Personal DeFi Thesis
Earlier in my career, I've worked at a string of traditional fintechs because I believed that their goal of opening up access to financial tools, access that has in the past been curtailed by gatekeepers, was a noble and necessary one.
I have since come to view DeFi, when executed faithfully to the principles of decentralization, to be the true torch bearers of “democratization of finance” - the very same battle cry of Fintech 3.0.
So I’m exploring the DeFi space from the perspective of adapting traditional financial instruments to a decentralized ecosystem.
And I’m looking into DAOs and smart contracts from a governance perspective because I’ve always had a fascination with the US constitution and the ability to have started from a clean slate and architected a society’s mechanisms. I think when we iron out most of the kinks, DAOs will allow a group to experiment with various forms of organization and governance and perhaps find a more efficient way of managing and supporting human societies.
Another motivation is my inability to picture a future that is not a federated patchwork of IoT devices communicating peer to peer across multiple blockchains. There’s a reason we started, slowly and painfully, upgrading to IPv6 30 years ago. Four billion is not a big enough namespace. IPv4 to IPv6 is a migration at the protocol layer. Concomitantly, we will need a migration at the application layer - a need that will become increasingly clear as more nodes and identities are added to the overarching graph of human society.
When every car listens and talks to every other car, building and intersection, and every item has an embedded AirTag equivalent, and every article of clothing contains conductive threads holding the wearer's or brand's decentralized identifier, and every store shelf and door interacts with the network, and every item leaving a factory can be traced across the earth (and soon satellites and other non-earth objects) - no single company, no Skynet or FAANG can manage that with credible neutrality.
And the only way to federate and maintain fairness and incentives that lead to growth and not collapse has so far been through the various consensus algorithms at the heart of the different blockchains competing for our mindshare.
Donald Knuth stopped using email in 1984, before most started - it's easy to forget that every new tech had decades of trial and error before it figured out how to solve the problem of its domain space so effectively that we no longer notice it. Crypto is in a very public gestation period that reveals the stretch marks - but it will push and pull and pivot until the kinks are ironed out. It’s already forcing us to have very tough and worthwhile discussions and debates, as I’m sure the SEC will attest to.
Crypto also has demonstrably created wealth for early adopters. Marrying a new technology to such wild financial incentives is unprecedented. Imagine that when we discovered antibiotics, getting a PhD in molecular biology and pushing that science forth could somehow generate six orders of magnitude greater profit than a mainstream occupation. This is what's happening with crypto - a major human development that's in a positive feedback loop relative to wealth. Galileo had to risk his life and social standing to create progress in astronomy; Marie Curie spent 80% of her adult life in a lab exposed to lethal radiation to advance our understanding of radioactive materials; John Snow, the father of modern epidemiology, had to publish his theory at own expense and sugarcoat it so as to not antagonize fellow physicians. These are all negative feedback loops, as far as incentives go. Crypto is the direct opposite. If even a semblance of these incentives existed for genetics research in Mendel's time, today we would be master puppeteers of the entire genome and the world would be unrecognizably differently and, arguably, far more advanced.
I left my last full time position, at Canopy, in May of 2021. Since then, given the above thesis, I've been on a broad exploration of the many far-away corners of cryptoland. I imagine when Vasco de Gama set sail, or Alexander von Humboldt landed in the jungle of Oaxaca, they felt a similar exultation.
In 2017 I had supported a project called Urbit by purchasing two "stars", an identifier within the Urbit ecosystem that, broadly speaking, aims to rebuild the Internet from the protocol layer up. In the summer of 2021 I learned that Urbit had migrated its stars onto the Ethereum blockchain and essentially turned them into NFTs (Urbit had been using the concept of a non fungible token at least since 2017, before the ERC standard). I sold one of the stars for a quite ridiculous profit on OpenSea, which was a watershed moment. It's one thing to talk about cool Ethereum standards - it's another thing to see a five digit sum materialize in your bank account.
During that summer a few recruiters reached out about web3 positions. I didn't want to work at another company but those communications exposed me to Helium and Algorand, which I was unaware of at the time. I ended up pre-ordering a Bobcat miner to participate in the Helium mesh network, which to this day I believe is one of the best examples of a blockchain providing utility in the physical world, in a way that a centralized organization would not be able to.
In the fall, I started diving deeper into DEXs and the novel trading primitives some of them introduced. Being based in NY, however, is a double whammy given that many exchanges do not operate in the US, and of those that do, many do not operate in NY state. I wiggled through and around the restrictions and traded every instrument just enough to conclude that I'm more interested in the algorithmic design of the exchange, automated market makers, and innovations still being developed, such as Kwenta's peer-to-contract model that allows for zero slippage and infinite liquidity. It also became painfully obvious that consistent wins in the market require tremendous presence of mind, detachment, and discipline - profound qualities the development of which takes time. I did not intend to set everything aside to focus on this so I stepped away from trading until I'm in a place to commit to that path.
The conceptual foundation behind speculation left me intrigued, however, and I went down a rabbit hole to understand its origins. That led to Louis Bachelier's theory of speculation, the quantitative revolution set off by James Simons, and a number of other directions all of which converged upon a bottomless pit of mathematical analysis. I'm coming to the conclusion that this is an asymptotic game, wherein we can employ tremendously complex models but inexorably bump up against the fundamental limit to the predictive power of technical analysis. The market is not rational, the retail or even institutional investor is not rational, privileged information consistently invalidates perfect information assumptions, etc, ad infinitum.
