An Overview on Algorithmic Stablecoins

Transcript

um hey everyone um it's really exciting to be here i'm joey founder of faye protocol and uh i'm gonna be giving you guys what was that oh yeah so i'm gonna be giving you guys a an overview of algorithmic stable coins uh which have been like in the news a lot i guess recently um so the first question that you want to ask is why are we building more stable coins this is the the question that dimitri from 1kx uh is asking okay guys stop making new stable coins it feels like there's another stable coin coming out every

week and um the question is why and so i think there's a couple reasons for that the first one is that stable coins have diverse value propositions which is kind of a weird thing to say like what do you mean they're all just supposed to be stable and i think that what's important is like what happens how the stable coin mechanism is used for example a stable coin like maker dow is optimizing for the ability to get leverage against your collateral and a stable coin like fey protocol is optimizing for liquidity so when when people buy fey those assets are supplied to secondary markets to provide robust liquidity for exchanging the stablecoin and that's why

exploring new designs is important because it's not just about the stable asset it's about what else you can do with the remainder of the of the mechanism the second thing is this idea called the scalability trilemma there's three things that stablecoins might want to optimize for one is collateralization like maker dao like liquidy another would be capital efficiency like usdc where you can always buy and sell for a dollar or even some crazy uncollateralized alternatives that haven't really been successful to date like empty set dollar basis and um the third is decentralization so on the centralized side you have custodial

stable coins like usdc and tether and on the decentralized side you have stable coins like dye maker dow and liquidy and fey and frax so again it's collateralization capital efficiency which are different because it takes um you know 1.50 worth of collateral to make one die whereas with usdc it only takes a dollar and then you have decentralization and it's been said that you can't do all three and i think there have been a lot of attempts to try and actually achieve this and that's why people are continuing to design more in the algorithmic space is because that's how you're going to get all three and that's what faye protocol is trying

to do that's what um you know some other like stable coins in the space i think we're converging on more novel use cases and it's all centered around this scalability trilemma and one of the three that i mentioned was decentralization and i have it as a third bullet because it's so important um decentralization is something that it like doesn't matter until it matters and you know it's really easy to use usdc and tether and say oh they work but the regulatory landscape is changing constantly and the whims of of fate or the people with power might go against you and so decentralization it's why we're all here um and so building stable coins that can achieve

this while also scaling to meet the massive demand that d5 can capture super important so that's why we're continuing to build new stable coins the size algorithmic stable coins are small less than five percent of the stable coin markets the three largest currently are uh tara usd faye and frax and i think this number is going to keep going up um it'll be it like the flipping that i care about is when algorithmic stable coins pass centralized stable coins um but i'll also obviously i want east to pass bitcoin so uh this is this is not really like a definition of an algorithmic stable coin

but these are some common characteristics that they have it's really hard to really pin down exactly what an algorithmic stable point is because you can do so many different things but um these are the the properties that i would argue that they have so first is the mechanism should be fully unchained there should be no custodians except for oracles which are really hard to do on chain because even if you're using some kind of on-chain mechanism to figure out what your asset is pegged to if that asset mostly lives off chain you you need an oracle that's off chain to make that happen now an exception you might consider like reflexes rye um and they're not pegged to a fiat currency so they don't need

an off-chain oracle but if you are pegged to an off-chain asset you need an off-chain oracle other than that the whole mechanism should be on chain and then you have an algorithmic peg mechanism that helps keep the thing stable there's a ton of ways to do this and we're going to talk about a couple of them the last thing that is important is that there's either no collateral which i think is personally really risky and unnecessary um or in in like the the stable coins that are working today all have some amount of collateral and it could be under collateralized or it could be i mean again we'll talk about some more alternatives and notice i highlight without liquidations this is an important asterisk because if you have

liquidations or an over collateralized model that's something like liquidy or dye where you have a vault you put your collateral in the vault and you get stable coins out i don't consider those algorithmic stable coins some people do which is why i'm not saying this is a universal definition but um this is what how how i conceptualize algorithmic stable coins and so then the question becomes how do you manage the risk of this algorithmic stable coin because if you're not maintaining this over collateralization you need to have a way to transfer the risk of the volatility of your collateral or the risk of the fluctuating demand for your stablecoin

