Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Ava Labs x CBER Ep 11: Decentralized Autonomous Organizations
30 Plays1 day ago
This episode explores the governance of Decentralized Autonomous Organizations (DAOs). We discuss how ownership concentration and whale voting can distort incentives and hinder platform growth. The conversation examines mechanisms to embed long-term incentives directly into DAO economic design.
Guests: Jungsuk Han (Seoul National University), Jongsub Lee (Seoul National University; University of Florida), Tao Li (University of Florida)
Paper: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4346581
Recommended
Transcript
Introduction to 'Crafting the Crypto Economy' Series
00:00:06
Speaker
Hi everybody and welcome to Crafting the Crypto Economy. I am Silvia Sanchez from the Avalanche Policy Coalition and today we bring you a special podcast series in partnership with the Crypto and Blockchain Economic Research Forum, also known as the CBER Forum.
Season 3 Overview: Global Faculty and Blockchain
00:00:22
Speaker
This is season 3 of 5 new episodes featuring leading faculty from global universities exploring various elements in the blockchain ecosystem. From stablecoins to DAOs and so much more.
00:00:33
Speaker
These episodes are a bit longer from our usual ones since we will be getting deeper. But again, it's kind of challenging to fit in a whole research paper into just one hour. But don't worry, each episode will have its accompanying research paper posted on the website and episode description for further reading.
00:00:49
Speaker
Anyway, I'll hand it over to our moderators, Professors Fahad Saleh and Andreas Park.
Episode Focus: DAO Governance Discussion
00:00:55
Speaker
We hope you enjoy.
00:00:58
Speaker
Hello, everybody, and welcome to another edition of Crafting the Crypto Economy, a podcast presented by Fahad Saleh and myself, Andreas Park, ah together with OWL Explains.
00:01:09
Speaker
Today, we're going to talk about DAO governance, and we're very pleased to have Jung-Suk Han, Jung-Suk Lee and Tao Lee I'm here to talk about their work on DAO governance.
00:01:21
Speaker
ah Jung Subhan and John Sub Lee are both from Seoul National University and Tao Lee is from the University of Florida, the Warren College college of Business. um Welcome everybody to the pod.
Understanding DAOs: Benefits and Vulnerabilities
00:01:33
Speaker
um Let's dive right into it. um Maybe you guys can explain to us and to our listeners, um you know, a little bit about the background of DAOs versus normal companies and in particular,
00:01:45
Speaker
maybe lay out a bit the landscape of different DAOs and what you're really interested in here. Yes, thank you Andreas. um So I'm going to just explain about DAO and what we are studying in in our work.
00:01:58
Speaker
ah DAO is a decentralized autonomous organization. It's a recent development. It's an organization without the central leadership, um running a smart contract using blockchains.
00:02:10
Speaker
So um you can imagine organizations without management or or CEOs. um So it actually doesn't have um the traditional problems of um agency problems yeah in the usual, let's say, companies and firms in the past.
00:02:30
Speaker
So it has promises of transparency and the democracy. But it it is not actually um perfect. It comes with vulnerabilities, especially with the large token holders, well which we call whales.
00:02:48
Speaker
So ah whales can often hijack the ah the leadership of the of the DAO by dominating in the voting process ah because they can easily get ah large ah voting power on top of what they already have.
00:03:06
Speaker
and In that case, they can pursue their um ah personal benefit, ah destroying the the value of the organization. So in in our work, ah we study how um how this can impact the value of DAO and how we can design that DAO so that that these kind of problems could could be mitigated.
DAOs vs Traditional Firms: Governance and Decision-Making
00:03:29
Speaker
So can I interject maybe for one second? um So because I think it's important that we actually try to understand the landscape there a little better. um So... um So first of all, let's just take a step back, right? So when you have a DAO, you say that's basically spark contract based.
00:03:45
Speaker
um And so so, are you, so I, you know, i'm I'm not an expert on this, but so I just try to understand this a little better. So when you have a corporation, we have a CEO usually who has a, who's overseen by a board, right?
00:03:57
Speaker
um And then to some degree, the CEO takes decisions, many decisions autonomously, but he has some form of of supervision, right? um And, you know, as we know from the long history of corporate governance, there there can be some problems, you know, captive boards and all of those kind of decisions, right? So when a CEO basically picks people from the board so to mitigate or to minimize how he actually gets over overseen, right?
00:04:22
Speaker
And so you're kind of trying to say that in a DAO we can, because it's so much more democratic every, you know, there's there's particular voting mechanisms, we can circumvent that. But just...
00:04:34
Speaker
out of curiosity, and this may not apply to your paper, but I think in a DAO, it's also relatively narrow of what a DAO in a smart contract world can really do, right? So because you have an existing smart contract and there may be some parameters and you can vote on them, or ah you do you have also something broader in mind where there's, I don't know, maybe some treasury or funds that has been collected in a token offering, and then it's a question of how to deploy this. So but so what exactly is but What exactly is a DAO really doing but that's the same maybe as a firm? What's different about it?
Exploring Hybrid vs Pure DAOs
00:05:07
Speaker
Functionally, not not just in terms of decisions. DAO can be ah diverse and then it can it actually has quite a diverse spectrum, um ah kind of ranging from ah pure DAO with a hybrid kind of organizations with the some some actually so traditional part, like ah core team, um usually coming from developer developers managing some part of the organizations.
00:05:35
Speaker
So um the As you pointed out correctly, in the traditional firms, the managers or the CEO or board board sort of um um harmonize the the organization and the operations.
00:05:52
Speaker
So they they do actually ah contribute to the value of the company, but then often they have their own agenda.
00:06:03
Speaker
So, ah of course, everyone wants to maximize their own benefit rather than the organization's benefit. ah So ah despite the all all the devices the contracting on on on all the ah compensation scheme that sort of try to align the interests of all the um participants, it often fails.
00:06:25
Speaker
um And it is costly too. And and we we we know very well oh from the traditional firms. so um so So that's why DAO is promising. DAO has the advantage over these traditional firms because it doesn't have to involve this kind of a conflict of interest among the human participants. But as I pointed out earlier, ah DAO is div diverse in the sense that there are pure DAOs and then some with the hybrid form.
00:06:56
Speaker
So hybrid form, of course, will have this core team may actually have agency problems. But ah this may be at least the lower than the traditional firms. And then the pure DAO will will not have any of these human elements. So, i of course, um we will have the diver, that could be diverse, but then in our work, we kind of focus on pure DAO so that it is easier and when we discuss conceptually.
00:07:23
Speaker
But our work can be, of course, extended to somewhat the hybrid form of DAO. yeah because it it has at least some more fraction of the non-human part.
00:07:35
Speaker
can Can I interject here? I just wanted to get a little bit more context because i think you've framed ah pretty well sort of the traditional side and I think a lot of people probably are familiar with the at a practical level with sort of the incentive misalignments that can arise in that setting.
