Not your vote, not your lawyer: Delegation and the Legal Challenges

Yeah, welcome everybody. We are happy to present you today delegation and its legal risks. And me and Teresa, we are both lawyers and then we're working together with some crypto projects protocols. So we are happy to share with you today are what we have seen so far in the space and the risk that we have identified along the way. Okay, so for the agenda, first, we're going to start with like the basics.

So like, sometimes in crypto, we forget there's like a word out there that has explored this before. So we're going to talk about the legal basis. We're also going to be speaking about how it looks actually right now, like in different dolls and stuff like that. We're going to be speaking about the legal challenges and their risks associated with delegation and sort of like practice cases, use cases, and how we look at this looking forward. I mean, since we are both lawyers, it would be really inevitable that we would come to you with a disclaimer.

So yeah, nothing in this talk is legal advice or financial advice. So please do not take any of this, you know, as an advice. And this also does not create an attorney-client relationship with you. And everything we share here is purely informational and educational purposes. So it doesn't necessarily represent any perspective, goals, values of any of the protocols or projects that we are connected with.

And of course, if you have some legal questions, please talk to a lawyer. Okay, so as I was mentioning, okay, what we tried to do initially is just to like set up the context and like see how this delegation looks in other areas of law. So like, I tried to like relate it a lot, like in my mind, some people might think differently. But like in my mind, I just feel like delegates, as just like this, senators, or like people who work for the nation. So that's how I envisioned them.

But like, this is not the only way it can be seen. It also can be seen in the corporate governance perspective, either as directors or as representatives of the companies. I mean, like, this is just like going back to like the legal basis doesn't mean it is this is representation or delegation in the web space. So the way we can like see it is just like, for example, like when you have a corporation or a foundation, you would have individuals who represent this company, and are supposed to act in the best interest, either of the company or the purpose, in case it's a foundation. Or for example, on the part of director holder representation to a proxy, that would be who they're accountable to.

So like, one of the biggest questions we're going to be answering is sort of like who are delegates accountable to? And like in this scenario, this is who we're accountable at the real world, like outside of like what free space in just like the corporate and like political science perspective. So when it comes to the civil law perspective, basically contract law, I want to talk about like what is representation so that we can all have a similar understanding going forward. So in the legal world, one person can act on behalf of another for the other person. And this is called representation.

And the most common tool used for that is a power of attorney or in briefly, it's a PIA. And the person giving the power of representation to the other person is called principal. The person receiving that right is called agent. And how the PIA works. So basically, the agent can sign the contract, make decisions or access accounts for the principal, right?

Because the agent is doing something on behalf of the principal. And he is or she is powered with that. And we have like different types of PIA. So depending on the jurisdiction, they might have their male household like different kinds of than what we have listed here. But basically, a general PIA gives a really broad powers to another person.

Basically, the agent can do whatever he or she wants on behalf of the principal and for the principal. And when it comes to the limited PIA is basically specific for tasks that the principal wants to conduct with the help of the agent. And generally, it also has some sort of duration. And then the durable PIA is a tool for the cases where the principal is not able to decide for him or herself, basically in case of accidents or some sort of innocence. So depending on the jurisdiction, also the shape of the PIA might be different.

In many cases, you might see written PIA, but sometimes it can be also authorized and witnessed. And here you see the schema. Basically, on the left hand side, we have principal and on the right hand side, an agent and the PIA, which basically includes the terms of the representation. And basically, the principal is legally bound by the actions taken by agents. So imagine this, if agent signs a contract, it's going to be as if the principal signs it.

So you may understand that the agent has this power upon a person, upon the principal. So most of the cases, that's why agents have also fiduciary duties for the principal. So for example, we have duty of loyalty for agents. So it means that the agent must always act in the best interest of the principal. Then we have duty of care.

The agent must make decisions with reasonable care and competence, just as they would with their own affairs. And then we have duty to act within the scope of the PIA. So the agent shouldn't go beyond the scope of this PIA, otherwise it could be a breach of the representation rights. And then we have duty of transparency, because we are expecting agents to be transparent with their principals. And then duty of accountability, basically, the agent may need to report back or justify their actions.

