24 Oct Blockchain governance: what is it, what types are there and how does it work in practice?
You’ve found your way here so I’m guessing you’re already familiar with the blockchain concept and how it works. For the sake of this article, we will skip the part on explaining the technology. If you need some catching up, check our explanatory piece on the topic.
So, governance — what is it exactly?
Governance is a term that relates to decision making processes within an organization — who is responsible for what, how are major decisions being taken or implemented, is authority vested in a select few or distributed among all participants, etc.
All those questions have different implications on an organization’s operational system, whether on a macro level (e.g. a decision on a new feature or goal) or on a micro level (e.g. how much is spent on marketing or who is responsible for a certain task).
A successful governance model is usually what makes an organization click. Logically, there is no one-size-fits-all here — every model is situational and its viability largely depends on the specific circumstances. While a participative democracy could work for some, having powers vested in a central authority may be preferable for others.
Though as diverse as they may be, exemplary governance models are often centered around a set of qualities: transparency, integrity, effective performance and collaboration.
In the context of blockchain, the discussion on governance currently revolves around two sets of models: centralized vs. decentralized and on-chain vs. off-chain. The first duel is the classic blockchain paradox which calls into question contemporary authority structures. The second refers to human involvement and the extent to which decision making processes are automated.
Regardless of the governance structure, matters at stake typically involve network access, funding allocation, block size, reward systems, voting and decision making (e.g. reversing transactions via a fork).
Let’s dive in then and see how these are solved within the boundaries of the aforementioned governance models.
Centralized vs. Decentralized Governance
Blockchain’s main value proposition is decentralization — eliminating gatekeepers or middle-men in transactions to reduce costs and enabling parties to directly transact with each other, e.g. via smart contracts. On a large scale, this effectively means the removal of centers of authority such as governments and their institutions, central banks, etc.
Decentralized governance is therefore a natural response to the current ‘State above all’ world order that has been the go-to approach for centuries. In the age of information and personal empowerment, decentralization acts as a driver towards a future designed to vest authority in the hands of the people. For this reason it could even considered to be the digital extension of a representative democracy.
Decentralized networks are usually consensus driven and enable participants to take active part in decision making processes within a network. Just like in Bitcoin — whenever a transaction is queued, it requires the collective approval of inter-connected nodes to go through rather than that of a central operator or groups of such.
Similarly, DASH uses a consensus-based voting system to introduce changes to its protocol. Governance proposals within its network are submitted to stakeholders, who are in turn given a deadline to ‘announce’ their decision.
The DASH model is arguably more effective in delivering decisions due to its urgency detail, which incentivizes network participants to vote before a deadline rather than turning a blind eye and letting others do it instead.
Besides the case for direct transacting and bypassing middlemen or gatekeepers, a decentralized governance model has the obvious benefit of providing checks and balances on those with authority, hence increased transparency and accountability.
Transaction approval has to go through a number of stakeholders and vote calls are usually made available for the whole network to see. A central authority is unable to ‘cook the books’ or introduce self-interest changes behind closed doors as the books are distributed among the participants. This also eliminates the risk of ‘Single Point of Failure’ of centralized systems.
Moreover, decentralization can help improve efficiency and the rate at which decisions are taken on the macro level. In line with the DASH example, specific decisions can be directly submitted and thereby bypass traditional hierarchical structures native to centralized governance models.
The case against decentralization
Nevertheless, decentralized governance has its own issues. First, it calls into question the collective’s ability to carry out important decisions. If authority is distributed, one will naturally look at the distributed parties and argue against their credibility of adequately representing the collective.
Without a working, deadline-induced voting system, decentralized governance is bound to fail. Decision implementation on the Bitcoin mainnet may take months, even years, to complete in cases where the collective is simply unable to reach a consensus.
The benefit of a centralized governance model, assuming a responsible and knowledgeable authority is in charge, relates to the latter’s monopoly over decision making. Typically, a central authority will possess all relevant information to make informed decisions which serve the community’s best interests.
Furthermore, a single authority is usually more likely to possess the necessary expertise to ensure smooth operations and informed decisions. This is at least in an ideal scenario where the authority in question remains responsible towards its managing obligations. In such a complex environment like blockchain, centralized decision making will often be advocated for due to information asymmetries between central bodies and a network’s constituents. Given the fact that the central controller is probably the one who designed the network, it makes sense to vest authority in the creator.
Additionally, centralization could also have the benefit of faster decisions. As participation from others is reduced, a central body can execute faster without the need of subjecting every single decision to a combined vote. Having a pre-determined decision making process system will of course be an additional prerequisite to proper operations. The infamous DAO attack from 2016 with 3.6 million stolen Ethereum is a clear example of how slow decision making can have vast negative consequences on a network’s entire operability and functionality.
