Progressive Decentralization (Failing)
Growth strategies most of the time are just failed decentralization attempts.
This is the first article in a series on Progressive De/Centralization. To read more, click here.
In 2020, Jesse Walden published Progressive Decentralization: A Playbook for Building Crypto Applications. This article, alongside many others, extends the concept of DAOs from a niche crypto-industrial concept to a market of thousands of builders who are not in the tranches of crypto governance.
Walden's playbook was well-received and influential. It divides the development process into three phases:
Early product market fit (PMF), or pre-PMF.
Community engagement through governance activities.
Token distribution as regulatory shielding.
In early 2020, when the guide was written, the crypto space was very different. TVL of the entire industry was $500m, which is the same amount of TVL of Fantom today (DefiLlama, 2023). Governance alpha was just launched. Liquidity mining was not yet a concept. Retroactive airdrop was not an expectation from users.
However, since the playbook was published in 2020, the industry has evolved and the playbook's effectiveness has come into question. Early PMF metrics are inflated by farmers expecting future airdrops. Governance activities have not demonstrated a convincing ability to scale to more contributors. The SEC is seeing through the facade and suing DAOs with large token holder counts. Unfortunately, progressive decentralization has also been used by many grifters and scammers as an excuse to rug projects.
Clouded PMF Metrics
“Show me the incentive, I'll show you the outcome.” - Charlie Munger.
Most protocols, on a high level, adopt the growth strategy of “use my product and I will give you my tokens.” Based on the frequency of token reward distribution, two kinds of incentive schemes emerged - liquidity mining (block-by-block basis) and airdrop (retrospective distribution).
While the initial aim of this strategy tries to distribute ownership of the protocol and give users a voice in the decision-making process, the token’s liquid nature inadvertently shifted the strategy to “use my product and I will give you money.”
Liquidity mining has been criticized for introducing reflexivity in token prices and conflating reward APR with actual organic usage (Eva Wu, 2021; Nansen, 2021 ). This issue is most prominent in new ecosystems and protocols. People come for the high-reward APRs and leave when rewards die out.
Trader Joe, a decentralized exchange on Avalanche, had this exact problem. It had a daily unique active wallet (DUAW) count of more than 60k and $2.4 billion TVL in December 2021 when its token price is $3. Currently (March 2023), its DUAW is at 2k (-97%), TVL at $100 million (-96%) and its token price is $0.2 (DefiLlama, DappRadar). It was common that individuals and big crypto funds would come to the platform, contribute capital, earn and sell the rewards token, $JOE. Alameda Capital famously said to the Trader Joe's team, “GG for surviving us farming you for months” (Cryptofish, 2022).
Airdrop became popular shortly after teams realized the shortcomings of liquidity mining. Airdrop distribute tokens to past users based on their contributions to the protocol. However, once airdrop became popularized and common knowledge in the mind of the users, it suffered the same fate as liquidity mining. Users would register thousands or millions of accounts to interact with a protocol, expecting future rewards - a practice called “airdrop farming.”
One famous example is the Ribbon Finance airdrop. Ribbon Finance is a premier option protocol on Ethereum, backed by prominent VCs. Before its airdrop, one of the VCs used its .ens address to create numerous proxy wallets and deposited money into Ribbon Finance right before the airdrop snapshot. They later sold all the reward tokens and netted more than 600 $ETH, around USD$2m at the time (zEniTH Owl, 2021). It caused a public outcry and resulted in the dissolution (from the public) of the said VC. It also raised questions about the anti-Sybil nature of different airdrop strategies. A recent analysis by Hop’s community members demonstrated more than 33% of its airdrop were distributed to airdrop farmers (Hop, 2022).
The current token distribution strategy clouds PMF metrics and makes the process of product development more confusing for developers.
Governance activity does not scale
One of the visions of distributed ownership is to enable community participation and development in public. However, human coordination is hard.
However, the current governance structure based on voting has severe limitations in achieving this vision.
