When Bridges Burn
Vol #110: December 15th, 2022
The Do Lung Bridge burns over and over again. It represents distortion and chaos in the endless fog of war. Futility that rages on, purposelessly.
When I last wrote about bridges, the cinematic reference was the Bridge Over the River Kwai. The scene from Apocalypse Now has way different vibes.
The burning of bridges is a particularly violent act of war because it salts the earth for both victor and vanquished alike. It destroys links that have been forged, and regresses the benefits of interconnectedness. Lines of communication will have to be rebuilt anew, over the scars of violence.
Perhaps that’s a little bit over dramatic for some magic internet money fraud. Still, the impact of the FTX bankruptcy will ripple across the trading and credit space in perhaps even more significant ways than we’ve seen in past versions of this movie.
To build decentralized bridges, you need to have the tommy boy optimism of Alec Guinness. Blockchains are built on trustlessness, so any external link is a lynchpin. When they burn, their fire reflects in the hollow eyes of Colonel Kurtz, feeding on the weak and strong alike.
The major users of the blockchain have been traders, investors, gamblers, and scam artists. Let’s not sugar coat it. For now it’s a money game, and naturally attracts speculators of every type. The opportunity sharks prowl the ecosystem looking for any way to make a buck or pull a rug. With different opportunities on different blockchains, the rewards for good navigation are fruitful.
Sound a bit familiar? That is the modus operandi of any investment firm. Macro funds want to be able to trade at every global market center in every imaginable security type. I once spent a month updating Japanese municipal bond data for a fund that had never traded a single one. Even ultra focused product specific operations (e.g. electricity traders) need access to a broker network, hedges, or capital.
Prime brokers fill this role in traditional markets, but centralized exchanges are the main players in crypto. Players like FalconX have tried to step into this space, but the competitive advantage of offshore juridictions is not to be discounted. These venues were not just a marketplace, they were a credit facility, linkage network, and custodian.
Billions of dollars of customer money has been fraudulently evaporated from the space, and we’ve lost a major bridge center. With FTX’s incineration, it’s going to take some time to bring together the capital and intrepidness to rebuild. Especially in this chop licking regulatory environment.
The nature of market making in the crypto world means keeping big piles of money at a couple of different exchanges. This is not true of say equity options, where you deposit at Goldman or Merrill and clear your trades on 16 different exchanges. (Not to mention the underlying OCC guarantee of listed options trade.) When there’s a price difference between FTX and Binance, a $10 spread sounds like easy money, but has numerous moving parts and mucho capital in play.
To make $1000 on that trade, you need to do it a hundred times. 100 Bitcoin has a notional value of roughly $1.75 million. At a relatively thin 10% margin, you’re required to post $175k on each exchange, buying and selling. With $350k of capital you make $1000 if everything goes perfectly. The exchange has not only lent you the money, but they’ve provided the trading venue and custody for the product. They’ll take a cut of your profits in exchange fees and interest, and pay their own bills for server farms and stadium naming rights.
Keeping markets efficient is an expensive business to maintain, but when it harmonizes, it’s a beautiful dance. Customers, dealers, and infrastructure providers are all happier. Losing one of the supporting participants is a significant blow. The individuals and teams who were keeping markets efficient, just took a shot in the broadside. Bringing liquidity back to markets will be a much slower process as major participants are smarting.
In order to frame what might happen next, it’s interesting to look at both historical parallels, as well as some of the crypto native solutions that are possible.
Not all that long ago Bear Stearns had a major share of the investment bank prime brokerage business. Shortly before their collapse they were the number two player with over 20% market share. This contracted rapidly in the months leading up to their 2008 firesale, but in not so distant history they were praised for their rock solid credit rating and balance sheet. Supporting this were armies of traders that were the sharpest and most ruthless dealers on the street. (Again, sound familiar?)
When JP Morgan agreed to buy them for $2 $10 per share, a major consideration was acquiring this prime brokerage business. Sure they may have been nudged by the Fed to complete the deal, and at a better price; but that business line has been a valuable long term asset to both JPM shareholders and the industry.
