Unchained - The Chopping Block: Curve Crisis Concludes With a Whimper, PayPal’s PYUSD Raises New Questions - Ep. 529

Episode Date: August 10, 2023

Welcome to The Chopping Block – where crypto insiders Haseeb Qureshi, Tom Schmidt, Tarun Chitra, and Robert Leshner chop it up about the latest news. This week, a quiet conclusion to that Curve thing everyone was animated about. Plus, major developments from Web2 giant PayPal as it continues its journey down the crypto rabbithole. Also, is Maker’s juiced DAI Savings Rate the new Anchor? The gang discusses. Listen to the episode on Apple Podcasts, Spotify, Overcast, Podcast Addict, Pocket Casts, Stitcher, Castbox, Google Podcasts, TuneIn, Amazon Music, or on your favorite podcast platform. Show highlights:  how The Chopping Block got that sweet, sweet Red Bull money what to make of the Curve hacker’s message after they returned the funds does Curve trying to dox the hacker proves that Arkham’s model is right? the lessons from the Curve situation and whether DeFi needs to change how DeFi protocols could prevent situations where a single entity holds a significant percentage of a token’s total supply how the enhanced DAI Savings Rate triggered PTSD from other stablecoins (ahem, Anchor) what the consequences are of increasing the DSR the gang’s prediction on how much PYUSD will be minted by the end of 2023 Hosts Haseeb Qureshi, managing partner at Dragonfly  Robert Leshner, founder of Compound Tom Schmidt, general partner at Dragonfly  Tarun Chitra, managing partner at Robot Ventures Disclosures Links Unchained:  $52 Million Drained in Curve Finance Pools Exploit Curve Founder’s Liquidation Could Trigger Chaos for DeFi Curve Exploit Results in Largest MEV Block Rewards in Ethereum’s History PayPal Launches PYUSD Stablecoin Built on Ethereum MakerDAO’s Spark Protocol Blocks VPN Users Curve Opens $1.85 Million Bounty to Identify Hacker      Learn more about your ad choices. Visit megaphone.fm/adchoices

Not a dividend. It's a tale of two-quan. Now, your losses are on someone else's balance. Generally speaking, air drops are kind of pointless anyways. I'm named trading firms who are very involved. D5.8 is the ultimate time. D5 protocols are the antidote to this problem. Hello, everybody. Welcome to the chopping block.

Every couple weeks, the four of us get together and give the industry insider's perspective on the crypto topics of the day. So quick introses. First you got Tom, the DeFi Maven and Master of Memes. Next we've got Robert, the Crypto Connoisseur, and the Tsar of Superstate. Then we've got Tarun, the Gigabrain, and Grand Puba at Conlet. And finally, I'm a sieve the head hype man at Driving Fly. So we are early-stage investors in crypto, but I want to caveat that nothing we say here

is investment advice, legal advice, or even life advice. Please see Chopin Block. That XYZ for more disclosures. So you might notice something a little bit different about us today. Turns out that we are now officially sponsored by Red Bull. In fact, we are so sponsored by Red Bull that they sent us to ruin when you tell us what we got in exchange for this incredible. Absolutely very official sponsorship.

So I believe what happened is me as the always drinking Red Bull on the show speaker had lamented the fact that, you know, I drink so much Red Bull, but they and talk about all the time and I'm not sponsored. And someone listening to this here show happened to be working at Red Bull or actually very close to someone who's working at Red Bull and marketing and sent over a little care package of, you know, four cases of 32 cans. of Red Bull, which had there not been conferences I was traveling, would have been kind of finished by now. But close. But in exchange for that, of course, we wanted to show our school spirit and school colors. That's right. I believe that that retails at Costco for about $40. But it's a bull market. It's a bear market in crypto. We have no other than that money is running dry. That's right. It's a red bull market at this point. So we're doing everything we can. Red Bull, if there's more where that came from, we're ready for it.

We're ready for it. RB is a set of two letters that is blowing up both in crypto and in extreme sports between Rollbit and Red Bull. Crypto is just filled with lots of these RBs. That's not financial advice as a show. We sell out immediately. So if there's any more Red Bull swag you can send us, we will go even harder than this. In fact, the number one thing, I think that means that you've it as a podcast or streaming site that loves Red Bull is the Red Bull Fridge. Yes. So we've seen that there are certain streamers who are like Twitch streamers who get this branded Red Bull fridge.

This is our bid to try to get a branded Red Bull fridge. So Red Bull, if you're watching, if anybody at the Red Bull corporate something or other, if you know somebody at Red Bull, send them this and let them know that we are ready to sell out even more if we can get a Red Bull fridge. You just ship it to Tarun, Brooklyn, New York. It'll arrive at Tarun. Just write that, just write that Tarun, Brooklyn, New York, and it'll show up. Everyone knows Tarun.

So, perfect. All right, well, that's enough for me. My head is going to get sweaty wearing this. Yeah. Until we get the fridge, we're not going to go to the full show with the branding. But if we get the fridge, we're going an entire show with Red Bull branding. I also loved on Twitter. Somebody said that Tarun was the first, what was it, the first intellectual athlete,

sponsored by Red Bull. That was very fitting. I thought that was very fitting. I think it might be the first Matthew. That's surprising. I would bet they already did that. Somehow, somewhere. They sponsor anything.

Like, have you guys ever seen the flug tag, which is a sport that I feel like they made up? It's where you make a like, a wooden plane that can't fly and you push it off a ramp over some water. And it's like, how far can you make this, like, thing that, like, can't glide? Hmm.

I mean, some of them are hilarious. Like, people get pushed off and the entire thing disintegrates as they're in it, and then they fall into the water. Yes. Oh, someone's in it. Oh, I thought it was just like what are these. Oh, my God. All right.

