If you would like to view the full recording of the live Crowdcast event, you can find it on our YouTube channel. You can also read the full Quarterly Report by Sean Butterfield here.
Participants:
- Sean Butterfield – Research Analyst, Messari
- Sami Kassab – Research Analyst, Messari
- Getty Hill – Co-founder, GFX Labs and Member, 2021 Compound Grants Committee
Sean Butterfield (00:07): We are live! It’s official! So yeah, thanks everyone for joining us on this Compound quarterly analyst call. I just want to let everyone know, if you guys have any questions feel free to drop them and we’ll try to address those towards the end of the call. Quick introduction, my name is Sean. We have Sami here, and we’re both research analysts at Messari. We’re here today with Getty Hill from GFX Labs. Very kind of him to be joining us here again today so thank you for that. If you want to just give a brief introduction of who you are, what’s GFX Labs about and kind of how you guys fit into the puzzle of Compound.
Getty Hill (00:49): Yeah sure thing, thanks for having me here. I’m glad to be invited back for a second time to go through the numbers here on Compound over the last quarter. So myself and GFX Labs have been involved in Compound now for definitely at least two years, solidly. We’re responsible for a lot of the fun governance stuff in the protocol, and one of the general movers and shakers on there as well. Definitely check out the other work that we’re doing in general in the space. We have Poppy, which is a neat crypto credit card product that we’re developing and the game Etherlands, but we love to dabble in DeFi and governance wholeheartedly as well. So you can see our work on Compound by checking out the forum over there or on Uniswap and whatnot as well. We’re in all places for the most part, we really enjoy participating in DeFi and the wild space of crypto.
Sean Butterfield (01:41): Awesome, yeah thanks for that overview. So yeah, basically the focus of this call today is really to kind of dig into some on-chain data, get a better understanding of what happened over the quarter and identify any changes in KPIs, some key changes to the protocol, and overall market trends. So, before going into that I just want to give everyone, maybe some new viewers, a quick recap of what Compound is. A leading decentralized interest rate protocol, serving as a marketplace for lenders and borrowers of various assets. Before digging into the macro, I kind of want to give a quick crypto market backdrop. Generally the overall crypto market in Q4 hit all-time highs, which largely drove a lot of the activity in the previous quarter. As we rolled into Q1 we saw some volatility, shaving $1.4 trillion off the peak back in Q4 to around $1.6 trillion in Q1. Revisiting those 12 month lows that we saw back in july of 2021. I think as we understand that a little better it’ll kind of help drive and appreciate the narrative of the overall Compound sentiment over the last quarter. It appears leverage increases as the markets experience bull cycles. So, as we mentioned, the market cooled down in Q1, and that’s a really good segue into going over some of these macro metrics for Compound. I’m going to share my screen here so bear with me. Are we all good here? Awesome. So yeah, you’ll see outstanding loans and deposits are really the driving factors of Compound. As you can see we saw a pretty substantial decrease across all indicators here. We saw about a 42% decrease in outstanding loans. Coming from about $6 billion to about $3.5 billion. Largely driven by the borrower appetite, the demand for leverage decreasing in Q1, and going down. Outstanding deposits dropped about 32% coming from $14 billion to just shy of $10 billion. Originations also took a pretty big hit, just over 50%, going from $11 billion to just under $6 billion. Subsequently, deposits also were down about 80%, a pretty nasty metric there. We saw $63 billion last quarter drop to about $13 billion and change this quarter. Really the only metrics that saw any light were liquidations, but at the same time probably not favorable for most of the users of the protocol. We saw about a 300% increase in liquidations from about $34 million to $134 million. I think that was mostly driven by the wETH markets on Compound. Digging a little deeper into the net interest, income, or the financial I should say. Total interest income was slashed by about 50% from $90 million in Q4 to just under $42 million. As the loan demand decreased, depositor demand decreased which ultimately decreased interest expense from about $80 million to about $37 million giving us just shy of $5 million on that interest income. Q1 there was a pause in the grants program so there weren’t any grants paid so we really just have an unchanged net income amount of $4.8 million. Going a little further down we have the token incentives paid. Similarly as the loans outstanding and the deposits outstanding decline, we’ll definitely see a decline across the token incentives being paid. So that fell from $55 million to about $26 million over the quarter. Giving us an adjusted net income improvement of about 53%, ending negative $22 million in the quarter. That was a lot of data thrown at you so I hope you were able to digest some of that. I wanted to ask you Getty, if you have any overall broad input on those changes over the quarter?
Getty Hill (07:06): Yeah, I think it definitely wasn’t by any means a strong quarter for Compound. I think similar though, it was a tough one in general for many platforms in DeFi so it’s not exactly Compound acting as an anomaly in this instance. I think generally speaking some of the key external inputs in the space, there are strong indicators to what the health of the ecosystem has kind of shown over the last quarter or so. Definitely slow, I mean, I noticed the other day that Tether had its last print on April 3rd and in February, late February was the last USDC mint. So things that for the most part in Q4 and certainly in Q3 were just chugging along printing huge amounts of new dollars and value into the space have substantially contracted. So I think those are interesting indicators to say okay, maybe there’s some external reason behind the shrinking growth in the platform over the last quarter. I do want to highlight one of the most important line items on this page here is aggregate utilization. After all, the protocol’s job is to take in assets that folks lend the protocol, and then lend assets out to borrowers and that’s a key stat to me on here. It’s a bit hard to say when you see outstanding loans in terms of a dollar value adjusting because there’s all these coin prices and whatnot. Aggregate utilization is an important one because of that. So it’s disappointing to see that it is down, but when compared to some of those other stats that are showing 41% shrink in outstanding loans, and the 32% shrink in deposits, I think it becomes a lot easier to stomach. When you understand the general bread and butter of the protocol here, sure it shrunk, but it really shrunk by less than 6% over the last quarter and that’s that’s the key component for me.