On-chain vs off-chain governance
On-chain blockchain governance, as you are probably guessing, happens… directly on the blockchain protocol. That is, any proposed changes or decision making processes must be embedded into the network code. This type of governance empowers the project developers (or node controllers) to be the main force behind company decisions and submit their vote on any new developments. Such a system does not usually follow a 1-node-1-vote principle, but rather depends on how many native coins every node is holding
On-chain governance is essentially a type of a decentralized governance models since it bears the feature of distributed decision making — nodes collectively vote and take decisions as opposed to having one central body in charge. Participation in voting processes is incentivized by on-chain rewards for the developers/nodes and leaves end users without a say when it comes to project management. Of course, the specific details on vote eligibility vary from one project to the other, but the main characteristic remains a code-based voting system for relevant stakeholders.
Besides being decentralized, on-chain governance bears the additional benefit of fast decision making, subject to a properly working design. As any proposed changes are imported on the project code and developers have incentives to vote, there is a sense of urgency to implement updates, especially in cases where the system is backed by a voting and implementation deadline.
Alternatively, on-chain governance may also result in low voter turn-out, specifically in the absence of an optimal reward structure for the participating nodes. In such cases the issue is magnified by the possibility of undemocratic voting — if some nodes choose not to vote, others with greater coin holdings will be able to determine the future course of the project. All in all, the system may potentially pave the way towards a pay-to-play decision making process whereby ‘richer’ nodes are able to eclipse their ‘poorer’ counterparty with sheer brute force.
The DAO hard fork of 2016, which we previously mentioned in the context of slow decision making, again makes a great example of how a dysfunctional voting system may result in a variety of issues leading to significant losses of funds.
Compared to its on-chain counterpart, off-chain governance is a lot more similar to traditional governance structures and easier to grasp. The degree of centralization is also significantly higher and the power of authority is vested in a select few, usually community leaders or stakeholders with a better bargaining position, for example large mining associations and core developers.
Off-chain systems of governance typically involve processes outside the blockchain or ‘code’ domain — only after having reached a decision, its implications are translated into on-chain action, such as for example a soft fork or investment allocation. Thus, this governance model makes use of pre-existing regulatory and compliance mechanisms to steer a blockchain network’s future. For example, a management board is faced with a decision to introduce a new project feature and subsequently carries out a vote (in the physical space) with a certain result. Said result, now documented manually, is imported onto the underlying blockchain infrastructure for execution — not much is done via technology here, simply the execution phase.
As you’re probably guessing, off-chain governance involves a higher degree of human involvement and little use of code for decision making purposes. In the context of blockchain, this can help reduce code-based and data import mistakes. With on-chain decisions one must be extremely careful not to write a faulty code and if they do, the network’s functionality is limited or non-working by design. Blockchain’s immutability feature therefore greatly increases the human error element, thereby making the off-chain governance model a lot more helpful for networks which tend to take less total number of decisions. Similarly, it could in theory provide a more ‘expert’ opinion from the party holding the most information.
However, off-chain governance is often tied to transparency issues. As decisions can be made behind closed doors and the network constituents are only able to see the output of those, one may logically question the validity behind decision making processes. Coupled with having a select few (usually a type of a board) in charge of steering the course of a project, you arrive at the initial point of concern that lead to blockchain’s development — stripping away central authorities and vesting power within the network and all its participants.
After having discussed the different governance models, we will now turn to their practical applications within existing blockchain networks. The examples given are some of the more prominent projects tied to governance-based discussions.
Who’s doing what? — Some more prominent examples of blockchain governance models in action
Bitcoin is the most prominent open-source blockchain network and most commonly used decentralized governance model example. Contrary to some opinions, the main decision makers within Bitcoin are the users and the miners, not the Bitcoin Core Development team or an active project lead such as Ethereum’s very own Vitalik Buterin.
It could very well be also argued that Bitcoin’s governance model is a mix between on- and off-chain governance, taking into account the role of community leaders outside the actual code.
The Bitcoin protocol is open and it does not have any special requirements for potential participants with the sole of exception of following the said protocol, most of which still follows the original one from Satoshi Nakamoto. Any ‘code-abiding’ citizen is therefore able to actively partake in Bitcoin’s development as long as they are able to supply extra computing power in return.
Those computing power suppliers (miners) are more or less behind Bitcoin’s governance, together with the ‘normal users’. Major network decisions are typically taken on a user peer review basis, followed by consensus of the most active miners — a type of a meritocracy if you like, where those with the largest input have a proportionate say in decision making. Under the moderation of the Bitcoin Core lead developer and other ‘higher end’ devs, the network votes on a Bitcoin Improvement Proposal (BIP). After the conclusion of the vote, the proposed changes are either implemented or repelled. For more info on Bitcoin governance, hard/soft forks, follow this link.
There is currently some debate about the level of centralization of Ethereum, the 2nd most popular cryptocurrency. Although it is often advertised as being decentralized, the network is still more or less dependent on the input of its founder, Vitalik Buterin, who is in charge of writing the Ethereum code’s major changes.
Nevertheless, Ethereum uses a similar governance structure to the one of Bitcoin where users can decide on network-wide software alterations by expressing their vote on Ethereum Improvement Proposals (EIPs). Having different types based on the decision in discussion, EIPs go through a number of stages (from Draft to Final/Deferred status) before implementation. All previous changes and current submission can be tracked in Ethereum’s open-source GitHub depository.
