In this short series of two articles we are going to examine a disparity between the financial empowerment that DeFi promised on the one hand, and where the freedom to transact with decentralized protocols actually has led us so far on the other hand.
In Part I, we set the stage and take a brief look at the popular use case of passive yield: A totally passive user would like to put funds into a protocol, forget about it, come back a few months later and somehow get out with more than they put.
In Part II, we will look at the opposite extreme: active participation by trading in markets.
We will argue that for both types, the very passive and the very active use case, a more moderated approach (in between active and passive) could be beneficial to users and would furthermore be a good match with DeFi when built on top of the right market structures.
The hypothesis is this: the DeFi use cases that will find product-market-fit are novel and engaging experiences offering a chance to grow investments and exchange assets (be they virtual or tokenized real world assets) with the help of tools that are
Let us briefly appreciate that with Bitcoin, Satoshi gave us a solution to a problem that many (perhaps himself included) had never really struggled with. Yet, even among those with access to
some people were instantly captivated by the surprising possibility of owning and transferring scarce digital assets without a trusted intermediary. Visionaries themselves, some people have then pushed to enable another layer: trustless computation.
Highlighting Vitalik here, his writings make me believe that he found early inspiration in the idea of enabling Robin Hanson’s futarchy, and prediction markets in general, on chain. As tools that could improve decision-making, these markets are very naturally paired with censorship resistance, permissionlessness and pseudonymity.
We will get back to this very OG idea of prediction markets below. But more generally, the direction was of course this: let us do something with those digital assets, let us run programs on the blockchain. In fact, let us build up a new and alternative financial system from scratch, bottom-up and open-source, with new and better cryptographic tools. Oh, and code is law!
And so, DeFi was born.
We moved fast, attracted many billions, and yes, we have also broken things.
But has DeFi actually empowered people — beyond enriching some leveraging the hype around the more basic powers of cryptocurrencies providing a possible alternative to fiat of as a medium of exchange and store of value — beyond Bitcoin, stablecoins (and perhaps Monero)?
We need to build protocols for “normal people”. The successful degen is well capitalized power user with that builders want to like their products, but does not represent a group where we can expect massive onboarding towards “real adoption”.
“Normal” people are often passive. They have a busy life and do not understand the intricacies of finance. They are unable to write a custom bot to manage their concrete tasks. Typically, they do not like maths either. And finance maths is particularly hard! 1
Nobody minds getting rich quickly. And given such prospects, many are still ready to spend some of their time when it comes to their investments to work towards this goal. A few curious souls are even willing to try out new tech, especially if there is a chance of getting paid via an airdrop.
But by definition, becoming rich (like the successful degen) is a competitive sport and unrealistic for most (at least when the whole space stops expanding massively). Even the seemingly passive strategy of investing early into the right assets and then just holding means pretty much the opposite of being passive: Being early because of more than just luck requires work.
A far more realistic goal for most users is to make some returns without much work. Such products have been quite successful in terms of onboarding retail (with Terra’s Anchor Protocol a sad but prime example). Yield, higher than in TradFi. Passively! Just deposit here, and earn 20% APY on a stablecoin. Or 500% on these other coins.
The elephant in the room is risk: Their passivity hurts users even for these kind of seemingly passive gains.
Some protocols that promise high yield come with high and potentially hidden risks and tend to suddenly collapse. One has to somehow predict this very carefully and be ready to react. Being passive only seems to work until it doesn’t.
Another important example is liquidity provision in AMMs which comes with the hidden cost of withdrawing less of the good coin and more of the bad coin in the end, losing against the most basic “buy both and hodl” in the absence of extra payouts (like trading fees). Related ideas are at least theoretically and eventually likely to lose their depositors money against more active investors (once those who are making the right calls have arrived).
Needless to say, compensating such losses of passive users with “farming tokens” that need to be dumped on ignorant buyers is never going to work in the long run.
Staking yield is low risk and comes closest to a passive yield source for normal people. This is a big and growing industry indeed, with liquid staking and restaking and staking via exchange interfaces, and so on. But it is rather limited and exceedingly boring: all it does is not inflating stakers but only others in an asset that hopefully goes down in price less than the staking rewards. There is no steady-state solution here: Assuming that ultimately pretty much everyone is staked, staking rewards are equal to the inflation rate, and the market cap has to keep growing for stakers to profit - not a steady-state solution after all.
The problem is simply: There is no free lunch.
But TradFi is better at this. Ultimately, its offer is this: A myriad of advanced, sometimes really hard-to-manage services with various risk-and-reward profiles for everyone to express themselves with minimal work, maximal effect and lowest catastrophic risk. This includes things like insurances, pension funds, indices, bonds, structured products and so on. Users can even profit (maybe) from the power of law enforcement (in other words, the old kind of contracts might still be useful). The effect is increased capital efficiency and lower risk.
Instead, what a smart contract does is: none of these, pretty much. For example, it cannot process the latest health data and policy projections to reprice life insurances. Thus, in the end, very often in DeFi things are instead about reshuffling money from those who want to make it to insiders who were early. Or to those who wrote the bot. Or both, if it is a sniping bot or a MEV bot or an arb bot, but more on this in Part II.
One obvious way forward is to follow TradFi’s example and introduce more active management of the funds of a user. This could perhaps be accomplished by leveraging a combination of more powerful execution environments, solid oracle solutions, off-chain components, zero knowledge proofs, and trusting counterparties (that will likely become more regulated as time progresses). This path risks becoming so similar to TradFi (albeit with various tradeoffs) that it ends up being rebuilding a potentially worse version of it: more clunky, far more expensive overall, and likely still more risky.
Another way forward is to increase active participation. This can be achieved by focussing on new applications that match a user need — build the apps that users want under the constraint that they are a natural fit for DeFi.
Our hypothesis at Contro is that users have an interest in investing, and are willing to make fine-grained decisions about the nature of their investments as long as
| this is | this is not |
|---|---|
| easy | frustrating |
| interesting | stressful |
| fun | time-consuming |
| fair | extractive. |
Concretely, users who come back a few times a week for only a few minutes should be empowered to have a chance of financial success. A trick to make things interesting and meaningful without being too time-consuming is to make the product about something that already exists in the real world.
Consider the prediction markets mentioned in the introduction. A defining property is their strong connection to the real world by promising fixed payouts at expiration based on an external contingency in the real world. Furthermore, they can ideally be had on any topic, with users creating their own market, which can lead to a very social experience.
In the next post of this series, we will explore the problems with speed-focussed markets and will explain why so far prediction markets were not liquid enough to find massive adoption.
It took Einstein to understand Brownian motion for the first time in 1905, although admittedly, he managed it in the same year as publishing Special Relativity and this idea about light quanta that he won a Nobel Prize for. And yet, it is a good guess that even Einstein would have been too lazy to optimize his strats! ↩
