DAO Treasury Management, Performance and What Comes Next

Transcript

further Ado let's bring on our first Speaker uh as I said he will do a talk called Dow treasury management and performance so give a huge round of applause for Andy Pavia hey everyone super happy to be here at eth Barcelona presenting uh the Dow treasure and compensation um performance analysis uh this was an analysis created by the pal team before we get started quick introduction of what Pala is the Dow orchestrator uh we are are a One-Stop platform uh for Dow operations and tooling we essentially help uh web3 ORS manage saves Integrations automations and credentials all in one place um and we've essentially been looking at how dalls

are doing with what's happening out in the market uh We've definitely gone through a very wild ride uh it all started where uh Mr J Powell and the FED changed the monetary uh policy started going very hawkish then soon after that we had uh Terra collapse I think we all remember that closely after that we had FTX you know everything was gone uh with that a lot of things went down in the market not great for us and then after that we have Mr Gensler right here seeing Securities everywhere so that's been uh quite all right for us however even with everything that is out

there even with the market not moving um we're still out here being building stuff and the beauty about decentralization is that it doesn't matter what's going out there as long as the infrastructure is there the blockchain is there we're going to be here just building so that's fantastic now the really bullish thing is that if you analyze at um if you just look at onchain essentially data of instantiations of multi6 across a bunch of different even M chains you can see that there's more and more safes safe multi being instantiated on chain which is super bullish um before we Deep dive into the analysis like I just want to go quickly

over the methodology and some disclaimers I'm not going to go super deep into how we do things just because we don't have that much time um but essentially we're looking at onchain information um from safe multisig using tools such as dune chat GPT uh obviously a lot of Jupiter books uh python uh we're using debank for everything related to defi positions and uh we're essentially aggregating it all plotting it and running some statistical analysis on it uh so if you like like data I hope you're going to like the stock some disclaimers we're as good as uh dbank is for essentially defi positions as we know there's an ever expanding set of smart contracts out

there they're all their own Financial in instruments and they can only keep up with so much so small disclaimer there another small disclaimer is that we're treating each multisig as its own entity it might be that there's many of them connected we're not connecting that so that's just quick disclaimers now let's go over the first questions uh what are the assets that the Dows are holding um so that's interesting question but we're going to segment it a bit further uh the first thing is that we're going to separate it between BS that hold or multi sigs that hold 5 million to 50 million and the ones that hold more than 50 okay second thing is that we're going to essentially

Define um uh three Market Cycles bullish which is in green bare which is in red and then sideways which is in yellow um on all the graphs you're going to see the end date so when you see the end date you can think of that market cycle um then we're going to classify tokens according to uh four uh different categories first one is just ether native token to ethereum U then we're going to have native tokens so when when I say native tokens I want you to think of like okay if we're analyzing let's say balancer you are thinking of the balancer token the next one is stable coin so here you have usdt usdc die B USD and then everything else that is not

those three we're going to refer to ERC 20 tokens we know that stable coins and Native tokens are erc20 tokens but yeah that's the nomenclature so let's look at the first graph um this is essentially the token analysis so we're looking at the dollar values across the different Market cycles for entities uh or multi that have 5 million to 50 million um I would like to point on the on the bull market uh that a lot of it was essentially in Native tokens uh and they also had a lot of rc20 tokens and you can see how the dollar value goes you know super declines super heavily in the next two cycles if we compare this to essentially

entities that have more than 50 million uh I want to point out two um differences here the first one is look at essentially uh the native tokens the positions here are way heavier and also you can see that in the last uh sideways uh Market C they have smaller positions in ERC 20s that means that they're keeping mostly their own native token um some stable coins and some ethereum now let's look at the percentage in which each token uh category predominates on the other one we were looking at USD value here we're looking essentially if we grab a multisig and we see that it mostly holds its own native token we classified it as