Sean Butterfield (09:09): Yeah great, thank you for that input. Sami did you have anything?
Sami Kassab (09:14): No, I think Getty just made a good point related to how a lot of the decreases aren’t a direct relation to specifically Compound, it’s mainly related to the general financial market that we’re seeing going on in the crypto space. We had general market conditions trading sideways and so a lack of users looking for leverage is essentially going to lead to a lack of borrowing activity. Pretty straightforward here.
Sean Butterfield (09:50): Getty, quick question, you know I feel like these metrics are pretty easy to follow, is there anything outside of this that you typically look at and maybe does it change as the overall market sentiment changes? Whether we’re in more of a bull cycle or more of a bear cycle, are there different indicators that you’re keeping track of?
Getty Hill (10:13): I think there’s some specific users that you can look at that demonstrate a good wind vane essentially for the flow funds in the space. If you look at some of these monster accounts, I mean especially if you want to figure out who the big users are on the platform and track what they’re doing and say hey, these guys are sophisticated. They’re on AAVE, they’re on MakerDAO, they’re on Compound, they’re on all the money markets. They’re even on cross-chain platforms. Where are they putting their capital? How is their capital flowing? Those users are the ones I tend to look at when we want to get a bit more of a specific sample as to what’s happening. One of the reasons why Compound is what it is today is that there are a lot of users who used it once upon a time, a year ago, two years ago, and just never left. That makes it a little harder to look at the whole picture and get a macro sense of what exists here and what’s actually occurring in the trend. If you look at some of those particular target users who have a few hundred million dollar position on Compound but also on AAVE and also on MakerDAO, they paint a better picture of exactly why users are flowing capital from point A to B. Is it because MakerDAO is reducing their interest rates and providing a more capital efficient opportunity for those users? Is it because AAVE is offering some new assets that have a collective better overall margin opportunity for those users? Those are the details that you can’t necessarily encompass in a macro overview like this that can paint a bit more of a precise picture.
Sean Butterfield (11:48): Yeah I totally agree with that comment, and yeah we might high level touch on something similar to what you’re saying as it relates to the more sophisticated deeper liquidity folks who are participating on the platform. So, moving along, Sami, I’ll let you take over here. We’re going to go over some of the charts as it relates to these outstanding loans and deposits and I’ll let you take care of that Sami.
Sami Kassab (12:24): I appreciate that. So, starting off here with the outstanding loans. In general we saw outstanding loans declined by roughly 42% like Sean was saying. These levels are going back to what we saw around July of 2021 and again like we discussed, this is mainly related to market conditions trading very choppy for the past quarter. We really haven’t seen any new highs like we saw in Q4. Just a lot of volatility up and down which is what you see in a sideways trading market. The biggest contributors that we have to outstanding loans are obviously the stable coins and that’s where we saw the biggest decrease. We had DAI which decreased roughly 47% this quarter and USDC decreased roughly 41% this quarter as far as outstanding loans. Now, if we go to the next chart here we can take a look at the borrower interest rates. In general, last quarter these rates were averaging at about 5% in Q4. For this quarter, Q1, we saw rates averaging about 3.8%. A pretty substantial drop in borrower interest rates. One thing I found that was interesting is ever since Q3 you’re kind of seeing a minimization of volatility in general for borrower interest rates and that’s kind of excluding what we saw in October and November of 2021. We saw the volatility there take up mainly around when the market was peaking as far as the general market cap. One thing I wanted to ask you about Getty, the majority of borrowing demand on Compound we know is originating from people looking for leverage and also people seeking out yield farming opportunities. That’s really because DeFi in its state right now is more of a circular economy. It doesn’t really feed into too many real world applications. I was curious if you had any views on where you saw future new demand, new borrowing demand, originating from? Do you have any thoughts on that?
Getty Hill (15:01): Yeah, I think in the immediate future there are a number of collateral assets the platform is looking to support and I think those will be the most immediate things to drive our demand. Particularly something that GFX Labs has been working on is getting MATIC added to the platform, Polygon’s native currency. I think there’s a huge amount of interest in getting that supported and on somewhere on mainnet because currently it’s not actually available to borrow against anywhere on mainnet, in size at least. I think that’s on the immediate forefront, and then from there I think the longer term view looks like the industry is still extremely nascent in the whole sense. I don’t think necessarily external motives from the traditional world are going to be influencing the platform anytime soon. I do think as we look to expand beyond basic ERC-20 tokens as collateral and some more sophisticated positions, new opportunities will rise to the platform. People have been talking about getting stETH added to the platform which has been quite popular. Other versions of staked ETH also, and then from there other derivatives like some of these AMM LP tokens and whatnot that you can safely assign collateral values to. It’s a matter of time, and building out the infrastructure for those components, but I think that’s what will be drawing borrowers in over the long-term.
Sami Kassab (16:18): Yeah, that makes sense. Then when we think longer term as well, once Compound goes multi-chain that’ll definitely bring new borrowing activity as well onto the protocol.
