Bitcoin is illiquid. Proof of work is appallingly slow and expensive, and the fragmented cryptocurrency market structure means that there is a chronic shortage of liquidity. Furthermore, all transactions are visible to everyone, which makes markets volatile and invites front-running. Consequently, using Bitcoin for high-volume transactions is a nonstarter.
This is not my opinion. It is Blockstream’s.
Blockstream’s comment is in the “Strong Federations” whitepaper, which underpins its Liquid sidechain. The actual statement is in geek-speak, of course:
Bitcoin currently facilitates remittance and cross-border payment, but its performance is hampered by technical and market dynamics. The high latency of the public Bitcoin network requires bitcoin to be tied up in multiple exchange and brokerage environments, while its limited privacy adds to the costs of operation. Due to market fragmentation, local currency trade in bitcoin can be subject to illiquidity. As a result, many commercial entities choose to operate distinct, higher frequency methods of exchange. These attempts to work around Bitcoin’s inherent limitations introduce weaknesses due to centralization or other failings.
Here is a plain English translation of the sentences I have highlighted:
- The high latency of the public Bitcoin network requires bitcoin to be tied up in multiple exchange and brokerage environments
It takes forever to move bitcoins around, so they pile up in stagnant pools. (“Tied up” = “illiquid”)
- Due to market fragmentation, local currency trade in bitcoin can be subject to illiquidity
Exchanges can run out of money. So you not only can’t move your bitcoins, you can’t exchange them for real money either.
In short, as I have said many times, Bitcoin is disgustingly illiquid.
Blockstream’s Liquid sidechain aims to make Bitcoin more liquid by enabling large volumes of transactions to be taken off the main chain. But it isn’t possible to improve liquidity while maintaining full decentralization and trustlessness. So Blockstream cheerfully compromises both of these prized features. It appoints a small group of trusted institutions to validate transactions and submit them to the main Bitcoin chain.
A “Strong Federation” is a cooperative of market participants – called “functionaries” – who collectively validate transactions. Unlike Proof of Work, where miners drop in and out depending on their view of the market, the number of functionaries would not change; the paper talks about replacing the “dynamic miner set” with a “fixed signer set.” But it isn’t clear from the paper how the signers would be appointed. Who decides who should be a signer? Why should we trust them?
For me, this is a huge hole. Blockstream’s Liquid is at present the only commercial implementation of a Strong Federation. Just look what the whitepaper says its signers are:
Liquid is a Strong Federation where functionaries are exchanges participating in the network.
Unregulated, manipulated and insecure cryptocurrency exchanges are the “trusted functionaries” in Blockstream’s Liquid. This is the full list of exchanges and other institutions participating in Liquid (from Blockstream’s press release):
Altonomy, Atlantic Financial, Bitbank, Bitfinex, Bitmax, BitMEX, Bitso, BTCBOX, BTSE, Buull Exchange, DGroup, Coinone, Crypto Garage, GOPAX (operated by Streami), Korbit, L2B Global, OKCoin, The Rock Trading, SIX Digital Exchange, Unocoin, Xapo, XBTO and Zaif.
Would you trust this lot to validate your transactions and manage your liquidity? That is what Blockstream wants them to do:
By moving the bitcoin-holding risk, intrinsic to the operation of exchange and brokerage businesses, from a SPOF introduced by a single institution to a federation of institutions, Liquid improves the underlying security of the funds held within the network.
Instead of holding all your funds in one corrupt and hack-prone exchange, a bunch of corrupt and hack-prone exchanges will link themselves together so that it is much easier to move your funds from one corrupt and hack-prone exchange to another. This also makes it more likely that if one goes down, so do the rest. Didn’t the authors learn anything from 2008?
Of course, I am assuming that exchanges are both corrupt and risky. Perhaps I am being unreasonable – after all, they are not banks (or so they say). The whitepaper suggests ways of ensuring their honesty:
Incentives can be aligned through the use of escrow, functionary allocation, or external legal constructs such as insurance policies and surety bonds.
These measures would have to be extremely expensive to outweigh the potential gains to corrupt exchanges from controlling all the funds in a sidechain. Perhaps I am cynical, but multi-million dollar fines are simply a “cost of doing business” for conventional banks. Why should exchanges regard surety bonds and the like differently?
