This Topic is to run a retrospective analysis on the Governance Mining Program, introduced by the Treasury League back in June 2022.
We took initiative after the Leagues M1-Report, where @Teo mentioned:
Governance mining: we will release, in Jan 2023, a report that analyses the last 6-months of the program to understand its efficiency with regards to vote participation, governance forum activity and other DAO metrics; based on the results and the feedback from tokenholders, we will decide how to stir the program to continue improving the DAO activity.
This analysis was made by the IdleDAO delegation of Bocconi Students Blockchain Association, in particular from:
Governance Mining Program Retrospective
We have been given the opportunity to participate in the following discussion regarding the ever-evolving governance structure of this thriving DeFi protocol, namely the Governance Mining Program.
Before we introduce our methods of assessing the effectiveness of the Governance Mining Program, let us remind ourselves of the goals of the Program. The Governance Mining Program was introduced in the governance forums in June 2022, rewarding “the contributors who are more active in the governance processes and add value to the community”. The goals, while not necessarily lined out in the governance post, are obvious: Idle, a yield protocol, thrives on different-minded people coming together and proposing their ideas and suggestions so market demand is best captured by their product. Voting is one of the ways to gauge the thoughts of such stakeholders. Another one is drafting proposals, where ideas are put forward to be implemented into Idle’s protocol, either to improve the existing product or to introduce new features.
Thus, we can define the final goal of the Governance Mining program to be two-legged:
- Better engage existing ecosystem stakeholders (not necessarily limited to IDLE/stkIDLE holders) so a more well-rounded view of proposals can be consistently achieved
- Attract new members into governance discussions and voting
Keeping those in mind, we can now discuss how we went about measuring the effectiveness of the Governance Mining Program.
To assess the efficacy of the Idle Governance Mining Program, it was necessary not only to look at the quantitative aspects of the Program ex-post but to also qualitatively evaluate governance activity. More specifically, we dove into the parties that were active in Idle governance and asked the critical question of whether their activity was at all affected by the presence of the Governance Mining Program or whether it was redundant. Furthermore, we compared DAO governance activity in the entire crypto space to Idle DAO’s governance activity in the period since the introduction of the Idle Governance Mining Program to gauge whether the data on Idle DAO’s governance activity was influenced by market factors rather than the Governance Mining Program. This analysis assumes that the value added by non-market factors can be entirely attributed to the Governance Mining Program, thereby potentially missing Idle Finance’s marketing efforts, for instance.
Firstly, we will analyze data that we find relevant to describe the governance activity during the time period, for instance, the number of votes, the number of views, and the number of replies to a proposal during the time period beginning at the point of implementation of Governance Mining at the end of Q2 2022 (June/July). More specifically, we take a look at off-chain votes as they are usually the most voted-on proposals and we are of the opinion that they best represent governance activity.
Below there are three figures representing in order: the average number of votes proposals received, the total number of votes received by all proposals, and the minimum number of votes for a single proposal for a fiscal quarter. We chose to report the minimum number of votes because it is a good representation of the number of active members who vote on proposals no matter what.
We can see that in the last year, while the average number of votes remained constant, the total number (and so the frequency of new proposals) decreased, and this led to the increase in minimum votes (fewer proposals, less probability of a low number of votes). Not considering this influence, we could argue that the Governance Mining Program, done at the very end of Q2 2022, forced more people to vote for the small number of new proposals presented just to obtain the reward. On the other hand, one can argue that as voting is an activity that one can do without putting in a lot of effort, no one can prevent someone from voting without properly analyzing proposals, therefore voting just because it makes it more likely that they will receive the Governance Mining rewards. Indeed, for some of the months, it was enough to simply vote on Idle proposals regularly to become eligible for Governance Mining rewards, not contributing at all to discussions on the Discord channel or on Governance forums.
Now, let us focus on proposals, this time looking at data of views that proposals received.
The decrease in the number of views for each proposal over time is clear. In particular, if we classify as “important” the ones that are outliers on the boxplot (the dots) for each quarter of the year, we can notice that also the number of “important” proposals has diminished in time, where the last one, the outlier in Q3 2022 is the one corresponding to the proposal of governance mining.
Below are reported figures of the views and replies received by proposals on the forum.
Even though the average number of votes for each proposal has remained the same after the governance mining, from these graphs we can notice that overall activity has diminished.
Looking at the proposals a little more closely, we can immediately see that the governance mining program didn’t attract new contributors as all of the proposals since the introduction of the program were made by Treasury league members.
Finally, we briefly take a look at governance activity across the space to assess whether Idle DAO governance activity was influenced by market factors rather than by actions taken by Idle itself. Therefore, we take a look at some DAO governance data since June 2022. More specifically, we use data from Deep DAO as a proxy for activity in the entire space.
As one can see in the above charts, DAO activity has trended upwards or flat, depending on how you weigh the two charts against each other in terms of importance. When compared to Idle DAO, one can argue that Idle DAO’s governance activity does not display any strong correlation with the rest of the market. This, while not at all disproving the notion that Idle governance activity is affected by market factors, weakens the argument that market factors explain the trend in governance activity. Thus, we can conclude that while market factors did not help to improve Idle DAO’s governance activity, they certainly did not weaken it either as there was a floor in governance activity that was upheld, but not greatly exceeded.
It is important to note that we do not know what effects the Governance Mining Program could yield when market activity substantially picks up. Such a situation would introduce not only more capital into the ecosystem, but more participants that are potential community members that see the long-term potential of becoming active Idle DAO contributors.
Conclusions and Action Steps
Overall, we find that the Governance Mining Program did not have any convincing effects on governance activity. Coming back to the goals of the Program as defined earlier: 1. The Governance Mining Program did not meaningfully increase engagement within the existing community and 2. did not contribute to onboarding new members into the community.
Thus, the Governance Mining Program should be revised since we believe the token allocation may be better spent on other marketing activities that benefit the DAO at current market conditions, such as onboarding new users that want to maximize their yield. We believe that a more likely success scenario is the one where governance activity increases after an increase in usage of the protocol by new users as opposed to the reverse occurring. This may be because putting in the effort to participate in governance discussions does not produce any great short-term benefits (apart from Governance Mining rewards which are not very large) and long-term benefits are difficult to gauge. Such aspects, though, are best left for further discussion.