Archegos Was Too Busy for Margin Calls
By
Matt Levine
- Juli 2021, 19:12 MESZ
One important thing that investment banks do is lend money to hedge funds to buy stocks. This is risky: If the stocks go down, the hedge fund might not repay the loans, and then the bank might lose money. So a central question of risk management is whether the hedge fund has posted enough collateral — that is, that the stocks it owns are worth significantly more than the money the bank has loaned it, so that if the stocks go down the bank will not lose money.
One aspect of this question is: What is the right amount of collateral? This question can be reduced to questions like: How much will the stocks go down in a plausible worst case? How quickly can we sell the stocks, and how much more will they go down due to our selling? You can look at numbers about historical and implied volatility and correlations of the stocks in the portfolio, you can compare the position sizes to the stocks’ daily volumes, you can plug these things into formulas and run scenario analyses, you can get some numbers. These are well-known problems that have received a great deal of academic, regulatory and practitioner attention, and you can read papers and books about the best practices for figuring out the right amount of collateral.
But another, underrated aspect of the question is: Once you know the right amount of collateral, and you call up the hedge fund to tell it to post more collateral, and the hedge fund says “I’m busy today let’s talk tomorrow,” and you call them tomorrow and they say “hey this week got away from me but send me an email,” and you send them an email summarizing your collateral demand and call them next week and they say “oh I haven’t had a chance to look at your email yet but I will very soon,” and meanwhile the right amount of collateral keeps ticking up … what do you do about that? There is a theoretical and contractual answer, which is, if the client doesn’t post the collateral you want then you terminate the swap, but you are a person and the hedge fund does sound really busy and surely it can’t hurt to talk to them tomorrow? Plus if you terminate the swap you lose their business, and your whole job is about doing more business.
Here’s an absolutely majestic paragraph about investment bank risk management:
On February 19, 2021, the PSR analyst sent a dynamic margining proposal to the Head of PSR for internal review, noting that he had made the terms “about as tight” as possible, yielding an average margin of 16.74% if applied to Archegos’s existing swaps portfolio and leading to a day-one step up of approximately $1.27 billion in additional margin. This was less than half of the additional initial margin that would have been required if Archegos’s Prime Brokerage dynamic margining rules were applied to Archegos’s swaps portfolio.On February 23, 2021, the PSR analyst covering Archegos reached out to Archegos’s Accounting Manager and asked to speak about dynamic margining. Archegos’s Accounting Manager said he would not have time that day, but could speak the next day. The following day, he again put off the discussion, but agreed to review the proposed framework, which PSR sent over that day. Archegos did not respond to the proposal and, a week-and-a-half later, on March 4, 2021, the PSR analyst followed up to ask whether Archegos “had any thoughts on the proposal.” His contact at Archegos said he “hadn’t had a chance to take a look yet,” but was hoping to look “today or tomorrow.”
The first few sentences there are about technical questions of the right amount of collateral, etc. But once you have the right amount, you call the client. And the client is busy! And then what?
In this particular case the answer to “and then what” is that, a couple of weeks later, the stocks that Archegos Capital Management owned went down a lot, and it did not have enough money to pay back its loans from Credit Suisse Group AG (whose Prime Services Risk analyst had called Archegos to ask for more margin), and Credit Suisse lost $5.5 billion because it turns out it did not have enough collateral against Archegos’s positions. So, not great. Credit Suisse was very sad about this, and its board of directors commissioned a report from the law firm Paul, Weiss, Rifkind, Wharton & Garrison LP about what went wrong, and today it published it, and the quote above is from page 116 of the report.
The report is … look, tastes will vary, and I concede that I am a weird guy, but it is so, so good? It is thrilling reading, as good as anything you will ever read about the management and sociology and processes of big investment banks. And that, it turns out, is the story. There is not some fancy finance thing that went wrong, some clever trick that Credit Suisse missed, some interesting problem in the math or the legal regime that caused Credit Suisse to lose so much money. Nor is this a story of individual stupidity or greed or recklessness; people generally had the right facts and were trying to do the right thing and kept each other in the loop. It’s just that they sort of kept each other in the loop as a substitute for actually doing anything. The processes were all moving along nicely, which gave everyone a false sense of security that they would produce the right result. Unfortunately the processes were slow and Archegos blew up quickly.
