# Superliquid Mechanisms for Decentralized Stablecoins by Ariah Klages-Mundt | Devcon SEA

- Speakers: [Ariah Klages-Mundt](https://streameth.org/speakers/ariah-klages-mundt)
- Channel: [Devcon](https://streameth.org/devcon)
- Date: 2025-10-07
- Duration: 25:34
- Watch: https://streameth.org/watch/yt-TdAa95XDFSw
- YouTube: https://www.youtube.com/watch?v=TdAa95XDFSw

## Description

USDC and USDT outpace decentralized stablecoins in large part due to their liquidity. This talk covers the theory, data, and risks of stablecoin liquidity innovations. This will include mint/redemption mechanism design, liquidity pool design, rehypothecation, and protocol-owned liquidity. The analysis will distill how the flexibility of decentralized stablecoin issuance mechanisms can safely be used to their advantage over centralized stablecoins, which Gyroscope v2 is putting into practice.

Speaker(s): Ariah Klages-Mundt
Skill level: Intermediate
Track: Cryptoeconomics
Keywords: Mechanism design, Economics, AMMs, defi

Follow us: https://twitter.com/efdevcon, https://twitter.com/ethereum, https://warpcast.com/devcon
Learn more about devcon: https://www.devcon.org/
Learn more about ethereum: https://ethereum.org/ 

Visit the https://archive.devcon.org/ to gain access to the entire library of Devcon talks with the ease of filtering, playlists, personalized suggestions, decentralized access on Swarm, IPFS and more.

Devcon is the Ethereum conference for developers, researchers, thinkers, and makers. 
Devcon SEA was held in Bangkok, Thailand on Nov 12 - Nov 15, 2024.
Devcon is organized and presented by the Ethereum Foundation. To find out more, please visit https://ethereum.foundation/