That's intense. Oh, I see. Got it. Got it. All right. Well, I don't know if we're ready for that. I think about extreme as we get is like sitting with wearing merchandise. But if that's attractive to you, Red Bull, we are ready for it.

Okay, so let's do a quick recap. Last week, we talked about all the craziness that was going on with the CRV attack. I think we were lamenting the state of smart contract security and talking about what this means and how Defi is potentially going to recover from it. Well, it's been about a week since the CRV hack took place, and there's some updates. So one is mysteriously, the exploiter of the CRV bug is, we now know a little bit more about them because they turned out to return a little bit of the money. So the story here is that one of the pools that was hacked was from a protocol called alchemics. And alchemics, they are, what is it?

Self-repaying loans. Self-repaying loans. That's what it was. I remember. It was very, very popular kind of DeFi 2.0 project back in the last cycle. So they had some kind of, you know, stable-to-stable pair, and that got hacked because of this old viper vulnerability. So the attacker ended up sending back all of the AL-Eath that was hacked, and they sent a message to, at the same time that they sent this money back, and this message, it was about $12.7 million, including the ETH and the AL-Eth that was returned. And the message they sent was the following. I saw some ridiculous views, so I want to clarify that I'm refunding you, not because you can find me. It's because I don't want to ruin your project. Maybe it's a lot of money for people like you, but not for me.

I'm smarter than all of you. Fuck. That was what I said. I don't know what you guys thought of this. It sounded just the way that he was talking like very Russian or something. I feel like it's a very Russian way to talk of like you guys are idiots and then an expletive.

You're projecting. Am I projecting? Am I projecting? I can't read any nationality from that statement. The only thing I can read is like a personality disorder. And I've met people like this in crypto where they just, you know, think that they are the smartest person on the internet because they can exploit

a contract. Guess is actually not, yeah, not ethnicity or ego, but under 22. I would be willing to bet like three to one, three to one odds like under 17. It's definitely solo. It's not organized if you're sending messages like this after a hack. But it did feel to me just like the way it was written that this is clearly not a native English speaker, maybe I'm being racist. The other weird part was the funds were not returned to Curve.

They returned them to the Alchemics, like, wallet, alchemics down. So it's not like the curve LPs or the one's getting refunded. It's like, Alchemics, you go figure it out. I like you, you're cool. Everyone else gets, gets fucked. Oh, that's an interesting angle. I didn't think about that. Okay.

But I assume Alchemics will return it to the curve users that it was taken from. Yes, yes. They're like, you go figure it out, but it's like, almost like, I don't, like, curve is, you don't have the responsibility anymore. Like, you know, Alchemics is the only one I love and everyone else, you know, sucks. And like, you know, they just piece that. It was very strange. Yeah, because he didn't return funds to any other of the hacked protocols. So there, and there were a few others.

So, yeah, maybe this is like Alchemics is cool. I like D-Fight 2.0. I don't like Curve. I don't know. Strange. So, so another side of that was that, so Curve actually put a bounty on. the recovery of these funds. So the curve basically announced that if you return these funds within a week, we are going to give you a 10% bounty on all of the hacked funds. You know,

there was something on the order of 60 million total that was hacked. And so, you know, $6 million, no jump change for an attack of this sort. But of course, those funds were not returned in whole to curve. And so curve at the same time as they announced that, hey, we're going to give you this kind of bug bounty window that if you return, you know, all things are off, we're not going to pursue you. They said, if we don't get the funds returned within a week, then automatically we are going to place a bounty on basically figuring out who you are. So doxing, whoever the attacker is. So that doxing bounty is now live. So it's about $1.85 million for the doxing of the hacker and or of successfully achieving a conviction on the hacker.

So now the race is on. I don't know if there are crypto bounty hunters out there who are capable of unmasking this guy. The only thing I find kind of funny about this is some of the people who are extremely vitriolic against Arkham for docs to earn, which arguably is not that different than this, are really promoting this. And I'm just like, I don't know, like, pick a thought. If you're going to be like, you know, don't do docs to earn, then like, you can't kind, like, I get that this is more morally, you know, less dubious than generic docs to earn. But I did think it was kind of funny to watch like people's sides flip for this. particular case. Because we talked about Arkham in episode recent.

What's your take on Docs to earn generally, Turin? In terms of like a normative stance, do you think docs are generally fine, generally not fine, special circumstances, okay? I mean, I totally get it for the security reasons. And I, you know, I don't hold anything against people doing it. I just wouldn't want to be participating as a bounty. So there's a difference between like, do you view the someone else doing it, anyone doing it as acceptable, versus do you view yourself doing it as acceptable? I think I'm much more in the camp of like, hey, look, for the security stuff, fine. It makes a lot of sense for someone to do it. But I do think like the generic thing can get unwieldy. But I do think the people who are very vitriolic

against it were very anti-vitriolic here and being like bounty hunt this person. catch them, like, in a way that was kind of like, well, maybe the Arkham people proved something about you to yourself, you know? Robert, do you have a view on this question? Yeah, I mean, we had the Arkham episode two weeks ago, three weeks ago. I think, you know, what they're trying to do is get law enforcement information to find a hacker that stole funds from the users. You know, I think in this case in general, like, they're free to use capitalism to what I

think is creating more public good in the aggregate at the detriment of one individual who acted maliciously to steal the funds. So I think in this case, especially, you know, with the wild west of crypto, it's going to be more effective than them not doing this bounty and solving the problem. Yeah. Tom, what's your take? Yeah. I mean, I think it's kind of like a proof of innocence versus proof of guilt kind of thing, right? Like at this point, we know that this person packed the protocol so you can rightfully place a bounty on his head versus like, you know, requiring everyone to prove what they are, if, you know, even if they maybe have not done anything wrong, or it's purely for some

other self-motivated reasons why you might want to know what an address is. And so there's a little bit of that component to it too, but I think you used to learn more kind about the forensics that they're using to identify something. because I feel like there has been a pretty high success rate with doxing addresses in the past. And some of it is obvious, right? It's like somebody offboards through like a exchange. And it's like, well, the exchange has your KWIC. So it's like obviously that that's not going to work.