Getty Hill (16:31): Yeah absolutely, those new chains will be really important. It’s something that AAVE has done massively well on is succeeding in onboarding new assets on those chains and it’s something we have to play catch up on at this point.
Sami Kassab (16:46): Yeah so another question I wanted to bounce your way is related to recursive borrowing. For the viewers here that don’t know what that is, can you kind of give an overview of what recursive borrowing is and then also provide your view on whether that has a positive impact on the protocol or if it’s more of a negative or frowned upon impact?
Getty Hill (17:10): Sure thing. So there’s two particular users on Compound right. There are users who supply capital because they’re interested in borrowing some other assets. So they supply one coin and borrow a different coin. Then there are the other users who just supply capital to the protocol in an interest to earn yield. Now within those there’s a sub category of users who just traditionally supply whatever given asset to the protocol, and then there are these recursive participants. So some just supply, some put on this bit more sophisticated strategy where you will borrow, or you do this positive collateral. Say DAI for example here, you have 100,000 DAI. Then you can go and borrow up to your maximum whatever the collateral factor is against DAI on the market and borrow the maximum of it. You now take everything you just borrowed, you deposit it back into the protocol, and you repeat this process over and again to get that recursiveness. Essentially, what it does is it amplifies your dollar notional position on the platform while maintaining the same amount of capital that you’re invested into the position. Why people do this is because at times we’re incentivizing participation in the protocol through COMP rewards. COMP is getting paid to the pro rata share of participants on the supply side and then the pro rata participation on the borrower’s side. So users can amplify their pro rata participation in this pool by doing this recursive kind of setup here. I think that’s the quick overview of what it is. This has been going on for coming up on two years, you know? But yeah, it’s been going on for a long time. There has been some recent debate about how sustainable? How important are these for the platform? What do they actually mean for the platform? All that is happening here from a purely stats standpoint is the user is depositing capital and the interest rates will largely remain unchanged and the utilization of the protocol remains unchanged. Regardless of if this user is a traditional supplier of the protocol and just lending that capital, or recursively borrowing it. The amount of capital getting deposited is the same and it doesn’t affect the change in utilization because they’re just perfectly looping it through. So, in general the only change in the protocol is changing TVL a little bit, changing interest rates a little bit. It doesn’t really provide significant capital to the protocol by any means beyond just a general supply. Essentially these things function identical to what a normal lender would do to the protocol, the only difference is that they’re amplifying these stats because there are some other reasons to make the stats bigger. But for the most part they just function like any other loan to the protocol.
Sean Butterfield (19:58): Gotcha, so it sounds like it’s not adding any additional risk to the protocol and so essentially they’re just farming out these COMP rewards. If we ended up seeing in the future at some point these COMP rewards being removed from the market would you essentially see these recursive borrowers leave as well?
Getty Hill (20:22): I don’t think they would necessarily leave, but I definitely think there would be a decrease. Essentially what’s happening right now is the protocol is subsidizing rates. Just to make up a number, if the protocol is paying out 3% interest rates and the COMP rewards add an additional 25 or 50 bps to the protocol, hey that’s 25 or 50 bps. In theory if we pull those out then some capital should drift away from the protocol to better opportunities. But the core premise that folks need to remember is that these guys are looping because they’re interested in getting the best interest rate available. Their job is to make money ultimately and so if we decrease our rates further by removing this extra incentive but we remain to be the best option for their capital, their capital will stay. So I don’t think we would lose 100% of capital that’s relying on these rates because at the same time, the general problem exists at AAVE as well and at any of their alternatives to where they can stick their capital. At this point the rates have been on a solid trend down since the very day they started, but particularly in the last half a year or so. You could look back to even five or six months ago, or at the very beginning of the year. You could still earn 12% doing a recursive loan on DAI which most viewed to be a very safe, stable rate to be earning capital and today it’s substantially less than that. I think we’ve already seen a solid progression of those rewards decreasing and of that participation decreasing along with it.
Sami Kassab (21:56): Perfect, I really appreciate you breaking that down. So let’s go ahead and move on to the next slide where we’ll talk a little bit about outstanding deposits in general. Wrapped Ethereum continues to maintain its dominance, it’s the largest deposited asset on the protocol. This quarter we saw DAI and USDC kind of neck and neck competing for that second largest deposit, they’re very close as far as actual dollar amounts. The interesting thing is that we saw DAI’s outstanding deposits decreased nearly 50% and then if we look at deposit volume for DAI in the quarter that was down 90%. Sean was actually pretty curious about why this DAI deposit volume had decreased so much, so he worked with one of our data scientists here and they were looking at the on-chain metrics and it was suggesting mainly that there were a few market makers that were basically arbitraging between AAVE and Compound. Like you said, looking for the best interest rates. It seems like they closed out some pretty large positions in Q1 and that goes back to what you were saying in the beginning. Looking at the macro doesn’t really tell you the whole story. Looking at some of these whales and some of these users of the protocol also gives you additional insight for what’s going on on the protocol. One thing that got me curious as well, last quarter we saw Coinbase’s DeFi yield integration. Essentially, if you had DAI they would pay you an interest rate and they were taking that DAI and then inserting into Compound. Last quarter we saw a large inflow of DAI and it seems that this quarter we saw a 50% reduction so it makes me think did we get a little bump in outstanding deposits from that integration that long-term wasn’t really sustainable, it just depleted slowly. It makes me think because the protocol has MakerDAO’s direct deposit module coming up which is essentially going to do something similar. So, I just kind of wanted to get your view on, do you think these integrations with different products have a long-term effect on the protocol? Or is it sometimes short-term?