00:07:49
Speaker
ah But you alluded to briefly sort of ah private benefits in the DAO setting. And that is, of course, a significant part of of of your work here. So could you just give kind of like a a concrete example of of what that might be?
00:08:05
Speaker
Yeah, actually, as Andres said, you know, DAOs can make many you know different decisions, right? Sometimes like a maker case, you know long to value ratio, the parameters could be, you know the main topic they want to debate and they want to improve.
00:08:21
Speaker
And often, you know, the taxes, they need to choose which are liquidity pools, which pair of tokens they need to consider. And definitely, you know not every device are fully decentralized.
00:08:34
Speaker
So they are actually a group of people who have a very concentrated, you know, voting power. And maybe some of them, they possess their own, you know, tokens somewhere else, so that if they want to promote you know some of their private token positions, that they may propose you know some unusual pair of tokens you know to be trading on their DAX.
00:08:57
Speaker
In such a case, maybe you know such liquidity pool may not benefit all the users, but it might actually benefit a specific user like whales, those who have concentrated token powers.
00:09:10
Speaker
So in such a case, there could be diverging you know you know interest among users, even on the DAO space. So you know that's kind of example you know we have in our mind as a private benefit of some powerful token holders.
00:09:26
Speaker
So if i if I may interject here, so um so I think it's really important maybe also to differentiate um also for the benefit of the audience. it's like um When we think of firms, who at corporations, they take a lot of different decisions, but we take the perspective of shareholders and ultimately the way we see it there is then the only decision variable or the only output variable that you that you care about is is how much money you make in terms of ah as being a shareholder. right So how much cash flow comes into the firm.
00:09:54
Speaker
and So you know this is really, i mean, it's reductionist, but really this is how we think of we think of shareholder value and ultimately this means profits of the firm. i um and But in a DAO, I think the way we we think about this possibly it is ah is a much bigger spectrum of possible ah interest that could exist, right? um so And I think that's probably important to point out because there's oftentimes when people talk about DAOs, it's too much aligned with you know corporate shareholder value, right?
00:10:25
Speaker
so uh i i'm just i just thought of indian maybe you you disagree with me on this moment i think maybe this this kind of helps you frame also what exactly it is that a whale could extract here right yeah so you know this is this is certainly true i mean you know i guess um sort of when actually we were developing our paper we were more thinking of you know our DAO as if the DAO is a business.
00:10:50
Speaker
you know This business generates cash flows, <unk> etc. But you know as so you know and as you said correctly, that you know there were there were there were a lot of DAOs out there. For instance, there were social DAOs. These are you know kind of ah engaged in community building and a shared interest. ah you know there were There were even some non- non-profit DAOs in that space, there were also investment in DAOs, etc.
00:11:16
Speaker
And that there were DAOs deciding whether they're giving grants to certain people or not. Yeah, so, um you know, that is certainly true, but to stay more focused in our paper, we sort of kind of focus on a particular type of DAOs, which is implicit.
00:11:33
Speaker
Well, I mean, since you have private benefits, you can also think of the DAO really as a collection of all all private benefits together, right? And so then the value maximization is the benefit of all of them, right? so um But anyway, so like maybe we should just go to your paper, right? ah so So you have a whale in your in your paper, right? And lots of small shareholders. I saw shareholders, haha, token holders.
00:11:57
Speaker
So maybe you just go ahead and and tell us what your what your what your thesis is and what you find. yeah Yes. um So um let' let's say there's a whale um who want to wants to ah pursue own private benefit. For example,
00:12:14
Speaker
um ah that they kind of ah lower their own, um on the let's say, service fee for their particular ah service in in the at the at the cost of other users, more for example.
00:12:32
Speaker
and Then they will make this value-destroying proposal. um But then, of course, other users will not be happy about it. So to... to to and To make sure this works at the voting, ah what this whale will do is ah hold enough tokens, and the the voting voting power, voting tokens, and then pass this ah value destroying proposal.
00:13:04
Speaker
Now, ah this means that we we actually have a nonlinear relationship between voting power, the concentration of the ah the the tokens, the token holdings, and then the likeli ahhood likelihood of ah value destroying ah changes.
00:13:22
Speaker
So if you if you imagine that the whale already holds so much um tokens, What happens is so ah destroying the the the platform or the the DAO um by lowering the value and then pursuing private benefit, this will this will cost them so much because they already have too much at stake.
00:13:44
Speaker
So burning the platform will be too costly. So they may not want to do that. Now imagine they have a very little bit of tokens. Then it'll it'll be ah very hard for them to acquire enough voting power ah because it'll be costly due to transition cost.
00:14:03
Speaker
So at the both end, it'll be unlikely that the the whale whale or the and the large token holder will disappear. pursue this value-destroying scheme.
00:14:16
Speaker
But there's actually a sweet spot in the middle where they have enough tokens. Now, they they are not too much invested in the sense that the destroying the platform, the burning the platform, isn't too costly, but then it is not too hard for them to acquire extra the voting power to to change the decisions at the at the voting process.
00:14:38
Speaker
So we we actually have so-called home-shaped relationship between concentration of ownership and then the ah the likelihood of value-destroying um um the the proposals. So so that that's one one ah kind of finding we have, which is that kind of true in our theory and and then our evidence as well.
00:15:02
Speaker
so So what you're basically saying is if somebody has... enough skin in the game, right? So then they would not act in a negative manner. And if they have too little, there's nothing they can do. So it's really something about there's a sort of a trade off between having too little skin in the game, but enough power to to be damaging. Is that what you what you're essentially saying?
00:15:25
Speaker
Okay, so it's a little bit like, um it feels a bit like, you know, the arguments when you have families, right? Family ownership or dual class shares in in in corporate governance, right? So if somebody if if you have a family that has very large voting stake, you know, you would argue that they still don't they still don't want to take decisions that actually destroy the overall value of it, right?
00:15:47
Speaker
That's kind of what you're after. So if you if you have to be big, then you have to be sufficiently big. Okay, I see. Okay, interesting. All right, okay, sorry, i interrupted you there. I just wanted to kind of bring that out.
00:15:57
Speaker
And so then the yeah you had a second result um that comes with... thats That's right. So we have them now um ah kind of this i kind of idea about the liquidity and the illiquidity in terms of these results.
00:16:16
Speaker
So usually in traditional governance, liquidity is a good thing. And then in financial market, liquidity is taken as ah as a good thing because it's easiness of trading the shares.
00:16:30
Speaker
So here, ah what we talk about liquid is liquidity of the tokens, how easy it is to trade tokens. So of course, it is it is good to have liquidity for token trading, but ah What we are finding here is illiquidity may help um ah that help ah preventing these value destroying ah changes.
00:16:55
Speaker
So how is that? So ah when the whale wants to acquire more ah tokens to change the ah the the decision,
00:17:07
Speaker
they they face this transition cost. But then this transition cost increases with this illiquidity. So illiquidity gives a natural shield, sort of natural protection to Dow in the sense that the whale is discouraged by the illiquidity.