But some of them feel a lot more nuanced. And so I don't know what's going on in the back end. And I think there was a there was like the NICL 911 telegram channel that got announced this week of sort of a collection of white hats where you can report that your exchange has been hacked or your perorals been hacked. But I'd be curious to learn more about what their recovery process looks like and what they're doing beyond sort of the most vanilla stuff. Yeah, it does seem like most of the time they are triangulating with exchanges. And this is one of those places where it really is kind of a improvised set of rules. Right. There is no, you know, these exchanges are acting internationally.

There is really no binding constraint that forces them to cooperate. They just do because there is a certain moral authority when a big, important defy protocol get tacked and everybody comes together and is like, hey, now's not the time to be kind of stopping and asking, is this okay to do this or not? Let's just go roll up our sleeves and help. So I think it does feel quite different than just a generic like Arkham. I mean, the problem with something like Arkham is that you can really kind of use it for anything. You can use it to blackmail people. You can use it to threaten people. You can basically be like, hey, do this or I'm going to go put your name on Arkham and try to dox you and blah, blah, blah. I think there's obvious room for harm in the

same way that like in a in a prediction market like obviously prediction markets can do a lot of great things you put an assassination market together and also all of a sudden quote unquote capitalism can have really nefarious side effects on individual people and so I think it's wise to have some you know I don't know again I haven't spent that much time thinking about ARCA maybe they've already thought about all this and they're like yeah we take down we have some kind of moderation committee or something that takes down requests that seem to be malicious but it's hard to tell in principle right of like hey here's an address tell me who it is. But this case is one of the that's very clear.

And I think one thing you see over and over again is that pretty much everybody in crypto, whether they're in exchange, whether they're a stable coin issuer or whatever, like they will all rally around, you know, fuck the hackers. And this is one of the most obvious cases where this, whoever this hacker is, is clearly kind of a loose canon. And people, I think, will be very quick to try to bring them to justice if they can. The upside of this whole story is that it seems like now the worst is over. We were talking a lot last week about.

some of the side effects and some of the collateral damage, so to speak, downstream of Curve, like in Ave and in Frax. At this point, the CRV price has recovered significantly. It's now back at 60 cents. I think it hit a bottom of something like 40 cents, so it's up more or less 50% from the bottom. You know, all the loans that Michael Igorov has outstanding are pretty safe with respect to their collateral ratios.

And the TVL and Curve has also been recovering. So it's up to something like $2.5 billion. from a low of 1.5 billion. So it seems like it's more or less over, although it's always possible for something, something to happen again to trigger these fears. Tarun, you were spending a lot of time kind of making recommendations to some of these protocols.

Now that the dust has cleared, is there something that you feel like needs to change in Defi? Like, is there a kind of broader lesson behind the lesson or sort of a meta lesson that we need to learn from this, beyond just okay, you know, Viper compiler bugs, blah, blah, blah. Yeah, I mean, I think there's a lot of lessons that are kind of have to be learned about doing some of these things in the decentralized world. In the same way that, you know, the bug bounties are a little bit unstructured, I think the responses are unstructured. I don't think there's like an immediate set of things to say right now.

We spent some time writing a post-mortem and just posted it actually a little bit before the show. And I think one important piece to understand is every DFI mechanism has some amount of flexibility to either the users

or to the community at large that control it. I'm understanding the limits of those and knowing which protocols have which limits and which protocols don't have other limits is very important. And I think oftentimes users or voters and DAOs don't quite pay attention to those details. And so then they're like, you know, maybe don't necessarily view certain things as risky as they should. But I think hopefully over time that changes.

I think there's there was certainly a lot of things where particular protocols were like, hey, we completely solve this in all situations. And it's not, life is not so simple. There's always some tradeoffs that you make in these designs. And I think as long as the community is marching towards just greater understanding amongst all market participants in a way that's easy for them to understand over time, I think this will get better. But yeah, there's not like some like, hey, here's the God solution that solves everything type of answer, if that's what you're kind of hinting at. Is there not something around just, hey, having a single borrower, like thinking about a single borrower, like thinking about a single bar. I feel like that was a lot of the story was that like one dude basically had what like a third of all the outstanding CRV supply.

I think it was like 40%. 40%. Yeah. Is that not the story? That feels like a big part of the story. That should be the story. I mean, there's certainly that part of the story, but you know, there's also a lawsuit around that 40%. So, you know, I guess maybe you should just follow the lawsuit and you'll find answers. But I would say the bigger thing is just more like one thing, and this is probably true in all lending institutions, not just in Defi.

People are reticent to remove customers that earn them money. And removing assets, removing things is very hard for a community sometimes because they have sort of either a sentimental attachment to it, either a like hey it's actually generating it's a position that technically generated a lot or generates all the revenue and again i'm not just saying this is a defy thing this is true in centralized land too like look at jp morgan paying these like huge fines for lending to epstein right like archa goes like like my point is like it's not it's not something that's it's just that i think people in defy maybe some of the participants might be newer or hadn't quite thought through some of these hazard issues.