the markets move in waves and there's trend trading and it's volatile so you need to have the natural order is volatility your assets are going to be volatile this is how it works and so you're fighting against nature to make an algorithmic stable coin and i see three primary mechanisms to absorb this risk or transfer it to a third party one is to have a volatile supply one is to have a secondary volatility token and the third is to use leveraged traders or insurance to hedge the volatility of your token and we're gonna talk about each of those three options

the first way to absorb the risk of your algorithmic stablecoin is to have a volatile supply and this is uh maybe the most controversial of the three options because it hasn't really been done like super successfully to date um the first would be rebasing where you literally just changed the balances of everyone who has the token some notable examples of this are yam and ample fourth and i would argue that ample force is doing pretty well it's a it's hard to integrate because everyone's balances are changing but they are really committed to their vision they're actually pegged to the 2019 us dollar which is really cool

because they're protected against inflation and they're they have a proposal out to list on ave and so in general having a rebasing stablecoin is really hard because like i said everyone's balances are changing but if you're borrowing that token and you immediately sell it for some other asset you don't really care what's happening to your balances because you don't need it so having it listed on the lending market is um a great idea and pretty cool to to see how that works in practice um coupon based stable coins like esd and basis cash these essentially say okay when the demand is high we're just going to mint

new stable coins and we're going to give them to this equity share holder and that's how the supply is volatile we're printing more stable coins and when the demand is low we're going to sell future growth at a discount and these have not worked in fact all the on-chain options of this are trading around like five or six cents right now um so not very stable and that's because it's really hard when the demand falls to get people to like take on like every additional person who says oh i'm gonna buy future growth eventually no one else is going to buy it's like if there's a queue of like 100 million people who are in line to get another dollar

the next guy is not going to come by it so um they work until they don't and i actually don't think they've ever worked very well and i was pretty involved in the empty set dollar community and i kind of saw the writing on the wall and that's when i started thinking about other ways to build an algorithm like stablecoin and fey originally used direct incentives as a way to have collateral but also incentivize the peg and for similar reasons this was also challenging because for traders specifically the price was changing as they would buy and sell and it makes it hard to know what the true price of the stable coin is because there's this invisible spread in the market

so if you go to sell and the the spot price is 98 cents but there's a four percent penalty then the real price is 94 cents and that makes it hard to make oracles and hard to integrate for similar reasons to rebasing stable coins and so faye protocol actually does not use direct incentives anymore and i'll talk about what we do use in a second so the the second option is to have a second a secondary token that absorbs the volatility of your stablecoin and um so this this this this idea is basically that the

party that's taking on the risk is like the governance token or some volatility absorbing token that when the demand for the stable coin goes up you can purchase one of these tokens for one dollar and that means the demand for the governance token is going up so you're sort of profiting off of the um demand for growth in the stablecoin and then that governance token would be burned from supply so if you think about it i want to make another and we'll use tara for an example because terra is a pure it has no additional collateral in the

system they have their stable coin ust they also have i think a korean one stable coin and then they have their governance token luna and i so the ticker is luna and the name is tara or vice yeah the ticker's luna and the name is tara so i don't actually know which to call it but the idea is that if you want to buy one ust you purchase one luna off of an exchange or on chain and then you burn it and that means if the market cap stays constant the price goes up a tiny bit for everybody so they're profiting when the system um increases in demand and then to redeem your stable coin you sell one of your tara for a luna and that luna or you you're one of your

ust for a luna and that luna is minted so it's created and given given out so this is similar to i don't know if any of you i'm sure some of you are familiar with like if you have a company you have a certain amount of shares and when you want to do a fundraising event you would create new shares and everyone who has existing shares gets deluded there's a very similar concept and this is great because like it's kind of like printing money you're like hey we have all these new shares and now we're going to sell them the problem is that if you do that forever people are going to be like i don't want to hold those shares anymore they don't have value anymore and this is the risk of a death spiral

this happened to titan as an iron stable coin and titan was the um was the governance token and they had so many stable coins in circulation and they all went to redeem all at once and they minted so many titan that the price of titan went to zero this is a real risk and it's something that needs to be controlled for and that is controlled for based on how you manage the collateral and if you have additional collateral which is really important because when the no additional collateral is the token in your system it's very self-referencing um this is not this is not a knock on tara i think that dokwan