00:17:27
Speaker
so and So what we find there is, yeah ironically, this ah costliness of the trading can can help the the improving the governance of Dow.
00:17:41
Speaker
So actually governance token could be illiquid and then this could and could be ah sort of the natural protection ah for the yeah ah small users. So there's conflict between small users and then in their wells, and then small users are often protected by by this kind of metro protection.
00:18:00
Speaker
That's another result of ours. So can I ask, um how about a borrowing and lending market? I mean, i mean not that doesn't apply to all tokens, right? Because decentralized lending is ah is still relatively few tokens that are affected by it. But you can imagine in the future, this can be expanded.
00:18:18
Speaker
but So if I can borrow tokens, um then that's the, i mean, in the DAO world, that means essentially that I have the voting power of them, right? Because I actually have the real tokens, the real thing.
00:18:29
Speaker
Is that not something that is that, is that fall under your liquidity, generally speaking? um I'm asking because you need probably a lot less capital to borrow tokens rather than, ah then in and and in particular, if you borrow the tokens, you don't even have to worry about value destruction, right?
00:18:46
Speaker
You're absolutely right. So you're right on the spot. So this this leads to our third result, which is another important element of of our ah work. So up to now, we're we're just saying it is important that ah we discourage this short-term opportunistic behavior of big players.
00:19:07
Speaker
So that that's why this illiquidity and then ah the skin in the game is coming in. But then now you're touching on the important part about the ah pursuing long-term interest.
00:19:21
Speaker
So this as you say, ah there could be, for example, a flash loan attack. ah They can just borrow massive amount and then ah execut execute that with the same...
00:19:36
Speaker
same contract, just with the enough ah time, maybe it could be very short amount of time to just give the voting in the middle of that process, then they can just um ah change the protocol of the platform, for example, ah so that they just benefit from that and they can just dump and then leave right away, and which which actually happened in and in some unfortunate cases of the past.
00:20:02
Speaker
So ah to prevent that, ah we can think about, say, long-term commitments such as locking up tokens before they can they can ah vote, or ah the voting power could be ah proportional to the amount of time that they have locked in their their tokens and so on.
00:20:20
Speaker
So we indeed find that ah this becomes a sort of kind of targeted illiquidity to the the today large token holders, whale, ah because it'll be difficult for them to get in and out in in right amount time.
00:20:40
Speaker
So it is kind of in line with our sort of previous result, but then this is kind of more about long-term commitment can help greatly with all all these loopholes that could possibly happen.
00:20:55
Speaker
Yeah, so that' that's kind of the a award or one of the the third main result, I should say. So can I just ah clarify? um I mean, so this is it's a little tricky of how you actually implement this, I think, right? Because um having held the token for sufficiently long to be making ah a voting decision is not really skin in the game, right? It has to be that you have to hold them after you make the decision, right? So that there is essentially some form of accountability, right?
00:21:24
Speaker
Now, I can see the issue for whales, right? There it makes sense. um But when you have... um You know, when you have a, when you're a small token holder, essentially have no voting power, so but you have the cost of, you know, if somebody takes a stupid decision that you're basically, um you know, that you have no influence over.
00:21:45
Speaker
right So I think this is always a bit of a problem that if you have no um if if if your decision has no impact on the future of the firm or or whatever of the Dow, right? So then you're kind of creating a disincentive or poor incentive there.
00:21:59
Speaker
um So what exactly have your mind do you have in mind? How would you model the lockup? Is it exposed and is it applied to everyone? Is it nonlinear? What's going on there? I think but we need to little bit clarify the skin in the game though, right? So I think whale, if it destroys the ch current platform and if the platform is mature and also has a very good platform value, then, you know, the implicit cost of you know destroying the platform for personal reasons and it could be quite costly.
00:22:31
Speaker
So it's not only about the you know the fraction of tokens whales hold, actually, it's also about, you know, how valuable the platform is as of now.
00:22:42
Speaker
So that if you think about the you know young platform, you know whose market value is very low, and then it's actually handled by a small you know group of you know large token holders.
00:22:54
Speaker
And that platform actually is very risky because in those wells, even if they destroy the platform, not much to lose, right? If there are some, you know, other personal benefits, you out of the destruction, then they may want to pursue.
00:23:10
Speaker
And that's actually one important aspect of the the economic skin in the game, right? So the platform's value is very important. And also, know, regarding your, you know, token borrowing, that's very interesting. then the liquidity, if the whales easy to accumulate their, you know, votes and then easy to dump them, then, you know, manipulation is more likely. So broadly speaking, if the whales can actually have another channels to borrow tokens to strengthen their, you voting power, then, you that could be also regarded as you know, part of our liquidity.
00:23:47
Speaker
And we believe actually, you know you know, certain tokens that are famous and well recognized, you they are handled by those borrowing pools. So I think as the platform matures and then the tokens are more, it becomes popular, becoming popular and popular, then the liquidity situation, know, could be better. Better means like easy to get and easy to dump.
00:24:12
Speaker
In such a case, if the platform value is not strong enough, then I think the whales, know, could be deviating, you know, to pursue their own private benefit. So I think the in our model, we don't actually explicitly say what kind of liquidity we are talking about, but just generally speaking, if the whales are, you know,
00:24:31
Speaker
if they can easily, know, accumulate their voting power in the open market transactions, and if they can easily dump the shares, then, know, their personal agenda could be a threat, you to the small token holders.
00:24:47
Speaker
But if they are actually counter, you know, counter of forcing the counter, counter enforces like ah large platform values, or some induced large platform values, you know, through the optimal contract. i think the Jungs of us tried to say, kind of lock in effect, you know, from the, you know,
00:25:07
Speaker
the the long-term incentive type thing, that actually has a lot of you know desirable endogenous effects. For instance, if the users see, oh, these whales are committing to these platforms' long-term growth, then they can participate more and they could actually endogenously increase in the platform value that could reinforce these whales not trying to destroy the platform, given the large skin in the game, the economic skin in the game.
00:25:37
Speaker
So I think that things are quite intertwined and we actually modeled this endogenous relation between platform value and then the wealth long-term commitment incentives and then kind of the rock pool type, you know, the risk, know, private benefit, pursuing risk.
00:25:54
Speaker
And these three things are intertwined and then, you know, we made some, you know, theoretical theoretical projections, predictions about, you know, how things can, you work to strengthen, and then make this you know platform or DAO more resilient and then strongly protected against the whales short-termistic behavior.
00:26:15
Speaker
So I think that that's actually what Zhang Sub tried to say, you know building the right incentives and having the right incentives to the whales is very important. because there are many spillover effects the platform's value and overall participation level of the users, etc.
00:26:34
Speaker
So can I just ask, um I mean, I assume in particular because you you were talking about the um you know the presence of the whale and and you know the signals that are coming with this. um Are you assuming that you can actually spot the whale, which I think is true for some DAOs, but it's not generally true given the pseudonymity of blockchains, right? is that How does that work in your frameworks?
