Battered Crypto Hedge Fund Three Arrows Capital Considers Asset Sales, Bailout
wsj.com
wsj.com
This weeks lesson is that in an up market its leverage that makes you money and in a down market its the unwinding of leverage that kills you.
They are also learning that interconnectedness will hurt more in a downturn. You can be perfectly delta hedged( don't care which way the market moves) and its the counter party risk that will sink you.
Now as these firms start to unwind they spill over to the next one. The collateral taht you lent to the failed firm is now gone and your once well hedge d and collateralized position is now in shambles because someone else can't pay you.
TradFi normally solves this by having a firm have oversite on everyones positions and have everyone pay money into a pot to help bail out failed counterparties so they don't spill over to other firms.
Crypto has tried to solve this by strict liquidation rules that tend to make things worse by turning firms into forced sellers when they may have been able to ride things out, thereby forcing markets down even more. To the point where its a profitable trading strategy to intentionally force the market down to liquidate leveraged longs.
As more crypto firms fail, they'll end up selling their best collateral first(BTC and ETH) forcing down the price even more.
$5,000 BTC and $200 ETH are in site if Tether starts to fail. Tehter has lent money to almost all these firms but doesn't mark down its positions from what it originally held them at.
If tether holds we may get out of this with $10-15,000 BTC but if Binance/FTX walk away from Tether then we'll be down another 80% from here I bet.
TL/DR its going to get alot uglier than it currently is due to the interconnected nature of the crypto markets. We'll find out that pretty much everyone was linked to everyone else as there are only so many ways to make money in crypto and everyone piled into them.
Defi is about to find out how hard it is to make money when no one wants to stake.
In crypto, there is no buffer. Auto liquidation of undercollateralized positions occurs. All of those humans in the loop, “unfair rules”, and settlement delays crypto proponents complain about are the very things that make traditional finance stable. This is why traditional finance doesn’t take seriously the idea of immediate settlement, and why end of day settlement is as good as it’ll get.
Is auto-liquidation a bad thing? It leads to more volatility in the short-term, but it seems that there might be long-term benefits of ensuring that players with unresponsible business practices will be forced out of the market.
"Traditional" crypto holders will be affected too (the value of their investment goes down while people are forced to sell), but this looks more like a temporary impact. As long as they don't trade on margin they won't be forced to sell and as long as they hold their own keys they are unaffected if exchanges are crashing.
From a "price" perspective the current events might be bad for the crypto ecosystem, but from a "health" perspective I think that they could be positive, as only the responsible players with healthy business practices are going to survive.
A stablecoin that backs its assets in the currency that the stablecoin issues (for example a USD-based stablecoin that is backed by US treasuries) won't be affected by any disruption in the market. An exchange that is fully backed and does not allow margin trading won't be affected either. Regular holders who hold their own coins or use responsible exchanges also aren't affected.
Yes - there's quite a bit of propagation. But all entities affected by that propagation didn't seem to have proper risk management and used business practices that work well when the market does well, but that fall apart in case of high volatility. Maybe I'm just biased because I'm not a fan of margin trading (and similar practices), but I don't consider it a disadvantage if an ecosystem is hostile to those types of business practices.
I think that's the thrust of the argument. We're about to find out how many DeFi companies are utilizing bad business practices.
The core might all be completely self-deluded or intentional scammers, but the second-degree people who invested/loaned to them intended to some degree to take a bit less risk than being directly involved in the scam, the third-degree people who rely on the second-degree people thought they were allowing a little innovation instead of missing out by playing it too safe, the fourth-degree people aren't even aware that they're invested in the core scam or delusion at all, and make fun of it online. There's a fifth-degree of people who have no idea that whether they eat depends on the scam (because the fourth-degree was a municipal bond issue, the city is now drowning in liabilities, so they had to cut food banks.)
edit: The most recent crypto peak was largely fueled by a ton of normies who put their retirements into something that has gone down somewhere between 40% and 70% in value since they bought it. Other people depend on those people.
Not if you pretend everyone is an island and economies aren't interconnected.
There are 9-digit lending positions in $ of crypto collateral that will be forcefully sold into the market if BTC/ETH keep dropping. Unless more collateral will be deployed, which lowers the forced liquidation price.
In a case where there's contagion [549] the banks or even the government can work to negotiate what's happening, and slow down the collapse (or even prevent it) - an example being the subprime mortgage backed securities which got to the point that nobody knew how to value them, so the government bought them all (and eventually actually "made" money by riding it out).