And, you know, even when presented with some of that information, kind of didn't, you know, versus like, well, this is, you know, our best customer or whatever. So I, it's a quite nuanced thing. I'm definitely not trying to, to say like, hey, like, X, Y, Z fucked up and this causes why, does it cause W? I'm just trying to work without the. I feel like you're being, you're being uncharacteristically diplomatic about this whole situation. Like, I feel like normally I would hear you like swearing and talking a bunch of shit

and like just kind of throwing your weight around. And I can feel you tiptoeing through this political landscape that you're, that it seems like you're in the middle of. So anyway, that's just a comment. I know that you, I know that you're, I know that you've got a bunch of forces. Maybe let me, let me ask Robert. We didn't hear from you last week. Obviously, the interesting thing was compound was the one lending market that did not have this problem. what's your take on the takeaways maybe what Tarun can't say? You know, my takeaway is, you know, and I'll use the Archegos analogy, there shouldn't be a situation in which one person has 40% of a token

and is borrowing up to pretty much the maximum capacity against 40% of the asset, whether it's a token or a stock, you know, in the traditional markets, whatever. Like there's so many examples of, of some entity that gets in horrible trouble because they borrow against a massive, massive, massive position. And it creates this reflexivity for the asset in general. Like, icon enterprises recently was in trouble because, like, he was borrowing against, like, his entire stock.

All of the Hindenberg research shorts have actually just basically been identifying public equities around the world. Where one person owns, like, a crazy amount of it and is levered to the hilt. against that asset. This is a fundamentally unsound practice. And like, I don't know how any system allows it, frankly. And I think there needs to be more thought that goes into designing systems where you can't just borrow against 40% of the supply of an asset, whether it's ICON or, you know, any other stock targeted by Hindbergh or whether it's curve. Like, there needs to be more checks put in place. And like, you might view them as a good customer, but it's a good customer until it goes

horrifically wrong. And so, first off, nobody should own 40% of a crypto asset. Second of all, even if they somehow do, like, no protocol should allow some of the bar against the entire stack. Let me, let me ask, there are a couple different ways to try to parse what you're describing there. So one, is the problem that there was 40% of the supply of CRV that was being borrowed against period, or was the problem that that 40% was owned by a single person whose actions were going to all be basically correlated instead of being a bunch of different borrowers? Great question. If it was a hundred different borrowers, borrowing against 40% of the supply of CRV, they would all have different liquidation thresholds, they would all have different

preferences, they would all have different vacation schedules. They wouldn't all be functioning at the same type of risk. That diversification actually will create. some safety. And like, it's not 40% of the supply that would be at risk if it spread across 100 borrowers. It might be like 5%. Okay. And when it's concentrated in one person, what you see is you see market participants hunting them. Like people on Twitter were joking around about this, but like very seriously, you know, there were market actors that were trying to liquidate eager of, right? They were doing so because one of his positions on Fraxland used an interest rate model that was honestly, frankly, quite interesting, but also dangerously led to people

trying to hunt his position by max borrowing against Curve in Fraxland to juice the interest rate model. Now, unlike most protocols where the interest rate models are like some chart that's a function of utilization. On Fraxland, this particular interest rate model was like a PID controller type model, where as a function of utilization over time, the interest rates, could go parabolic. And so on this market, the interest rate to borrowing this curve was going parabolic. It went from like 15% to like 75% over a day or so. And it was projected to keep on increasing parabolic to the point that Michael would have been unable to finance his position. He would have been liquidated. And there would have been this crazy potential cascade risk

if there was a liquidation on one market because it could have just triggered like a global decrease in the price. And so what wound up happening was he had to essentially start selling and unwinding his position, starting with Fraxland because that was the subset of the overall position that was most at risk. And so this really only happens when a position is so large and concentrated in one person. Like if it was split up amongst many people, who cares if one of them would have gotten liquidated, it wouldn't matter. But if it's one Uber whale with 40% of this flight, you're jeopardizing the whole asset. So here's the tricky thing, though.

So I agree with you, and that makes perfect sense, that you don't want the entire supply of an asset moving in lockstep in crypto, in defy, where we actually, we don't know. If Igorov had 100 addresses, there's no way in principle for compound or Ave or Fraxland to be able to tell that, oh, this is all the same guy, right? So how do you control that risk in defy? And in CFI, it's straightforward because, you know, you could ask the beneficial owners, you figure out, okay, even if you have some SPV, I'm like, okay, I know this is really you. How do you prevent that in Dify?

Well, the easiest way is for an individual protocol to cap or limit its exposure to anyone asset, like at a contract level. So, like, a great example is I think the current version of AVE allows this. I think the current version of compound allows this where you say, oh, the protocol's only willing to accept 15 million of CRV or like some amount that's safe enough that even of things went horrifically wrong, the protocol could shrug it off. That's the easiest way to avoid the majority of the downside risk. But the other way to do it is like, you know, you could basically, but I guess that's the easiest way to go about it. There's more complex solutions, but I think

the easiest is just limiting exposure to one specific asset, especially if, you know, it's not incredibly liquid. I mean, another point is like how, you know, correlated are the liquid prices for all the positions, right? There's kind of no difference between 100 borrowers at exactly the same liquidation price and one borrower. I mean, gas, fine. M-EV

also goes up, so, but roughly speaking, they should be roughly the same. And so how distributed a liquidation prices, how is the volume liquidity adjusted at each of those price levels?

This is the type of stuff that we spend long time analyzing. And so like it is it is there are ways to do it. You know, unfortunately, some of the early versions of protocols didn't have caps. And so that's kind of one of the reasons we got into this position. But I think, you know, in general, people are obviously hopefully learning a lot from this. I think, you know, like I said, I think community members in general probably hopefully have, you know, this is their hopefully archegos moment. and they understand that. It's definitely a very teachable moment.

