Ain’t no party like a counterparty (cuz a counterparty has risk)
Ain’t no party like a counterparty (cuz a counterparty has risk)
Plus 2 Fast 2 Fungible and Elon Musk is sued over Dogecoin
In this issue:
Ain’t no party like a counterparty (cuz a counterparty has risk)
No such thing as a free lunch (reader submitted)
Elon Musk is being sued for his Dogecoin nonsense
The hot new trailer for 2 Fast 2 Fungible
Ain’t no party like a counterparty (cuz a counterparty has risk)
In September of 2008 the Lehman Brothers investment bank filed for Chapter 11 bankruptcy after facing unprecedented losses in the subprime mortgage markets. Bankruptcy forced the Lehman Brothers to sell their (~$680B) assets at distressed prices, pushing asset values down even further. It also meant large losses for the banks that had loaned Lehman Brothers money. But most importantly it damaged the market’s confidence in all the banks.
The Lehman Brothers’ bankruptcy was contagious — other banks rushed to close loans, withdraw accounts and reduce exposure to each other, increasing pressure on each other as they did and making the situation worse. The risk to Lehman had become systemic risk — when the crisis happened the damage wasn’t just limited to Lehman. It affected everyone in the entire banking system.
In May of 2022 the TerraUSD stablecoin and accompanying Luna governance token violently collapsed, wiping ~$60B worth of value off the market. The collapse forced the Luna Foundation Guard to sell their (~80k BTC) assets at distressed prices, pushing asset values down even further. That was obviously a disaster for holders of Terra/Luna but it was also a major problem for many other banks and pseudobanks in the industry, since many were just depositing customer money into Terra and keeping a cut of the interest earned for themselves.
We talked last week about Celsius, a centralized crypto lending/staking firm now in the slow, painful process of admitting that it is insolvent. Babel Finance (a similar company based out of Hong Kong) has also frozen withdrawals and UK competitor Finblox has limited withdrawals to a maximum of $1500/month. Broker Invest Voyager is taking loans to "safeguard customer assets." And it seems increasingly likely that crypto-focused hedge fund 3 Arrows Capital (~$18B assets under management) is already insolvent.1
According to The Block, 3AC has had at least ~$400M worth of liquidations from major lenders. They spent ~$560M to buy locked Luna that is now worth ~$670 (assuming price holds up until the tokens unlock). They were heavily tied up in the stETH/ETH trade we talked about recently. And apparently they have been swiping cash from wherever it was available, including shared trading accounts:2
As with Lehman Brothers the first order impact will be to the markets for the assets that 3AC will dump for cash. They have large token stakes in $BTC, $ETH, $AVAX, $KSM, $MINA, $NEAR, $POLK, $SOL, $LUNA, $AAVE, $BAL, $FIDA, $DYDX, $LDO, $JOE, $SYN, $MULTI, $MINA, $AXIE, $RAIDER, $IME, $NYAN, and $MVS so it seems likely those tokens will face heavy sell pressure as they unwind.
Probably the more important question is who invested with 3AC or loaned them money. Rumors are that 3AC asked for (and regularly received) undercollateralized loans — meaning that 3AC was effectively ~4:1 margin long on their portfolio and they were rumored to be recycling their collateral to get even larger loans. Crypto lending platform Nexo released a statement assuring customers they had $0 of business exposure with 3AC, but both BlockFi and Deribit have made oblique statements about recognizing losses from distressed accounts.
Scrutiny in particular has fallen to BlockFi, which was caught on the wrong side of the GBTC premium arbitrage we’ve written about before, has sustained a ~$100M SEC fine earlier this year and only ~17% of their loans are fully collateralized. Most tellingly they raised a round of funding at a lower valuation (~$1B down from ~$3B) meaning they both needed more money and had to admit to investors that the business was less valuable than they thought during the bull market. They also (like many crypto businesses) recently announced ~20% layoffs.
BlockFi has confidently announced that customers funds are totally safe, but it is hard to say whether BlockFi is feeling confidence or just projecting it. Alex Mashinsky (CEO of Celsius Networks) was on Twitter berating journalists for FUD only hours before Celsius froze withdrawals and began exploring bankruptcy. The risk here is that health of the company itself is only one factor — fear of a possible bank run can become a self-fulfilling prophecy. Maybe BlockFi lost enough to 3AC to materially damage its business — or maybe enough people will worry about that possibility that it becomes materially damaging anyway.
