Decentralisation is a simple enough concept to define, but a very complex one to understand. I daresay it’s misunderstood to the point that quite a lot of crypto enthusiasts think that centralisation is evil, when in reality, it’s just a different mode of governance. Indeed, decentralisation can be vastly inefficient when compared to centralisation; with the lack of a hierarchical structure, one tends to suffer from an inverse tragedy of commons phenomenon. Why should I execute a particular task, if at the end of the day, I’m still going to be rewarded for it? Resources are therefore not necessarily equitably distributed, even though this should have been one of the points to be resolved by DAOs. This, in turn, leads to the conundrum of incentivisation within a DAO.
Payment is the most known form of incentivisation, although not necessarily the best one. In this empirical opinion, I will be exploring the various incentives that one can have in place to promote activity and participation in a DAO. Activity and participation are two fundamental ingredients of governance, and both of these ingredients require incentivisation.
First of all (and, I suspect, much to the weeping and gnashing of teeth), you do not need to have a token to represent, much less incentivise, governance. This does not mean that tokenising governance rights is not effective; on the contrary, it is a model that works well for scaling DAOs that are viably oriented around the right projects. However, a lot of projects out there adopted the concept of a governance token simply to elegantly resurrect the manic 2017-18 era of ICOs, and raise funds under the pretext of doing so in order to “distribute value to the community”. But that’s a topic for some other time; what I want to lay out clearly is that governance does not require a token, and that governance requires active participation, which in turn requires incentivisation. Indeed, at times in this article I will inevitably intertwine governance and participation, as the latter is a fundamental sub-set of the former. What comes next is a classification of the different modes of incentivisation that I have seen as prevalent in the space and dissect each and every one of them in turn, hopefully with the end result being that of settling on the best way to incentivise long-term participation in DAO governance. Do note that when using the term “incentivisation”, I am not necessarily referring to financial incentivisation, and that my views are mostly non-economical.
Incentivisation through financial governance tokens - FIG
Governance tokens were one of the catalysts behind the recent bull market, with such tokens being especially popular in most DeFi projects. Initially, waters were tested with the granting of pure voting rights to token holders via early governance systems such as Compound’s Governor Alpha, quickly evolving to include revenue sharing mechanisms that emulate, to a certain extent, the equity model (sans certain key elements, notably the ownership rights granted to equity holders). The recipe here is simple: users contribute either money, or time/effort, in return for the right to participate in the governance of a particular protocol. By contributing either/or, they would be given governance rights of varying degrees depending on the project in question, with such rights represented in a token that has market value. Therefore, for the sake of simplicity, we’ll call this financially incentivised governance via token representation (‘FIG’).
FIG works very well in bootstrapping a community, albeit a potentially mercenary one. We know the drill – create a product, drum up some hype (more in this point later, as it merits a separate section), launch your token, attract liquidity through unsustainable returns, and concurrently bask in the glow of attention by having a lot of interest in terms of governance… until the glow fades off, and all’s that left is a few can-kickers on Discord vehemently insisting that the project’s token is undervalued and underappreciated.
In short, FIG on its own via the distribution of a token in a widespread and unclassified manner is unsustainable and, frankly speaking, undesirable. I wouldn’t want a pack of rabid degens that ape in on anything that remotely smells of money, telling a potentially talented group of developers what they want, when, and how, just for the sake of ensuring they can get paid on the pump and leave on the dump. Neither do I want to have the evident systematic risk of a governance collapse due to a massive migration of mercenary would-be voters and which could result in a halt of the project due to, say, not enough interest left to even meet the quorum for proposals to be passed. Therefore, FIG, on its own, is not sufficient to incentivise long-term DAO governance participation.
Structured FIG (SFIG)
Let’s try and upgrade FIG by introducing sustainability’s shy sister: structure. Structure effectively underpins the corporate world, and at the end, structuring a thing involves a delicate act of balancing between not structuring something well enough and it falls apart, and structuring something to the extent where it’s excessively red-taped, and therefore suffocates and dies. So together, let’s structure FIG in a way which neither results in a cold soup and Goldilocks gets sick, but nor does it result in a soup that scalds her.
By structuring FIG in a way that incentivises, and selectively rewards, those who actually contribute towards the betterment of the project in question, one can both lessen the efficiency loss of distributing a token to a user who will either not participate or sell the token immediately, as well as ensuring that the active participants have a higher chance of being genuinely interested in exercising their governance rights due to their having actively contributed to the project prior to the distribution of the token.
It is interesting to note that selective rewarding of participants can take place without a token. I’ve seen this happening in Concave, where thousands of users competed to obtain Discord roles in the hope of being whitelisted for the eventual token sale. Yes, a token features in this example, but much to my surprise, users started settling in their assigned roles without the token being in place for months, and executed their roles to a greater or lesser extent. Roles were also removed from those who underperformed or went AWOL. It was an interesting experiment to watch, and one which reinforced my belief that incentivisation can take place with non-monetary means.