It's one of those times where I think people are able to actually learn a lesson when they see how kind of close to the brink everything came in DFI. So I'm hopeful, and I think I'm very positive, that people are going to learn from this. There was another interesting experiment in DFI that just took place this week, which was with MakerDAO. So MakerDAO, their stable coin is called Dye. There's this yield that Dye pays out called the Dye Savings Rate.

The idea is that if you lock up your dye into this vehicle, call it, which is the die savings rate, you can earn a yield just passively sitting on your dye. There's a few other kind of cute things like Chai, which basically allows you to tokenize a dye that's sitting in the die savings rate. But basically the die savings rate for the last, however long, has not been very popular because the amount of yield that the die savings rate was paying was very low. I believe it was like, what, 1% or sub 1% for a while? And they recently started jacking up that rate.

So that rate, treasuries obviously now are north of 5%. The die savings rate was up to, what was it, like 2 to 3%. It was like 3.3%. And then recently they started jacking it out. It went up to 8% about like a week ago or a few days ago. And that's kind of what caused this entire sort of situation now, I guess, with the DSR. Yeah. So basically there's this thing called the enhanced die savings rate, the EDSR.

And EDSR is this experiment to what happens. if we jack up the interest rate on die to 8%. Now, that is a, that's a rate that's sort of at the margin. As more people start depositing, and very quickly you had, I think, over half a billion dollars of diamonded so they could capture the yield in the EDSR. That yield has come down significantly.

Now it's, I think, closer to 5% as opposed to 8. But my understanding of this experiment was it was a way to kind of see if we offer more yield to borrowers, What is the elasticity of demand, such that, you know, how much more capital are we going to get into the protocol and more diamonded and potentially more liquidity created for our Sablecoin? Just by offering, you know, a relatively small amount of yield that we don't have to pay for a super long period of time. And so I think this is, in a way, Maker Dau kind of experimenting and trying to model out what happens if we start offering one element of the endgame plan for MakerDow, which is this kind of big master plan that involves potentially yield farming. if we start offering additional yield or juicing the dye yield, what's going to happen to the dye supply?

And how can we model that out? And it seems like Maker is sort of poking and trying to see what happens if we do this. And if we do if we do this, we want to make sure that we're being intelligent about how much we're spending on the yield and attracting more deposits and trying to make a play at becoming this super stable point, let's say. Any thoughts on what Maker's doing here? Well, I think if they really want to go for it, should just, you know, take Anchors 18% interest rate and start there. Because 8%, I mean, yeah, that's not going to get you the $40 billion of stable coins. Well, the difference is that

anchor guaranteed 18% regardless of utilization, which is completely insane. Maker was only offering this for the marginal dollar. And as more people came in, the yield went down. Of course. I was being slightly facetious. But just to be clear for the audience because I think a lot of people would hear that be like, oh, shit. Yeah, it is, it is very true that one thing I've realized being in crypto for the last five years is there are still lots of people I talk to who don't know the difference between USDC, die and UST, and they're all just like, oh, they're all stable coins, all bad, all good. There's no, like, there's no nuance to like splitting the, splitting them into good basket,

bad basket. So, yeah, it is, it is kind of funny. I think the idea of doing these elasticity experiments is interesting, although I think this one was kind of obvious that it was going to be dominated by large players because the key metric is the spread between borrowing against your dye versus sort of minting new dye. So the idea is like, hey, look, if I can borrow dye and pay 3%, but then I can earn, and then I can use that to create new dye and earn 8%, there's a 5% spread, and I can recursively do that a bunch of times. I think the end games goals were not necessarily to like reward the largest holders,

but of course, just structurally, such incentives generally tend to be harder to capture for the smaller users. So it's interesting to see they're already like kind of changing and, you know, we'll see what the outcome of this experiment is. Yeah, it feels kind of like a wash. Like basically the amount of dye in circulation went up by almost exactly the amount of incremental dye that was added to the DSR. So it was definitely like to run like these whales who come in see this carry trade where you can borrow

a die at 3, 4%, you know, deposit it and then 8% and like, you think it's I think the idea is not wrong, right? It's sort of like, you know, a bank offering like an interruptory rate that obviously goes away after a period of time. But it's not really having the intended

outcome of like actually introducing you know, meaningful incremental demand for, for dye and sort of people actually being able to treat this like a savings account in some way. Yeah, it does appear like the financialization of DAI is mostly going to appeal to smart traders or going to arbit away pretty quick. That said, it did, I don't know if this was the intended effect, but it did get a lot of eyeballs with the headlines of, oh, my God, die is offering 8%.

I don't know if that's good or bad because I think most people it's pretty obvious that that's a negative interest rate margin for die. I mean, for some people, they might be like, ooh, maybe I should start paying more attention to the Maker-Di ecosystem. I think for some people, it's like, oh, God, I guess we didn't learn any lessons from 2021. So I'm a little bit mixed about, like, whether,

even just like the headline of, you know, whether or not it's a short-lived experiment, the headline of die is offering 8% now in the DSR is actually good or bad net for Defi. Yeah, I agree. I think a lot of people have PTSD to like Tara and anchor and at a time you say stable coin or any yield related to a stable coin, it kind of flashback to that.

I think like the DSR is also, I think, kind of strange. Like, die, you know, the end goal is right. You want this to be an actual currency that is accepted that people use, that they save in, that they engage in commerce in. And that's really only possible if it's liquid if people actually use it and it's out in the world and there's demand for it. Yeah, with the DSR, yeah, they have chai, which sort of this like wrapper on top of the DSR that creates a tokenized version of your deposit.