Getty Hill (24:32): Yeah I think it’s tough to say. We don’t really have a long history or really any history of protocol integrations materially changing the landscape of what the platform looks like. I’m really excited for the D3M module though, MakerDAO’s direct to buy deposit module, to come about. I think that’s a really efficient way for the protocol to get capital and also to symbiotically work with MakerDAO in an uncompetitive environment where everyone wins. So I’m very much looking forward to that. I do think that though it’s a bit hard to say what it all looks like. There’s a strict balance in my mind as a bit more of a traditional DeFi person, that protocol integrations are great as long as you don’t add on too much systemic risk to the space. Because ultimately some of the games that folks are playing with Compound and say well, Compound is in a great position to succeed if AAVE explodes one day, god forbid it ever happens and vice versa. So everyone needs to make sure we also don’t stack on too much systemic risk in the space which could lead to the demise if something truly awful were to happen.
Sami Kassab (25:40): Moving on to depositor interest rates, in general we saw rates decrease as borrowing demand declined over the last quarter. Q4 we had the average rate for depositing be around 2%, this quarter it’s around 1.5% on average. One thing interesting to note is that we are seeing the normal depositor interest rate, and the depositor interest rate with incentives, start to converge. Not only for depositing, but for borrowers. This is partially attributed to COMP basically halving COMP rewards, and I’m sure we’ll talk about that quite a bit in later on, so I’ll just kind of mention that for now. But, related to all these new integrations I have another question for you. Do you see the source, the main source of deposits on Compound in the future coming from professional liquidity providers? Whether it’s funds, VCs, institutions, is that what the protocol is targeting rather than the normal DeFi user? This might be a general DeFi question, are these protocols starting to focus more on institutions that have more deep liquidity rather than the person who’s yield farming in their mom’s basement?
Getty Hill (27:19): I think we need to keep in mind when we’re talking about depositor rates, we’re largely thinking about the stable coins on the protocol. DAI, USDC, Tether, and I think largely speaking, the folks who are leaning on those markets the most, representing the largest portion of those markets, are institutional players or players with extreme high net worths who are looking to park dollars for safe steady rates for a long period of time. I know for certain that some of the larger participants in the borrow/lend ecosystem in crypto, like the BlockFis, the Celsius, or the Nexos of the space, certainly look at general dollar rates. The primary stack historically in crypto for dollar rates has been the funding on the Bitcoin perp. But certainly DeFi dollar rates are creeping into the methodology as well as like hey, how much do my dollars earn and where can I put them? We’re slowly developing a bit of our own DeFi based LIBOR (London Interbank Offered Rate) across all the ecosystems.
Sami Kassab (28:23): Yeah that makes sense. Thanks for breaking that down. Last slide I’ll be covering right now is going to be the utilization ratio. If we can go to that slide, Sean? This slide basically summarizes what we’ve been saying. Basically the outstanding borrows have decreased in relation to outstanding deposits. The only interesting thing I can comment about this is that historically we’ve seen 40% almost serve as sort of a support level and we haven’t seen that rate drop below 40% for a significant period of time. But Q1 we have seen it fall below 40% for an extended period of time. I think that pretty much wraps it up here, I’ll pass it back to you, Sean, to take over the next few slides.
Sean Butterfield (29:35): Cool, thank you. So we kind of touched base on this a little bit earlier in that macro table as it relates to interest income. Obviously as you can see there’s a substantial decrease quarter over quarter. I think it was about a 53% decline from $90 million to $42 million in interest. It’s directly correlated with the overall demand for loans, repayment on those loans, interest that essentially flows through the protocol. One thing I wanted to mention is looking at the data, it looks like USDC actually led the way in the deterioration and interest income. It accounts for about 62% decline with USDT and wBTC fighting for second in interest income deterioration with 50.8% and 50.7% respectively. Moving on to protocol income, the narrative is very much the same across the board outside of liquidations. Protocol income, same thing, was halved in Q1 a little over 50% and the same thing, USDC led that reduction in income. I’d be curious Getty, I know there have been previous conversations related to improvement in protocol income. How important is it to the overall Compound protocol and community at large? Instead of me talking about some of these points I’d be interested in hearing some high level remarks about what could contribute to the improvement of protocol revenue, and maybe how far we might be away from some of those implementations?
Getty Hill (31:32): Yeah so I think I’ll preface it with let’s keep in mind Compound’s governance has been around for less than a year and a half at this point so everyone’s still learning what’s the best way to organize and everything. That being said, the reserve factor is what determines how much the protocol takes from interest paid by borrowers before it reaches the lenders. That number is a set parameter on every single market on the platform. Generally speaking, stable coins have a much lower one, altcoins have a much higher one. Particularly when we’re talking about the big income earners for the protocol we’re thinking about DAI, USDC, Tether. Now, USDC and Tether both have a 7% reserve factor which means we get 7% of interest paid from the borrowers there. Whereas DAI notably has 15% which means, I think the thought was this thing’s a bit more volatile so we’ll take a bit more when it was originally set. That being said, back to my original governance comment here, the protocol hasn’t really done much to play around with these or do any type of research really about should we be choosing a higher or lower number and so we have rather limited data to actually say what we should be doing. The protocol needs to get in a spot where we’re willing to experiment a bit more with those rates to actually understand how income can be optimized for the protocol and take those revenues to something a bit more sustainable than what we’re currently at today. So it’s a bit of a complex problem that generally needs a bit more time and testing before we really speak to it. Hopefully that’s something we get to start doing here in the near future.
