FTX: Inside the Restructuring - [Business Breakdowns, EP.153]

0:03This is Business Breakdowns. Business Breakdowns is a series of conversations with investors and operators diving deep into a single business. For each business, we explore its history, its business model, its competitive advantages, and what makes it tick. We believe every business has lessons and secrets that investors and operators can learn from. and we are here to bring them to you. To find more episodes of Breakdowns, check out joincolossus .com. All opinions expressed by hosts and podcast guests are solely their own opinions. Hosts, podcast guests, their employers or affiliates may maintain positions in the securities discussed in this podcast.

0:45This podcast is for informational purposes only and should not be relied upon as a basis for investment decisions. This is Matt Russell, and we are releasing a bonus episode of Breakdowns today. While we typically love to cover businesses, we thought this was an interesting opportunity to cover a special situation around a business, and in this case, the FTX bankruptcy. I was joined by Erin Broderick, head of U .S. cross -border restructuring and insolvency at Evershed Sutherland, and Erin represents the ad hoc committee of non -U .S. customers for FTX. So that's given her a front row seat and a hands -on seat to everything that's unfolded at FTX since they entered Chapter 11 in November of 2022.

1:30We tried to cover some of the basics around bankruptcy proceedings using FTX as a lens to compare to other restructurings. And Aaron helped explain what the key negotiating points have been in this particular case. We'd love your feedback on this episode. So if you're listening on Spotify, drop us a response there, or you can respond to any of our social media accounts linked in the show notes of this episode. Now on to the conversation with Aaron on FTX. All right, Aaron, thank you very much for joining us here in what is a different flavor of episode for business breakdowns. We're typically diving into a specific business that's an ongoing entity and basically picking it apart.

2:14here, we have a business at the center of everything, but we're going to talk about what happened with FTX and specifically with the reorganization restructuring of FTX. And to kick us off, it would just be helpful to hear your own background and how you are tied into the situation at FTX. So maybe we can start there. I'm a partner at Evershed Sutherland. I head up the U .S. cross -border and international restructuring department. I've been a restructuring attorney for almost 20 years now. And I became involved in FTX in representing the Ad Hoc Committee of Non -U .S. Customers, which is currently comprising about 75 members holding over $2 .5 billion in claims against FTX.

2:57Not an insufficient number or small number there. Maybe we just start at the beginning. In November 2022, slightly prior to that, things got very intense with FTX. allegations came out. And what ultimately ended up happening was a Chapter 11 bankruptcy. So my understanding with Chapter 11 is that there's a reorganization, you're not intending to liquidate the assets. There's some business behind it. And there's some restructuring effort to keep this business going. And that's different than Chapter 7. But tell me where I am wrong in that assessment of Chapter 7 versus Chapter 11, and what the ultimate process was here in terms of going into bankruptcy?

3:42Sure. Well, Chapter 11 is known as a reorganization proceeding, whereas Chapter 7 is a liquidation. In practice, that's a bit misleading. Really, the key distinction is that in the Chapter 11, you have a debtor -in -possession process, which means the company's management and board can stay in power and in control. And in the Chapter 7 proceeding, it's turned over to a court -appointed trustee. So I think what's confusing in the FTX case is obviously the company's insiders could not be trusted to remain in control. So there was a shift over to John Ray as the company's new CEO and new appointments made for the board.

4:24But it is a debtor in possession proceeding with the new management and new board in control. There's reorganization. Selling of assets is often part of the Chapter 11 process. Indeed, in FTX, there's been a monetization of the token portfolio as well as part of the ventures portfolio really from the outset of the case. With the Chapter 7 being the court -appointed trustee, just curious why it didn't go that route. My initial assumption is there's not the expectation that the court would realize the highest value in terms of that liquidation. So there was some effort to say whoever the debtor is in possession of this process is going to be able to extract more value for the overall pie of lenders or creditors in this particular case.