00:27:13
Speaker
Yeah, I mean, this is actually, you know, this is of course not perfect, but nowadays there are quite some platforms that would allow you to trace back who is the real identity behind those, you know, different wallets.
00:27:29
Speaker
So sometimes government agencies and, you know, individuals are able to use those platforms to kind of trace the real ownership. So, but, mean you know, of course, in in our paper, because our paper is more on the on the theory side, we are kind of, you know, being more naive when we do empirical work. So so we were treating each wallet as a different wallet.
00:27:54
Speaker
But even by doing that, it's very striking that we find that the top three wallets already control more than 60, sorry, more than two thirds of the voting power.
00:28:05
Speaker
So it's possible that maybe only two or just one entity controlling the entire dial. So, yeah. So can I, ah so one thing that occurs to me, if you think of, um I mean, as you say, right, so this is something that is visible, right? And so if you have a non-optimal contract where you worry that the whale actually will cause a certain degree of damage in a,
00:28:31
Speaker
If you think of a traditional corporate governance or corporate finance perspective, you you would imagine that if there's ah if there's a value-destroying whale, there's also a way to ah buy up the remainder of the Dow, right? So essentially do ah do an M&A, do a takeover, and then just get the maximum value back out. Is that is that something you... I mean they mean, that's outside of your model possibly, but you know we we usually think that the forces of capital have...
00:28:57
Speaker
of capitalism have some have some bite here. Yeah, I mean, you know, I think this is really interesting comment. um no as far As far as we see in the real world, sometimes there were venture capitalists who were very early on you know investing in the in the in the platform and then later on the platform became a DAO and you see that you know I actually have not seen cases where a venture capitalist or a whale would be taking over DAO, but you know I somehow can imagine scenarios like that when a platform becomes you know profitable enough this you know but versus the current valuation is you know much lower than the um future valuation, right? Then the potential valuation, you could imagine cases where a whale is going to take over the platform.
00:29:50
Speaker
using its vast loading power. but By the way, just to clarify, it because you mentioned illiquidity earlier, so isn't isn't this a case where sort of illiquidity could again matter in the sense that... ah So, I mean, I think that maybe if I'm if phone understanding Andreas' question correctly, I think part of the point is that like there could be somebody who had no holdings to begin with, as opposed to somebody who was locked in with some initial holding,
00:30:15
Speaker
And that person sees that, so, you know, this this whale is doing value-destroying activity and decides, okay, let's just buy out enough to sort of correct that um and it'll pay off because the because the enterprise will be worth more.
00:30:27
Speaker
And so in that case, though, I would have, like, essentially the need to sort of acquire a meaningful share for somebody who starts with nothing is going to be higher than somebody who already has something.
00:30:37
Speaker
And so here it seems like illiquidity could actually cause... ah sort of essentially the friction that that stops you from achieving sort of the optimal outcome. So I guess part of the question is like, can illiquidity sort of cut the other way potentially also, or is the is the channel maybe a bit different than that at least I'm currently thinking of it as?
00:30:58
Speaker
um Yes, farha ah that that's so that's correct. um so um This illiquidity is a friction that prevents value destroying whales from acquiring enough enough tokens to to affect the voting result.
00:31:17
Speaker
So so so yeah that that is that is why i I mentioned this natural shield for small users. ah So getting back to Andreas' audio question, can it doesn't matter if we cannot identify where this whale is coming from because the whale may hold potentially multiple wallets and so on.
00:31:40
Speaker
So this actually doesn't matter in the sense that ah the ah whether the whale has multiple wallets or not, the incentive is the same whether the whale only uses one wallet or multiple wallets because the Ultimately, whale wants to have enough tokens to change the result, but then due to illiquidity, the whale may not be able to do that because it becomes too costly.
00:32:04
Speaker
But then if the illiquidity is not not too much of the friction, maybe there's still some costs, but then the whale is already quite close to that the threshold of they're in affecting the voting, then whale might just go on and do that, ah whether the whale actually has multiple wallets or or not.
00:32:23
Speaker
So ah theoretically, it doesn't matter. But then, as Tao mentioned, the empirical data is a challenge when we actually test it. But but then ah also we find that in the data, it is quite concentrated. So at least we we know that there are quite a bit of concentrated whales so affecting the result.
00:32:43
Speaker
And then indeed, the illiquidity could be a really helpful um ah shield ah when when the whale is at least ah having some some buffer toward this critical threshold of changing ah the voting outcomes.
00:33:01
Speaker
So also adding a little bit, you know, so given given whales can, you know, hide behind the multiple wallets, so the the empirical measurement of the degree of concentration of the ownership in our paper will be maybe underestimated.
00:33:18
Speaker
so that, you know, the the measure we propose, you know, may, like, a properly, you know, depict the true degree of centralization, because maybe whales can be, you know, hide themselves behind multiple wallets.
00:33:32
Speaker
And I think also that in Fahad's question, actually, illiquidity has a lacking effect so that you know these wells are responsible for the longer time period pursuing you know long-term platform growth.
00:33:46
Speaker
But maybe at the same time, this is a good internal governance mechanism. But if you think about the kind of M&A in hostile takeover by other companies, Potential token holders who you know may attempt to acquire tokens and try to take over this platform from the You know bad whales Maybe that aspect, you know, that's what I call what we may call like ah external governance mechanism whatsoever, right?
00:34:13
Speaker
Maybe you know the the the the meaning of the illiquidity, you know different but we haven't thought about that M&A effects or the market for you know, Dow control, like a market for corporate control, like a market for Dow control.
00:34:29
Speaker
We haven't thought about that economics, but I think the liquidity issues, know, could be, you know, related to the many different aspects of the governance mechanisms. But here, actually, our main focus of the, you know, illiquidity is actually it creates the lock-in effect, and it creates the, you know, more long-term mystic,
00:34:47
Speaker
you know, the incentive to the whales so that they hold longer, their voting power grows, but that means that they care about the long-term value of the platform. But there might be you know other illiquidity facts you that there could be you know associated with DAO M&A. Because we actually had a general corporate finance in review paper. And then I think we broadly review you know where the DAO literature is and then what are the related issues regarding DAO operation.
00:35:18
Speaker
DAO takeover, DAO M&A, more active involvement of the crypto VCs, you they are emerging. So I think that maybe all those questions from you, comments from you could be more and more relevant as the DAO space is evolving over.
00:35:32
Speaker
um So can I ask one question though? um So I want to go back to the to the argument that you that you started with where you say the if the whale is sufficiently big, has enough skin in the game, then they would not act against the interests of the DAO. But if somebody is sufficiently large but not large enough, then that is essentially where the problem arises. so Now I'm wondering if this is not also something that we can solve with a voting power scheme. right So for instance, if there is a highly convex voting scheme where um if you are in this intermediate range,
00:36:06
Speaker
then you actually don't have enough power to move the needle enough. is Is that a way how we can think about this? Is that another way to solve the problem? yeah Yes. um so Indeed, ah quadratic voting is the kind of the scheme yeah you are talking about. so It becomes increasingly costly, right convex way.