(Credit Suisse responded to the report by doing a bunch of things, including adding new risk officers, getting rid of some executives, and clawing back $70 million of bonuses.)
When the Archegos story came out this spring, there was a sense, from the outside, that the banks had missed something, that there was some structural component of Archegos’s trades that caused the banks to underestimate the risks they were taking. For instance, there was a widespread theory that, because Archegos did most of its trades in the form of total return swaps (rather than owning stocks directly), it didn’t have to disclose its positions publicly, and because it did those swaps with multiple banks, none of the banks knew how big and concentrated Archegos’s total positions were, so they didn’t know how bad it would be if Archegos defaulted.
But, nope, absolutely not, Credit Suisse was entirely plugged in to Archegos’s strategy and how much trading it was doing with other banks, and focused clearly on this risk. Its Credit Risk Management (CRM) unit, which was in charge of managing Credit Suisse’s overall credit risk, got detailed information about Archegos’s portfolio, its concentration, and how long it would take to liquidate it. From pages 17-18 of the report:
In December 2020, Archegos reported to CRM that its top five long positions represented 175% of its NAV; moreover, Archegos held two positions that represented between 5 and 10 days’ DTV [daily trading volume], six positions that represented between 2.5 and 4.99 days’ DTV, and another nine positions that represented between 1 and 2.49 days’ DTV in those respective stocks.
In January 2021, in connection with its 2020 annual credit review, CRM downgraded Archegos’s credit rating from BB- to B+, which put Archegos in the bottom-third of CS’s hedge fund counterparties by rating. CRM noted that, while in prior years Archegos had estimated that its portfolio could be liquidated within a few days, Archegos now estimated that it would take “between two weeks and one month” to liquidate its full portfolio.
And the business unit that managed the Archegos relationship, Prime Services, had its own risk team (Prime Services Risk, or PSR); when CRM downgraded Archegos, the PSR analyst asked why and summarized the answers as follows (page 113):
CS sees a vertical slice of [Archegos’s] book, meaning there are not any hidden names we’re unaware of
So names like Viacom, Tencent, Discovery all > 3 DTV [in the CS portfolio], if there is an issue, all brokers would be looking to exit simultaneously
And CRM further warned Prime Services the next day (page 114):
The CRM analyst covering Archegos escalated the same concern that the PSR analyst had elevated to the PSR Head the day before: namely, that Archegos’s concentrated positions with CS were likely also spread across its other prime brokers. The CRM analyst told his supervisors that, while Archegos refused to answer specific questions about its holdings at other prime brokers, Archegos had told him that, “as they leg in to positions, they ideally prefer to do so pro rata across their core [prime brokerage] providers,” including CS, although that was not always accomplished. The CRM analyst noted that CS “should assume that [Archegos] potentially ha[d] additional exposure” on the same large, concentrated names “away from [CS].”
So there is just no truth to the idea that Credit Suisse didn’t know what was going on. It knew that Archegos had large concentrated swap positions with Credit Suisse, and that it replicated those positions with other big brokers; it knew that “if there is an issue, all brokers would be looking to exit simultaneously.” It knew that if things went wrong they’d go very wrong, and the right people cared about this risk.
Similarly, there was a sense from the outside that the fact that Archegos did its trades on swap, rather than by owning the stock directly and borrowing from Credit Suisse using margin loans, reduced the amount of collateral that it posted to Credit Suisse. This part is true, for sort of a dumb reason. If you buy $1 billion worth of stock and are required to post 25% margin, you put up $250 million. If the stock goes up to $2 billion, you have to have $500 million of margin. At that point you have $1.25 billion of equity in your account (the $250 million you originally posted plus your $1 billion gain), and you can take out $750 million of it, leaving you with 25% of margin.