5:10Is there anything more to it beyond that? Yeah, I just think there is a practical expectation of the limitations of a Chapter 7 trustee, both in terms of expertise and of resources. So while the first day filings painted a pretty grim picture of the lack of financial records and what could be ultimately some pretty dismal recoveries, I think from the ad hoc perspective and many that are in the crypto community realize that there is a lot of value here. And there needs to be experienced professionals to make sure that that value is not squandered and that it's realized the maximum recoveries to creditors.

5:50If you turn it over to a court -imported Chapter 7 trustee, those efforts would not have been undertaken.

6:00Understood. Typically, what I was used to in distress situations was you had creditors involved with a lot of experience going through this process. There were often several classes of loans or bonds, and you had committees or groups formed where certain groups in the top tier of security joined together versus those lenders that were maybe a notch below. And everyone was fighting in terms of their efforts to get the maximum recovery to their specific group. How did it come together here? I don't imagine that you had the same setup in terms of the traditional lenders or funds that were involved, not even the amount of time to get those funds involved, at least in the beginning of this process.

6:42So can you just talk about what this process looked like relative to others that you've gone through? Sure. I think big picture. At the outset, it was pretty clear that the FTX .com silo with the foreign customers was really the piggy bank for the entire enterprise. And yet you have these distinct silos from the dot -com exchange, the U .S. exchange, and then Almeida and Ventures as separate silos. So there was a recognition that if you were to keep those silos separate, you have this group of customers whose funds were stolen. Yet there's not a token portfolio or cash sitting on the dot -com debtors balance sheet.

7:23that's in various other silos, particularly Almeda was invested into some of the ventures portfolio or other third parties. So there was a recognition right up front that there was going to need to be a substantive consolidation of the estates to make sure that the value elsewhere that came from customer funds was allocated in a fair and really maximum amount that could be given to customers. So that's the first point. And then the second point about the capital structure is it wasn't unique here. There are purported secured lenders at the Almeida silo, but their loan arrangements were often secured by FTT or other tokens.

8:02And again, we have that overlay of some of the collateral that may have been posted was actually customer funds. So at the end of the day, what we're really talking about is you've got the customers and you can throw in the US side here because I think their claims are so small that they're really not at odds with the foreign customers. But you have these customer claims and many of the customers believe that they have ownership of these assets, that they should be superior to everyone as owners instead of mere creditors. But the question becomes, well, do we need to prove out the ownership arguments with tracing and identification of assets if we can just get the customers a maximum value of the estates because the general unsecured creditors here, besides the subordinated government claims, it's maybe around $2 billion at the end of the day.

8:56Many of those claims, there's counterclaims and settlements being worked out. So it's really not customers against general unsecured creditors as much as I think that's in the news. I think it's more making sure that the government agencies recognize that their claims are really for the benefit of customers and seeing any recoveries on those claims go to customers. I'm going to just restate this so I think I understand it properly. There's one road that you can take where it's the customers proving that they have the most senior form, maybe not of actual lending, but they're actually owners of assets that were then taken from them.

9:33But if there's a way to just make sure that they're paid back in full, you don't actually have to prove that that was the case? Is that a fair way to represent it? Yeah.

9:46It comes down to the property estate arguments that have been brought up in other crypto cases. In cases where it was a clear cut ruling one way or the other, you actually had assets that were properly held in custody that were segregated. And here, there's no dispute that Almeda and And other of the debtors and insiders misappropriated billions, probably north of $10 billion in customer funds. The DOJ, other governmental agencies, and the debtors have been able to trace certain of those transfers into commingled customer accounts. So, okay, billions that were transferred out that were customer funds.

10:29and they can identify where they were invested into certain venture assets or went to buy a boat in the Bahamas, but not on account of individual customers. And there are also non -customer funds that were commingled there. So we could get into a very complicated exercise of tracing identification to say, on a pro rata basis, for the collective benefit of all customers, X went into the estates. we think that this appreciation in this particular ventures asset is due to customer funds. We could go through all that allocation. It would take months, if not years. We don't exactly know how it would end up.