00:36:26
Speaker
ah the The problem with the quadratic voting is ah it is subject to the so-called civil attack. Let's say a whale can hide in multiple wallets and then a behave as if and they are multiple different agents.
00:36:45
Speaker
And then because the quadratic cost is only kicking in with that you enough enough um ah ah voting tokens, they can have lots of many identities so that the cost is a very small per each top each each wallet or or each identity, and they can have more voting power and then affect the result.
00:37:10
Speaker
ah but If I can interrupt you, actually, I actually try to say the opposite, right? So you should give more voting power. So if you have an intermediate range of of votes, you actually have lower voting power, right? Because you're basically saying that if I own, let's say, and give you let's throw some numbers out. If I own 80% of the DAO tokens, then I don't want to destroy them.
00:37:29
Speaker
But if I have 20%, I may actually harm them. And so if I have 20% of the voting power and, you know, um Maybe if i if I could depress this instead of having 20% of voting power, only have 10, then that actually could solve your problem. you know it Essentially, it's the opposite of quadratic voting. right In quadratic voting, the more power I have, the less as and the more votes i have ah the the more shares I have or tokens I have, the less I actually get in terms of get to vote for it.
00:37:59
Speaker
is that me So it's ah it's essentially the opposite of what you describe. let Let me try to understand ah your question correctly. I mean, let let's let's try to understand because... not um So what I'm saying is, so ah you what in in our, um say, um ah work, what we are saying is, if they have enough ah voting power, they can just just go inch closer ah very easily and then and then change the result.
00:38:27
Speaker
Therefore, there's a sweet spot of value destroying kind of the ah changes. But then they are far away from that due to illiquidity, the transition costs, they are unable to do that.
00:38:41
Speaker
ah So you I think you're you're suggesting maybe ah some somehow this voting power could be um sort of change it according to the ownership so that that they are not able to manipulate the result?
00:38:58
Speaker
Well, there's many ways how you can structure voting, right? I mean, it doesn't have to be one token, one vote. So you already mentioned quadratic voting, right? So if I have, you know, say have a quadratic cost, so I basically have, you know, a number of tokens out there.
00:39:12
Speaker
And then in proportion to the amount of tokens that I have, there's a cost for the token. So um if I want to have 10 votes, I need to spend, let's say, 100 of my voting tokens. If I want to have five, i only have to spend 25. That's kind of what quadratic voting means.
00:39:28
Speaker
So the more vote the more tokens I have, the less power I have actually in voting. um and But you could also have, I mean, there's there's these votes where one suggestion is you have proof of humanity, right?
00:39:40
Speaker
So it's like, you know, I pay more taxes than some other people or somebody, you know, a billionaire pays more taxes. as Well, this is doubtful, but um a billionaire may pay more taxes than a normal worker, but they all have the same votes and and in a democracy, right?
00:39:56
Speaker
So there's also different ways that we think about voting there. um And what I was suggesting is, I mean, maybe this is a, sorry, yeah I don't want to, we're drifting off a little bit from your paper, but because you said early on, it's like somebody who's really big, has no incentive to destroy the platform, but if you're just in the somewhere in the middle, right, then you have enough voting power to influence the platform and to distract value, but you don't have enough skin in the game.
00:40:21
Speaker
to to kind of to to feel the pain or the value destruction that I have and that you have and so maybe it's worthwhile then you know trying trying to find that line where when you have enough skin in the game, then you get to vote a lot. But if you don't have the skin in the game, you don't get to vote to destroy value.
00:40:41
Speaker
so but but I mean, not sure if this is manageable. It's just a thought. I think I get it. So, yeah, sure. I think there may be some some room for improvement in terms of this designing sort of the or the ownership and then and the voting power.
00:40:57
Speaker
ah But ah what we're sort of so far studying is in terms of the standard sort of structure, it could be like that. But then i guess what what you are addressing too is somewhat ah closer to ah maybe... um sort of some dual class shares kind of structure. Maybe, you know, voting power is somewhat changed maybe with the ownership. And so like I guess it's maybe it's not the same thing, but then and so some kind of... and
00:41:31
Speaker
changes in the, in the not not the one one one token, one vote, but then ah somewhat like different, maybe it could be some function of the token ownership and so on. But I think it's worth investigating into. Well, I mean, you know, I mean, one of the beautiful things of the blockchain world is for ignoring, i mean, so I don't want to, you know, say, I mean, this is not, this goes way beyond what you actually discussed in your paper, but I think one of the beautiful things of the blockchain world is that we can actually try out new things, right? Or that people are willing to try out new things that have not been tried before.
00:41:59
Speaker
Although, of course, here I would say, you know, you kind of need, and before you try something out, you should have some kind of theory. and um And um I'm not sure what the right theory is here, but what assumptions, what knowledge you need to have.
00:42:12
Speaker
um But I think, Fahad, I think we had a discussion about this before too, or something about information. Maybe you want to chime in here. First, let me maybe just say something on on the on the previous discussion, and i'll I'll transition to the information point also because i think it's related.
00:42:27
Speaker
I think sort of at a high level, maybe ah kind of what Andreas was trying to highlight is that, of course, there's a distinction between sort of – voting shares and actual shares. And I'm using sort of the traditional language here to make that traditional analog, right? um and And maybe this is something that I think actually in practice isn't sort of thought about enough.
00:42:46
Speaker
um And and and your your framework is probably something that could be um enhanced to sort of to to capture that ah trade-off, which is a really important one because, again, in in practice, I think people haven't thought enough um about the distinction between ah between those things.
00:43:03
Speaker
um But ah sort of taking it to a slightly different place ah that Andreas was alluding to here, um you know on a prior episode in a prior season, actually, we had ah Jerry Souklus and Brett Falk on, and they were talking about actually things like quadratic voting, but they were talking about it in the context of um when the votes...
00:43:26
Speaker
ah ah when there is information, basically, and people are differentially informed about, you can think of it as the quality of the vote, right? um And so I'm just kind of wondering sort of how that sort of would fit in the way you're thinking about things in this work, in that um I guess one way of saying it is that ah you could have settings where ah the votes are...
00:43:55
Speaker
ah are in so are are the type of votes where information is very important in terms of achieving the right outcome versus a situation where it's not. And in the case that it is, then you might imagine something like endogenously, you would have whales because it's the people who sort of ah understand the the the context well, who are holding large shares.
00:44:18
Speaker
um And so, i you know, really, i guess this goes more to the empirical side of it, because it's possible then in those cases that seeing whales in the data is not indicative of something to be ah too concerned about.
00:44:30
Speaker
um ah but rather is just indicative of sort of, yeah, like the votes in this setting are um are things where information is quite significant. So if you could just talk a little bit about maybe how how you would think about information in in in the in the context that you're that you're studying.
