One of the major breakthroughs that blockchain technology affords is decentralization. Whether it be a decentralized payment network, decentralized storage, or other — the benefits of decentralization can be applied across a variety of applications. But with this breakthrough comes a new challenge — how to effectively govern these applications while retaining a decentralized ethos. The easy answer is to just stay consistent with underlying function and govern in a decentralized manner. Unfortunately, this solution is not without its own risks and definitely not the end-all.

Following FTX’s collapse and the destructive chain of events that occurred in the industry largely as a result of mismanagement from irresponsible decision-makers, it brings to light the risks associated with centralized governance. However, centralized governance, or at least a degree of such, is still going to play an important and necessary role in the future development of Web3. This is not necessarily a bad thing if done right, and reasons why are explored in this article alongside an introduction to on-chain and off-chain governance models.

Introducing Blockchain Governance

It is first helpful to briefly define the scope of blockchain governance for purposes of this article as well as the core distinction between off-chain and on-chain governance structures.

When I say “blockchain governance,” I am referring to the governance of a blockchain-related project, whether that be a Layer 1, DeFi protocol, or other application. Some of these companies, like the publicly-traded Coinbase, are structured in a traditional manner, with a CEO, board of directors, etc. However, for this article I am focusing on projects or protocols that either have a decentralized underpinning and/or seek to be managed in a more decentralized way.

In this realm, governance is all about the processes guiding how decisions are made and who makes them. Blockchain projects sometimes comprise very large and diverse ecosystems with numerous stakeholders — e.g., core developers, node operators, speculators, users, service providers, application developers, and so on. The question becomes how to structure a process that effectively balances the varying interests of these parties while at the same time ensuring the success and longevity of a project.

Some processes are structured “off-chain” (i.e., off the blockchain network), where decisions are made on an informal social level before being enforced, and others occur “on-chain” where any rules for instituting changes are actually hardcoded into protocols. Most of us are less familiar with the latter, where smart contracts lay the foundational rules and autonomously execute collective decisions and votes from a community.

Framing the Governance Decision

Fully decentralized, on-chain governance has its place, and the rise of decentralized autonomous organizations (DAOs) partly proves that. There have been many successful implementations of DAOs and other forms of on-chain governance to date, such as that utilized in MakerDao and Aave. Certain advantages like being able to reach consensus on major decisions rapidly versus more traditional bureaucratic operations are immediately apparent. As are the benefits of taking centralized decision-making authority away from potentially corrupt actors. More on these benefits and also downsides to come later.

However, what about for complex and systematically-important base layer blockchains like Bitcoin and Ethereum? Could they similarly be governed and run in a decentralized fashion? They could, but it would seem almost incredulous to place the future of those protocols, upon which the entire foundation of Web3 currently lies, in the hands of the masses that most likely do not have the technical know-how required to effectively contribute.

This is part of the reason why the governance of Bitcoin and Ethereum primarily occurs via an “informal” process off-chain. Although these protocols are still obviously designed and operate in a decentralized manner, it is impractical for technical decisions such as those relating to Ethereum’s core implementations (e.g., transaction validity rules, etc.) to occur via processes like coin voting which might be susceptible to attacks from whales whose interests may or may not be aligned with other community members and users. The same can be extended to other protocols and projects, whose best interests and future direction might better lie with more informal / centralized governance structures.

An Illustration — Ethereum’s Governance Process

For an example of how an off-chain governance process can be successfully incorporated into decentralized protocols, look no further than Ethereum. This section is not meant to suggest that a similar process should be applied to other projects out there — rather, it serves as one (very important) illustration of how informal processes can be implemented in Web3. After all, they somehow managed to migrate the entire ETH network from proof-of-work to proof-of-stake via the recent ‘Merge’ without any hiccups or any CEO leading the way, which is an incredible feat.

Overview

To begin with, despite Ethereum’s off-chain governance process, do not forget that the network itself remains decentralized and permissionless — i.e., there are no set rules for who can or cannot build an application or send a transaction on the network. However, to make changes to the core protocol (which these other applications / transactions run on top of), there is a very coordinated process in place to ensure any changes are secure and widely supported by the broader community. This is an ‘informal’ process and occurs via social discussion — if a change is approved, it is then implemented in code.

Ethereum Improvement Proposals (EIPs)

An important process used in Ethereum governance is the proposal of what are known as Ethereum Improvement Proposals (EIPs). In simple terms, just think of these as written proposals from community members suggesting ways to improve how Ethereum is run. If you see a need for change or way to enhance the protocol, propose an EIP and potentially have it included in a future network upgrade.

The catch — unless you have a very strong understanding of blockchain tech it is highly unlikely that any EIP you propose will be seriously considered. There is a high technical barrier for submitting a well-formed EIP as well as a strict vetting process that requires peer review, gathering of community support, and presentation to protocol developers, all which are outlined in detail in EIP-1.