However, structure is quite difficult to lay down without centralised input; indeed, there was an evident centralised coordinative effort from the people behind Concave, so as to assign roles, strip them from people who underperformed, etc. Therefore, inevitably, the major issue here is that it will be at odds with the concept of a proper DAO. However, centralisation and decentralisation represent a range at opposite ends of a ruler. In order to push the slider towards the decentralised end, a DAO can be split into siloed sub-DAOs or groups that are headed by one or more persons; each and every sub-DAO is responsible for its own subject area, and the participants in each sub-DAO are either remunerated at a predetermined flat token distribution, or remunerated on a project-basis, with the latter normally subject to voting within the sub-DAO itself via tools such as Coordinape.
Another issue with the SFIG model, apart from that of centralisation, is lack of scalability unless the sub-DAO model is adopted. However, one needs to be careful with the sub-DAO model, as one wants to avoid the issue of empty siloes, i.e. sub-DAOs which do not perform as well as others. Therefore, SFIG is likely to be viable mostly for small or mid-scale projects. There is also a shared issue with FIG, albeit on a smaller scale – the issue of mercenary participants.
Termed Incentivisation through a Locked Subscription Model (LSM)
So, let’s turn our attention to modes of incentivisation that are not necessarily profit-driven, even though financially based. Typically, if you want someone to do work for a predetermined amount of time, you ‘lock’ them in through a definite contract. This principle applies to the veToken model ushered in by Curve, where one needs to stake the governance token for a determinate period of time in order to be granted governance rights. In theory, this signals the intent to commit one’s fidelity to the project for that period of time, including the participation within governance matters.
It can also be described as a subscription model, hence the title given to this mode of governance. It works very well in conjunction with a variable quorum model that changes depending on the amount of locked tokens, and incentivises participation since it would be in one’s own interest to ensure that, by the end of the lock-up period, the project has developed to an extent where the unlocked tokens are now more valuable in price. In other words, it incentivises added-value participation, rather than participation for the sake of it.
I am quite a fan of the LSM, but there is the problem of excluding those that would want to participate in certain governance matters, without wanting to render illiquid their governance token in a market that is known for its volatility. Plus, the LSM does not really tackle, at heart, the issue in most DAOs out there – how does one spark and maintain fervour?
Incentivisation via Hype & Anticipation (H&A)
In a world dominated by instant gratification, holding off that sense of gratification can work immensely in one’s favour. There were various multiple hyped launches in 2021, with two of the most memorable ones being TempleDAO and Solidly. Let us pick the TempleDAO example for the sake of expanding on this incentivisation model. For those who are not familiar with their project launch, it’s worth taking a look at how they structured the so-called Fire Ritual and Opening Ceremony for a masterclass on how to drum up hype and expectation on a project, while building a community. If you want a ‘tl;dr’ – they set a series of tasks to complete for those who wanted to participate in the token sale, some of which were gamified within Voxels, and some of which were semi-sadistic tasks set by appointed community members. It was a mammoth task, but one which created a sense of hype and anticipation that made people want to be part of the project. Hawk-eyed moderators and core team member spotted those which were more vocal than others in their support and active participation, and brought them on board as Disciples and Initiates.
The sense of hype and anticipation (H&A) is a neat trick to use to make someone long for something strongly enough that they would do pretty much anything to get it, with the presumption that once they get it, they will hold on to it for dear life (see what I did there). In other words, make them long for it, then earn it. This serves as a filtering process to select those that show their intent on being part of the project, as well as testing their resolve and, in TempleDAO’s case, logical & puzzle-solving skills.
As you might expect, I will now present the issue here as well, and it’s one that’s very much linked to human nature. Gratification does not necessarily result in gratitude, and in most cases, gratification is short-lived. Once we get something that we want, we quickly stop appreciating it. Therefore, incentivisation by making a participant earn their rights is a great way to both vet that person’s will to participate, as well as to do away with pure mercenaries, but it cannot stop there. Once the fish is hooked, it needs to be reeled in.
Cult-Oriented Governance (COG)
H&A can further honed to create a community that can almost be described as cult-like (and saying this without any sense of denigration). Cults are normally autocratic in nature, so inevitably, once again there is the problem of centralisation here. However, before going into the problems, it is interesting to explore an example of one of the strongest crypto communities that we’ve seen last year, and some ingredients that led to the successful creation of a cult-like environment.
OlympusDAO is one of the best examples that comes to mind – the poster child of so-called “DeFi 2.0”. Bonding, insane APYs, a massive treasury, and a community can best be described as one of the most rabid ones I’ve seen. The community rallied around a Greek god and his disciples, coining terms such as OHMies to identify themselves, praising the novel tokenOHMics laid down by the Greek god, and generally making themselves very well-known across all four corners of Crypto Twitter & Reddit. The ensuing fork-storm that we saw was remarkable, with one fork even managing to create its own community and story worthy of an Oscar-movie (looking at you, Wonderland).