00:44:47
Speaker
so So maybe I can make quick comment and then ah you know ah my colleagues of mine can also make a comment on empirical side too. So one one ah one comment actually from theoretical side is that ah this illiquidity is naturally related to this informational side.
00:45:05
Speaker
So if there are um more information asymmetries, of course, ah we we know that illiquidity will be greater because investors want to protect themselves from adverse selection and um ah work from information asymmetries, right?
00:45:21
Speaker
So ah as you mentioned, quadratic voting ah can can help mitigating the the effect of the um ah the ah large token holders potentially want to change the result in in an adverse way.
00:45:41
Speaker
ah What we are saying illiquidity is kind of ah natural equivalence to quadratic voting. So as I said, quadratic voting, it could be actually vulnerable to civil attack.
00:45:54
Speaker
But then ah the just the usual token ah ah to and voting ah with illiquidity could be just like quadratic voting because as you acquire more and more tokens to change the result, ah the result that the the the cost could go go up ah very ah high. ah in In fact, contextually, if there are actually more even more information estimators, let's say ah the the whale knows something and then wants to change things in their own way and then just um um ah pursue private benefit, then this could lead to even extra illiquidity.
00:46:31
Speaker
And then, so so you you can you can as as I said, you can imagine this illiquidity as as but kind of some form of quadruple voting in a natural sense.
00:46:43
Speaker
sense So that's one one thing yeah one one one then you can think think about in the theoretical side. But then maybe, as you mentioned, this is also empirical question. So maybe um colleagues of mine can comment on this empirical.
00:46:59
Speaker
Yeah, so I think ah you know on the empirical side, it'd be interesting to directly test right in which cases, in which votes, information symmetry is more more of issue versus versus the other ones.
00:47:14
Speaker
However, you know i mean we we we sort of conclude that it is it is not difficult you know it is somewhat difficult to classify those different platform you ah votes because different platforms have you know different sets of proposals.
00:47:31
Speaker
Some platforms like DeFi platforms, like lending platforms, MakerDAO, their proposal is mostly about you know risk parameters, which are very technical, versus if you think about a generic, a smaller DAO,
00:47:46
Speaker
the proposals could about all kinds of issues, right? You know, I mean, some proposals are even but about the community issues. So, you know, we just decided that X-Entee, it is not that easy to classify all those different proposals based on information symmetry. So, you know, what we did was we just look at all the proposals dropping out you know, really irrelevant ones, right? Proposals that only had one line, proposals only had somebody's names.
00:48:16
Speaker
So, but even if just looking at the average proposal, we we do find ah strong evidence supporting our see theoretical predictions. For instance, we do find that, you know, concentration, ownership concentration proxied by Herfendaal index is negatively related to platform growth going forward.
00:48:39
Speaker
And we also find that this negative effect is mitigated by by larger platform size and more token inequity. And we also find that when whales are able to pursue long-term interest, you know, proxied shifting from, you know, regular governance token to two staking platforms, you see that platform valuation jumps.
00:49:06
Speaker
So, but, you know, this is, of course, very coarse, right? Because we're just looking at the average proposal, but going forward, and we can potentially, you know, use a good algorithm, use a good argument to sort different proposals and see which proposals would be fitting into ah more informational frictions versus others. So, you know, this will be very interesting future direction.
00:49:28
Speaker
let Let me also add back to the Fahad point. I think that's very excellent point. We actually, our paper, I think maybe this goes back to, know, my paper with Tao, you know, earlier, the information cascade or, you know, some consensus dynamics in the Tao proposal voting.
00:49:46
Speaker
and For sure, the informed whales, how they actually communicate about the proposal, you know, that's very important because think about the small users, they may not be equally informed, you know, as Wales.
00:49:59
Speaker
So, you know full democracy by small users with no substantial information may not be an ideal situation compared to you know, some three or four whales, you know, who are very well informed, you know about the impact of the proposal.
00:50:17
Speaker
So, you how, you know, and the the proposal, you know, game starts and ends, you know we haven't looked at it in our paper, but for sure the blockchain, you know, marks every, you know, dynamic transition of it how, yes, or no vote has been accumulated, you know.
00:50:35
Speaker
so So I believe, you know, maybe not the democracy democracy, you know, of users who are not informed, you know, that may not be the most ideal you know how outcome, but maybe some sort of a the, you know, plutocracy, you know, some three or four whales who are very well informed and they obtain consensus.
00:50:55
Speaker
So that could create some sort of, the you know, information cascade, you know, during the dynamic voting game. that voting dynamics could be leading to a better outcome.
00:51:06
Speaker
But in our paper, we just take the snapshot of the voting outcome per se, per proposal, as Tao said, and then just construct a HHI type concentration index and see how that end the game or the snapshot of the voting outcomes are correlated with the ah future platform growth.
00:51:26
Speaker
But actually what I'm trying to add is how we get there, how we get to the voting outcome, you know and how agents are interacting, communicating, obtaining consensus in on the ongoing proposal. I think that may add more value.
00:51:43
Speaker
and But in our current paper, we are not looking at it, but maybe we can take a look at that in in our subsequent research, for sure. Yeah. Yeah, so i mean that that brings me to another topic here. So is um there's actually a lot of delegating of votes going on too in DAOs, right?
00:52:00
Speaker
which, um you know, so if you think of information aggregation, that that's one thing. If you think about not having the time to think about I mean, some of these DAO proposals happen at very high frequency, right? So they happen every other week or so, or even much faster, right? so you you really can't take, even if you have an interest in the platform and the loose ends succeeding, you don't have the time to actually, you know, do all of these.
00:52:24
Speaker
So how would that affect your results? and Does that take away some of the power of the whale or because So, I mean, in fact can we just take a step back, generally speaking? I think the participation in DAO votes is often relatively small, right? So to reach even quorum can be hard for them, right?
00:52:43
Speaker
So ah there then if they if people actually are not even exerting their power to vote because it's not worth their time, then, you know, whales can easily take over. with With vote delegating, that's actually harder, right?
00:52:56
Speaker
So do you have any sense of how this would work, how that affects your results? I thought, Andrea, I thought denigation would even increase whales' concentration because, for you know as you said, smaller holders have very little interest in voting because because the votes are not going to influence the outcome.
00:53:16
Speaker
So smaller holders are more likely to denigrate with you know larger holders, and those larger holders we classify as whales. um you know But, you know, but i know of course, there's a there's a host of issues here because, you know, you you actually could imagine ah some and some smallholders are not naive. So they are they are looking at which whale is actually doing a you good job, voting in a way that's consistent with with with the smallholders' values and, you know, stuff like that. And then, you know, they are more likely to maybe
00:53:54
Speaker
um dedicated with you know better whales per se. Right? So, I mean, this is sort of the has, no i mean, interesting, I think this has a link with the ah traditional corporate governance in terms of voting choice programs, you see a lot of the, you know, big asset managers like BlackRock, Vanguard, and now are coming up with those voting choice programs um in the sense that because they are criticized by you know, other investors like pension funds or or even governments, that they they actually have too much power, right? You know, if you think about in the U.S., those big three asset managers,
00:54:34
Speaker
control more than 30% of the average stock. So um then they are now open up at the gate to smaller holders. You know, if if say you and I invested in the BlackRock ETF, we are likely to have voting power now.