## Transcript

[Music] hi everybody I'm Mariah clausman I am a co-founder of gyroscope which is a decentralized stable cone protocol and liquidity protocol uh and today I'm going to talk a bit about uh how to make decentralized stable coins more Super liquid and this talk is going to be a mix of uh our work building out gyroscope as well as a background of academic work uh coming from our phds uh that went into actually building gyroscope and it's going to be a largely high level talk uh focusing on the concepts but also trying to ground it in some uh some real examples so let's uh let's take stock of the stable coin space and kind of start at the top so there's something like a stable coin Renaissance right now uh but at the same time uh usdc and usdt are largely dominating uh the the the space and uh what we see is there's a surge particularly in centralized stable coins coming to Market so this is like new contenders like payal's PUSD uh as well as Athena which is a delta neutral stable coin and other real world assets mainly backed by by T bills um and at the same time there's also an increasing maturity uh in decentralized stable coins so one example is there's no more Tera there's there's no more Fay uh and there there's new Innovative mechanisms that are coming to Market things like curve USD uh gyroscopes guid which which we built obviously uh and several more examples but as a category decentralized stable coins haven't really grown in the last two years and in fact they've kind of shrunk in that time um so the main question we want to ask is is why and at least one reason uh which is the focus of this talk is that scaling liquidity at Peg is is very hard and actually this applies to to any stable coin uh but particularly to decentralized stable coins as they're set up now so one primary challenge is attracting any liquidity in the first place and uh basically new stable coins they have to attract new liquidity uh and it Compares with like usdc and usdt in many ways it's like an inherent disadvantage uh because usdc and usdt have entrenched use cases and a lot of natural trading uh demand and pairings which basically means that uh usdc and usdt LPS are independently profitable from from trading volume but when you start a new stable coin you have to start from zero you don't you don't have that it's basically like a chicken or the egg problem and it's very costly um because the the cost to attract stable coin capital is basically between 15 and 30% these days uh and also borrowing stable coins uh is also similarly expensive the second part of the challenge is competing with entrenched liquidity that's that's at Peg and uh the scope of the problem becomes pretty clear when you start looking at historical price deviations uh and this is something that also affects all new stable coins so the chart here is showing new centralized stable coins that have come to Market and showing a histogram of uh price deviations from Peg for for those new stable coins and these are pretty major players including like Pi USD and and Athena uh and and one thing to note is just uh how regular the deviations of like 20 Basin points of or more uh actually occurs for these pretty major stable coins which adds quite a lot of friction in terms of the usability of these systems so from the usability angle uh it one thing that's very important is how easy it is to move in and out of a new stable coin especially with with low slippage being very important and so 20 basis point deviations or more can be uh make it a lot more difficult to actually use the stable coin uh this is how decentralized stable coins compare so actually you see a lot of similar problems with a lot of uh uh kind of like spread out pricing uh but something that's uh that's that Sparks some optimism is that uh there are actually two decentralized exceptions here um so there's both D and gyroscopes gyd um I guess the die is a little hidden in the in the graphic but it it shows up very similarly um uh and this basically just presents some some ideas that maybe you can actually do better by using some advantages of of of decentralized stable coins and then the third part of the challenge is that it's just very difficult technically especially for decentralized stable coins and walk through a few of these reasons um when when a stable coin's price is expected to move around uh it becomes harder to get deep liquidity for it and this is just a practical consequence uh you need to spread liquidity over more prices um and and you need to convince people to to do that and take the risks of Market making over that range of prices um and why does this does this happen it usually comes down to frictions about the stable coin design itself so for instance in minting or redeeming or otherwise issuing the the stable coin uh which can come down to things like fees and speed of settlement uh one thing to note is that like offchain settlement which is what you basically have to do for centralized stable coins can often be very slow as can kind of cross chain uh kind of mechanisms where maybe if you mint on one chain but you have to bridge to another chain that can lead to kind of delays uh in the worst case up to like possibly a week uh another another thing that comes into play quite a lot is that uh you often have to balance the interest rate uh between uh for a leverage market for for when you issue stable coins in in a borrow Market uh and the the reason here is that the main use case is leverage for for these sort of markets uh people usually borrow the stable coin and then sell the stable coin uh and the interest rate basically has to keep the stable coin on Peg by balancing the supply and the demand in in that market or else the stable coin D pegs uh and then you can also have kind of like shocks to underlying assets and uh various risks uh that we actually did quite a lot of work in in our in our research on on this uh but but that's kind of not the focus of this talk so uh we'll kind of leave that to there's another good talk at at previous deves actually if you're if you're interested in that but let's walk through one example here so this is a case study of go which illustrates both of these kind of frictions coming to uh uh coming to fruition at different points so the price of go has been pretty volatile since it since it launched this is the the pricing from this year um and there are two main reasons to this the first reason is that in the the first part of this chart uh basically there were frictions to the interest rate borrowing the the the balancing the borrow Market uh basically the go to market strategy uh relied on providing below uh below equilibrium interest rates to kind of like incentivize Supply uh but that basically means that there are too many people trying to leverage and sell and you end up just with a with without a balanced market and then the later part of this chart uh is is another problem so they they they fixed uh the the the interest rate sort of issue and go or or changed the the rates uh but uh there was a kind of an excess of incentives at one point uh they did have a like onetoone mechanism to to Mint new go against like usdc and usdt but unfortunately this was exhausted and so you basically see kind of like above uh above Peg happening uh in that realm but then they they fix that problem but this kind of illustrates like where some of these pricing issues can come up okay let's take one step back before we go any further which is why should we care about decentralized stable coins anyways um so I mean first you should think about like from fundamentals I think a lot of us here care about actually like building decentralized things we care about uh bringing too much counterparty risk in or censorship risk or Geographic risk or kind of like tying to particular custodians or kind of