During the course of learning about speculation I came across Austrian economics and delved deeper into that. This is probably the single most eye-opening set of ideas I've learned in the 10 years since college, rivaled only by the Buddhist meditation and kriya yoga I started doing right after graduation and that has permanently changed my perspective, and life. In the past I had been tangentially familiar with Keynesian economics and never much questioned it; quantitative easing/tightening I viewed as a hallowed set of tools in the armory of our great steward, the Federal Reserve; cheap capital was clearly the secret sauce behind America's exceptionalism as it exemplified the government's willingness, nay, imperative, to stimulate innovation and value creation by providing funding that America's entrepreneurs and visionaries require. There was one snag in my sanguine view of economics - Bretton Woods. I was familiar with the outcome and monumental decisions that were made, but something felt off - I felt like I was missing a part of the story, an invisible part to be read between the lines. Chasing that down led to Austrian economics. I felt like a blind man receiving the gift of sight because for the first time the mechanisms involved made sense, could be understood and explained (it passed the mom test) empirically, and did not require parsing through the sophistry that so many bureaucratic disciplines inherent to modern government hide their inefficiency and frankly, disutility, behind. Understanding the stock to flow ratio allowed me to finally understand why gold has been a currency among numerous societies for millennia. Understanding sound money naturally leads to an understanding of unsound money, which in turn illuminates the root cause of endemic inflation and cycles of recession that have characterized the 20th and 21st centuries. Replacing the free markets with a central bank reflects a level of hubris not found since Icarus thought he was a bird, for two reasons.
In a free market economy prices communicate information, and are crucial to determining how to allocate resources. Prices are an emergent property of free markets, an epiphenomenon that distills the infinite complexity inherent to 8 billion actors making thousands of decisions per day (not to mention the sheer entropy of atmospheric, tectonic, and orbital activity), into one variable - the price. A central bank with the power to exert price controls (adulteration of labor market) and modulate the interest rate (adulteration of capital markets) must indeed embody the hubris of Apollo if it believes that an office building of bureaucrats can accumulate enough information to make decisions about where goods and capital should be allocated, and at a better clip than the free market would have.
Secondly, granting the central bank the power to increase the monetary supply willfully and woefully ignores millennia of historical precedents which show beyond the shadow of a doubt the perils of devaluing a society's currency:
When Nero started clipping the aureus, he set into motion the slow but sure decline of Rome. Ironically, Caesar's creation of the aureus 75 years prior is what enabled the empire to survive the transition from a republic and even thrive.
Irony is a bitch, and while, after splitting off, the Byzantine Empire survived and flourished precisely because Constantine committed to and set a precedent for maintaining the solidus coin's gold content at 4.5 grams, Byzantium ultimately succumbed to the temptation of currency debasement, receiving the coup de grace from Mehmet II in 1453.
The Weimar Republic's ill-advised albeit not entirely unjustified über-printing of the Deutschmark led to an infamously disastrous cascade of events.
To continue with the theme of irony, as soon as I had emerged out of the Austrian economics rabbit hole, the Fed raised interest rates for the first, but far from last, time this year - in March. I've passively watched the Fed raise or lower rates since Bernanke's tenure and this was the first time that I clearly saw, and saw through, both the mechanisms and the tragicomedy of monetary policy.
On a related note, I read the white paper and started a deep dive on Luna a week before its depeg. That was fascinating to witness, like watching two celestial bodies collide light years away and thus locked into history, powerless but to observe. My personal take on it is that
the proximate cause, as many have said, can certainly be attributed to undue centralization of the governing body, which, because of its cabalistic nature, was free to make decisions misaligned with investors' interest, most egregious one being the encouragement of institution-scale deposits in Anchor. The influx of idle capital skewed the balance towards saving, which in turn reduced the protocol's ability to rebalance in times of peg deviation. Ultimately, Do Kwon yearned for stablecoin dominance, with particular vitriol aimed at DAI. He made the choice to reach that goal by any means necessary, and those means turned out to be large institutional investments. The downside went from increased systemic risk to death spiral in a matter of months.
the ultimate cause is choosing the wrong model for the algorithm to emulate. The Terra protocol as outlined in the whitepaper did a tremendous job of recreating the functions of the US Treasury and Federal Reserve. The algorithm was equipped with levers that can respond to multiple stressors and kick off feedback mechanisms that, via the auxiliary Luna, can optimize Terra for the 1.00 USD mark. The problem is that the most efficient system for pricing in all factors and allocating capital is the free market, not the central bank. If anything, the swift rise and swifter demise of Terra may, in retrospect, serve as a prophecy of events to unfold in the actual US economy, and eventually the world. Terra was a faithful facsimile of the system that has steered the ship since 1913, and its regrettable outcome should not be assumed to be aberrant. In fact, and this may be my looniest take, Terra shouldn't have been tracking USD in the first place. An unsound currency being bled dry by the Fed's trigger-happy printing on a death spiral of its own that happens to play out over decades instead of weeks, like the frog contentedly boiling to death, is not an asset to peg to. Gold would have been a better option, or Bitcoin, which is an arguably sounder form of money.
Most recently, I've been exploring the implications of Soulbound Tokens and Decentralized Society. There are a number of fascinating applications that come to mind once you consider the possibility of a native web3 identity and trust layer. The one I'm most interested in is undercollateralized lending. A functional implementation of basic loan primitives would go a long way towards bringing DeFi to feature parity with TradFi. An MVP that focuses on building a system of correct incentives and constraints, and that issues microloans with a very short expiry date, would be a relatively low risk way to test this concept. I'm still in the early stages but find this intersection of web3 utilitarianism and human psychology (the creative ways people will try to game the system are limitless) particularly close to my heart. ∎