and that team does a phenomenal job of managing this risk and i think ust is one of the most interesting stable coins out there right now um but this is sort of an existential risk to the model and so i think models more like fey and frax are going to be more prevalent where under normal circumstances you're paying either one dollar worth of external collateral or or maybe partial collateral like with fracs um and the difference here is that you're holding exogenous collateral like ether or some other stable coins in your reserves that you're using to do the exchange and then whatever the difference is that's the part that's minted or burned like in the case of tara luna

but it's mostly backed by additional collateral in the case of fracs they hold usdc in their reserves about 85 right now and the remaining 15 percent is covered by this senior edge model where you're minting and burning the governance token with fey we actually target a hundred percent collateralization and then if the collateral appreciates you're over collateralized right now phase over collateralized because of the eth run-up that happened in may um and then on the way back down a lot of the tokens were sold so we were left over collateralized and then if fay becomes under collateralized we could do the minting the governance token and selling it and so in the case of terra the governance token is profiting off of demand for the

stablecoin which is sort of self-referential in the case and similarly in fracs in the case of fey the governance token is profiting off of the collateral itself so it's more like um when east does well tribe does well instead of when demand for fey goes up tribe as well um that's sort of the idea behind the secondary volatility token this is a pretty confusing one so if you have any questions after definitely let me know um and the third which i think is the most interesting or they're all interesting but this is like where like sort of it's sort of a trend like an evolution of thought with these algorithmic stable coins and this idea that um yeah some people want stability but

other people want volatility they want the leverage and i think that that's what's making crypto so volatile it's very fueled by leverage as an industry so this idea that risk is actually a product and so is stability so maybe you can make a market for these two things and that's what perpetual futures are it's literally a market for oh i want volatility or someone else wants to hedge and then you have the longs and the shorts that are kind of matched off and if one is in more demand than the other you get a funding rate that's paid from the side that's in more demand to the side that's in less demand um so the this idea of the the perfect coin is a stable coin where it's

collateralized by ether that's then short ether against some oracle price so when ether goes up your collateral goes up but your notional value goes down due to the hedge so you're perfectly settled at one dollar and the reason why this doesn't work all the time and why there aren't like a bunch of stable coins like this is that demand for the leverage is fluctuating so there might be people more people who want stability or more people who want leverage and some of the guys from angle finance are in the room and they're they're doing a very interesting model that is uses perpetual futures at the core and has some other stuff to help meter out this this demand problem

with the the leverage um and i think we're going to see more stable coin mechanisms that are involved with perpetual futures we're thinking about some similar stuff a protocol and it's similar to collateralized stable coins like die and liquidy where people want that leverage and they're doing it using an over collateralized position here you're doing it with the perpetual futures that's sort of the idea here so where are we going what like are we going to keep making stable coins um i think that we're going to see more and more ways to transfer this risk of the algorithmic stable coin to third parties so better algorithms um like there's innovations happening all the time

um number two is uh improved stablecoin liquidity so stablecoin optimized exchanges like curve balancer v2 uniswap v3 um saddle finance these are all places where you can get very low slippage trades on stablecoins and i think we're going to see more liquidity for stable coins more stablecoin optimized amms and these are going to help a lot to bring more stable coins to market and the last is better user experience for stable coin swaps i'm imagining a world where you go to your favorite d5 protocol or even just some like web 3 website that uses d5 in the back end

and you bring your favorite stable coin and you take advantage of the improved liquidity to seamlessly convert from the stable coin you like to hold into the stable coin that the application you're interfacing with requires and you don't see it happen you're just you put in your fee you get out it's which it converts to lusd in the back end to get into whatever platform you're using and then when you withdraw it converts back into the stable coin that you want there are some new innovations and ethereum that allow this and i think we're going to see products like this as the liquidity improves as the demand for different long-tail stable coins improves

and the last thing is i think there's going to keep being more and more stable coins i think there's going to be stable coins that are specific to different products that different ecosystems prefer certain stable coins to other ones and we're going to see a continued trend of more innovation more stable coins and because of all the liquidity and because of the ux it's going to be a good thing it's not going to affect the market and you're going to have robust sort of liquidity and applications so that we can really try to tackle that scalability trilemma we mentioned in the beginning and make sure we're decentralized and yeah i i'm excited and i'm really happy to be a part of it

and uh so i like i said i'm joey from faye protocol you can follow us on twitter faye protocol my um twitter handle is joey underscore underscore santoro s-a-n-t-o-r-o and um yeah i'll be here answering questions thanks again for listening to this overview of algorithmic stablecoins questions what do you think is the in the long