blocking into particular vendors and uh this is a risk if you use too many decent or centralized stable coins that you're basically embedding this into the what we're building in in Defi and what wasn't the point of defi basically to make things as decentralized as possible but if you're still not convinced then you can just also take the uh the Practical Viewpoint which uh there there's a very good example when usdc deeg in the svb crisis it basically took down defi for for a few days because this is how defi was uh was constructed at the time um and in principle we can build things that uh that don't have this susceptibility and and that's another reason to kind of like build decentralized protocols that can manage the risks in a decentralized way without relying on a a centralized institution to do it okay so the rest of this talk is going to go through a bit of uh how do we make decentralized stable coins uh super liquid and we're going to run through quickly what the state-of-the-art is including some examples of what doesn't work and then go through uh what we think is a better way actually to to to solve this so the the current state-of-the-art focuses on two parts the first is around designing primary markets for the for for a new stable coin uh and so primary Market basically means minting and redeeming uh for a decentralized stable coin in particular onchain mint and redeem um and one benefit to these decentralized stable coins here is they can do this onchain which means that there's potentially a lot lower frictions than for a centralized stable coin so in particular you can do mint and or de atomically so you can actually like uh kind of balance markets using flesh loans instead of requiring market makers to take inventory uh and this is actually why gyd and D in in one of the earlier pictures uh was actually achieving a much tighter pricing than than even Pi USD because it has these advantages in its effectively mint and redeem market and another idea is you can actually segment the uh uh uh where you need liquidity using mint and redeem you can basically let the primary Market handle the extremities of the pricing region and then you can concentrate uh the secondary markets on the the areas that are most important within kind of the minting Redemption bounds uh a slight caution here is that there's a lot of risk to to primary markets and and building them in the right way and this came about as as an example also in like the svb crisis um there's a whole other Devcon talk from last year actually about how do we build these uh in the best way uh so I encourage you to check that out um but but yeah we'll just kind of give a quick nod to to that here uh you should of course be conscious of the risks that you're building into these systems and one other topic about primary markets that I want to touch on is uh what's another version another flavor of them that's called Os or poos meaning algorithmic market operations and P meaning protocol owned liquidity so these are effectively uh where you have something that's like a pre mined or not yet issued uh stable coin that's been uh deposited into a liquidity pool and basically somebody can issue the uh the stable coin by buying it from the liquidity pool um and this is basically equivalent to a mint and redeem sort of flow but where the amm price uh the price curve is actually providing the the the pricing for for mint and redeem and uh it also has the benefit of being able to kind of bootstrap secondary markets because uh you're basically depositing into something that gets blended with secondary Market liquidity and how does this affect liquidity at Peg well actually it comes down quite a lot about specifically how it's implemented and the mechanism design around it uh for for gyd it actually works very well and there's an example here in this uh uh an illustration here uh it's able to help bootstrap liquidity at Peg very well um but there are also cases where this doesn't work very well and it comes down to a bit about either bad design or possibly kind of like miscalibration about exactly how the how the flow Works um so a quick case study about that is basically digging into what happens in fraxos so in a frax Amo uh there's some injection of like this printed frax into curve pools um but frax tends to trade below Peg like kind of despite this mechanism trying to to support it and what you end up is with is basically pools that aren't optimized for kind of the pricing of of fracks which you can see here you can see something about like the density of uh liquidity in in these liquidity pools versus uh the actual histogram of where prices are are trading the other side here is secondary markets and I'm kind of going to Breeze through this a bit um but the main idea is that you can kind of optimize liquidity density um by instead of picking default calibrations for a liquidity pool making something that's much more customized for the actual prices that you want to incentivize as a stable coin and the graphic here kind of illustrates the concept that if if you can that you can basically squeeze more liquidity out of your liquidity pools if you can uh kind of identify the right bounds and also like focus in the right ways um and we've done this quite a bit as part of of gyroscope there there are a few examples um but I'll Mo mostly leave these here for for later reference this these came about from uh uh our work calibrating gyro pools um but there's many examples where you can just get a lot better kind of liquidity profile uh by making a much more customized pool than by using kind of like a a default stable swap pool and this chart also kind of walks through that it actually works by kind of like comparing Capital efficiencies but we'll also leave this uh you can check out check uh check on it later if if you're interested what I want to spend the rest of the presentation on then is what we think is a new way to solve decentralized stablecoin liquidity and kind of the motivation here is that there's a there's a unique ability to decentralize stable coins and it's that you can issue in new ways that wouldn't really have been possible in central stable coins and so the classic example here is something we just saw actually uh for like protocol owned liquidity and andos uh a centralized stable coin just can't do this it can't issue new stable coins in this printed sort of way uh so so that shows that uh that it exists but the idea we want to pursue is like can you build on this flexibility in a new way that might be more efficient and might also be safer and so this is where we want to announce what we're actually building next in gyroscope uh our V2 um this is actually the first time we're talking about this mechanism so you kind of follow us if you want to hear kind of more about where this is going uh but I'll try to give you a little flavor quickly so uh it builds on a concept that we call duplex yield and I'll illustrate this what this means in in the next couple of slides but this slide gives a general overview about like what sort of Market is is being set up so on one side of the market we have idle asset holders so this could be something like Bitcoin holders lrts uh or other kind of like top yield bearing positions already these are earning what we're calling this duplex yield and it can be also be earned on top of other yields the other side of the market here is uh asset asset issuance protocols so stable coin issuers being one example you could also consider like LST issuers uh and what they're trying to get out of this mechanism is cheap scalable liquidity for their assets uh and possibly also unlocking a new Revenue line um but an important thing at the same time is that you also have this third group actual people who are using your stable coin and we want to make sure they're