The next group teetering on the brink of disaster are the Bitcoin miners. It is hard for miners to get good loan terms because everything they own to put up as collateral is heavily correlated with the business. If a restaurant fails the building it was based in is probably still valuable — but if a Bitcoin mining company failed it is probably not the best time in the world to be selling bitcoin or Bitcoin mining equipment. That means lenders demand a lot of collateral from Bitcoin miners, which means Bitcoin miners have a very thin margin for profitability.
As the price of Bitcoin stays depressed or even drops further the weakest Bitcoin miners will start to go bankrupt, selling all their Bitcoin as they go. As they do stronger Bitcoin miners will buy their facilities, contracts and equipment — but they will need to sell their own bitcoin to do so. Most Bitcoin miners will either lose their bitcoin going bankrupt or spend it buying up the ones who fail.
Publicly traded miners have ~40-50k Bitcoin in their treasuries — a little over half as much Bitcoin as the Luna Foundation Guard sold when UST collapsed. Until those coins have been sold the market probably isn’t done dropping yet.
No such thing as a free lunch
“I was reading your thread-of-Twitter-threads and I saw your thread about decentralization always being expensive. But it seems like the analogy about highways doesn’t really help with blockchains, where we can add orders of magnitude more capacity. Demand eventually grows but those improvements can last for a while! For example when I was at [company] our notification system was pretty limited at first. Then we improved the pipeline throughput ~100x and we didn’t have to think about scaling for another few years. You just have to stay ahead of the growth in demand.” — FP
Congestion pricing is counter-intuitive and fairly different from capacity planning in other contexts. You can’t actually grow capacity faster than demand in a congestion constrained system because the demand is already there. That's what makes the system congested. New demand doesn't need time to develop — it is waiting just out of sight ready to consume any new capacity as soon as it emerges. Adding more capacity to a congested system is like drilling holes at the bottom of a water tower: bigger holes will let more water through but the water pressure will stay the same.
Economists call this phenomenon Jevons’ paradox. The only way to escape this trap is to have more capacity than the entire system can make use of — but the demand for blockchain transactions is effectively infinite. The cheaper transactions get the more transactions people want to make. There is no meaningful limit. The water tower can never be emptied.
As a concrete example there used to be a service built on Bitcoin called Satoshi Dice that let anyone do provably fair micro wagers directly on the blockchain. It was quite popular! At one point Satoshi Dice transactions represented >50% of traffic on the network. Nowadays Satoshi Dice no longer operates on Bitcoin because transactions are too expensive — but it hasn’t been forgotten and it wouldn’t need to be reinvented if transactions became cheap again. It would just turn back on.
Every transaction on a network creates work for validators to confirm. As transactions get cheaper there will be more of them, which means it will take more work for validators to secure the network. Everything that makes transacting cheaper makes validating more expensive by definition. So if you want a decentralized network (i.e. a network that is cheap/easy to validate for yourself) transacting on that network will always be expensive.
No such thing as a free lunch!
Other things happening right now:
Elon Musk is being sued for flagrantly manipulating the price of Dogecoin. I don’t know how likely the case is to actually succeed but I can’t emphasize how much I support the effort. If anyone knows the legal team I am happy to provide expert testimony pro-bono.
In a fit of pique this week play-to-earn focused DAO called Merit-Circle voted to "refund" one of their early investors, YGG (denying them a significant profit on their investment) for not sufficiently contributing to the DAO. Outside the money, that is. According to the proposal author: "Decentralization is about freedom. The freedom to do as one desires, unbound by anything and everything." That genuinely might be the stupidest thing ever said? It is unclear if the agreement YGG signed in investing leaves them any legal recourse but one suspects other VCs in the future are going to be more careful.
Presented without comment:
We mentioned 3AC once before back in August of 2021 — at the time they had just finished a ~$6M shopping spree for 104 Cryptopunks.
In bird culture this is considered a dick move.