Similarly if you buy $1 billion worth of stock on swap (that is, Credit Suisse buys the stock and gives you the economic exposure via a swap), and you are required to post 25% initial margin, you put up $250 million. But if you bought $1 billion worth of stock, on swap, from Credit Suisse, when Archegos started putting on these trades, that initial margin was “static”: It’s 25% of the initial value of the contract, not its current value. So if the stock goes up to $2 billion, you still only need to have $250 million of margin in your account. So you can take out your full $1 billion of profits, leaving you with 12.5% of margin. 1
And that is basically what happened: Archegos did lots of swaps with Credit Suisse, and the stocks went up a lot, and it took out all of its profits, leaving Credit Suisse with very little collateral.
At some point Credit Suisse realized this was bad, and many pages of the report are devoted to Credit Suisse’s efforts to get its swaps business onto a “dynamic margining” system that would allow it to require more margin as positions expand. But these efforts were stymied by the two great permanent obstacles to good risk management:
The business people didn’t want to ask the client for more margin, because they worried that they’d lose the business; and
When they did eventually ask the client for more margin — less than the risk managers wanted, but something — the client was busy and promised to get back to them.
We covered Point 2 above (“he ‘hadn’t had a chance to take a look yet,’ but was hoping to look ‘today or tomorrow’”), but it was sometimes subtler. In some internal discussions, Credit Suisse took some comfort from the fact that its contracts allowed it to terminate the swaps or raise margin requirements quickly. But it would be hard for Credit Suisse to exercise those rights because the client didn’t want them to. Page 109 includes an email from CRM to PSR with this piece of bitter sarcasm:
Need to understand purpose of having daily termination rights and ability to raise margin [with] 3-days notice on swap if client is not amenable to us using those rights.
If you have a contractual right to raise margin requirements, but your client would get really mad if you used it, that’s better than nothing, but not much. You can’t raise margin requirements while things are good, because you don’t want to make the client mad. When things are bad, you stop caring about making the client mad, and you can raise margin requirements — but by then it’s probably too late.
On Point 1, here is page 19 of the report, again about February 2021, a month before Archegos blew up:
CRM suggested that the business develop a precise timeline for transitioning Archegos to dynamic margining, that it return to monitoring Archegos under the more punitive Severe Equity Crash scenario, and that Archegos be required to post $1 billion of additional initial margin with CS. The business forcefully rejected the idea of requiring Archegos to post $1 billion of additional margin, saying it was “pretty much asking them to move their business.”
It turns out that asking Archegos to move their business would have been a really good idea! As Credit Suisse found out a month later. But that is the sort of thing that business units only ever discover in hindsight. Nobody gets to run a business unit at a bank by cheerfully sending clients away.
That’s why you have risk managers! Unfortunately (page 13) 2 :
Around [August 2020], the CRM analyst covering Archegos raised concerns to his supervisor about PSR’s overall management of counterparty risk, including, specifically, with respect to Archegos. He observed that the PSR team in New York (covering Archegos) was not “adequately staffed to be reliable”; experienced PSR employees who had left CS had not been replaced; everyone he would “trust to have a backbone and push back on a coverage person asking for zero margin on a heaping pile” was gone; “the team is run by a salesperson learning the role from people” he did not trust to have a backbone; and PSR was not “the best first line of defense function anymore.”
The way of a bank is that coverage people ask “for zero margin on a heaping pile,” and the risk managers have backbone and say no. Or they don’t, and you get a $5.5 billion loss and this report.