Sean Butterfield (33:13): Sure, recently I was scouring the forums and it looks like there’s been mention of building out an asset onboarding framework as it relates to why we should onboard a specific asset. I understand demand would be a very critical factor in that but maybe if you can kind of touch base on any input you might have on that and how you think that would move forward?
Getty Hill (31:32): Yeah, that’s a bit of a long point so I’ll try to keep it brief here so that I don’t brick the rest of the deck here that I know we have. But you know, when it comes to listing assets you really have to think about it in a few different categories, or at least the few different points you need to assess. One is the general interest from folks who would be supplying that as collateral and folks interested in actually borrowing this collateral. One of those two things needs to be substantially true for the protocol to even really assess it. From there it becomes the security risk and understanding that. A lot of the assets that folks are interested in adding to the platform right now, other than MATIC, are FRAX, LUSD and RAI which generally speaking are a bit outside the normal range of where the protocol operates. Especially in part of the complexity and risk spectrum when it comes to asset listings. So it’s a delicate balance of we want to add things to the protocol that users are interested in and want, but we don’t want to stack on too much protocol risk. Because, not to be dramatic, but if we list an asset and we give it a collateral factor and it has some type of error in it, the classic example is the infamint, then all the funds in the protocol can be drained if that thing has like infinite, you know has any collateral factor greater than zero. So, the protocol tends to be quite conservative with listing new assets especially because when you look at the stats as we are here, wBTC, ETH, DAI, USDC, the protocol would be not too different if we didn’t have any of these other assets in the protocol and we wouldn’t be going through this spreadsheet if you just imagine we only have those four coins. It would be largely the same stuff to be talking about so you really have to be careful about the marginal improvement by adding a new asset.
Sean Butterfield (35:30): Yeah, that’s great. Sami, I don’t know if you had any questions or input on that?
Sami Kassab (35:38): I just think that’s a great point that’s getting made. It’s really just stables that have the majority of borrowing demand and it kind of goes back to the multi-chain approach that is being talked about. Just having one collateral asset or one deposit asset I believe which is just going to be USDC, right? We’ll be talking more about that multi-chain approach as we get towards the end of this slide deck.
Sean Butterfield (36:09): So yeah, I’ll pass it back to you Sami as it relates to liquidations.
Sami Kassab (36:15): Yeah, so like you pointed out Sean, in the beginning. Liquidations was one of the only metrics that kind of had a good quarter. We saw a substantial increase in liquidations. Question to toss your way Getty, can you give some thoughts on liquidation revenues importance to the protocol? Do we see this as healthy revenue?
Getty Hill (36:38): I think this is a newer thing, so it’s not recent but I think in mid-2021 we changed the liquidation system to split part of the liquidator incentive. So historically there’s been about 7% or 8% that was distributed to the liquidator and then the protocol was like we’re going to keep three of that for ourselves and we’re going to give five of that to the liquidator. Now this is something similar to what MakerDAO does. They get some revenue from liquidations as well as I believe AAVE, but I could be wrong on that one. So you know it’s a weird one, my personal opinion is that protocols shouldn’t be incentivized to have liquidations. The inverse of that opinion to play devil’s advocate quickly, is just simply that well the protocol is taking out risk here you should be similarly compensated for that loan. My opinion is that risk comes in the factor of the interest rate and that reserve factor we were talking about before. But yeah, I think it’s a good thing to be experimenting with especially we all have to remember, as I get stuck in the trenches of DeFi quite often, that really we’ve only been around for two years so there’s a lot to be learned and experimented with.
Sami Kassab (37:49): Yeah, exactly. So, moving on to the next slide here we’re going to have a chart on token incentives paid. I guess this is where we can talk a little bit about some of these governance proposals that have been passed. So, one of the most critical or major governance proposals that we saw passed this quarter was proposal 92, which basically reduced COMP rewards paid out by 50%. This seemed to be a pretty heated discussion as far as in the forums. There seemed to be people on both sides and I guess the plan was to kind of slowly step down to zero rewards. I think proposal 100 aimed to do that but recently, I believe two days ago, we saw proposal 100 fail. So, what is the plan? I know the community has to really talk about this but as of right now we’re just kind of sitting at 50% rewards, right? Is there a plan that the community is talking about? Is there a reason why this first proposal went through and now all of a sudden it seems like the community might have changed their mind? Because I believe it was outlined how initially you guys were going to step down to 50% and then eventually to zero. So can you talk a little bit through maybe both sides of the argument related to removing rewards versus keeping them?