DEFI has algorithms that lock up the capital and will automatically liquidate it if certain parameters are met - which others can use to "attack" it - if the oracle sees bitcoin fall below $20k in a flash crash, it triggers the selling of ten thousand coins, say, which floods the price down further, you snap them up, the crash is over, you slowly sell at $25k or whatever. Bitcoin itself is pretty resistant to this, but the other coins, not so much, especially the side coins.
As a side note, most US home loans do NOT have a "call" clause and the bank can only initiate liquidation after you've missed payments and gone delinquent for a certain number of days; this was instituted in the US after calling mortgages contributed to the Great Depression. Some business loans can be called after a period of time for any reason (usually, interest rates have gone up).
Often, the lender is also the exchange through which people hold their crypto so they don't even have to contact the holder before they sell their holdings.
1. The price of an asset need not be a nice continuous line. It can jump.
2. That's why liquidation occurs not at -10%, but at some point before that, to manage this gap risk. Might be at -7%, or whatever. This reduces, but does not eliminate gap risk. For that, the crypto exchanges have insurance funds.
TradFi - Your word is your bond.
Of course few ppl actually look at the code, or sometimes the code is just bad and you get bad actors willing to exploit this.
That being said: In practice and real life there are a bunch of dodgy companies and badly written eth-contracts(code).
I've seen many a times, "business ppl" just go with lowest cost when wanting some software done. Try going for the "lowest cost bidder" when getting a web3 contract coded !
You right, bugs do and always will exists and we are all but human, but some bugs have a much higher cost than other - even if they are simpler.
So as a borrower, today, if you were to borrower 8000 USDC and deposit 10 ETH as collateral, the AAVE protocol would automatically sell your collateral if the ETH price went to 941$. So your position is "auto liquidated". https://aavecalculator.com/
AAVE currently has ~$9.9B in assets locked into its protocol.
Part of it depends on what the margin is used for - margin for VTX is going to be handled differently than forex margin.
Because they don't. Because it's a wild west. TradFi has it's failures as well - but that's "whataboutism".
DeFi is collapsing in real time and we'll probably see oversight and regulation around this in the near future.
From the perspective of people who support the mission of DeFi and aren't just in it for the money, this is a feature, not a bug. We want these dodgy, extractive institutions to go under and not have to bail them out.
Government regulation of retail finance is a feature, not a bug.
It's not a bug unless you believe adults should be treated like children and not be able to take the risks they desire or face the consequences of those risks. Anyone investing in DeFi should absolutely have known the risks; it doesn't justify giving the government even more power over people's finances.
So, the only reasonable move for an individual is, simply don't participate in the system in the first place, keep your money in the mattress. (Crypto equivalent: hold your own coins).
The problem is, credit actually creates real economic growth. A society that encourages credit via regulation will outcompete a society that leaves it to an unregulated marketplace (and thus disincentivizes individuals from participating).
Are these people adults? Yeah. Should they know better? Yeah.
The question is should we do anything about it? The Super Bowl is the most watched television event every year and there were lots of crypto ads this year. Do people need to be protected from themselves?
We have regulate the traditional financial industry. We put warnings labels on cigarettes. My view is that we need even more regulation of the traditional finance industry and apply those same regulations to crypto.
This is literally how traditional finance works as well. If you have a leveraged position at a futures or options exchange, you absolutely will be liquidated as soon as your position gets close to being in the red. I 100% guarantee you that the CME or Goldman will not let you "ride things out".
You don't have to take my word for this. Look up the history of what happened to the energy hedge fund Amaranth.
...The way the traditional system works is that if you have a margin loan, and the value of your collateral declines, your broker calls you up and says “hey we need more collateral,” and you either post more collateral and it’s fine, or you don’t post more collateral and your broker liquidates your position. Or sometimes you say something like “hang on I’ve got another call, I’ll call you back after lunch,” or “sorry I don’t have the money right now but I am a good customer and you know I’m good for it,” or “I don’t have the money right now but I’m pretty sure prices can only recover from here so why don’t you let it ride for a day or two,” or whatever, and you don’t post more collateral but your broker doesn’t liquidate your position because you’re a human and your broker is a human and everyone is embedded in a repeated social game with fuzzy rules
> In the traditional financial system, very few things work like this. One thing that mostly does is a margin account at a retail stock brokerage: If your stock declines, you will get a margin call, and if you don’t post margin within a defined and fairly tight time frame your broker will sell the stock, and this really might all be done by a computer in a pretty formulaic way. But if you have a big enough account — if you are a big hedge fund or family office — it doesn’t work that way. When Credit Suisse Group AG decided that Archegos Capital Management did not have enough collateral in its margin account, a Credit Suisse representative called Archegos and asked it to post more collateral, and Archegos said, sorry, we are really busy this week, let’s discuss next week. In theory Credit Suisse could have liquidated Archegos’s positions, but in practice that would have been rude, so it didn’t. Credit Suisse did not extend credit to Archegos based on some defined formulaic function of the value of its collateral; Credit Suisse extended credit to Archegos based on some fuzzy holistic relationship-based function of how much business it hoped to do with Archegos, how much the Credit Suisse people liked the Archegos people, how its traders felt about the collateral, when its risk committees had meetings, stuff like that.