The Strong Federations white paper seems to imply that because the exchanges are part of the Liquid network (and therefore not third parties), there is no risk:
Strong Federations such as Liquid improve privacy, latency, and reliability without exposing users to the weaknesses introduced by third-party trust.
So you can eliminate the risk to sheep arising from wolves outside the sheepfold by bringing the wolves into the sheepfold and giving them full control of the sheep. What could possibly go wrong?
Of course, in prehistoric times, humans did exactly this. But it took thousands of years to train the wolves to round up and control the sheep instead of eating them. Exchanges are wolves, not sheepdogs. I forecast that any sheep venturing into Liquid will not survive for long.
And nor will weaker exchanges. Cooperative platforms only work if the participants are cooperative, and unregulated exchanges hate each other’s guts. They are intensely competitive and their aim is to drive each other out of business. I do wish that starry-eyed technocrats devising schemes that rely on cooperation between entities that want to eat each other would pay more attention to psychology. Any scheme that relies on fiercely competitive market participants cooperating for profit naturally tends towards monopoly. Just look at Bitcoin mining.
But despite the (considerable) risk of becoming tomorrow’s lunch, there is apparently a good reason for sheep to join Liquid:
By moving business processes to Liquid, users may improve their efficiency and capital-reserve requirements.
Apparently existing market provision is not good enough. Well, that is certainly true of cryptocurrency markets, but then they are extremely primitive. But the authors ignore cryptocurrency markets completely, and instead launch into an extraordinary attack on prime brokerage in conventional markets. They start with describing how prime brokerage generates market liquidity:
….liquidity provisioning is the primary business model of Prime Brokerages and Investment Banks. Fund managers commit their funds to a single location’s custodianship under the premise of reducing costs associated with investment management and improving access to both investment opportunities and liquidity. Third-party broker dealers then grant each participant access to the liquidity of their respective counterparties; a function of aggregation of capital under a single trusted third-party custodian. This system offers investors a means of preferential access to liquidity by enabling customers to buy, sell, and hedge trades with their respective counterparties in a single location. These centralized systems provide convenience to market participants, but are not without risks.
This is accurate as far as it goes, though as there are many prime brokers in conventional markets, a prime broker is hardly a “single point of failure” (SPOF) for the entire market. However, it is fair to say that prime brokers are weak points for their customers. Post-Lehman, of course, we have introduced regulations to protect their customers from losses. The authors completely ignore this.
But the writers then go on to offer, as an example of prime brokerage failure, losses sustained by the Eurosystem due to Lehman Brothers’ collapse in 2008:
One realized example of these risks is that of the Eurosystem following the global financial crisis, in the wake of the financial default of the Lehman Brothers. The effort of the Eurosystem to liquidate assets collateralized by 33 complex securities took more than four years, and resulted in over EUR 1 billion in losses.
Eh?? This was not a prime brokerage failure. The Eurosystem is a central bank, not a fund manager. Central banks don’t place customers’ funds with trusted custodians. They lend their own money to banks in order to maintain market liquidity and manage interest rates. The Eurosystem had lent money to Lehman, in return for which Lehman had pledged securities as collateral. As the value of those securities collapsed, the Eurosystem lost money.
Losses due to asset value collapse can happen in any market system. If the value of bitcoin falls, those who hold bitcoin – or have accepted bitcoin as collateral against (say) a USD loan – lose money. Making the system more liquid actually makes losses of this kind more likely, because people can dump collapsing assets more easily. Bitcoin’s illiquidity helps to support its price.
But the authors’ unfortunate choice of example inadvertently reveals the real issue with this paper. Rather than disrupting prime brokerage, as the authors seem to intend, the paper’s solution to Bitcoin’s liquidity problem in fact replicates the interbank market.
The interbank market pools and redistributes liquidity across market sectors, just as Liquid aims to do. And it has key “functionaries,” known as broker dealers, whose job it is to maintain market liquidity and act as gateways to the payments system. Without a functioning interbank market, transactions can be very slow or even fail, and banks can literally run out of money. Just like cryptocurrency exchanges, in fact.
Far from reinventing the financial system, the cryptocurrency world seems to be gradually forming itself into a simulacrum of the existing system. Perhaps this is inevitable. After all, the existing system is the way it is because of thousands of years of trial and error. With all its faults, perhaps it is simply the best that humans can do.