11:06So what the ad hoc's position has been is let's try to get the maximum amount to customers pro rata for everyone's benefit of the estate value. And if I just take that example and compare it to maybe a bank is the most simple comparison. If you were to see a bank go through a bankruptcy process. How would that differ just in terms of proof? And I think what you were keying in on there is specifically the custodial nature of funds. But how would it be treated if a bank were to go bankrupt? And I have a second example, but maybe we could just start there. Banks can't go bankrupt because they have regulatory regimes that allow for their liquidation.

11:45And there's protection on customer deposits. And actually, depository relationship is one in which you actually do have a claim to the bank's assets. It's very different here where there's an argument. I wasn't putting my assets on loan to the exchange. I own the assets. I was free to withdraw them. And I thought there'd be a one -to -one ratio of assets held by the exchange and what was recorded on my customer ledger. But in a bank liquidation situation, you would essentially take all the pooled assets and distribute those out to the depositors. It happens in pooled investments. When you look at MF Global or other cases where there were pooled investments and similarly situated investors, they have allowed to have a distribution of the pooled assets parada to everyone according to the debtors' books and records.

12:41Again, the complication here is we don't have the nice little pot of customer funds. It's all mixed together with some non -customer funds and customer funds like. That makes a lot of sense. And that's where I was going next was something more in the exchange like business category. And I think you answered it well there with MF Global. In this particular case, I think there were 130 related entities that along with FTX filed in this process. If you compare that to other cases, would you say that is a traditional amount of entities that you typically have with one of these? Is that uncommon? It feels like there's a lot going on there.

13:18But how different is that versus what you would see in your other cases that you're dealing with? It's a substantial amount of debtors. But I think the general practice, particularly in Chapter 11, is you want to get the whole global business into Chapter 11. It's something that's unique about insolvency proceedings in the United States versus other jurisdictions, as there's very broad jurisdiction to file in Chapter 11. And if one debtor can satisfy the jurisdictional and venue requirements, you can bring in any other affiliate. That really helps to just streamline administration of the cases to prevent infighting with the corporate entities.

13:58Again, particularly here where corporate distinctions were not respected, it made a lot of sense to file all the entities together and really prevent all the assets from being dissipated across the world. In this process, how many groups ended up performing? You mentioned representing the ad hoc committee. Were there a lot of groups that ended up coming together? What did that look like? No, not really. I think that speaks to the nature of the common customer experience, that they had terms of service that said, you own these assets. And whether it was fiat or digital assets, when the bank filing hit, everyone lost their right to withdraw those funds.

14:42We have not seen formal groups come up on a splintering of the customer class. That could certainly occur, but I think the best result would be to have the customer class united and just trying to get the highest recovery for what is currently one customer class under the debtor's plan. And did that form fairly quickly just in terms of gathering? Again, I feel like usually you have these lenders who have sizable chunks of the debt. So they represent a large percentage. In this case, I would imagine it was very fragmented in nature. So just corralling all of that together would be a challenge. But what did that look like in terms of getting a united committee together?

15:32Ad hoc officially formed on December 5th. And we had three members in the first week. And it was very difficult to collectively organize because the customer lists were also confidential in this case. So it wasn't like your typical case where you could look at the schedules and you could identify large holders. and everyone was very concerned about privacy and having to be disclosed as a member of an ad hoc committee. But it was really grassroots and contacts that I had around the world that my colleagues had reaching out and wanting a voice in the proceedings and not really knowing where else to turn because they weren't getting a lot of information from the estates and didn't feel like the official committee that has fiduciary duties to all unsecured creditors in the case really represented them, particularly on these ownership arguments.