term future of stable coins when fiat currencies no longer exist and there's no u.s dollar to peg against what is the most likely what does a stable coin look like in that time totally uh this is an awesome question it's a deep philosophical question i embedded in the question is there's no more usd i think there's always going to be a usd maybe maybe not always but i think there's going to be usd for a long time and maybe it won't be as attractive of a place to store value but it'll always be there because you need to pay your taxes and stuff but um what does a stable coin look like well you can peg a stable coin to any oracle feed or any oracle price and that oracle price could change it could be based on some kind of bonding curve or some kind

of pid controller like what ry uses and i i think rye is awesome it's a really cool project um it's kind of hard to explain like wait it's not pegged to anything the peg changes but there's going to be these interesting algorithms and i think there will be a power law distribution with the non-pegged stable coins like there's going to be some big winners in those markets because it's really hard to tell everyone hey you're going to use this like random unit of exchange that changes sometimes it's hard to coordinate around that so i think the narrative is probably going to latch on to one or two of these and then those are going to be very popular i think fiat peg stable coins are always going to be popular

because most people live in countries and are down for that but i think that like the the growing demand for like non-pegged stable assets is gonna um really pick up and i guess the third option there is peg to commodities or pegged to commodities um i think sam sam kazamian from frax finance is working on something called the frax price index which is basically going to be an idea of true stability like the consumer price index and um they're going to try to peg a token to that like fey is going to be involved reflex is going to be involved maybe olympus dow and yeah there's going to be these new these new indices that we're pegging to like a consumer price index

that represent like inflation protected cost of living sort of stability um that's a really big idea and i think it's going to be really important thank you that's a great question hi uh so you mentioned that you don't you that fey is not using direct incentives anymore yes uh is it like fundamentally float design or you will come back to that like with better mathematics behind it in the future totally uh i don't think it's fundamentally flawed design although um some of the some of the curves that we use were were pretty uh like i guess strong in the beginning so it led to and also like with the launch there was kind of a very high supply but very low

demand so it created kind of this weird perverse incentive but um yeah i think direct incentives are interesting i'm not sure if we're going to roll them back out again but i don't think that they're like off the table i think putting them on top of uniswap was probably not the right idea because like i said it makes what you see on uniswap not the truth and i think that's really like i didn't realize how how confusing that was going to be because like the more people that use your stablecoin the less they're going to know like about direct incentives or anything like that so you want to make it as simple as possible for your users so maybe if there are direct incentives it's going to be more like a dynamic minting and burning fee which other stable coins do like

liquidity has this will probably have this um so it's kind of where you put them to answer your question yeah thank you uh i was totally what your thought was sorry yeah i was wondering what your thought was on makers like protocols uh but collateralized by governance tokens like synthetics if you have any any thought on that okay is there an example of this that you could point to synthetics like synthetics it's collateralized by snx the model is like maker they have votes right yeah i'm familiar with synthetics it's uh it's interesting i think um they're 600 collateralized for a reason it's it's very risky to again use your governance token as collateral and so like especially

when you have so many price feeds like synthetics has so it's very capital inefficient um and i think that's why we see synthetics usd has not really has not really even come close to competing with dye even though they've been around for about the same time um those protocols are really cool and i think they're much more important for like exchanging between price feeds i don't i don't know that they're gonna make the best stable coins you you mentioned the flipping as is as in is flipping btc and becoming number one then in your opinion what keeps btc at number two does it even stay in the top ten oh yes so uh so i guess when i when i mentioned that flipping it was more saying that like

to to me as a builder in the algorithmic stable coin space i care a lot more about how the algorithmic stable coins becoming more used than centralized stable coins um but i think because of decentralized finance because of the ability to make stable coins and put them in your applications ethereum's going to flip bitcoin and i think part of that is because of stable coins um what's going to keep it there i'm not i'm not totally sure um and i think bitcoin's awesome for what it's worth i'm i'm pretty much an ethereum maxie but i also like bitcoin a lot so i hope that like i said there's going to be a ton of stable coins there's going to be a ton of blockchains there's going to be a ton of cryptos and it's going to be a power law distribution i think ethan bitcoin are

clearly the winners for the medium term and hopefully for the long term but you never know the industry is crazy so i'm not sure i could really tell you what's going i definitely can't predict the future sure yeah i'll be here yeah thanks everyone