on Alameda’s right to leverage because: (a) they were overconfident in the ability of Alameda to manage risk; (b) they never thought that Alameda’s leverage would dip into the amount other customers needed; and/or (c) they assumed incorrectly that some form of legal segregation of accounts would prevent the technical removal of customer accounts from happening.
I don’t know what the right answer is to any of these questions. What I do know, having looked at the inside of collapsing financial institutions, is that it is far far far more complicated than most people think. It is one of the reasons why there was no criminal prosecution in the MF Global case, despite what seemed to me far clearer evidence of wrongful conduct.
Re: prediction markets, they are interesting, and I hadn’t seen them before. Slightly moves my needle towards actual fraud, maybe back to 75-25. But I’m not sure this is a scenario where we should expect prediction markets, at least in the short term, to be particularly reliable, due to herding effects. See here.
I would also add, as a long time criminal defendant myself and someone who is very much against mass incarceration, that I may have a bias here, as I do not want to see anyone sent to prison. Even my worst enemy. Jails and prisons are cruel places, and there is relatively little evidence (if any) that they serve any rehabilitative purpose. That may be biasing my assessment of Sam because I do not WANT to see him (or anyone else, really) in prison.
The issue is that as a user of FTX, you were supposed to be able to choose whether your money was being lent out or not—e.g. there was a “lend/stop lending” button in the interface. It seems totally reasonable to me that FTX loses your money if you lend it. But my current impression is that the amount lent to, and lost by, Alameda was much more than the amount that users agreed to have lent out. Agree that segregation of funds, if implemented properly, would solve the problem here.
Here is a screenshot of the (now deleted) Tweet where SBF claims FTX does not invest customer deposits:
Here is the relevant section of the FTX ToS:
I think what Sam is saying here is true, if customer deposits are being used as leverage and not as investments.
It’s arguably a bit misleading, if Alameda was a customer with a seemingly unlimited line of credit. Bu t what Sam is saying is still true.
I respond to the TOS points above. But to me—and I could be wrong on this, as I have not been in securities litigation for many years—the terms of service aren’t really dispositive.