00:54:51
Speaker
because voting power will pass through to us. So, I mean, in in my opinion, the, you know, delegates are sort of like those menus, ah different voting policy menus adopted by those big three asset managers.
00:55:08
Speaker
So ah then small investors, smaller token holders are choosing their preferred delegate. in terms of casting their vote. So I think this is a really interesting topic. This is really not in our current paper, but going forward, you know, it would be really interesting to study that potentially relate to our current model.
00:55:28
Speaker
And also I think regarding Tao's point, in the in the repeated del vote delegation game, I think these whales, you know, tell they they need to care about the the small users' feedback.
00:55:42
Speaker
And I think the vote delegation mechanism should encompass, you know, and how they are dynamically evaluated by the small users or small token holders who delegate their votes, you know, to these whales.
00:55:56
Speaker
So if they, misvote or they make the wrong judgment on but on behalf of the small users, then there must be some way to penalize those, you know,
00:56:07
Speaker
misbehavior. Then I think, you know, eventually these whales care about the small, you know, the voters, small users, you the incentive or there their goal so that vote delegation achieves, you know, its desirable goal. Otherwise, maybe they can tentatively deviate from the you know global objective just to pursue their private benefit.
00:56:31
Speaker
So, But I think that how to design the vote delegation mechanism, I think it itself is very important. So I think one important thing here is the the landscape of the Dow at the moment.
00:56:44
Speaker
So stock market is heavily institutionalized now. So the small investors are ah the vote through delegated voters and so on. And then and there are large institutions um managing individuals' money and so on.
00:57:02
Speaker
ah they They vote on behalf of the those small investors. But Dow is not as heavily um dominated by institutions yet.
00:57:13
Speaker
So they are mainly um ah joined by ah many so small individuals. ah ah so So that's why Wales impact is much greater ah than the stock market.
00:57:26
Speaker
Of course, on the way forward, maybe there will be more institutional ah participation and then things may change and there will be more delegated voting and so on. So, um yeah, ah so we'll we'll have to see how the landscape changes in the future and the future evolution.
00:57:44
Speaker
Yeah, I mean, one thing that I remember, um I mean, Tao and I were were at the conference um from John State University at the weekend, right? And there was actually a discussion about that a number of venture capitalists have funded DAOs.
00:57:55
Speaker
And so AZXC is one good example. They actually um deliberately stay out of DAO voting, right? um Just for, I think, political reasons or for just conceptual reasons.
00:58:08
Speaker
um did I mean, as we know from corporations in in in the US in particular, there's a lot of things we have to ah worry about, let's say large pension funds or large funds ah having a lot of power in multiple different boards because that leads to actually anti-competitive behavior.
00:58:23
Speaker
um That's obviously too much for this particular um for this particular ah podcast, right? Because it's a completely different topic, but I think this is something maybe also worth mentioning here.
00:58:34
Speaker
Farhad, you have a question? Yeah, so um I don't want to take us too far afield, but I but i feel like I need to ask, given this last ah little bit of the discussion. um So Jungsook made reference to kind of how in traditional markets that you know there are a lot of, ah let's call them retail investors, uninformed investors who are just sort of delegating to you know whoever their financial intermediaries are and so on.
00:59:03
Speaker
But kind of one of the cool things about blockchain is that there's so much information directly on the blockchain, right? And of course, we wouldn't anticipate that retail investors can process all of this information by themselves.
00:59:20
Speaker
ah But there are now you know legitimate players in the AI space that are thinking about things like, well, can't we essentially create a tool that essentially uses the information on the chain to help ah users do various things? And so I'm not i'm not talking specifically necessarily about DAO voting, but that would be one application in that broader umbrella of like, okay, look, there's a ton of data here.
00:59:46
Speaker
We can essentially create a tool that will... that that will ah help users do things that are that are otherwise too sophisticated for them to do. And so a bit of a spacey question, but like can you maybe, thinking forward,
01:00:00
Speaker
i what do you think is best practice for, uh, sort of retail investors who are generally uninformed about these voting proposals?
01:00:12
Speaker
Is there direct, and in some sense, like as we, you know, as new projects launch, should they be internalizing sort of the idea that maybe some of these, uh, retail investors might actually be able to access tools like that?
01:00:24
Speaker
Um, but yeah, so if you could just talk a little bit about kind of how you see the positioning of what we might think of as uninformed, uh, users going forward given other things going on in the technology landscape.
01:00:37
Speaker
So that will be true a perfection of decentralized organization. So if we actually do delegation, ah that that's actually going against the decentralized to to some extent.
01:00:52
Speaker
um in Although it achieves some efficiency, um there is now some some layer of agency problem. yeah the The delegated the party may not exactly...
01:01:05
Speaker
um ah represent your needs. I mean, everybody may may have very different needs and then and they have to somehow kind of trade off between ah achieving some benefit at the cost of giving up some very individual needs.
01:01:24
Speaker
So maybe with this AI-based or or you know very data-driven decision-making kind of tailorize for each individual can push things toward the idealized decentralized organization where every and individual ah participant's needs are reflected, although they may conflict with each other. At the end, it's aggregating everybody's needs, kind of um pushing toward the um and sort of the improvement of aggregate welfare.
01:01:58
Speaker
So ah that my view is that that maybe with that, that we ah we may ah improve the the result of the ah the voting outcomes.
01:02:08
Speaker
So, you know, Fahad, to me, Your you're mentioning of those AI agents in terms of aggregating information, for instance, for voting, these, in my opinion, are sort of like those different menus offered by those big asset managers. right you know Imagine that we actually could have two AI agents. One is ah for you know Democrats, one is for you know Republicans in terms of how they vote on a particular proposal.
01:02:40
Speaker
And then, um you know, people sort sorting to those different AI agents and, that you know, the the ai the AI agents would come up with, you know, a you know different set of votes for for the same proposal.
01:02:55
Speaker
um So I think this could potentially to avoid, you know, whales getting even larger due to vote denigation. So in the sense that if we can have enough of those AI agents or enough of those menus,
01:03:09
Speaker
know this could be a desirable outcome. now However, the the design of those a agents would be also tricky, right? you know how you How you design those things you know in terms of what the information you actually want to incorporate, right? Stuff like that. So, yeah, but it does this is just kind of thought.
01:03:29
Speaker
All right. um Maybe it would be useful at this point. um if you If you could just, I mean, we have talked about, we've covered a lot of ground on possible, you know, ah you know, but what things that go way beyond your paper, but maybe it's just useful for the audience if we could just, um in ah in a sense, if you could just tell us from your perspective, from your finding, and in and based on the particular setting that you study, what do you think is the take home, particularly on a policy dimension?
