CIP-1694: I have a concern re: the current proposal for voting incentives

Governance
1

Hello All –

I’ve just watched the really fantastic CIP-1694: An explainer video on YouTube, and it is my opinion that Voltaire is shaping up quite well, but I’m concerned that requiring voter delegation only upon withdrawal of one’s staking rewards is far too passive, and that a more active incentive will be necessary, and perhaps even critical…

MakerDAO

I spent a few weeks deep-diving into MakerDAO last year, and was startled to discover that the state of their governance was what I would consider to be an absolute train wreck… Since I first learned about MakerDAO years ago, I had this picture in my mind that they must have their governance structures down to a science… The crypto community in general seems to have a fairly high confidence in DAI, and it was because of that, and DAI’s ability to weather bear markets without suffering major failures, that I believe I had this impression…

So you can imagine my surprise when I learned (amongst other things):

  1. A while ago there had been some major trouble with DAI, and to save it there had been an emergency measure put in place (through a vote) to allow DAI to be at least partially backed by USDC as well as ETH, and that once that particular “Pandora’s box” had been opened, not only was it then never closed/repealed, but more and more USDC had been brought into the system since then and community members had been heavily campaigning to allow even more types of ‘real world’ assets in to back DAI… Which, in my opinion, is far too risky and also strays far too far away from what I would think of as one of DAI’s founding principals, namely securing an on-chain asset pegged to the value of USD using only other assets (namely ETH) that are not tied to USD itself, but that is only my opinion and not what I’m here to debate and so I digress…
  2. Much more to my point: MakerDAO voter participation was* egregiously low, and that was due at least in part (I would say nearly entirely) to the fact that they had absolutely no voter incentives in place… And, in fact (IIRC), increasing voter participation with some form of incentivization plan (be it passive or active, direct or indirect, etc) was widely being discussed, but it was a very hot-button topic and it seemed like it was very difficult for any of the very few people who actually were active in their governance to agree on anything…

* I see that they’ve just moved themselves into what they’ve dubbed the “Endgame” phase… And I forget if that included voter incentivization and, if so, exactly what that looks like, and I’m unwilling at this moment to dive back down that particular rabbit-hole, as it is fairly deep and dark as I recall…

My Concern & Proposal

IMHO an active voter incentivization model is crucial to an on-chain governance system, and my concern is that only requiring voter participation at the time a potential voter chooses to withdraw their staking rewards is far too passive, and will result in voter participation that is too low for too long too early in the age of Voltaire due to either:

  1. People knowing that they’ll need to delegate their vote to withdraw, but since they don’t plan to withdraw anytime soon then they’ll feel no pressure to bother doing the work necessary* to pick a DRep, or…
  2. People who long ago “set and forgot” their stake delegation won’t even know they should be delegating their vote, because they probably only check-in on their staking ROI every-so-often, and are otherwise not paying enough attention to Cardano itself…

* And I would argue that picking a DRep, and continually checking-in on one’s chosen DRep, is at least as much work, if not more work, than picking/maintaining your choice of stake pool… A stake pool can be chosen (and that choice maintained) largely by simply looking at its ROI… In contrast, you must pick a DRep you feel has principles and opinions (etc) that closely align with your own, and then you must actually audit their voting and other participation history yourself in order to determine whether or not to remain delegated to them, etc…

That being said, I do not at all understand why we are not considering rewarding voter delegation more directly, in the exact same way as staking delegation, namely by simply just paying-out rewards, but I trust there’s a good reason to take a less direct approach, and so I’ll get on to my ultimate point…

My proposal is that when a wallet is not delegated to a DRep (including the ‘auto DReps’ for abstain, etc) or has otherwise not met the minimum requirement for ‘governance participation’, then the staking rewards to that wallet should be lowered (aka slashed) similar to how they would be lowered if the stake pool the wallet is delegated to was acting badly (by being offline* / not producing blocks, etc)…

* In fact, I think the argument could easily be made that if your votes are ‘offline’ then you are the bad actor within the governance system, which more than justifies slashing the rewards you receive from the system…

Additionally, this ‘slashing’ would (I assume) further benefit the network, since (again, I assume) non-participants’ slashed rewards would go to the same place as any other slashed rewards (I assume the treasury…?)

Also, it’s possible we could govern this mechanism via a network parameter which decides the severity of the penalty for non-voter-participation, the same way as there is for how much a stake pool’s amount of pledge affects its rewards payouts (which is currently set to a rather low severity I believe) so that like the pledge param, the voter param can be adjusted over time… But I’m less certain about that (having a ‘voter param’) than I am about the fact that slashing rewards, or at least some much more active and/or even direct incentive, must be put in place…

Thank You

Thank you for reading this, I welcome all feedback/rebuttals/corrections/conversation/debate, although I may have limited time to respond, as my work is heavier than normal ATM, and I stole time away from it to write this only because I feel strongly that this needs to be discussed, and the sooner the better…

2 Likes
2

I like your thinking on this and I think we need ways to incentivise voter participation but I don’t think the incentive should be some direct monetary payment or punishment for the following reasons:

  1. I recall listening to a podcast from a governance researcher some time ago where it was contended that studies had confirmed that paying voters to participate resulted in bad outcomes.
  2. Creating some sort of punishment or payment based on blockchain voting would invite tech savvy people to create an automated mechanism to fulfil the requirement. They would just create some automated web tool or other software which, in a way, allowed you to “tick the necessary boxes” without you having to actually put in any effort. I think everyone can then see how such a “tool”, with in-built bias, could skew voting results.

What we are really trying to incentivise here is people actually doing the hard work of thinking deeply about governance issues and then their subsequent votes actually meaning something. I don’t know how you properly incentivise people to do that without perverting the vote outcomes. I think they have to want to do it themselves for the good of the system.

I am very concerned that some sort of direct payment or punishment will result in bad actors gaming the system or somehow getting disproportionate power through vote buying.

Sadly, this seems very difficult to solve. I think we need advice from governance researchers that are deep in the weeds about this problem.

In Australia there is compulsory voting and if you don’t vote there is the potential to receive a fine. But, I believe the fine is nominal and is probably fairly easy to avoid with an excuse. Nevertheless you still get people just writing “Mickey Mouse” on their vote card, which, there is no way to prevent.

In effect, Cardano’s 1694 proposal to require some sort of delegation in order to access staking rewards is an attempt to make voting compulsory. Furthermore, the automatic DRep for abstain enables the “Mickey Mouse” voters an easy way to avoid the investment of time and hard thinking. This might be the best we can achieve?