shielded from risks we uh and and this is through an idea that we call duplex collateralization so let's dig into a little bit about what this means I'll give you a little intuition so uh the most efficient way to Market make is often to borrow assets as opposed to holding assets on the book and duplex yield essentially is coming from a more efficient way of Market making stable coins through kind of like an optimized borrow like mechanism instead of requiring LPS to hold it on the book and it works through something we call a secured swap line um so this creates basically an ability to exchange one stable for another and it's insured with with a collateral position and the idea of where lower cost is coming from here and efficiency is that uh you can actually through this mechanism use something like Bitcoin which has a much lower hurdle rate lower cost of capital something like three to 5% uh in to to Market make stable coins instead of requiring the uh the hurdle rates of 15 to 30% if you're actually holding the stable coins on the book and you can stack these yields on top of each other so we actually think the future yield stack will look something like this you have all of these different uh staking yields already and the most effective yield we think will add this duplex yield this Market making sort of yield on top and so the idea is uh I mean being more specific than the intuition is to streamline a sort of borrowing mechanism uh that is restricted to codified Market making um main idea here is that you can basically uh price discriminate between the leverage Market which uh ends up having very high borrow costs because it has to as we saw in the previous slide it has to balance the supply and demand of the of the leverage Market there's a lot of demand for leverage um but you can price discriminate uh between this leverage market and other markets like this uh sort of Market making market and that a decentralized stable coin can do this in a uniquely low way because it has this flexibility about uh about issuance and also wants to Foster liquidity for the asset so this slide walks through kind of how a secured swap line actually works um I'll only be able to give a quick flavor um but you can check back in on it later and there'll be more information coming out as well um there's kind of like three stages to it you start with the idle asset that uh that you want to earn a yield on so something like Bitcoin you have stable coin issuers who jointly issue against this Vault so uh these issuers May set their own interest rates possibly zero if they want to incentivize or possibly not if they want to uh kind of risk control and uh B could also be borrowed elsewhere by the way so you could also pair with usdc you just don't get as many advantages and then effectively what's happening is that the these A and B are being provided in a secondary Market people can swap through this um and at the point where somebody is swapping one asset for another uh these sort of like printed stable coins get backed by the asset that's swapped into the pool but they're also backed by the collateral and uh the collateral provider is being paid for this which is where this yield is coming from and the market making losses basically are being insured by the collateral uh and so not taken on by the protocol and then there's this is enforced by liquidations and then there are various properties you can get um so uh one nice thing is that you get all of this uh efficient liquidity but you don't have to trust kind of each other's assets as like stable coin issuers and you also don't have to trust market makers uh and you can also get conceivably very high Capital efficiency um so up to $20 of liquidity for $1 of tvl basically if you want to take on that risk and if you don't a protocol can set it up differently so that brings us to kind of the end of the talk um there's kind of a few main takeaways here um but I'll just leave this up and uh we can open it up for questions okay how Okay so for our favorite and the most important part Q&A please use the QR code and you can also upvote the questions okay the first one will be will fully algorithmic stable coins possible in the future so I think here it depends what you mean by algorithmic a lot of people mean algorithmic to be unbacked uh which is not the kind of stance we want to take um we think a stable should be backed and and fully backed uh but if you mean algorithmic in the sense of being automated then we think yes and that's how we designed gyroscope to be basically as automated as possible but yeah I wouldn't be very bullish about uh algorithmic in the sense of uh being un asset-backed I mean I think especially after Tera that's just obviously a bad idea okay next question why so much hyper around Athena could you explain what Athena is yeah so Athena is basically a tokenized what's called like a carry trade so you're kind of taking a hedged position um and it it it takes kind of two sides to to to a leverage market and it's the leverage Market on centralized exchanges which get like a lot of demand and you're basically earning a yield through Athena when there's excess demand for leverage and you're kind of helping to facilitate that leverage uh and so Athena yields can be very good when there's really high demand for leverage but they can also be low or possibly negative when the the demand for leverage goes down a lot okay and then when issuing a printed way it's like latent liquidity meaning you stock up the place smart contract with the tokens that once bought by a user become real so the preting idea is uh basically an extension of how uh how Amo work today I suppose you can fill a pool um with basically UNC circulating stable coins and the idea is that they would enter circulation when somebody actually buys them from the pool which is why it's kind of parallel to like a a minting mechanism and can be thought of as a minting mechanism um but uh you get kind of the efficiencies of of being on a secondary Market um I I hope that answered the the question okay and who are the issuers what capital they need so the target issuers for the mechanism we're building for gyro V2 is decentralized stable coin protocols so the issuer is really like the the smart contracts or people like are free to almost issue themselves by using the smart contracts um it's possible that it could also be used by like some other issuers it's just that like centralized uh stable coins in particular usually don't have this flexibility and so they may not even be able to entertain it um but then the stablecoin the decentralized stable coin issuers what they really want is to like get very good liquidity without taking on a bunch of risk for for their protocols uh which is why we kind of made this like duplex collateralization mechanism where the aim is to get kind of like the very good properties you might be able to get from something like OS and protocol own liquidity even if it's on much more risky pairs but where the protocol and the the normal users of stable coins get an extra layer of protection um and if something happens uh it's uh insurance that kind of like covers the costs okay so our last question will spin your government governance token ear part of fees that's the killer question I guess isn't it um and there are routes I suppose to doing this uh but it depends on kind of governance selecting the right route okay we have 10 more seconds do you think CX integration in cctp like mechanisms will help adap and stability of decentralize bit coins yes uh so sex Integrations I think will be very good in terms of sourcing new users because people come through Sexes and CCT across chain uh easily because it's a kind of an issue if you have to wait a month month or sorry a week all the time to kind of like Bridge okay thank you very much Ari