Okay, so like, the way DALs are represented into like, internally, into the real world, they might be a little different. And mostly in the part of like, how these entities are being represented, either by their directors, or their supervisors, or their representatives or mandates outside true legal entities. Now we're going to speak mostly about like, how it works internally. And like, I do have like, here pointed out two really important questions that are like the core of this presentation, who are that are, okay, who are these parties actually be accountable to? And what happens when there's a conflict of interest?

Okay, so if you want to start with the delegates? This is a really hard question that I always question myself. Okay, so like, are the delegates that are being delegated tokens to represent the token holders to take certain decisions and voting proposals? Are they accountable to like the delegates, to like the delegators, I mean, or are they accountable to like the whole network or the whole ecosystem? This is like a very important question that we're going to be like, addressing through this workshop.

And then the other question is like, what happens if there is being a conflict of interest? Like, I mean, like you as a delegate, you're supposed to disclose, okay, so like, this proposal is actually of interest to me, and I should like abstain to vote. But like, we don't really have other than those mechanisms are just like, sort of like a code of conduct or sort of like this different things in this pace. How do you actually, like enforce that? So like, that's a really good point.

And then on the external part, oh, no, second point. Okay, this one is like a bit more clear. So what happens when there are members of a committee that have been appointed by the DAO, like by the token holders, by the delegates, or by which other parties? I do think here, it's like much more or less complex, because like, here, you're understanding your responsibility is towards the ecosystem in general. And on the other part, this is like what I do in my little practice.

But in the other part, we have to like understand that this representation, like this vote that you're having on Shane, or up Shane or whatever, they're actually probably not in all cases, going to also have a reflection in the outside world. So there's going to be a contract and agreement, there's going to be something that's going to be happening. And then for that part, this correlation between the votes that are happening within the DAO, and then how is that correlated to a legal entity that are going to be acting on behalf of those individuals, have to have like, certain like, vinculo, like a certain, how do you say that in English, like a certain like, bond. And there has to be like, counterbalances. And like, they have to make second balances between these both parties.

But at the end of the day, a director of a legal some of the decisions that are requiring the up Shane, I'm sorry, like the real world scenario. So let's talk about some decision mechanisms in DAO. Well, we really like this tweet. So when governance changes in real life, people die. But when governance changes in Web3, it's a Tuesday, we're still early.

But this space is still experimenting with different methods, actually. So here you can see different decision mechanisms that we kind of like identify for this talk, but obviously it's not, it's not exclusive list. So in the on the first slide, we have this like direct token voting where you have token holders who vote, and the execution happens on chain via smart contracts. Most of the time, the pros is that like if you have a transparent system, but at the same time, it can lead to voter fatigue or veil dominance. And then we have delegated voting.

And then you have the delegates who actually vote on proposals. And this can be again, off chain or on chain executed. It is more like scalable, at least this is what is now thought of as and it's also expected that more informed people, more people who are active on these on these networks actually join. But on the con side, we may have delegated centralization and low accountability. And then we have snapshot voting, which is again, somehow exclusive to token holders, but it's a kind of an off chain mechanism, off chain voting mechanism.

So the execution requires basically an off chain signal, then sometimes on chain execution, like depending on the topic and how the system is created. It can be really cost effective, because like you don't have to actually spend gas. But it's not enforceable if you don't have any follow up contracts, if you have like on chain execution, so someone needs to push the button. Then we have like multi tick execution, basically, it's more exclusive to Alexis signers. And the execution can be manual off chain or on chain as well.

The pros would be like, you have a secure treasure, treasure control, because you know who is controlling what. And you have this like hybrid flexibility. But it may not be again, fully decentralized. And sometimes it can be slow. Also, this space is experimenting with like layered bicameral voting system where you have like a mix of voters.

And the execution can vary again, based on the structure and the pros, it could be inclusive or balances of power, because then there's like token holders, and then they are actually selecting some other people to vote on their behalf. So you have this like bicameral system. And on the con side, it's quite a complex design, and it's kind of harder to coordinate. So the delegation, the topic of today, then why do we really talk about delegation, because I've been in the space like late 2017. And I've been hearing all the time, how do you make people vote?

So the delegation is the process by which token holders assign their governance rights to someone else named the delegates. And these delegates are usually individuals or teams who are more informed, active, or trusted to vote responsibly on behalf of others. So this is their, this is the expectation. And then why the delegation exists, as I said, we don't know how to make people vote. So maybe delegation will actually be a like, in terms of turning the low water turnout to like a higher turnout, delegation might step in here.