native token and then we create percentages um so this one is the one that is 5 to 50 million and this one is the one over 50 million and I want to show essentially one big difference is that native tokens on the larger ones predominate heavily if you compare that to the previous one now if we look at a stable coin analysis um we also find a big difference here this one is the one that is 5 to 50 million you can see that usdc predominates here quite heavily usdt is not as prevalent if we look at the larger uh entities you can see that a lot of them are using mostly usdt so

couple of theories here might be there's just not enough volume for usdc however the smaller entities are probably looking as usdc as a less risky stable coin due to you know what's happened with tether in the past now the next question how have Dallas endured the last Cycles from a p&l perspective and when I say pnl think uh Val Val difference within the the multisig um so this is essentially a box diagram so what's within the Box uh the upper Edge and the lower Edge are essentially the 25 and 75 uh percentile uh and then the flags at the top are essentially the lower and the top

percentile and whatever is outside of the box is essentially outliers um so you can see that from the 5 million to 50 million um on the bull market most of the entities essentially made money uh so you you can see there in the value difference uh value difference think uh if it's 500 is 5X um during the bare Market the majority of them essentially lost money there's a couple of outliers that essentially did really well as you can see by the red dots um and also during the sideways Market um a lot of the entities are making money however the 50 percentile is 3% so and and then you have a distribution that some of them are losing money some of them have

lost all their money if we look then at just the net sum value Difference by period so this is if you grab all the essentially uh the multi uh and then you aggregate them by total profit or loss uh and you essentially aggregate together you can see that in aggregate during the bare Market the smaller entities lost $8 billion which is insane they also made you know some money during the bull market but compared to how much they lost during the bull market it's nothing if we compare now the box diagram between uh the larger uh entities so the ones that have more than

50 million we can see that essentially both of them if you look at the 50th perile they're both very close to losing 85 84% of their treasuries which is catastrophic um and if we look at the sideways Market essentially the bigger entities are doing much better if you look at the 50 percentile 33% of them are up compared to the smaller entities the 5 to 50 million entities uh so that means that for some reason the larger entities are able to recover or do much better in this sideways Market might be kind of like a power law essentially the bigger right now everyone is slowly getting into the market and they feel

more comfortable injecting money into the larger entities and if we aggregate the 50 entities right so if we aggregate essentially the money that they made or they lost we can see that they lost $30 billion the larger entities compared to 8 billion of the smaller entities smaller entities is a much larger set um now the next question is there a correlation between transaction volume and and assets under management um so for this we made a little ratio uh you know we have yearly transacted volume so onchain transactions over the total value locked

so we created essentially this ratio to try to figure out how active these entities are uh and we were asking the typical question of like okay is the entity that is trading a lot more often and it's moving positions a lot more often are they actually more performant or not does more activity lead to more losses and is there any sort of connection here so this is a distribution of the r of the ratio that I just described uh you can see that you cannot have anything below zero obviously um and then we see a wide range here on the sort of ratios that we have um funny enough and I thought this was very curious the red one 2021 that's the bull

market obviously a lot of activities uh on the smaller entities we have some that were essentially trading 45 times their tvl which is very very uh aggressive uh but during the bull market there was a lot more sorry during the bare market so the the the green color they were even more active which I that I I think that is pretty pretty curious uh and then during the side sideways Market they're not trading as much this is essentially a k means cluster analysis where on the y axis we have the value difference or what we're saying

p&l and on the x axis we're looking at the essentially the ratio again the ratio the limits uh how aggressively they were trading uh and you can see that at least from the 5 million to 50 million entities the ones that didn't trade as much were able to have have a better performance in terms of p&l and there's a lot of them that traded quite a bit and then they lost money and then if we look at the entities that have more than 50 million um there's one difference here is that essentially they're all trading way less if you look at the scaler of the ratio

uh on the previous slide it got up all the way to 150 here the r the the scale is 45 uh so they're just essentially trading a lot less probably because they're holding liquid assets or assets that are not as liquid so they cannot trade as often and if we look at the K means analysis there's there's one green one over there uh that you can see that it traded a ratio of 10 times so it essentially uh moved 10 times its own uh total value locked and it did essentially uh uh 10x so this it's an outlier of someone that's really or a group of people that are really trading and still being able