But the die and the DSR isn't out there in the world, right? Like if all the die were in the DSR, it wouldn't really be a very good stable coin because it would just be locked in this contract and not really tradable. And if you sort of split up between chai and dye, then they're not really, you know, liquid either. And so I understand sort of the incentive, but think from a product perspective, it's just not super usable. Like it doesn't have the right ergonomics for what I think maker is trying to,

trying to do with it? It seems like the DSR is kind of like the risk-free rate or something in Defi, where if the DSR increases, it kind of forces interest rates everywhere to follow in lockstep. So it is, I mean, it's not literally risk-free for obvious reasons. I mean, it's tied to a particular Defi protocol. But it seems to have that effect in some way on the yields across Defi. And it's interesting because I haven't thought about this very much. So I'm curious Robert Turun, how you think about this.

to Tom's point, if you raise the rates of the DSR, right, that basically kind of incentivizes people to create die, but then kind of lock it up. So essentially just like lock up ether, or lock up, you know, whatever they're minting die with, USC, whatever, and just have it sit there and do nothing. Whereas if you just instead lowered the interest rate on vaults, you kind of have the same effect,

but you increase the circulating dye instead of increasing the amount of collateral that's just sitting there doing nothing. But why would you want to increase the DSR and not just lower the interest rate on vaults? Well, I think you could. Like, I think at the end of the day, what they're creating is a system in which, you know, interest passes from those who are borrowing to those who own the asset. And it really should be the same roughly to increase the interest rate that you're paying or lower the interest rate that you're charging to get to the same.

equilibrium supply and demand dynamics, roughly. But I think in this case, because the savings rate was so low, it just makes more sense to start on that side. One tiny thing to note in that regard is interest rates are strictly positive numbers, and so they're sort of going down has slightly higher, you could think of as higher weight than going up if you have equiproval. ups and downs. Partially, this comes from the fact that sort of like the geometric mean is less than the arithmetic mean. Maybe that's too much in the weeds on this. But the idea is like

when I'm looking at things that are ratios and I can only go to zero, like zero is a point of infinity and infinity. Like I'm always a finite distance from zero, but I'm always an infinite distance from the maximum possible rate. So I'm sort of like biased. And so that sort of means that your lever going down, which is sort of the collection rate, is like less effective than the increase rate in some way. It's like less elastic, which is why you might see the experiment behave better by doing, you know, increase the interest rate paid by 10% versus decrease the borrow cost by 10%. It's unfortunately this like tiny, there is some tiny bias to those two, but the technical detail.

Yeah. This Muni Supply, who's involved in Maker Governance, which talks about this sometimes of like, you know, there's sort of this mental model of how we think Maker works here. We think DeFi works of market is very efficient. Capital is very fluid. People move to, you know, normalize rates across different venues. In practice, like, it's actually not that elastic when you actually adjust interest rates.

a lot of people still stick around that they don't even realize that the rates are moving. There's a lot of lag and it's not quite as responsive as you would imagine. It would be in like a truly sort of efficient, optimal market.

So maybe some evidence to Turen's point. We just need more AIs involved in this process who are making all these decisions for us instead of human beings. Well, I hear the AI hype has taken a summer vacation a little bit.

Is that right? At least that's what all the, the API and views numbers seem to suggest. Okay. Well, let's stay on the stable coin subject for just one more story. So the big story this week

that a lot of people are talking about is the new launch of a stable coin by PayPal. So this stable coin is called PYUSD, PiUSD, which somebody called Python USD. So this is a stable coin that's issued by Paxos, much like BUSD, Binance's stablecoin. Paxos, you might remember,

kind of took a beating last year, earlier this year, I think, earlier this year, for their involvement with Binance. But they're back at it. So they're launching this stablecoin. It's going to be primarily, it seems like it's going to be primarily used by PayPal itself. I haven't seen any exchange listings yet for PYUSD or any other distribution partners. The fees on PYUSD look to be pretty high.

So if you're buying PYUSD or swapping through it on PayPal itself, you pay one point on a swap that's north of $1,000, which is pretty rough. That's like, you know, that's coin-based fees, basically. It's like a pretty aggressive fee schedule. But apparently that's like par for the courts for PayPal. They seem to charge pretty high fees for consumers. I've seen very mixed reactions to this story. I think some people are like, oh, look, this proves that, you know, Web 2 companies.

Obviously, PayPal is the largest fintech, as far as I know. I think it's still worth like $70 billion by market cap. So it's a very significant company. PayPal doing this is a vindication. that, hey, you know, traditional fintechs see the importance of crypto. They see the importance of stable coins. And they take this industry very seriously. Tom, can you pull up a chart of the coin itself and we can see some of the numbers on.

Wait, is it actually live? I thought it was just announced. Well, I know the token has been deployed. Yeah, so trying to contract that solidity developers have enjoyed flogging. Yeah. So there's been a lot of, there's been a lot of smack talking on Twitter about the particulars of the contract. Apparently the contract was written in solidity 0.4.2, I believe, which is a very old version of solidity. 20,000. 29,000? 26 mil. Also, only has six decimals, which people were annoyed about.

And it does to make it ergonomic with USDC, you know? Yeah, I, no, no, I actually generally am like, all right, who cares? It's not like these stable coin contracts are, like, pushing the engineering limit of any programming language. They're always like the most. most boring contracts you could look at. So who cares that they're not that great? Okay. So there's 27 million tokens right now, PayPal U.S.Es. There's nine holders, though. Yes, nine holders. One address is 25 million. That's probably PayPal themselves, no? Yes. Yeah. I assume. Treasury and then a bunch of that. That's got to be. That's got to be PayPal on behalf of everybody.

