You might have heard the term “DeFi” making the rounds in the cryptosphere. Like us, you may even have tried to understand what it was all about but given up due to the complexity, opacity and jargon surrounding its key aspects.
Others have tried to shed light on its mysterious operations. The FT’s Miles Kruppa and Hannah Murphy, for example, noted as far back as the heady pre-pandemic days of December 2019 that DeFi operates as a broad umbrella term for blockchain projects that aim to eject human involvement in financial services by using smart contracts. To accomplish this, DeFi depends on the creation of “liquidity pools” and overcollateralised repo-style arrangements.
Even so, DeFi, which stands for “decentralised finance”, remains a murky if increasingly hyped up corner of the crypto world.
To be blunt, FT Alphaville’s initial knee-jerk reaction — based on what we should confess was not that much research at all — was that the very idea of decentralised finance being something new or exciting was hugely naive.
The story of financial markets, after all, is one of continuous decentralisation, innovation and engineering to synthesise seemingly risk-reduced returns and free lunches, which usually turn out to be anything but. Crypto is just another part of that story.
From repo markets to eurodollars and commodities, the story always starts with decentralised frameworks arising quite organically, being heralded as amazing wealth generators, then turning risky, then blowing up, then having regulatory forces battle to contain them — usually by centralising their processes and slowing down their growth by forcing expensive checks and balances on the systems.
Were the alchemic solutions being offered by DeFi really immune to that age-old pattern?
We somehow doubted it.
It’s for this reason a new paper DeFi Protocol Risks: the Paradox of DeFi, by Nic Carter of Castle Island Ventures and Linda Jeng, a visiting scholar at Georgetown University Law Center piqued our curiosity when it crossed our radar.
Neither Carter or Jeng are openly hostile to crypto, in fact quite the opposite — Carter is a well known advocate of bitcoin and crypto. That their incredibly well researched work should arrive at conclusions that fitted with our base assumptions seemed noteworthy, therefore.
To be clear, Carter and Jeng do see potential in DeFi bringing down the cost of financial services and broadening access. But they are also incredibly realistic about the challenges and risks. Notably, they are also wise to many of the inherent paradoxes of financial systems that we at FT Alphaville are so fond of flagging.
The paper is a dense read, but of course it would be: it’s DeFi. Nonetheless the authors do a good job of explaining the background of the movement, laying out the current landscape and then setting the scene for the risks.
But it’s their concluding points that really should not be missed, since they speak not just to the heart of the issue that faces of crypto but also to fintech and financial innovation more broadly. Basically, that there really is no free lunch. And, as Grossman-Stiglitz highlighted as far back as 1980, there is also “a fundamental conflict between the efficiency with which markets spread information and the incentives to acquire information”. This means prices will never reflect the information which is available “since if it did, those who spent resources to obtain it would receive no compensation”. Which de facto means if you remove the incentive (a.k.a. cost) to hunt for information, you will get bad prices and thus incur risk.
Here in any case is a similar concluding point from Carter and Jeng (with our emphasis):
Many of the risks described above stem from the decentralized nature of blockchains. The goal of automating the delivery of financial services and reducing human dependencies also has the congruent effect of reducing oversight and control. Disintermediating traditional intermediaries reduces high fees and entry friction, but also creates new opportunities for new types of intermediaries. These new types of intermediaries require the sufficient economic incentives and, thus, could be potentially more costly and risky than the monopoly rents extracted by today’s centralized intermediaries. Ultimately, this new host of intermediaries in a decentralized financial ecosystems could stymie the drive toward the twin goals of democratizing financial services: lowering cost and improving access.
They also observe just how difficult it really is to disintermediate a system from its base source of value, which stems from the core system. (This of course is something the eurodollar and MMF markets also found out in 2008):
External dependencies on traditional finance, namely banks, is another important source of risk as well as a transmission channel for risk. Although one of the goals of DeFi is to create a new kind of financial system without traditional intermediaries, the irony is that as DeFi struggles to make itself more useful in the real world, its dependency on the established financial system grows. Its reliance on traditional finance is not only a source of risk but can potentially serve as a transmission channel for risk between traditional financial and DeFi systems. DeFi leveraging stablecoins’ backing by fiat or other financial liabilities is an excellent example of this type of risk. Another dependency is that the crypto industry still needs to bank with commercial banks in order to conduct the cash legs of their transactions.
Last, here are some insightful observations about why it is so much harder to remove back-office personnel than many appreciate:
Wholly new risks are introduced by DeFi stemming from the reliance on open protocols and the fact that the underlying infrastructure is un-owned. Removing the back office and human oversight results in many efficiencies, but they also introduce risks. Thus, it is up to the end user or contract administrator to monitor the risk of the protocols themselves, and many would not want the burden. These risks are amplified when financial primitives collide with automated, hard-to-intervene contracts. Here is where all the chaos in DeFi is really from – systems that are built to be scalable and automated but that are underspecified or not understood by their creators. In sum, blockchain technologies bring many benefits. But the tools or processes used to disintermediate or gain efficiency also have costs in recourse, reversibility, risk management, etc. – the ‘paradox’ of DeFi.
In short, where there is a burden associated with processing information or familiarising yourself with an opaque, jargon-heavy new system that promises to be safe as houses and efficient but demands you to just trust it on that, there is and always will be a market for a trusted third party to do the due diligence for you.
And that, to a large degree, is why the Financial Times has a business model.