to capture uh a very good p&l however if we if we look at the correlations there is no statistical uh essentially the measure is not statistically significant and we can't say that trading more leads to making less money um so yeah that's not something that we can say that is statistically significant yeah so some conclusions um you know in terms of like activity uh the average ratio was consistently greater than one for the entities that are between 5 million and 50 million

indicating that there is high activity uh the activity for the larger multi6 was less so and we cannot say there's a clear uh correlation between essentially the ratio and performance um the multi6 themselves are highly highly highly correlated to the market from a token composition um there seems to be increased focus in Risk Management in the the later periods and you could say there's some diversification however that diversification from a correlation perspective on the p&l doesn't really hold true but just more from like each token has its own uh risk operational

risk you got different teams so essentially if you have different tokens you are decreasing that that um individual risk for each token um and we've seen that essentially usdc among the smaller entities is more widely used and with that uh we're about to launch a report where we essentially go a little bit deeper into the methodology uh and the sort of tools that we use and with uh essentially with more conclusions um yeah so please scan it um would love to hear what you guys think about the report and if you can also um give us a follow we do a lot of essentially analysis into uh Dow tools

uh what Dows are doing performance of the market uh and essentially different techniques that um Dows or Dow tools are using yep and with that would like to open it up for questions hello so in the bull market we saw many a lot of those um increasing their treasury value mainly because of the increasing prices of the Native token correct and we also saw a lot of Dos disappearing because of in the bare Market or in the bull market they didn't

manage to convert their treasury position more into a stable in a good balance yep and that was the main reason why many of them disappear so well or or some of them they didn't survive at the same time when you have a a dow and your token is not as liquid as you would like to you cannot go and just simple swap them into Stables one correct what is your advice or what would you say if tomorrow we have a next bull market and there are many dlls again um maybe mainly maybe doing the tokenomics and they really want to have this treasury

issue that you are very exposed to your native tokens what would you recommend to not to or what would be your lesson learn in this bare Market yeah so that's a really good question I would say two things uh the first one is essenti create a future prediction of what your expenses are going to be right calculate for the traditional Market cycle so let's say five years make sure that out of those five years you have stable coins to essentially be able to pay your expenses and then leverage your treasury in a smart way so you can still kind of reduce the supply out there of your own

native token because that's how you incentivize your people and that's how you incentivize the team and that's how you incentivize contributors so I think it's essentially uh you got to grab a chunk just to play it very conservatively and make sure that you can keep your operations running and then be smart about when you can essentially essentially get some liquidity from your old own token without um affecting the the vector of your token in terms of price exactly exactly yeah it's a tricky one yeah yeah yeah yeah yeah so I think kind of dividing treasury into chunks uh and kind of

getting people that know how to get yield on certain opportunities where you do the proper third party risk analysis uh I think it's the right thing to do yeah meing Which hypothesis when you were measuring the pnl of the different saves Which hypothesis did you do in terms of the cash flows that come in and come out you just look at the value at the beginning with everything and the value at the end with everything or yeah yeah right now it's value difference I think there's a lot of things like let's say you need to well first of all we're taking a

couple of assumptions first of all uh the this ones are Dows so they're not moving money outside of the unchain perspective which for a portion of them it can be not true right it's still a model because you will need to essentially be able to classify deposits transfers and withdrawals to be able to compute a true p&l that's why when I say pnl I mean value difference uh but I mean you would need to essentially be able to create the whole mapping from a financial perspective to say this is a true p&l right so essentially what we're doing is we're looking at the beginning of the period end of the it what's under the control of your multi and with that we're grabbing all

this data so if there's new Investments or there's payments of salaries that's all included there so it could come from external for yeah exactly exactly which we could say that the p&l of your organization also takes into account the expenses um but if let's say if they're not Dows and they're using multi6 and then they're putting things into centralized exchanges we don't see that right but I we still think it's a nice model to kind of from a high level what's going on and what are some behaviors that we're seeing across uh these two different sets of entities that we're analyzing all right well guys thank you very much for uh joining hope the data

was useful