Most recent case is that Archegos guy over-leveraged at Credit Suisse not to mention the LME or GME fiasco. The "traditional" market is rigged/bended by various individuals ALL THE TIME.
https://www.reuters.com/markets/commodities/elliott-associat...
You aren't liquidated immediately the moment your needs more liquidity.
We wait till the end of the day short of absolute emergencies.
I know Amaranth well being in Canada and Calgary specifically at the time:)
And for tradfi a typical hedge fund gets a call and is told to post more collateral and they make real efforts to allow the fund to post collateral. There is almost never an immediate liquidation as this tends to cause more trouble than it fixes, even on the CME or CBOE:)
Also generally the amount of collateral that needs to be posted is in the millions of USD so people tend to understand that isn't instant, especially if those funds are currently in money market funds or bonds held elsewhere.
Ho ho ho - no it's decided as soon as your lender things that it needs to liquidate you, and if it does it will know that you're not going to be able to go to court to fight afterwards.
A lot of nasty scenes start with someone on a phone shouting "you can't do that for another 2 hours".
Yup. Yup we can.
Depends on your CSA (Credit Service Annex/Agreement) and what the joint agreement was. Also depends on the type of position (daily settled via exchange vs contractual)
Amaranths positions were sold to Citadel and JP to wind down gently. That's pretty different from forced liquidation at the exchange.
Your prime brokerage is.
That is indeed generally the case. There are, however, monthly collateral agreements in some cases where you can ride it out for a month before getting margin called. Also, there are sometimes credit-rating based collateral calls. The most famous would be AIG's -- they eventually had margin calls of over $100B when their credit rating fell and the housing market also fell and liquidity dried up.
Funny thing in such cases -- when the borrower owes banks those sums of money, it ceases to be the borrower's issue and becomes the bank's issue
Who are these fully capitalized crypto firms?
I have no love for these firms so I really don't care if they are solvent. For those that took major risks with leverage and mismatched their risk on duration, let them fail. Some in this thread seem to be saying that having humans involved in liquidations is preferable because it prevents contagion. To me, that sounds like letting someone get away with bad decision-making and safe-guarding them against the previously defined consequences. Sounds like the same kind of philosophy that makes some firms in traditional finance "too big to fail." This kind of thinking kicks the can down the road instead of addressing the problem and feeling the pain now.
I say liquidate them. Maybe that's easy for me to say because my only exposure has been through defi protocols that I interact with directly and with risks that I've considered. These Cefi companies take retail money, use it to take risky positions in defi, then take a large cut of the profits. Works great until a market downturn and people start asking for the their Eth back and, whoops, we don't have access to it until 6M+ post-merge (Celcius). That's called piss-poor risk management and I don't want these kinds of actors in the market long-term. I feel bad for the retail that is getting screwed in all this but in the end, we need regulators / DOJ to come down very hard on these companies. If we clean house now, the space will be stronger in the future. Yes it will be painful in the short term.
Do you have a source for this?
The points you're looking for:
> Secured Loans (none to affiliated entities) $3,149,732,368
> Other Investments (including digital tokens) $4,959,634,446
https://www.coindesk.com/markets/2022/01/26/tethers-new-acco...
???
Tether has been saying they would produce an audit since July 2021, that's a year ago!
https://www.cnbc.com/2022/05/17/tether-usdt-redemptions-fuel...
This is long but fascinating. Tether is a scam, it just hasn't fallen apart yet.
https://crypto-anonymous-2021.medium.com/the-bit-short-insid...
[1] https://protos.com/tether-papers-crypto-stablecoin-usdt-inve...
IMO, pre-2017 prices for BTC (<$1000) returning is inevitable. If Tether fails, then pre-2014 prices (<$100) are more likely. BTC was at $5K a little over 2 years ago, anyone who thinks the bottom is at that level is deluding themselves.