Getty Hill (39:39): So, first I want to set the stage a bit and say when we think about rewards and how they’re distributed to users of the protocol we need to think about them in token terms and in dollar terms. Now ultimately we’re assuming that the folks who are interested in these rewards the most are here for the financial incentives, that these guys are running and interested in money. So ultimately for their experience, all they care about is the dollar notional value of these tokens that are coming to them. From the protocols perspective, we probably care about the dollar notional value of those tokens but we also care about the general number of tokens as we’re essentially entering new tokens into the supply, the circulating supply, and diluting all the existing COMP holders. So it’s a bit of a difference there that I want to highlight. If we look over the last year or so, certainly in the last eight months, as COMP prices declined with generally the rest of the DeFi ecosystem. Those dollar notional rewards or distributed users have kind of declined itself. From the people who are interested in using the protocol’s perspective rewards have already been cut, right? The inverse of that would be for us to maintain a dollar value and increase the notional amount of COMP we’re distributing to maintain the actual dollar notional rewards that they’re achieving, but that’s not the case. So from their perspective, hey these things have already been going down for a while and you know, that’s what they’re used to and we already have a lot of data to support what those users are going to be doing as those rates decrease across not only Compound, but the blanket of DeFi. Now particularly with the most recent and 92 and 100 proposals, I think everyone is on the same page that these rewards aren’t sustainable for the most part. I shouldn’t say everyone but there’s certainly some agreement around that these aren’t sustainable. Things need to be reformed, and I think that’s why 92 is a bit more easy to stomach. It was just a general, we’re going to cut these in half. It was quite clearly cited in 92 that there was going to be a proposal 100 that was going to cut them to zero. I think folks largely either didn’t read that or were just like, all right well we’ll deal with that problem when we get to it. So 92 was rather not controversial. For the most part folks kind of just voted and supported it, it seemed pretty straightforward. The folks who I think represent the protocol most, happily understand that it was a good decision. Then 100 ended up being extremely controversial. I mean I haven’t gone through the data because there are a hundred proposals at this point to go through, but I think it is the most controversial governance proposal that has occured on Compound, and certainly one of the most controversial ones in DeFi in general. In the sense that we had, for our protocol, very good voter turnout, about half a million on each side and it came down to 7,000 votes. Which really when you go down the list of people who are set up to vote in the protocol it was a pretty good population of folks who actually turned out to vote. There’s a number of people who I was disappointed not to see but it was relatively good. Disappointing outcome from my perspective because I really think rewards at this point are doing very little to support the system and largely just hurting existing COMP holders like myself, but you know that’s that’s the plight of governance.
Sami Kassab (42:55): Yeah, I think it’s really interesting how basically depending on how you interact with the protocol leads to how you would view this decision, right? So if you’re a big whale that would be using this protocol, obviously it’s not in your incentives to want some of these rewards to be removed. But, if you are a VC, or a contributor, or just simply a holder, it’s in your best interest to have these rewards eventually go to zero. Because you’re not getting crowded out anymore, you’re not getting devalued essentially. So it’s really interesting to see everybody in the community put their opinion in like you said. I mean the forum post was just popping with everybody, I think it was the one i’ve seen with the most comments. Like you said, that 7,000 vote delta was extremely close. Anything to add to that, Sean?
Sean Butterfield (44:03): Not a whole lot, but as an observer, I’m not a participant of the protocol, it makes a lot of sense from a sustainability point of view. Plenty of mercenary capital that’s soaking up all the rewards, dumping the tokens, applying sell side pressure on the COMP token price. It makes sense that this isn’t sustainable long-term and making those steps, strategic steps, to try to keep everyone on the same page, keep everyone happy, but also maintain the health of the protocol. Definitely in agreement with the conversation there.
Getty Hill (44:44): I’ll add a quick tidbit there, so while this proposal was going on, I was doing some quick research to figure out how many of these COMP rewards are roughly going to folks who are just immediately selling the token. What are these recursive participants doing? Because generally those are the folks who are doing it, and what are they immediately doing et cetera. If you just look at the top 10 suppliers of USDC to the protocol, the top 10 suppliers of DAI to the protocol, and you only look at those accounts. DAI and USDC alone account for over 77% of total COMP rewards distributed in the program. So largely those two alone account for pretty much everything. Then from there you can look at these users and see what their activity is and what they’re doing. You’re going to count right there between all of them that they represent these recursive users who are just here simply with mercenary capital, which is fine, I’m totally fine with that. They get the COMP, and then just immediately sell the COMP. That accounts for about 41% of the COMP rewards getting distributed and that’s just the sample of the top 10 users on those two markets alone. I was trying to get some quick data in like 30 minutes together and that’s what I quickly found. I was like well if it’s 41% with just this set, the number is definitely a fair bit higher. All we’re getting out of this is a little more capital in the protocol and a higher TVL number and you know, for what? To continue diluting all the existing holders and accepting kind of mediocrity on this program. To me it’s a pretty black and white issue that I did not expect to be so controversial in proposal 100.
Sean Butterfield (46:29): Yeah, you bring up a lot of good points and it’s not that we don’t want participants on Compound but we also need to kind of create a healthy floor and stabilize the overall protocol so kind of shaking out some of the mercenary capital by removing some of those incentives makes sense.
Getty Hill (46:51): Yeah I mean we have a great user base. You guys have the data here to show that there is billions of dollars of regular, natural usage of the protocol. Folks who are supplying ETH and wrapped Bitcoin, and borrowing dollars for the core example of leverage on their assets. This program that we have here to distribute incentives, best case is adding like $1.5 billion, really best case is adding $1.5 billion. Most likely it’s adding somewhere between about $650 million, $800 million to the protocol. It’s not significant to the actual day-to-day operations of the protocol or fundamentally changing anything about the protocol. It’s really just in my mind a leach on the protocol at this point. Compound existed well before the COMP incentives ever existed and proved that it had a natural use case in the market, and the data still supports that it does today. So can we get rid of this thing already please?