Yeah. And the rest is like 20 bucks, 16 bucks. These must be like tests. So right now it seems like only Paxos and PayPal themselves. I say we challenge ourselves to play a game live on the air where we predict the total minted supply of PayPal USD on December 31st and the winner gets the case of Red Bull. Yeah, yeah, a lifetime supply of Red Bull that they're going to send us. I hope PayPal is listening and that PayPal is willing to fund this endeavor because I promise it will only bring you great success to your future stable coin. agree okay in order to do this that we have to do a blind so does everyone have like a piece of paper

that they can write their guess on okay everybody type it into the private chat okay just then we all can enter on the count of three okay okay okay hold up hold up all right um okay three two one okay I'm gonna read off the I'm gonna read off the guesses not very bullet highest to lowest, okay. Hasib, Hasib has guessed that PayPal USD will end, 2003 at $499 million. Tom has guessed $250 million, Tarun $100 million, and Robert 70 million. Okay, I guess we'll check back at the end of the year. Viewers make a note in case we forget that we have a lifetime of Red Bull supply

that we got to supply if, to whoever, wins. Whoever's closest. Whoever's closest. Whoever's closest. Okay. Interesting. Are we doing, are we doing geometric mean or are we doing, no, I'm kidding. I leave that to the room. All right. Okay. Interesting. All right. So why do you guys think it will be so low? I guess, Robert, you had the low assessment. Why do you think it'll be 70 million? I don't think there's enough incentive for people to pay the one and a half percent fee to get PayPal USD to take it into token form to use it. It's like, it's not like they're going to be like, you know, know, incentivizing some curve pool and, like, creating some crazy yields on, like, PayPal

USD. Like, so frankly, I think it's going to be really low because I don't think by year end there's going to be that many use cases on chain or reasons to hold it. And so people, like, experiment with it, but, like, no one's going to be, like, dropping the big bucks into, like, PayPal USD. My logic is just more that the only announced use case they had is that PayPal and Venmo will both have wallets and you can transfer between PayPal and Venmo via PiUSD. I didn't see any other use case, but again, I could have missed something from the announcement. That use case, I'm sure that

they will fill with some amount of money as like their normal business process instead of them doing it via bank accounts. They'll do it via Pi USC. And they'll maybe waive the fee for themselves effectively because they're paying themselves. I just can't imagine that they'll do a huge amount of volume. I don't know. I was trying to guess what the volume of that is by end of year. And that was my guess.

Yeah, it's very fair. Just a point of clarification on the fees. So far, the understanding is it's for going from any crypto to PiUSD, it's 1.5%. But there's no fee, obviously, on dollars because then you just kind of create this weird gap and spread. And so it's similar to Coinbase in that regard, right? You can just go through the Coinbase app. You're paying like 2% or something crazy.

But it's free to mint and redeem, you know, USDC. I think the issue is just, I agree with everyone else. There's like not really an incentive to use this thing. And I think overall, like, crypto has not been a huge winner for PayPal. I pulled their 10K from last year. And they listed their AUM at 600 mil for their crypto product. So overall, like very small. And scaling off of that, I don't really know where the incremental demand is going to come from from a new stable coin that doesn't really seem to have any benefits.

Yeah, I mean, I largely agree. I mean, we all pick small numbers, right? Nobody picked a big number. Not you. Well, see, you have the biggest guess so far. Yeah, you have the biggest guess of $499 million. I was hoping somebody's going to pick $500 and I was going to get right below. Yeah, exactly.

But unfortunately, I ended up picking just a very, very large number. And so I think Why did I choose this number? Honestly, like even really crappy stable coins have managed to achieve reasonably large numbers in crypto

because I don't know why. Like they just happen. They just do. And it's like hard in ex ante to think of the reasons. I don't think this will be on finance, which might be my reason for why I don't think we have to think. Maybe not.

Maybe not. I mean, was GUSD listed on Binance? was, I mean, obviously true USD was in Gemini land though, right? Like, yeah, yeah, yeah, that's yield farming. They also paid people to mint it in the early days by selling it at a discount to all the OTC desks and trying to like jumpstart. I just don't say PayPal doing any of that type of stuff. Like it's just harder.

So maybe, maybe, maybe. Maybe they won't. Maybe they won't. But I think for PayPal, I think they find this business line clearly more palatable, especially in this market environment, than crypto itself, right? I don't think PayPal was ever going to go hard on crypto, qua crypto,

but I can absolutely see PayPal going hard on stable coins because they see the market. They see that this market is big. They see circles getting a lot of revenue from sitting on all that USD supply. And I think that when it comes to targeting enterprises, as opposed to targeting consumers, per se, I can imagine them having a lot more success,

not a lot more success, But I can imagine them making a dent in what Circle is going after, which is like trying to get corporate treasuries and other things to use their stable coin. So I think there's a story there. 500 million is not a success story in stablecoin land, right? $500 million is a tiny, tiny stable coin by most measures. But it is what happens when a really large, very, I mean, to be clear, like PayPal is three and a half times as valuable as Coinbase. It is a massive, massive company.

So if they want to really, you know, like grab the shovel and make a move, they could. Well, we're going to see who gets that case a Red Bull and how sweet it will be. We're doing a case or we do a lifetime supply. The irony, though, is that like the utility of the three of you combined for this is, like, infinitely lower than mine for lifetime supply. But a lifetime supply. Let's do case, then. Let's do case. I was the one saying lifetime supply.

I don't even know how you measure lifetime supply. Isn't it a lifetime supply like 30 years? I don't need to be in like those commercials in the 90s. It was like technically this is 30 years or something. I finally saw the movie you guys recommended to me or TV show Pepsi wears my jet about like that like, I bet you there's some court case on this lifetime supply things because no one ever says it anymore. I feel like it was like until sometime the mid 2000s and then it like disappeared. And I kind of feel like someone like actually went after someone for not giving them their real lifetime supply. How much would a lifetime supply of PayPal USD be?