Everything in the report is like this. This report is not a bunch of lawyers identifying a bunch of problems and characterizing them, in hindsight, as “red flags.” Everyone saw all the problems here, evaluated them reasonably, came up with sensible solutions and then didn’t do them. A ghostly recurring character in the report is the CPOC, the Counterparty Oversight Committee, a fancy committee that Credit Suisse started after losing a bunch of money on swaps with another hedge fund, Malachite Capital Management, in 2020. 3 CPOC grew out of “‘Project Copper,’ an initiative to ‘improve [CS’s] ability to identify early warning signs of a default event,’ and ‘enhance [CS’s] controls and escalation framework across functions during periods of stress’” (page 15). And at the very first meeting of CPOC, in September 2020, it discussed Archegos, identified the key problems, agreed on good solutions, and then forgot about them forever:
The meeting materials observed that Archegos “makes substantial use of leverage relative to peer [long/short] equity funds and exhibits a highly volatile performance pattern”; that Archegos “has generated some of the largest scenario exposures in global [hedge fund] portfolio”; and that Archegos had “[c]hunky single-name stock exposures (a number of positions > $750 mm and > 10% GMV) albeit in liquid names.” At the meeting, participants recall that members of CRM and the Head of PSR co-presented Archegos. The Head of PSR noted that the business and Risk had already agreed on actions to address Archegos’s limit breaches and observed that Archegos had never missed a margin call, even in the tumultuous markets earlier in the year. While the minutes reflect general discussion of Archegos’s concentrated positions and the “desirab[ility]” of an automated margin add-on for concentration, we have seen no evidence that anyone called for urgent action. Indeed, the “Action/Decision ” for Archegos was for CRM to “notify of any changes with the counterparty and revisit the counterparty at a future meeting.” CPOC did not set a deadline for remediating Archegos’s limit breaches, for moving Archegos to dynamic margining with add-ons, or even for reporting back or revisiting the status of Archegos at a future meeting. CPOC did not discuss Archegos again for nearly six months, until March 8, 2021, at which point Archegos’s risk exposure had increased dramatically.
In some very abstract sense, the way to manage your exposure to a large counterparty is to talk about it in a meeting of a senior-level committee and agree on good solutions. Because someone will walk out of that meeting and say “ooh this powerful committee plans to check back in on this counterparty at a future meeting, and if I haven’t fixed the problem by then I will look dumb and lazy in front of my bosses,” and then they will go fix the problem. Usually that more or less works! It’s not ironclad though.
On March 8, a couple of weeks before Archegos blew up, CPOC discussed it again. It again correctly identified the problem (page 20):
The CPOC discussion also highlighted Archegos’s “[s]ingle issuer concentration,” including a $3.3 billion position representing “more than 8% outstanding float (next five largest are in the range of USD 1.2bn to USD 1.5bn).” CPOC discussed the difficulty of potential liquidation given the size of these positions.
But then it proposed an inadequate solution with a deadline that turned out to be too late:
Notwithstanding the red flags relating to the size, concentration, and liquidity of Archegos’s portfolio, CPOC concluded: “Action/Decision: Move client to dynamic margining with add-ons for concentration and liquidity within the next couple of weeks. If no traction perceived by the middle of week of March 15, request an additional USD 250mn margin from the counterparty.” The Head of PSR was designated the “owner” of the action item and given a target completion date of April 2021. Notably, that $250 million request was less than one fifth of the amount that would have been required as a day-one step up under the dynamic margining proposal PSR sent Archegos just two weeks earlier (and one twelfth of the day-one step up that would have been required if Archegos’s dynamic margining rules for Prime Brokerage had been applied).
And then the business people went off to implement the solution by asking Archegos for more money, but unfortunately Archegos continued to be very busy (page 21):
The business continued to chase Archegos on the dynamic margining proposal to no avail; indeed, the business scheduled three follow-up calls in the five business days before Archegos’s default, all of which Archegos cancelled at the last minute.