To give a high-level idea of what the process may entail:

Like anything, the best way to actually understand the Ethereum governance process is to participate — particularly when the structure is as fluid as it is here. You might also be wondering what some of the more noteworthy past EIPs were. I am sure you have heard of a few. For example, EIP-20 was introduced when the network was less than a year old and provided a standard upon which the launch of countless alternative tokens were based, and EIP-721 provided something similar for NFTs by advancing a standard interface for a young NFT market in 2018.

What is interesting is that despite Ethereum’s governance process occurring off-chain, it would still be exceedingly difficult for a select group of individuals to cause harm to the network through irresponsible decision-making. For example, everyone knows Vitalik Buterin and might believe he is in control, but he does not have nearly the amount of power to influence governance outcomes that many might assume. The governance structure limits Vitalik’s power, just as it limits everyone’s power.

On-Chain vs. Off-Chain Governance

Now that we have a good grasp of some of the basic considerations surrounding blockchain governance, let’s take a little closer look at the benefits and risks associated with on-chain and off-chain governance in particular.

On-Chain Governance

Benefits. Some of the benefits of on-chain governance were mentioned before. For one, it helps make sure a process is consistently followed which can increase coordination. This allows for quicker decision-making compared to more traditional bureaucratic operations which are not defined by the rule of code. It also importantly takes authority away from potentially irresponsible centralized actors — we are all well aware of the reality of this risk given recent events.

Drawbacks. Some of the drawbacks of on-chain governance arise from issues associated with coin voting, which has historically been one of the primary ways that decisions are enforced on-chain. Here, protocols generally issue governance tokens which can be purchased by others for a say in the future direction of a project. When a vote occurs, actions are taken only when the majority (or other prescribed portion) of token holders agree.

One issue is that coin voting might lead to misaligned incentives between token holders and other users. As noted before, some protocols and projects are made up of diverse constituencies that have many different values, visions, and goals. Coin voting, however, only gives power to one constituency (coin holders) who have a vested interest in over-emphasizing the goal of making the coin price go up even if that involves harmful side effects for other community members or for the future sustainability of the project as a whole. There are ways for projects to mitigate this risk, such as through allowing delegation of voting by minority interests to trusted community members, but the risk is something to be aware of.

Another issue concerns outright attacks through various forms of vote buying (see here for an example). If the number of coins voting dictates the outcome of decisions, what if a wealthy individual or group scoops up a controlling interest? This risk is particularly relevant for projects within DeFi, where application-layer smart contract systems often directly control external assets (and are thus governed on-chain). This smart contract control cannot be forked away, so if a DeFi protocol with user assets locked up is captured by an attacker, if the community were to fork and create a new version of the project they would subsequently lose all the coins and other assets that are stuck in the existing platform. Preventions against such attacks are thus imperative.

These drawbacks highlight the fact that like anything put directly into code, there is a chance that on-chain governance can be exploited and/or gamed to the extent there are any flaws in structure.

Off-Chain Governance

Benefits. The primary benefit of off-chain governance is that it allows for more flexibility in governance. Rules are not prescribed in code and decisions are not conducted by coin voting, which negates some of the main risks associated with on-chain governance. Recall Ethereum’s governance process as an example, which allows for fluid conversation and feedback when deciding on future upgrades. This less rigid structure decreases the chances of any gaming of the system, assuming there are sufficient checks in place on those ultimately making decisions.

Drawbacks. Of course, the last assumption noted above is a big one. If the informal decision-making process is not structured correctly, we once again are presented with potential incentive misalignment and also run the risk of corruption from those in charge. This risk even applies if well-intentioned with putting the most educated and responsible individuals in positions of power — e.g., intellectuals might care more about abstract philosophies and solutions than real day-to-day concerns like user experience which affects the masses.

Other Models / The Future

After reading this piece it should be clear that both on-chain and off-chain governance models have their benefits and risks, and that no one solution fits all. A primarily off-chain structure might be more appropriate for a Layer 1 like Ethereum, whereas a DeFi protocol will want to incorporate more on-chain aspects. Success in either case is determined by the details of how exactly each structure is set up, which encompasses an ever-evolving body of thought not limited to either/or.

For example, some have called for hybrid models, where on-chain voting is limited to only certain aspects of a protocol or only serve as one data point in the broader governance equation. Vitalik has advocated along these lines.

“The approach for blockchain governance that I advocate is ‘multifactorial consensus’, where different coordination flags and different mechanisms and groups are polled, and the ultimate decision depends on the collective result of all of these mechanisms together.” — Vitalik Buterin blog

Others support on-chain governance but want to move beyond coin voting as it exists in its present form. Some non-coin driven voting alternatives include proof of participation (e.g., POAPs) or proof of personhood (one vote per human) methods. The possibilities are really quite wide-ranging, and which methods will stand the test of time we still do not know. There is also an obvious place for off-chain governance models in the future despite recent events, so do not blindly assume DAOs and other forms of on-chain governance will rule the day entirely despite the buzz and calls for the downfall of centralization.