DeFi: The Road (Back) to Censorship Resistance

As the year draws to a close, the term “decentralized finance” might evoke visions of a collapsed casino that transferred generational wealth to North Korea.

North Korean hackers are understood to have stolen $1.2 billion in crypto since 2017, mostly from DeFi. (Decrypt)

If we look past the various protocol exploits and illusory-value applications, however, decentralized finance has delivered primitives that hold the promise of a radically-improved financial system, based on the following qualities:

1. efficiency
2. transparency
3. resilience
4. inclusion
5. self-sovereignty via censorship resistance

Surveying the landscape today, the space appears to have executed successful proofs-of-concept on the first four dimensions. Using protocols like Aave, Enzyme and Uniswap, anyone with a crypto-wallet can — respectively — save, invest and trade instantaneously with no custodial intermediaries. Activity is governed and recorded transparently on the blockchain, and blue-chip protocols have shown more resilience than many centralized counterparts in adverse conditions.

On Aave, users can borrow against collateral and earn interest on their deposits. Essentially – it is an autonomous, borderless and permissionless bank.

On the final dimension, however, the movement is falling short: resistance to censorship, the pillar of self-sovereignty, depends not only on the success of the application layer, but also on the healthy construction of the asset layer.

Today, if we consider Ethereum, the most relevant chain for decentralized finance, key assets have censorable “backends”:

1. USDC, USDT, BUSD and other fiat-collateralized stablecoins managed by entities that can freeze users’ on-chain balances.
2. WBTC, whose underlying Bitcoin reserve is held by U.S. custodian Bitgo.
3. DAI, a synthetic USD issued permissionlessly against a base of collateral consisting in large part of fiat-collateralized stablecoins.

Ethereum’s stablecoin majors are all directly or indirectly dependent on censorable assets. (Courtesy of CoinGecko)

While these monetary assets certainly bring value to the on-chain economy, they also undermine the sovereignty of Ethereum by competing with its native asset. To understand the importance of monetary premium for any blockchain, listen to this podcast by EF researcher Justin Drake or read my previous article. But for simplicity, consider the following scenarios using Ethereum for illustrative purposes:

1. The mass adoption of Ethereum happens, but transactions are mostly denominated in stablecoins like USDC. This undermines ether’s adoption as a monetary asset, leaving the chain without sufficient “economic security” to resist a state-level attack based on stake acquisition.
2. With Metamask tracking user IP addresses and large share of economic activity denominated in regulated stablecoins, balances can be seized “at whim” by the authorities.
3. Ethereum is undergoing a contentious hard fork, and stablecoin issuers must choose where their coins are redeemable. As stablecoins play a significant role in the on-chain economy, the issuers’ decision — or those of the applicable regulator(s) — will have a decisive impact, circumscribing the sovereignty of the chain.
4. The U.S. government decides that DAI, a permissionless and unregulated representation of the U.S. dollar, should not exist. It asks actors like Circle to freeze any balances in Maker CDPs. DAI depegs precipitously and an (un)healthy dose of on-chain chaos ensues.

These scenarios are not likely, but neither are they outlandish. All defeat foundational purposes of cryptocurrency, namely financial self-sovereignty and the separation of money and state. If a public blockchain must be “sovereign” to realize this vision, its native currency must succeed as a monetary asset. And today, rather than taking this vision forward, segments of “decentralized” finance risk acting as a Trojan horse for U.S. financial dominance through entities like Circle.

This does not mean that the current suite of applications lacks utility. On the contrary, anyone with a cryptocurrency wallet can now access a full range of financial services, and this is monumental. Whether it is “decentralized”, however, is a more complicated matter. If it is not decentralized, it is simply better categorized as a “fintech” innovation, not a monetary, political or moral one.

In a certain sense, this highlights the value of Bitcoin. Its focus on being a monetary asset, rather than a global computer, makes it insusceptible to “capture” by decentralized finance. However, smart-contract blockchain ecosystems can mitigate the risk of capture with a few strategic efforts:

1. The ecosystem should unite around a stablecoin collateralized solely by the native asset, de-prioritizing those relying on centralized intermediaries. This not only removes the censorship vectors, but also promotes the usage of the native currency as collateral.*
2. The ecosystem should manage upgrades without hard forks via on-chain governance. If there is ever ambiguity on the canonical fork, stablecoin issuers are effectively in control. This attack vector is mitigated if an on-chain political system exists to offer a clear framework on what constitutes a legitimate evolution.
3. Linking the points above, the on-chain political system can be used to subsidize liquidity on the native asset and ecosystem stablecoin, prioritizing the “national” currencies. Read about liquidity baking on Tezos as a related example.
4. Ecosystem applications should advance the native asset and/or ecosystem stablecoin as the main transactional currency/ies.

RAI is an example of an Ethereum stablecoin backed solely by ETH.

And in general, the cryptocurrency industry might benefit from better terminology — for instance, “open finance” versus “self-sovereign finance”.

Naturally, the above steps require a cohesive socio-political layer capable of collective action. They are not intended as a conclusive manual, but as a tentative mental model to (re)build better.

Thank you for reading!

_________________________________________________________________

* Discussing the right design of algorithmic stablecoin is beyond the scope of this article, but — in a nutshell — it should be over-collateralized, have self-adjusting interest rates that can turn negative and depend on an oracle system with minimal censorship vectors.

Not investment advice. Interested in discussing these ideas? Reach out to helvantine@pm.me.