16:30So it continued to build and build and build. And as the underlying customer class has changed and had more secondary hedge funds in the customer class, the same is true with the ad hoc committee. In number, it's about 80 % original holders, but in claim amount, it's more the secondary hedge funds at this point.

16:56On the confidential point, did that have to do with this being a crypto asset or is that just the nature of this type of business where you're dealing with something like an exchange? There were two lines of arguments to keep the customer information confidential. The first was the damage that it could do to individuals by revealing their identities, particularly those in certain jurisdictions that would be identified as being wealthy. We've seen some personal stories there. The second line of argument was that the customer list is a asset of the estates or a reboot of the exchange or any other type of sale transaction involving the customer list.

17:39You would lose that value by revealing the customer identities. So with many Chapter 11s, you keep the business running, keep the, in this case, it would have been the exchange running. Didn't seem to happen here. Was there any portion of the business that was kept in operation? The exchange was shut down. So it has not been operating throughout the bankruptcy proceedings. The debtors recently announced that they are not in the process of continuing with the 2 .0 or exchange asset sale process. That said, they haven't taken it off the table. There was a robust marketing process, and I think there's still value in the customer list through a separate exchange.

18:21I think where there is just a misunderstanding in the customer base is the value that was attributed to the actual exchange. where I can say this publicly, I've gotten to go ahead to say it, is that no bidder in the marketing process ascribed material value to IP or infrastructure or underlying technology. So I just think as upset as some customers are that there hasn't been a concentration on a reboot, you can't control the market. It's interesting because I think the headlines Minds have suggested that the exchange business was a profitable business and much of the misdoing had to do with misallocation of funds into other endeavors, some of which you referenced before.

19:13That would lead me to believe that that was not the case, or maybe it wasn't as strong of a business as maybe some were led to believe. Do you think that's misrepresenting the situation in how I framed that? Look, there was value in some of the trading products. for sure that there is also heightened scrutiny from a regulatory standpoint on some of those trading products. So to say that the exchange had no value, I don't think that's the case. But in terms of a restarted exchange using the existing infrastructure and security and all the like, there was just not that ability to do so. Yeah, I guess it's impossible to compare it to what was the exchange before this, because obviously what happened here had major implications for the entire industry and much of what happened in the year leading up to this.

20:06So impossible to ever go back and make a like for like comparison there. Getting back into the process itself, how things came together, you referenced some of the secondary hedge funds coming into these claims, just into this market. Can you talk a little bit about that, how that was in terms of evolving and developing? These claims get traded on the market again in traditional bankruptcies. I would imagine it looked a little bit different here. Did that have a major impact on the overall proceedings? And was there anything unique about the process evolving with maybe some more mature players and those that operate in this world?

20:44I don't think that the participation of the secondary hedge funds is unique in this case, where they think that they can drive returns, they will get involved. I think in terms of the actual claim expositions, there are some unique features in FTX. For example, not dollarizing claims at the outset, but scheduling all the liabilities in terms of the token and then having the estimation motion that would actually set the value for those tokens. Not heard until the end of January. That was unique in terms of pricing. But I think the market built in the right assumptions there. There's a KYC process that original holders have to undertake in order to receive distributions in the case.

21:35The portal that was set up and all the processes around that were geared towards original holders and not a secondary buyer. So that's added some logistical complications. I think there is a misunderstanding from the original customer perspective that is not familiar with Chapter 11 proceedings or the typical players therein, that there's something nefarious about secondary buying claims from original holders and that there's a big distinction between their claims. There's not, really. At the outset of the cases, when preferences were being focused on, you could argue there was distinction because secondaries were able to diligence possible preference exposure.

22:24But as we can get into with announced recoveries, it's likely that preferences will not be pursued. And so there's really not a distinction between the original and the secondary claim holdings at this point. And that's reflected in the treatment under the plan as well. With the claims, just from the logistics side of things, I'm curious with some larger bankruptcies, you typically have corporate customers that might have claims and you'll have some traditional hedge funds that will purchase those claims. Here, a very different customer base. In terms of those claims actually exchanging hands, they're not just listed on the New York Stock Exchange in terms of buying and selling.