Okay, I'm looking up on Reddit. Apparently it says the lifetime supply is determined by the rules set by the company, different companies do different things. All right, so here's what we'll do. We'll find a commercial from the 90s that specifies a lifetime supply of a drink, and that's what we'll use to adjudicate this bet. So we'll do a lifetime supply according to 90s rules of Red Bull. also receivable also receivable in dollars or in PayPal

not in dollar not receivable in PayPal USD payable in PayPal USD or in Red Bull those are the that's why PayPal that's why PayPal has to pay for this because they're getting minters hey hey that's true that's true they need it they need it they need it this is the yield part I will we will accept red we will accept PayPal hats for how much we've been shilling PayPal minting I don't think we're shilling it I think we are giving an honest analysis and critique. That's true. Actually, my question is like, okay, let's maybe zoom out a little bit. If we look at like the most successful non-exchange public company that sells crypto,

it's like probably block, I guess, slash square. Yeah. Because of like, so one question I've always wondered is like, why have, why has PayPal or other syntax players like in general? Like I know a bunch of them left, like Revolut just left the U.S. market and stuff. but why hasn't anyone been able to beat block at the Bitcoin sales? Is it really just because Jack of Jack's marketing? I can't be that.

Well, okay. So it's because the cash app Ux for buying Bitcoin is really quite simple. And like I know a lot of normies that are like, yes, it's easy to buy Bitcoin using cash app. And like that's it. It's easy to buy Bitcoin using cash app. But it's just. easy now using Venmo and PayPal, right? Like, they added all the same features, but they don't seem to be getting them. People actually use cash app as their bank. Like, if you're young, like,

you just keep cash in there. Rappers are not wrapping about PayPal. A lot of rappers are rapping about cash app. PayPal, PayPal, Red Bull rappers, two R's, the two R's of success for you. Yeah, I'm looking at the defilemma list of stable coins. And there's some very strange ones towards the bottom. Bring up the page. What do you got? Yeah, great. Come on. We got to get We got to get this on video. We got to see. UnstableCleance.com. I should have done another data quiz.

Can you zoom in? All right. What do we got? Val U.S.D. Bob. Davos protocol. Dollars on chain. One of my favorites. L-U-G-H.

I believe that's Lug. $6,000 outstanding. Wait, is that this real cello dollar? No. It's just no way. The data is wrong here. It's got to be. Yeah, I was like, wow.

If it's that. no way. There's no way there's 160,000 outstanding. Yeah, this is 42 mil. I don't know when this data is wrong, but. $0.98. Neutrino. Ooh, I remember neutrino. Oh, yeah, that's trading at four cents. Wait, how, at what, actually, we should, we should look through our guesses. When do you delist? Hold on. I want to know how many of these are off peg. Spice USD 28 cents, meme dollar. 0.02 cents. When do you delist a stable coin as like not a stable coin?

Before you go up, we still have another way of having a bit of gambling slash guessing, which is each person has to guess what rank the first coin that is over their number is. Right. So like say like the 50th, say the 30 first coin was the 71 million. This is too hard. No, no, no. What do you mean? Guess it's the rank. All right. I guess. Fine. I guess that 70 million is staple coin number nine. I think 15. I would guess $500 million is $6, $5, $5, $5, $5, $5.

I guess $5. $6 for $250? Yeah. All right, let's go. Let's check it. Let's check it. I was close. Okay, no, you're eighth.

So if PayPal dollar's eighth, then I'm right. They just have to be above Pax dollar. That does not seem hard. To be like at the same as Pax dollar or Gemini dollar? That seems very doable. That seems very doable. I think I'm going to win this. Also, I think one.

I think it's not to call it. I think I kind of got some lot. If you guys want to buy out, I will, I will accept partial payments if you guys want to buy out of this. I don't even think you guys want the prize. Only I want the prize. It's cash settled. It's cash settle. It's cash settleable.

Whoever wins can sell it to Tarun who's willing to pay the highest price. Okay. Okay. Wait, wait, one thing we forgot to talk about. I remember the first time the PayPal stablecoin news came out, maybe 2021. Wasn't it supposed to be on like another chain, not Ethereum? Or was it supposed to be on Ethereum? Always.

Sorry, I thought it was supposed to be on an L2 and they made the switch to Ethereum. Okay. I think maybe not an L2. I was reading this story as well. I forget where they were supposed to be. Yeah, something changed in this announcement. I read this announcement. I was like, something's odd, but I can't figure out exactly.

Oh, it was supposed to be on Salana. Was it? Okay. I think so. Yeah, yeah, yeah, yeah. I thought I read it was originally supposed to be on Solana. And after the fall of FTX, they pivoted. Interesting. But I'm literally remembering the tweet. Yeah. You thought one is a weird choice if you want a mainstream stable coin, but, um, I don't know. Is it? Doesn't seem like a weird choice. For PayPal, for people going retail heavy. Yeah. Prior to FTX, right, so much USDC was on Solana.

It's not so... Yeah, but I mean, somebody on Twitter made this point. Like, you mined something on Ethereuml1. It's going to find its way everywhere. You know, if people actually are using it, they can move it on Arbitram. They can move it on. I think that wasn't as true in 2020, like mid-2020, because everyone got very afraid of bridges. Like, there was, like, a bit of reflexivity to that.

So it's not obvious. Yeah. Okay. Well, I think we've successfully gone full D-Gen on the show. We now have a bet. I'm going to do a bit of research on what 90s lifetime supply rules are, and I memorialize exactly what the content of the bet is this week on Twitter. Well, we should just get Red Bull to agree to give us what they define as a lifetime supply.

Totally, totally. But assuming they don't do that, assuming they don't do that, then we're going to have to settle it ourselves. We have to pony up the line. That's right. That's right. So for now, we'll check back in next week. That's it.

Thanks, everyone. Thank you.