The current governance structure, which is based on voting, requires most product updates and experimentations to be proposed in the public governance form, discussed in public, and voted in public. This form of governance is common in almost all protocols (Uniswap, Gitcoin, Aave, MakerDAO). Although certain groups within each protocol can make dynamic changes without a full on-chain voting process, most of the actions happen through the full governance process.
This process does not scale as the protocol grew larger and more contributors join. My conclusion came from working within and with DAOs and talking to numerous core contributors for different protocols. In the current governance form, as the protocol grew larger in terms of revenue and mindshare, more contributors would come and propose ideas. These ideas span from partnerships to new product offerings. While not all are adopted, some nonetheless pass governance and become part of the protocol. However, very rarely are proposals checked retrospectively. This causes the protocol to keep on growing and wasted efforts become hard to cut off. No one wants to be the person to tell their friend that their product is not working.
Moreover, for any contributor to propose a meaningful change or product experimentation in a given protocol, the contributor must garner support from a vast voter base. Whipping votes is not something most people are comfortable or good at. This overhead causes fatigue and slowly drives contributors away from the protocol.
One example is Index Cooperative (Index Coop). It used to be one of the most active and structured DAO before talents left due to internal disagreement and governance overhead. Index Coop offers a suite of index products on Ethereum. At its peak, it had more than USD$500m across its indexes.
Started by two “parent” organizations, DeFi Pulse and Set Protocol, Index Coop quickly grew to be more independent and had its own full-time contributors base (IndexCoop, 2021). To organize its contributor base, Index Coop sets up its governance structure. The governance structure is separated into a few components.
On the top level, there’s the Index Coop Council (ICC). At its core, the ICC is just a delegate proxy address. However, the ICC not only has immense social power but also holds immense voting power backed by Set Protocol and DeFi Pulse.
One level below the ICC is Nests. Nests are specific working groups focusing on different verticals. Index Coop had a Finance Nest, a Product Nest, a Governance Nest, and a Community Nest. Each of these Nests has a level of autonomy to spend and hire for their given goals.
Because it is a DAO, the concept of “everyone can freely contribute” makes the organization hard to grow larger. People started having different ideas about the direction of the DAO. To convince others of their vision, people formed internal groups and alliances. However, to the average contributor, this would be very distracting since office politics are always tiring.
After realizing the inefficiencies of the public governance process and the overhead it posed to contributors, Index Coop is currently going through the process of reshaping their governance structure.
Index Coop is not alone, however. While Index Coop tends to discuss their disagreements more publicly via governance forums, other prominent DAOs either handle these conflicts privately or informally centralize control. Nonetheless, all these findings point to the fact that the current governance format does not scale to fulfill its purpose.
Decentralization as a regulatory shielding
While there is debate over whether all tokens should be treated as securities, most tokens are indistinguishable from equities in practice. In the past, people have argued that tokens are not equity as long as it has a large token holder base (as a proxy for “sufficiently decentralized”). However, in the recent Coinbase insider trading scandal, SEC classified nine cryptocurrencies as securities even tho there are thousands of token holders for some of the tokens (SEC, 2022).
Within the case, the SEC argued that these tokens’ value is dependent on efforts from the founders or core team instead of a decentralized community. They cited reasons including how the core teams held a large amount of voting power and utilized this voting power to pass the majority of the proposals while voter-turn out is low.
The assumption that a large token holder base constitutes a sufficiently decentralized network is no longer tenable. A large token holder base does not constitute a sufficiently decentralized network. Therefore the strategy of distributing (liquidity mining or airdropping) tokens to many different addresses might not provide enough shielding in the current regulatory environment.
This could significantly push back the timing of token issuance because protocols are generally centralized in the early days due to small team sizes and the lack of a broad contributor base.
Coming next…
Our current approach towards progressive decentralization falls short of what is required. In the next article, I will go back to square one and re-examine what decentralization means in the dApp layer. Read it here.