23:01What does that process look like if I wanted to go out and purchase some FTX claims? Do you have a sense of what that actually entails in terms of getting my hands on those? Yeah, I mean, look, there's a number of different ways that buyers and sellers and some intermediaries have connected the larger claims. It tends to be your more traditional outreach through a broker to the larger distress funds. And some of those have minimums that they're buying. And there's also been a number of exchanges set up so that parties can connect to buyers very quickly. And because there is this unique customer code identification in the FTX schedules, and now that the dollarized amounts are known associated with the schedule tokens, it's been a lot easier from a pricing and diligence perspective.

23:55So I do think that there's still a bit of a shutout of your smaller claims in the market on both sides from diligence and cost. But you are seeing smaller claims being traded through some of these online platforms that make it very easy to diligence a claims. You've been mentioning dollarization a few times here. It's worth jumping into. I'll take my impression of what the issue is and we can answer that. and then you can tell me whether I just had the wrong impression to start with. But if I am purchasing Bitcoin or Ethereum on FTX, I am typically sending in dollars to make that purchase. And there comes a question of, should my claim be calculated in terms of the dollar amount that I use to purchase that Bitcoin or Ethereum or the actual Bitcoin or Ethereum that I purchased and how much it was and where it's trended over time?

24:50So is that a fair representation of what's going on with dollarization of claims and the challenges in terms of determining that? Because obviously the price has swung quite dramatically in terms of where Bitcoin and the lesser extent of Ethereum have gone over that timeframe. Because I think that there's some principles, because I think what you're seeing both in FTX and hearing through Genesis confirmation trial that began today is there's the bankruptcy code requirements that you can't get around. And then there are all parties' acknowledgement of equities and how do we get the equities within the confines of the bankruptcy code.

25:28502B bankruptcy code, it requires that claims be valued as a petition date in U .S. dollars. That's done purposefully for equality of creditors so that you don't have different denominations or different time periods in which to value claims. So you really can't get around the dollarization of claims. In terms of in -kind distributions, if you have an unsecured claim, you get paid back in dollars unless you agree to some sort of equity interest. With the ownership arguments here, there is a equitable basis to try to make in -kind distributions if you could. In Genesis, very different from FTX, So we're going to hold on to our crypto assets to try to do a in -kind distribution.

26:18The case as of the petition date of the bankruptcy filing, there was such a shortfall in the token portfolio at the exchange that I think it was decided that's not going to be the best route and it's going to be extremely costly. we're going to go for a value -maximizing prudent monetization of the illiquid assets for the benefit of customers. And so that's where we are. And certainly, I don't think anybody wants the estates to go out and buy crypto at today's market prices to make in -kind distributions. So I think what we're really talking about is the appreciation of the assets that the debtors are holding?

27:01Who should get that? Should it be that customers just get 100 % of their petition date, dollarized claim pay? Or should they have the benefit of appreciation that they would have otherwise had, but for the bankruptcy filing, given the ownership arguments? And then there's further arguments of, is there a distinction between someone that held BTC as of the petition date versus someone that held fiat. Because again, these weren't contracts nominated in one asset type. You were freely trading in and out. So I may have liquidated my token portfolio a couple hours before the bankruptcy finally, because I wanted to get my funds out.

27:40Should I not get the value of the cost of capital time value of money if I would have put that fiat right back into the market and seeing the same returns as a crypto denominated claim. So what we're really talking about is you can't get around the dollarization of claims and value in as a pitch date. We just need to figure out what's the best way within the confines of the bankruptcy code to make sure that that appreciation ends up in the hands of customers. In terms of the assets that actually stayed within the business. Have they been held there where there's some large balance of Bitcoin that has thus appreciated over this period of time?