And then, in a truly incredible move, while it was ignoring Credit Suisse’s demands for more money, Archegos was demanding money from Credit Suisse, and Credit Suisse was giving it the money:
Moreover, during the several weeks that Archegos was “considering” this dynamic margining proposal, it began calling the excess variation margin it had historically maintained with CS. Between March 11 and March 19, and despite the fact that the dynamic margining proposal sent to Archegos was being ignored, CS paid Archegos a total of $2.4 billion—all of which was approved by PSR and CRM. Moreover, from March 12 through March 26, the date of Archegos’s default, Prime Financing permitted Archegos to execute $1.48 billion of additional net long positions, though margined at an average rate of 21.2%.
This is my favorite thing about the Archegos story, and I still don’t know what to make of it. Basically Archegos did a bunch of swaps with a bunch of banks on a handful of stocks. Then those stocks went up a lot. This meant that Archegos was in the money on all of its swaps: It had huge profits, which the banks had to post as “variation margin,” crediting Archegos with money for the profits on its swaps. And in March of 2021, days or weeks before it blew up, Archegos cashed in those profits: It “swept” its variation margin, asking its banks to send it the money that it had made, leaving only the minimum “initial margin” in its accounts. 4 And because the swaps had gone up so much, the margin Archegos had left at Credit Suisse represented a tiny percentage of the total amount, roughly 9.4% of the exposure (page 127).
And then in late March, Archegos’s stocks went down, and Credit Suisse asked for the money back, and Archegos said, whoops, no 5 :
Archegos’s concentrated positions had dramatically appreciated in value in the months leading up to its default. During the week of March 22nd, the value of these positions began to fall precipitously. Archegos’s single largest position, ViacomCBS, dropped 6.7% on March 22 and continued to fall in the days that followed. On March 23, Archegos had over $600 million of excess margin remaining at CS but, by the next day, that excess margin was wiped out by market movements and Archegos owed CS more than $175 million of additional variation margin, which CS called, and Archegos paid. That same day, March 24, while the ViacomCBS stock price continued to fall, another of Archegos’s significant long positions, Tencent Music Entertainment Group, plummeted 20%. CS determined that it would be making a $2.7 billion call for variation margin the next day. Given the size of that call, the matter was escalated to the Co-Heads of Prime Services and the Head of Equities, who scheduled a call with Archegos for that evening to inform it of the upcoming margin call. Archegos’s COO informed CS that Archegos no longer had the liquidity to meet either CS’s or any of its other prime brokers’ margin calls on the following day. That evening, CS’s IB CEO and Group CRO were informed about the Archegos situation; it was the first time that either recalled hearing about Archegos.
On the morning of March 25, 2021, CS issued two margin calls—one for Prime Brokerage and one for Prime Financing—that together totaled over $2.8 billion. That day, Archegos reiterated that its cash reserves had been exhausted by margin calls from other prime brokers earlier in the week. While Archegos claimed it was committed to making its counterparties whole, it explained that it was only slowly liquidating its positions so as not to disrupt the market. That evening, Archegos held a call with its prime brokers, including CS. On the call, Archegos informed its brokers that it had $120 billion in gross exposure and just $9-$10 billion in remaining equity. Archegos asked its prime brokers to enter into a standstill agreement, whereby the brokers would agree not to default Archegos while it liquidated its positions. The prime brokers declined. On the morning of March 26, CS delivered an Event of Default notice to Archegos and began unwinding its Archegos positions. CS lost approximately $5.5 billion as a result of Archegos’s default and the resulting unwind.
Archegos took $2.4 billion out of Credit Suisse in mid-March. In late March, Credit Suisse asked Archegos for $2.8 billion back, and Archegos made a comical show of turning its pockets inside out to demonstrate that it didn’t have a penny remaining. Where did it go? One possible answer is that Archegos plowed all of it into margin for new swaps, and clearly some of that is true. (Archegos did $1.48 billion of new swaps with Credit Suisse in late March, posting 21.2% margin, and rolled over some other swaps; presumably it did similar trades with other banks.) Another possible answer is that Archegos came out of this trade better than its banks did. 6 The banks had to have committee meetings and consider the client’s feelings before asking for more money. Archegos didn’t have all of those processes and committees and sensitivities. It could just demand its money.
)
.