Roughly a year on, the community is definitely not dead with over 85,000 OHMies on Discord, and although Zeus is not as active as in the past, he is still seen as a leading figure around which the masses rally. It may be likely that most OHMies would fail to describe the economic model underpinning OlympusDAO sufficiently well, and probably have a lack of understanding of certain concepts such as the long-term sustainability of runways. However, they will vociferously defend their leader, and more importantly, they actively participate in governance, with OIP-93 recently attracting more than 1 million OHM.
Cults are powerful, and incentivise participation like nothing else out there. However, cults are often made up of individuals that do not always think rationally and independently, and have a major weak point – heavy centralisation in a figure that may just as well unravel the whole project if cards are played wrongly.
From cult to culture (and idealism) – Idealistically Motivated Governance (IMG)
If a cult manages to gather sufficient network effect and community participation that it outgrows its cult leader, then one can no longer talk about a cult in the proper sense. This takes us to the last system of governance, and one which so far is rare in the world of DAOs – governance revolving around idealism, which is based on a belief that may or may not be achievable.
The best example is, of course, Bitcoin. People tend to forget that, amongst other things, Bitcoin is a tokenless and permissionless DAO that is arguably the best example of a DAO to date. It has evolved to the point where it is nigh-on impossible to bring it down, as there is no discernible person or entity responsible for running it. Supporters of Bitcoin come in various tastes and flavours, but most can be said to ascribe to one common belief – that Bitcoin is ‘sound money’, and that it has ushered in a new era of financial independence. When one reaches such a lofty stage, participation is spurred by that idealistic motive alone, and that’s one hell of a long-term tonic to ensure that there are, at all times, people wanting to partake in the overall direction of a project. Of course, Bitcoin is also built on FIG in a sense thanks to mining, but it isn’t necessarily the primary mode of motivation.
Ethereum has likewise moved away from a heavily centralised governance system revolving around its demi-god Vitalik Buterin, and can be said to have achieved a sufficiently-decentralised level of governance. A point that is however common to Bitcoin and Ethereum is that endorsement of either is seen as a powerful statement; to go out there and state you’re working on Bitcoin or Ethereum, to wear their t-shirt, to have a say, big or small, on anything happening in relation to either network, places such persons in a generally favourable light in the eyes of the wider crypto communities.
So which is the ideal governance model, ser?
As much as it might disappoint you to hear this, there’s no perfect governance model for a DAO. IMG is arguably the strongest one in the long-term, but as stated above, it’s very difficult for a protocol to achieve a sufficient network effect and following to spark off an idealistic movement. Financially-motivated governance models need to be well-structured to be sustainable in the long-term at the risk of being tainted by pockets of centralisation, and COG risks being too centralised to properly merit the title of DAO governance.
However, I will share my personal opinion on what would be an ideal model for most DAOs out there, assuming that they are sufficiently big in size to warrant a DAO governance model. Siloed sub-DAOs (SFIG) with their own patrimonies/treasuries and leaders or committees elected via veToken holders (LSM) will work well for a lot of DAOs that I’ve seen over the years, with non-veToken holders being given governance rights on other proposals so as to still give weight to the opinion of the community at large. Indeed, veToken holders can be given extraordinary voting rights (EVR), such as choosing leaders/committees, the implementation of important technical changes (including to smart contracts), token emittances, etc. Conversely, non-veToken holders are given ordinary voting rights (OVR), namely on matters that help the DAO progress and develop such as expenditure of funds over a certain threshold, the selection of partners for collaborative efforts, and other incremental changes that still affect the overall structure of the DAO. This model can also work by having the LSM applicable to those wishing to exercise EVR, while OVR can even be exercised by non-token holders in general (therefore, having a governance system similar to the one proposed by Optimism). Since matters decided by OVR tend not to be as material as matters decided via EVR, this can work out quite well.
The sub-DAO leaders or committees would have the authority to decide on day-to-day management and running of their respective sub-DAO, so as to ensure the efficient operation of the sub-DAO in question, while still being shielded from the responsibility of having to decide on structural matters that would rest at the hands of both the normal and veToken holders. They would also be responsible for linking up, and collaborating with, the other sub-DAO leaders/committees, with the common goal of always driving the DAO forwards, and motivating the members within their own sub-DAOs, rather than facing the mammoth task of incentivising a whole DAO. The formation of sub-DAOs would also help ensure, as much as possible, that the DAO in general does not transition to COG since there would not be an easily identifiable leading figure with a heavy dose of influence and control over the project in question.
Speaking of a common goal, the most important element of a DAO is the clear setting of a common mission. Without that common mission, one has to question whether a DAO is required in the first place. If one can also then instil a sense of pride and ‘higher calling’ to work towards fulfilling that mission in a DAO, then this by far remains the best catalyst to incentivise long term participation.