Sean Butterfield (47:50): That makes a lot of sense. As I said, as an observer just looking at it really from a data perspective, hearing input from the community and yourself. Overall, you just want an attractive protocol that’s sustainable, that’s healthy, that promotes growth, and encourages users. So it’s a step in the right direction and I’ll be curious to see how the next proposal that might try to bring the emissions down to zero will go. As we move forward, we touched on deposit volume a little bit. An 80% decline over the quarter, from $63 million to $13 million in deposits. DAI accounted for a lot of that and as Sami was mentioning, me and another couple folks were digging into some on-chain data and it looks like it’s exclusive to a few market makers that were really trying to arbitrage between AAVE and Compound early or sometime in Q4. Not only is it super cool to see the on-chain data to kind of make those suggestions but it’s also interesting to see that’s really what drove a lot of quarterly deposit volume. Then also as we enter Q1 it looks like maybe a separate market maker was basically redeeming their funds from the protocol. Maybe in preparation for the reduction of COMP emissions towards the end of Q1. So yeah, really interesting to see as you look at the chart, it’s a pretty glaring drop in quarterly volume. As we kind of wrap up here, at least as relates to charts, originations, same story. Demand for loans declined. The demand for new loans will follow. So seeing a 53% drop from $11 billion to $5 billion. What I thought was interesting is four consecutive quarters of the declining originations. I think that’s largely related to the overall market sentiment but we might be curious if you have any kind of input or future outlook of originations moving forward.
Getty Hill (50:25): You know, stats like volume and originations are a bit hard to nail down. Just because it’s DeFi, capital is just extremely liquid. If folks want to leave and join the protocol for an instant they are welcome to do so with no penalty. So that makes these stats kind of hard to focus on and that goes back to the very first slide. The thing to really care about is that utilization number. So for me that’s the number I focus on. Less so the originations and volume stats.
Sean Butterfield (50:56): That makes perfect sense to me. I mean like you said, you could easily throw a billion in and take a billion out inside of a day and it still contributes to originations or deposit volume on a quarterly basis. So that makes a lot of sense. So yeah, moving on into governance. It looks like there’s about eight proposals that were passed in the quarter. I won’t touch base on all of them, but I’d say there were a few that were pretty critical changes to the protocol. Sami, did you want to take a stab at those?
Sami Kassab (51:37): Yeah, so we talked about two of the really important ones, proposal 92 and proposal 100. I guess one of the other ones I found interesting was proposal 89 and that was reducing the proposal threshold from 65k COMP to 25k COMP. Which I think is important to stimulate further activity into the governance system. One thing I thought that was interesting was that we really didn’t see that many voters on proposal 100, even though it might have been one of the more popular propositions or governance proposals. I think there were a total of 48 addresses. So Getty personally, have you seen that trying to get people to contribute to some of these governance proposals has been difficult?
Getty Hill (52:37): You know, it’s not so much the number for me as it is the amount of COMP that shows up to vote. These systems aren’t democracies you know, we run on a plutocracy system here where it’s based off of capital rather than who’s present. So for me, it’s what’s the COMP turn out in these things? The COMP turnout was decent, it was pretty on par with the other significant proposals that we’ve had historically. It certainly could have been higher, I would have liked to see a number of folks turn out that didn’t turn out. But I think for the most part it wasn’t disappointing as a specific proposal. In general it’d be nice to see that we had a larger percentage of the COMP turnout to vote. As you know we’re 6 million circulating supply, 6.5 million circulating supply. To only have a million turnout isn’t terrible, but it isn’t great either. It’d be nice to have two million turnout for these things as far as the number of voters goes. If you just check out on the Compound website there’s great databases for this stuff. Really, once you get around to voter 50 as far as the leaderboard goes, the amount of votes really substantially drops off. A lot of these things are largely decided amongst the top 25 people, just because that’s where the power is really contained as far as the number of votes, and who possesses those votes.
Sami Kassab (54:01): Yeah, any thoughts on potentially, I don’t know if any protocols actually do this, but incentivizing people to contribute to, or just to vote in general?
Getty Hill (54:16): I don’t think that active voting is really important. That’s kind of the weird part of this. So there should be more things to vote on in general. I think that’s important, that’s one of the reasons why I like the idea of 89 and that there’s about three facets of making proposals right now. You either have enough votes delegated to you that you can make the proposal, or you either have the COMP yourself, and you know the total number of coins. Or you’re on the white list. Essentially a thing that the community multisig in governance controls is the whitelist addresses with proposal power. The whitelist is neat but that’s largely a social mechanism and so the first two are still the most critical ones for the protocol. You cannot call yourself a true DeFi protocol that has governance unless you have completely on-chain control of all the code and all the contracts that make up that system and make it accessible for anyone to come and change something about that system without asking anyone else for permission. That is to me, the core tenant here. So lowering the proposal threshold from 65 thousand to 25 thousand materially doesn’t change the risk of the system, the quorum, the number of votes required to actually get a proposal to succeed, is remaining the same. Folks could actually argue to raise that number as well but the system should be more accessible to a larger populace to make those proposals. Not everyone is as fortunate as me. While I have a large percentage of my net worth in Compound, I don’t actually have the 75 thousand votes that we have on the platform. We’re lucky enough to have some folks who have those tokens that are willing to delegate to us since we pay a lot of attention to the protocol.
Sami Kassab (55:58): Gotcha, yes, proposal 89 really just increased decentralization and inclusivity of the protocol. So for the sake of time, I guess we’ll speed through this next slide here just on grants. Is there a date yet related to when the new grant program will be back?