What do you want to bet?
We go, "Wow, that's pretty interesting. Hope it doesn't happen" and we continue to eat our soup.
I enjoy reading stuff like this without the need for them to publish a study.
A belief in the long term value of cryptocurrency is very tightly correlated with ones current cryptocurrency position. The demand curve, then, for cryptocurrency, is extremely steep, and the "buy pressure" for cryptocurrency will not increase tremendously with falling prices, since the people who would generally be willing to buy at these reduced prices, have just lost their life savings.
Likewise, even the poeple who think that crypto is valuable long term know that you can't pay your taxes in it. If people don't believe that it's worth anything, then it isn't. They all know that the price falling could indicate that right this second is the most your cryptocurrency will ever be worth. The more the price drops, the more likely this is to be the case, and the more likely I am to sell. If the sell pressure is greater than the buy presssure at some price, then it is very likely to be the case at every lower price, resulting in catastrophic collapse.
This inversion event is more likely the longer the asset has remained in bubble status, since the optimism proportion of the inflated asset is strictly increasing with price.
Take increasing risks in search for higher yield? Check
Leveraged trading to ratchet up profits? Check
Margin called when the risky bet doesn't pay off? Check
Counter party risk? Check
Contagion + domino effect? Check
Fire sale of decent quality assets putting down pressure on its prices thus hurting the whole ecosystem? Checkhttps://www.ft.com/content/126d8b02-f06a-4fd9-a57b-9f4ceab3d...
I'm curious -- in traditional finance, there are often daily collateral settlements which force both parties to change collateral based on valuations shifts by way of the exchange. is there such a concept in crypto? Wouldnt this resolve a lot of the issues you're speaking about?
I have often described DeFi as Decentralized 2008.
https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...
What do you mean, may have been able to ride things out? They owe money they don't have...
Oh, you mean, if you wait long enough crypto might come back up and they're solvent again? Right, after they lost more than they had on the first bet, you wanna let them bet again that it might come back up?
"TradFi" at least has a real floor because "real" money (as in: currency that can be exchanged for goods and services) and "real" investments (as in: shares of companies that provide goods/services) do actually exist. Even if significant portions of the market are driven by speculation.
Are we reaching the point where people trading crypto get bored and just... stop?
But are we at least done pretending that crypto currency (at least, as it exists now) will ever have any practical purpose beyond being a fungible entity upon which gamblers speculate?
No, I don't think that's happened yet. I still know tons of people excited to build for "web3".
I don't think we need "rich" websites with javascript, "smart" phones in everyone's pocket, or "social media" at all.
But those all happened, and it seems to be a lot of the same vibe for "web3" builders right now.
There is no underlying value in having answers to a random cryptographic problem attested to you.
My entire point is that crypto is a unique asset class, so I think it’s here to stay as long as there are people willing to speculate.
The producers and consumers of soybeans are extremely likely to continue producing and consuming soybeans even if no one speculates on the value of soybeans. This is the "physicality" or "human interest" in soybeans.
Why would crypto be "here to stay" when there is zero use for it besides speculation?
I should also point out I’m not bullish on crypto. I just acknowledge it’s unique asset class, and I don’t think it’ll just disappear.
I'm saying people will continue to speculate on soybeans because they will always have some value.
With a purely speculative asset, the market will eventually run out of "bigger fools". At that point, any speculative trader will look at the asset and say "Why would I pay more for this than the 0 dollars it is worth?"
When you buy stock in a company whether is in during an IPO or new issue of stock that money is used to hire employees or build a new factory and make even more profit. That can increase your dividend and make the stock go higher. The same thing can happen when you buy the stock on the open market but the connection isn't as direct. Companies with good stock performance are able to attract better employees which will increase their profits which benefits the stock holder.
None of this happens when you buy bitcoin or ethereum. Some crypto exchanges make up their own crypto currency and offer high interest rates on those but as we are seeing most of those are turning out to be scams.
It was not a bailout for banks, it was for their account holders.
Good riddance.
PS: don't really think 3ac behavior is representative of crypto trading firms as a whole (Wintermute, Jump etc). A lot of them are highly professional and responsible operations who have weathered the storm just fine and in many cases, have generated nice pnl during this event.
How the fuck is this the subtitle? The firm's founders saying they believe this is evidence of nothing. A responsible journalist would not include it at all. It is journalistic malpractice to make it the subtitle.
> $5,000 BTC and $200 ETH are in site
I think you meant to write oversight and sight.