When Lehman failed, hedge funds got spooked and started to diversify. No one wanted counterparty risk, or capital tied up in a bankruptcy hearing. Funds had their assets chopped up across an average of 5 brokers by 2009. This diaspora quickly reversed though, as the only explosion of activity following the GFC was on the compliance front.
I don’t think this specific element will play out in crypto, simply because there aren’t any other real options. Binance is mega-dominant, but even they have been plagued with massive withdrawals. OKX is a reasonably sized player, but no one really had the listings and functionality of FTX.
Economics forced centralization as the smaller prime brokers couldn’t meet all of the operational and capital demands of a new regulatory regime. Today, Goldman Sachs and Morgan Stanley are head to head with about 19% market share, followed closely by JP Morgan at 16% market share - thanks to the Bear book. The next 7 are less than half that, clustered between 4-8% market share. Crypto participants would welcome this level of oligopoly.
The best way to win client market share is to best service their demands, and in the prime business that means access. Not to CFOs for a whisper print; but to markets, capital, and the technology to win. The biggest players naturally accrue competitive advantage in all of these domains. Crypto participants are no different here, they want cross chain bridges, diverse listings of futures contracts, high leverage, and good risk controls (for their co-participants of course, never themselves.)
Hedge funds are trucks built to drive through any window of market opportunity. They like to maximize returns by playing as big as the risk department will let them. Further, the fattest pitches tend to come in some of the more esoteric markets. Brokers want these size players because they pay the biggest fees.
Sales gets clients in the door, but risk management is what keeps you in the game. This wrecked Credit Suisse with their exposure to Bill Hwang’s outsized bets on illiquid and bubblicious technology stocks. Sam Bankman-Fried has pleaded that his own failings as a broker were those of poor risk and internal controls. He was certainly a bad risk manager, but he’s most likely a bad criminal too.
Today, FTX creditors (i.e. funds with stolen capital) are scrambling to find new places to deploy what remains of their funds. Others have pulled out all together, as witnessed by a dramatic fall in liquidity and trading activity.
Centralized exchange volumes have fallen to their lowest level of the last 2 years. Crypto prices have somewhat decoupled from their correlation to the macro inflation trade, but not in a particularly positive way. NFTs are chugging along, but the mania of thousand dollar gas fees on mints is gone.
The years following the GFC were doldrums of low volume, volatility, and an endless stream of new exams to take and procedures to follow. Trading firms in 2005 had a compliance officer that smoked a pack a day and lived on a houseboat. By 2010 they had members of a different bar.
With so few of the traditional model prime brokers available to crypto firms, there is an open door for decentralized protocols. Solutions that can effectively network capital between different blockchains will level up the ecosystem. How this works technically remains to be seen.
There are good reasons to be skeptical about the security concerns of a bridge in a permissionless world. Within a single chain the system of validators has a consensus architecture that is designed to be extremely difficult or impossible to hack. A bridge however, necessarily must be a trusted intermediary across the gap. It’s inherently a centralized point of failure.
The design of good bridges was a major topic earlier this spring before the months of never ending ripples across the crypto space. It is more important now than ever, as the rubble from FTX’s demise still smolders.
At the design level, there are very compelling experiments happening. Decentralized stablecoins were also an interesting experiment, so suffice it to say that prudence is warranted.
A bridge can choose to rely on a massive amount of external capital such that it’s “good for it” on both chains, and thus incur the security requirements of that size of a bounty. Efficiently using this capital - like with Stargate - can help, but it’s not a panacea. An alternative has been to create a native token that offsets against liquidity facing multiple different chains - the ThorSwap model. This has its own risks, broadly similar to LUNA / UST, but the thrust of the argument is noteworthy.
Following the GFC, there was an interesting power shift and pendulum swing in prime brokerage dynamics over the course of a relatively short period of time. Crypto has prided itself on speed-running the course of history, so evolution will undoubtedly unfold as the digital ink here dries.
On a separate note, The Till will be taking the next two weeks off to enjoy the holiday season with friends and family. I hope everyone gets their end of year rebalances done in short order, so they can do the same.
See you all next year!