Getty Hill (56:20): No, I mean I think it’s a matter of a personnel problem. Essentially you just need someone to come in and run it. That person, if someone’s listening here, listening to this recording is interested in trying to run a grants program for a protocol and might have some experience applicable to it. Do come to the forum, write a forum post, make your argument that you should do it. You can get paid quite handsomely to do this, this is by no means a side gig. Uniswap employs a team for their grants program and so do other protocols, and they all get paid quite handsomely. So I think it’s really just a personal problem at this point. Everyone’s for it but no one has the time or the bandwidth for the compensation that’s being discussed.
Sean Butterfield (57:00): Yeah, just a couple last points as it relates to the road map a little bit. If you could maybe share some details on where we’re at as it relates to the multi-chain strategy and then lastly, I know there are some limited details as well around Compound 3, but maybe if you could share what you can, what you know, and your general outlook on both of those.
Getty Hill (57:30): Yeah, absolutely. So, I don’t have any proprietary information. I’ll say that first, I’m just for all intents and purposes, a community member and I know for a fact that really largely all the VCs are in the same boat. The Labs team, who’s the core engineers behind the protocol, does a very good job of not telling anyone anything. So what we do know is really just the public information. So they’ve done two developer calls in recent times where they have discussed Compound 3 with folks. Generally it’s a much slimmed down version of the protocol. Instead of the generalized money market right now where folks can deposit and borrow anything on either end, you’re supplying a thing as collateral and you’re only borrowing USDC or whatever that reserve asset is. So the first version as they’ve told everyone is going to be with USDC as the reserve asset. In theory that reserve asset could be anything else, which makes the system a bit interesting. You could have a bunch of little versions of Compound 3 like little silos where you could have one that’s flavored to ETH where you’re depositing a bunch of ETH derivatives and only borrowing ETH as the reserve. You could be depositing a bunch of Bitcoin derivatives and buying only wrapped Bitcoin as the reserve. All sorts of fun things could be done with this and I think that’s kind of where they’re going with it. A bit more of a gas efficient, simpler protocol to swallow and just generally more efficient. Like all the original smart contracts at this point are well over I think two years, and so that’s the other main component here. Making it easier to deploy and run on the other chains. So frankly though we don’t have tons of information as far as timeline goes in development, we know what they’ve stated is that there is the plan to post the open source version of the protocol in late April. So maybe this week, maybe next week, I’m not holding my breath. Maybe a couple weeks after that and then we’ll get a testnet version of it the following month. The code is going to go to audit to Open Zeppelin and Open Zeppelin is the formal audit partner of the protocol so I think they’re putting on the finishes with that and it’s going to be going to Open Zeppelin here in the next few weeks, so I’ve heard through the grapevine. Then we’ll kind of go from there. I mean, I think everyone’s kind of hoping for a mainnet launch in July or something, or June. But everyone keep a massive grain of salt with everything I said because this is coming through about like 30 minutes worth of two different conversations, and everything is subject to change and I don’t know anything really in reality. So, we’ll see what comes. I think everyone’s excited.
Sean Butterfield (01:00:06): Yeah, that sounds great. I was able to catch some of those calls and it does sound like the testnet is supposed to kick off shortly so I’m sure everyone’s pretty excited about that. Real quick, it seems like the development’s been a little different as it relates to the multi-chain strategy. I know the initial idea was to kind of go with that Starport integration and it seems like there were some development hiccups. As the landscape developed over DeFi and multi-chain strategies compromised platforms. It sounds like they were kind of trying to pivot to a different model? Are you able to maybe touch on that a little bit?
Getty Hill (01:00:50): Yeah, I don’t know much. I think the general idea was let’s make the code simple and let’s make it easy to deploy and easy to manage. There have been hints at additional tools to help manage all these different employments and some new testing tools throughout these protocols. It’s been discussed that the new oracle solution that the protocol is going to go with is just going to be directly Chainlink. Which is advantageous for all the different chain deployments where there isn’t necessarily robust on-chain liquidity that exists on Ethereum. But I mean largely we don’t know. Again there’s a lot of guessing work going on right now so I think we’ll find out in the next few weeks here a lot more information. Then you even still have the product the next couple of months and it should be a fun summer.
Sean Butterfield (01:01:39): Yeah, sounds good, something to look forward to. I don’t know if Sami, if you had any other questions on our end? You know we’re pretty much rounding the hour here so I think the questions from the community have been submitted and I think we’ve kind of touched base on that. If any other viewers have any questions please feel free to throw them down, otherwise it’s been a good time. I’ve really enjoyed having you and hearing all your insight.
Getty Hill (01:02:24): For folks who are interested in following along with Compound in general there’s the forum at compound.xyz. There’s the governance action forum on Compound Finance, you can check out the proposals that exist there. Join the Discord as well. There’s a developer call that exists every Wednesday in the morning, well every other Wednesday. Then on the other Wednesdays that aren’t the Compound lab developer call I tend to run a community call on Twitter that’s a bit more casual than the formal developer call. In general for folks who are interested in following GFX Labs and our interesting governance on Uniswap, MakerDAO, and Compound, and the other awesome products that we’re developing over here. Which is in fact, what really I spend most of my time on, believe it or not. Do check out GFX Labs and Poppy Finance and Etherlands, all the rest of our main projects.
Sean Butterfield (01:03:04): Awesome, well thanks again. Sami, thanks for joining as well. I hope everyone has a beautiful weekend and we’ll do this again next time.
Sami Kassab (01:03:13): Yeah this was fun, thanks everyone!
Getty Hill (01:03:15): Thanks for having me, cheers!