Or, again, if you have like governance fatigue, if you have too many proposals, too frequently, and if you're expecting your token holders to vote on like every proposal, then you may create governance fatigue. So delegation might be a solution for that. And also delegation is expected to help with the scalability. On the benefits side, basically, you can also get much more like specialized expertise information through delegations. Because most of the time, people who sign up for delegations are quite active on, you know, on the forums, and they're basically following all the news and developments and everything.

And it's maybe more efficient, because, well, the delegates who are actually willing to vote, they will vote much more frequently, so it will keep things going. And again, scalability, because it will probably reduce bottlenecks. So in this, in this space, we see Uniswap, Optimism, ENSDAO, LedaDAO, use active delegation models to basically overcome these challenges that I just said. Yeah, just like as an example of the Bicameral, the Bicameral, we have like Optimism. I know like today or yesterday, they did change a lot of their things on this aspect.

But they do have like this two parts, where like citizens are on one end and token holders are on the other end. So they, I know like the one of the biggest changes was on the citizenship part. So like, if anyone here has some info on that, I want to share like in the Q&A, that would be great as well. Another thing that the space is experimenting with, and we found quite recently is the automated delegation, and it's coming from OVAL. So basically, once you have this like OVAL governance token, and if you do not delegate it willingly, then the system delegates it for you automatically.

So basically, you lose control over your voting power. But this is a kind of a backup system for OVAL because they don't want any governance tokens to stay dormant, and they want to keep them in the system. Okay, so like, one of the legal challenges we see in all of that, it's about the liability, liability in two aspects. Okay, so I'm delegating to you. So those tokens are actually still mine.

So like, I still have control over the execution, and I can have, I can have like that liability. And what happens is those who I delegate to are not acting in the best interest, or like they voted on a proposal that was considered sort of like, against the law. And then on the delegate perspective, it's like, okay, so like, I'm the one taking the final decision. Maybe I don't have like the final control of the tokens, but I'm putting my name out there. And there have been like certain, there have been like certain cases.

For example, in Uniswap, they were like, they were like, some of the delegates who were like receiving subpoenas, which are like notifications from like the US in certain parts of the regulation side in this sense. So they were like, okay, so like, I'm not actually a legal shield. And I'm receiving these notices in my house or like wherever. I don't know if they were in their house, but like they were seeing like this notices. And there's kind of like, okay, so like, how should I proceed?

One of the recent things I've seen that I mean, like, I'm not, well, we haven't seen the practice how it's going to be worked. But it's a really interesting point that it's particularly on the delegation part is with the CK association. They have an actual association. And what this does is like it has other functions. But in this context, it sort of like grabs up the delegates.

So you have to like opt in and like be a member of this Austrian association in order to participate in the voting of any proposal. So I was just like reading the bylaws. There was like no, like there, I thought there was going to be sort of like a lease of all the addresses that opted in into like this membership. It isn't like a visible list. So I don't really know how like shielding this is, but it's just sort of like an interesting perspective.

And another interesting point is, for example, like a lot of like the biggest PCs when they delegate, they actually have contracts with the delegates. So this is something they're using like also to shield themselves that this decision, they're not like participating directly. And so, yes. So when it comes to second legal challenge, we can talk about accountability and the risks arising from directly delegation itself. So the first is that we see the centralization.

Basically, one delegate can get the delegation from many token holders and kind of accumulate the voting power. And basically, it would lead to centralization. The second thing, the stealing of votes and bribery. Basically, behind the closed doors, some people may have some agreements with token holders to get their votes. And we have also seen LobbyFi and Dark Cloud as examples just to support that kind of model.

And as a third risk, we can talk about abuse of delegation powers. So in the space currently, to my best knowledge, delegates go on the forum and then they identify themselves, they introduce themselves, what they do, et cetera, just to gain the trust of the token holders. But there is no identity proof, like there is no identity verification. We actually are bound to statements of what they have told us. So in this case, we cannot really see if they have conflict of interest or other intention behind this.

So this can definitely actually lead to an abuse of delegation powers. And then we have conflict of interest. Basically, again, through the online material that the delegates have written, we are not able to find out the conflict of interest. So it needs to be disclosed, but there is no enforcement if it is not disclosed. And then regulatory exposure.