Tezos is famous for its on-chain governance model. The interesting thing about this project is that it has a cyclical voting system in place — one for each quarter of the year (so every three months).
Furthermore, every voting cycle has its own individual use-case, first one having to do with consideration, second one with actual voting and the final one being live update. Stage three (implementation or rejection) operates under a simple majority principle — the threshold is set at 80% and proposed changes that satisfy the requirement are introduced to the Tezos mainnet.
Tezos’ model is a fine example of on-chain governance. It also takes into account the exact number of votes needed to advance a change, which should be in the core of every voting system. The network additionally includes the possibility to vote on the voting system itself — a great and creative way to solve inefficiencies caused by dominant stakeholders, low voter turnout and others.
For more detail, follow the link.
At its core, DFINITY is one of the few networks that utilize an algorithmic governance model with its so-called Blockchain Nervous System (BNS). The BNS is comprised of …care to take a guess? Oh well, NEURONS! Each Neuron is essentially the network’s able-to-vote entity.
To submit an amendment proposal, Neurons must first choose a proposal category as DFINITY has categorized decisions based on their applicability — e.g. there are economic/policy/code categories. Every submission is additionally subject to a proposal fee which acts as a natural quality buffer against trivial or otherwise unimportant proposals without any real impact.
Individial Neurons may always submit proposals, but the fee requirement may naturally bar a huge part of the network from exercising any power over decision making. The bigger the fee submitted by each Neuron, the greater their voting power. This system is called into question, because it creates the reoccurring pay-to-play problem governance issue, whereby stakeholders of more resources are able to assert their stance on the sole basis of their network-native wealth.
The DFINITY governance structure furthermore includes an automatic voting system, enabling Neurons to replicate others’ votes. As helpful as this feature may be in cases where one Neuron is more knowledgeable about a certain topic (e.g. letting someone else vote on Policy since you don’t know much about it), its viability is questionable as it may create an environment of overreliance.
Decred is a network that has taken a rather atypical development path. They haven’t done an ICO and are working under a self-funding model, similar to the one of DASH. Currently under the supervision of a centralized holding vehicle, the Decred funds are meant to be decentralized in the future through the use of DAOs for full process automation.
Decred is one of the more well-known networks that utilize off-chain governance, at least for the time being. It is a project focused on privacy and features the Decred Assembly — a group of individuals who is in charge of decision making processes within the project. Assembly members are in turn elected by an Admission Council, subject to a 60% simple majority requiring vote which also applies for expulsion, though this one is carried out by a different entity — the Attrition Council.
Decred’s governance model strongly resembles that of a classic corporation with a Management Board of sorts. Major decisions are currently made by a select group in a real-life, non-code format with a designated future move towards a more decentralized system and the inclusion of other stakeholders’ involvement.
The EOS governance model is one of the most unique ones out there — it has its very own Constitution and a lead governing body — The EOSIO Core Arbitration Forum (ECAF). EOS governance features a 21-man strong executive body comprised of individual Block Propagators, who collectively make the most important decisions within the network.
The EOS functionality is premised upon the 21 core team having access to specialized hardware and a direct link between them, instead of having to sync in a whole entire network together to promulgate decisions.
EOS’ main governance structure, its Constitution reflecting the project’s roadmap and protocol, is subject to amendment by the 21 BPs. For a vote to pass, it requires a 15/21 majority for a continuous period of 30 days as well as user approval. In the case of accepted changes, EOS has introduced an implementation period of 7 days, after which every active node is expected to upgrade.
EOS is thus often comparable to a traditional government in the way its voting system is designed. Of course, the level of decentralization within EOS is greatly reduced when compared to other networks and its governance model has sparked some controversy among the crypto community. If you are looking to explore the topic on EOS governance in more detail, make sure to visit this more elaborate piece.
The future of blockchain governance
Balancing out the interests of all relevant stakeholders within blockchain networks is quite a difficult endeavor. As exemplified, there are various ways to design a blockchain governance model and its viability largely depends on a variety of factors, intrinsic to every individual projects. While some aim to denote the importance of informed, expert opinions (e.g. EOS), others focus on the community element and try to be more inclusive when it comes to stakeholder participation in governance.
The state-of-the-art of contemporary governance models will surely evolve in the future. Some projects are already trying to propose optimal alternatives such as cross-chain governance in order to satisfy the needs of the evolving crypto market and its community.
However, there is one thing that can be said with absolute certainty — blockchain has redefined the classic notion of governance and has shaken the very foundations upon which our state-first, ‘old’ world is premised upon.
The future will undoubtedly test the resilience of present day centralized institutions and assume a direction towards more personal empowerment and increased participation of all relevant stakeholders rather than that of select few in charge of decision making.
Even for entities more suitable to operate under a centralized governance models, checks and balances are pivotal in promoting transparency and accountability among decision makers. Creative hybrid structures of governance may prove to solve these issues, but it remains to be seen whether a space, characteristic for its tendency towards decentralization, will evolve in line with its inception.