01:03:58
Speaker
So, you know, you start with a situation, you have a you have a a whale, you may have a market where people could ah borrow NAND tokens or buy tokens and accumulate holdings.
01:04:09
Speaker
So what do you think is the problem and how do you think you can solve it given what you that did in your paper? What would you do in terms of design? um So let let me just summarize what what we have said maybe in in two lines first.
01:04:24
Speaker
First, um prevent short-term opportunistic behavior. Then the then So that's that's one one thing. And the second, promote, which is kind of the flip side of the the first thing, promote the long-term commitment.
01:04:41
Speaker
So in that way, we can we can resolve conflict between a whale, ah ah who may pursue private benefit, and the small users, who who who may be exploited by these big ah users or or the whale.
01:04:59
Speaker
So ah naturally, um the the policy implication could be ah ah promoting this long-term commitment of the big players.
01:05:10
Speaker
um but ah you all Earlier you mentioned ah asked, okay do does it have to be committing before the voting or after the voting? ah It could be various forms.
01:05:22
Speaker
um So, ah for example, already vote escrow is ah used and that could be one one form of that. And then ah This is related to, let's say, other form of voting, such as conviction voting.
01:05:38
Speaker
ah Let's say voting power is increasing with the time they committed. But whatever form they choose, um but the principle we want to emphasize is, as I said, two things, prevent short-term opportunistic behavior and then commit ah promote long-term commitment.
01:05:56
Speaker
And also adding to Jang Suk's point, and in our model, actually, we only have one whale. and then the the group of optimistic users. But as Fahad mentioned and Tao mentioned, I think maybe multiple informed whales, they and i actually exchange their ideas, they communicate, and they actually promote more informed decisions you know on this DAOs platform.
01:06:23
Speaker
yeah That could be the most ideal situation so that, you know, Full democracy democracy with zero information may not you know lead to the waste of a crowd so that DAO will not make any informed decision.
01:06:37
Speaker
maybe the vote delegation or in a couple of central news or central wallets like Wales, they are informed about the platform and then they might represent, you know, small users, you know, interest more broadly.
01:06:50
Speaker
And then if they can actually, you know, climb in their opinion on a couple of important ideas in the issues of the DAO's growth, maybe that's sort of the, you know, I don't know, don't want this is right term, you know, plutocracy or oligarchy, you know, type more like old politics type, you know, the things, you know, maybe, know, that could applied to the Taoist space. So, you know, a couple of, you know, group elites, they actually represent the others with the information,
01:07:24
Speaker
and then make the efficient decision, make test the you know the more practical that goes. But you know the the more research has to be done empirically, theoretically, and then you know we are actually you know gathering more data so that maybe our team can contribute more on this issue.
01:07:40
Speaker
Anyway, so that's that's actually quick addition of my thoughts to the John Sech's punchline of our current paper. And doubt maybe Dow can add more. Well, what I hear, if I may just interrupt for one second. So what I hear is actually, is is is it's sort of a certain tenor from um all of you, is that really a Dow should think very long term for if you if you're a token holder. if you if so I mean, there's lots of things with Dow. One of the things we saw, they give out tokens as certain types of incentive.
01:08:07
Speaker
um but oftentimes without any long-term commitment. and So if you give somebody an incentive, there should be actually something that you have something to do with what's coming out. And I think this is was probably one of the biggest failures.
01:08:18
Speaker
ah Really not a failure, but it's sort of like, it wasn't, you know, On the one hand, there was a lot of talk about that you want to get people involved and want to have them, you know, you give platform users sort of a skin in the game. But then it wasn't really a thing, right? It was because you could just flip the tokens and just get the money, right?
01:08:37
Speaker
So it was always a little bit more of ah you know, ah how do you call it? It was gaslighting people really about what's going on. So what you're really after, when you do DAO voting, you and if you if you want to be serious about a DAO, you have to think about the long-term path.
01:08:52
Speaker
It has to think about power of voting and also about long-term commitment to it. I think this is kind of, I think, a really big takeaway here for me. But maybe Tao, you want to add to that? I actually don't have much to add, but, you know, because both Johnson and Johnson actually gave, that you know, also, you know, Andrew, you guys all gave a really good summary of our paper. But, you know, I actually just want to say that at this point, it's not clear that DAOs,
01:09:21
Speaker
you know are there for long term because of various risks. you know you know if If you think about it, in terms of ni legal risks, we do have a lot of legal issues here. right A lot of the legal issues are not really to defined.
01:09:35
Speaker
I guess in the past few years, there were several landmark legal cases involved you know regarding um the, let's say, so-called fiduciary duty of users involved. And in some of those cases, ah in some of those cases, judges say that DAOs are more like limited partners and limited partnerships, meaning that users are users involved could be treated as limited partners.
01:10:06
Speaker
And I think this is sort of a shock to to a lot of people who really believe in DAO. and um you know And also other the issues you know involved and like how actually do you bridge off-chain world versus on-chain world especially you know when we talk about certain types of DAOs right? DAOs who invest in real assets and stuff like that. So I think all of these issues sort of threaten the long-term viability of the DAO space and hopefully the the you know current the current administration can come up with know more guidelines regarding DAOs and you know we we know that that they're coming up
01:10:47
Speaker
you know The SEC is coming up with guidelines regarding crypto investing, etc. But it seems like there's really not enough discussion about DAOs particularly. so Yeah, i think I think that's actually a very important point that you that you're concluding with there. which is so you We can all think about the economic perspective and and the benefits of it and and how to design it properly. but um I would argue that there was, i mean, especially about about ah from the last administration, there was a viewpoint in which you would try to take you know innovation and novel ideas and try to interpret them in such a way that you basically prevent them from unfolding their potential.
01:11:26
Speaker
and So if you just take the worst possible interpretation of what could come, and this is in many ways, I think it's a problem when you have relatively... lose legal frameworks where can be interpreted in a way by somebody who is just dead set against some form of innovation in a negative way.
01:11:45
Speaker
I think that's always a problem. This is why you need to have some form of clarity around this. um But I think as as economists, and I think our contribution can be, and and I think you made one there, very important one is to say,
01:11:58
Speaker
look, here's some some issues that arise in you know given the structure that you're aiming for, and here's how this can be solved. And in a way, i think it's maybe this is actually something I want to conclude with this.
01:12:10
Speaker
I think what you're proposing is the the solution that you're proposing has nothing to do. doesn't require regulator. It doesn't require anybody to write a law. It requires a design space in terms of how we want to organize our our organization, which is visible to all and is therefore also comprehensible to all.
01:12:27
Speaker
So you don't actually need somebody to and to to to chime in on that one. um It's just my take here, but by all means. um i think I think that's probably the way we want to conclude here, Fahad, unless you have something to add. um i want to I want to thank Junsook and Jongbu and Tao for your time and for explaining this to you work.
01:12:52
Speaker
We hope you enjoyed this podcast. Thank you for listening. As a reminder, you can find additional materials on owlexplains.com and can stay updated by following us on social media. That's all for today.