Teresa just said that some delegates at Uniswap received subpoenas in the U.S. Basically, it's a legal notice. And so with the delegation process, when you put yourself out on a forum and introduce yourself, basically, you become discoverable. And that will make you exposed to these kind of like regulatory risks as well.

And having no liability shield is also a bit concerning for the delegates. And here, for about the first risk, basically the centralization, not decentralization, we have a case study about OptimismDAO. So OptimismDAO has a delegated voting system as well. And on the token house side, we have one token one vote model. On the citizen house side, we have one person one vote model.

But if you see the chart on the right side, basically, six delegates are controlling, like each of them are controlling more than 5 million OP tokens, meaning that they have actually quite a big power in decision making. Like you can see it on the graph, which is below. The veils doesn't really represent the token holders per se. It's kind of like in the optimism ecosystem, it represents anyone who's able to meaningfully shape the decision making system. So delegates can be also called as veils.

So the key takeaway about OptimismDAO is that without the mechanisms preventing the centralization, basically, we are undermining the DAO's foundational principles and basically the principles of distributed governance. Yeah, so like in this part of like being accountable, I think like we want delegates to like be accountable, but accountable to who? And this are like things we can like add, like, for example, like one of my the most important point I see here is regarding the conflict of interest. I mean, like this has been like a lot of studies on this end from the governmental perspective, and like how important it is for like not participate or like avoid like situations in the ones you have a conflict of interest. So like that's like super important.

There are also like, I mean, we see a lot of this entities about this DAO's actually having like non-binding sort of like constitutions on that end. But like in order to like opt in, you should like be like, okay, so like I'm gonna like sort of like follow these rules. There's like this point of like, okay, they're non-binding. And like what happens and it's been like, it's happened before that like you breach certain like rules in one DAO and then just like, you go to another one, like maybe no one knows you there. So you're gonna start from scratch.

So like, those are like a little like some issues that we see. Another like really important point that I see on the part of accountability is like, who am I incentivized? Like who gives me like incentives to like participate? So like, if you say that the project itself, they give incentives to like the delegates to like build and act in the best interest of the DAO, that's one thing. But like, what happens when the interest from the delegates that are being participants come from particular projects?

So like at the end of the day, who me as a delegate should respond to? Like the individual who's like paying me or the broader community? So like, these are questions that come to my mind. And the next challenge would be that smart contract vulnerability. So it's, yes, it's a very generic vulnerability, like a risk that we see in all like smart contracts.

But within the governance context, basically, if we have bugs in governance contracts, it can it can lead to make lead to errors in contracts, and incorrect world tallies or delegation loops, or override bugs or inability to revoke or update delegations properly. These can be quite counterintuitive in the system for the system that that is built. Of course, we have exploits and attacks that can hijack the governance. And we can also talk about the governance manipulation here. Because delegates themselves can also collude, coordinate or even be bribed through the smart contract.

Without the off chain context, basically malicious proposals can be also passed, like even though they are socially unacceptable. And on the legal side of this, it's kind of like, hard to again, talk about the liability issues here. Because like, imagine like you have a bug. And we need to figure out first, like, is it really a bug? Or is it a future?

Or is it basically an abuse? And no jurisdictions have a clarity on that at this point. So when it comes to the mitigation strategies for that, like we can maybe talk about the thorough auditing, and maybe the use of pulse switches, time locks and veto powers, and basically the delegates responsibility frameworks. So the golden nuggets that we want you to take away from this talk is that delegation, it might be necessary, or maybe not necessary, it's up to you. But like, it's definitely risky.

And we have personal liability risks continue. And obviously, the design must be also balanced with the efficiency and accountability. So like, yeah, like, I think like this brings us with more like, questions than answers. I thought we're gonna have more time for questions or people for input. But I don't know if anyone has like any sort of like questions until now.

Oh, we have them in the screen. Oh, can we legally stay 100% anonymous while delegating our vote to be trusted by a party in a doubt? I mean, like, legally anonymous in what way? Like, because like, when you delegate, like, like, you're not like you're, like your information is not being made public. At the end of the day, you're like, you don't have to like KYC in order to delegate.

I don't know if it was like regarding on that end. And in case in the system, isn't that being forced voting? And how do you? How do we know the actual delegation happens? Actually, there was like someone who pointed out this question and, and in Twitter, and then like the level of the community, they responded back and they say they only use this for like certain forums, and the ones like the current made and