CoinGecko Podcast - Bitcoin & Cryptocurrency Insights

Building The First Fractional Algorithmic Stablecoin with Sam Kazemian, Founder at FRAX Finance – Ep. 24

January 26, 2021 Bobby Ong Season 1 Episode 24
CoinGecko Podcast - Bitcoin & Cryptocurrency Insights
Building The First Fractional Algorithmic Stablecoin with Sam Kazemian, Founder at FRAX Finance – Ep. 24
Show Notes Transcript

In this episode, Bobby Ong, co-founder of CoinGecko is joined by Sam Kazemian, Founder at FRAX Finance. Bobby interviewed Sam on the background of FRAX, its fractional algorithmic stablecoin, problems with stablecoins, plans to increase FRAX adoption as well as proposed FRAX bond.

[00:00:02] Intro
[00:01:23] Introduction of FRAX
[00:04:32] Use of FRAX Share token (FXS)
[00:06:30] Idea of fractional algorithmic stablecoin
[00:12:22] Thoughts on problems of stablecoins
[00:18:19] Thoughts on market absorption for FRAX
[00:23:29] Demand for FRAX
[00:38:35] FRAX bond proposal

Quotes from the episode:

“Right now we only use USDC. The protocol starts at 100% collateralized. As it expands, it'll slowly test lower and lower collateral ratios. And so what that means is the FRAX stablecoin is only backed by USDC at the collateral ratio.” [00:02:32]

“One of the main stories that I like to tell about how FRAX came to be after the whole Tether saga.” [00:07:24]

“One of the things that's interesting that we're exploring, not finalized, we don't really know if this is what we'll release or anything, but we were thinking of having FRAX be backed by $1 of LP token value.” [00:30:21]


FRAX Finance -
CoinGecko -
Frax (FRAX) on CoinGecko -
Frax Share (FXS) on CoinGecko -

Social Media

FRAX Finance:


Bobby Ong (00:02):
Welcome to the CoinGecko podcast. I'm your host Bobby Ong. Each week we will be interviewing someone from the blockchain industry to learn more about this fast moving crypto currency economy. If this is your first time listening, then thanks for coming. The CoinGecko podcast is produced each week to help you stay ahead of the curve. Show notes can be found at I highly encourage you to join our newsletter where we send out top news in the crypto industry every Monday to Friday. Come back often and feel free to add the podcast to your favorite RSS feed or iTunes. You can also follow us on Twitter and Telegram at @CoinGecko.

Bobby Ong (00:41):
Welcome to the CoinGecko Podcast. For today’s episode, we have the honour of welcoming Sam Kazemian, founder at FRAX Finance. Before starting FRAX, Sam was the Co-Founder and President of Everipedia, the first decentralized online encyclopedia on the blockchain. Sam graduated from UCLA in 2015, double-majoring in Neuroscience and Philosophy. I'm a big fan of FRAX. I have some position in FRAX, for disclosure. Wrote a tweet on FRAX and super happy to have Sam join us on the podcast today to talk and learn more about FRAX. So welcome to the CoinGecko Podcast, Sam.

Sam Kazemian (01:11):
Thanks Bobby. It's great to be here. Jason, head of engineering, we're just here cranking away nonstop. Uh, you know, we've been building since launching three weeks ago and things are good. So really appreciate being here.

Bobby Ong (01:23):
Yeah, I guess for the first question right Sam, maybe let's start with a basic introduction of FRAX. So tell us the listeners who are not familiar with FRAX, what is FRAX and how is it special?

Sam Kazemian (01:34):
Yeah. So we like to say that we get our name FRAX from the stability mechanism that we have invented. So "Fractional Algorithmic Currency" in terms of a stable coin. So what that actually means is, um, FRAX is the first time where we've introduced the idea of a market sets collateral ratio. So a stable coin that is only partially backed, but how much it's partially backed is set by the market at whatever rate that the stable coin is going to keep its target price. So for FRAX is $1, right? And, um, the idea with this fractional algorithmic mechanism is that it starts off 100% collateralized. And so we started the protocol with just USDC as collateral, super simple; USDC is a dollar. It is as you know, centralized stablecoin but you can take this idea implement multiple collateral later. In fact, we might, but right now we only use USDC. The protocol starts at 100% collateralized. As it expands, it'll slowly test lower and lower collateral ratios.

Sam Kazemian (02:43):
And so what that means is the FRAX stable coin is only backed by USDC at the collateral ratio. So for example, you know, 95% that means that it's only backed by 0.95 USDC and 5 cents of value is stabilized algorithmically, right, with no backing. And it'll change the collateral ratio, and if the price drops below a dollar, it'll increase the collateral ratio. And so that's kind of the unique coherent idea that we not only named FRAX after Fractional Algorithmic Currency but we kind of introduced this novel stability mechanism for the first time that I think a lot of people in DeFi are talking about.

Bobby Ong (03:25):
Cool man. I mean, FRAX is the first fractional algorithmic stable coin being created. This must be pretty awesome, right? I mean, inventing something that on mainnet, live in production that nobody has done this before, right?

Sam Kazemian (03:35):
Yeah. That's what's really exciting. It's also a little bit nerve wracking, right? Cause, cause we don't know how it'll work so far. It's worked great. You know, the entire protocol is around 84% collateralized right now and there's over a hundred million FRAX stable coins in circulation. That means there's about 16 million FRAX that are algorithmically stabilized. And it's been keeping the peg quite tight, you know, give or take one cent and you know, you can see the graph on CoinGecko. We actually use your guys' great and really reliable prices on our front end as well. Appreciate that. And um, it's been doing fine, but the thing about it is we don't know. Right? And so that's what makes it really experimental really interesting. And we'll see, but I think that it's going to turn out to be one of the better ways of stabilizing a currency or these algorithmic stable coins.

Bobby Ong (04:32):
Yeah. I mean, if we look at the price of FRAX, so FRAX, for those of you who are not already aware of it, so FRAX, there's two token in the FRAX ecosystem. The first one is FRAX, which is a stable coin peg with the US dollar. So if you look at historical chart, like FRAX has pretty much maintain it's peg around between $0.90 to $1.02 or so, during times of extreme volatility. Although it's not exactly at $1, it's been pretty good compared to all the other algorithmic stable coins which hover anywhere between $0.30 to $2.50, which is quite crazy considering that it aims to be a stable coin. The other token in FRAX is really the share, FRAX Share token, FXS. So maybe let's go in a little bit deeper into this FRAX Share token. And then what's the use of FXS and how does it come into creation and goes out to circulation? Maybe can explain a little bit.

Sam Kazemian (05:21):
Yeah. So basically in FRAX there's obviously a portion of the FRAX stable coin supply that's stabilized algorithmically, right? And so what algorithmic stable coins generally have in common is, there's either a bond or share, or sometimes both token mechanism that basically expands or attracts. You know, when it's expanding, seigniorage value is given to the shares and then whenever it attracts um, basically it's, uh, the stable coins have bought back to which back the supply. The FRAX share token is what's used to enact that monetary policy for the algorithmic portion of the FRAX supply. And the FRAX Share token gets all seigniorage value, fees from the protocol in terms of revenue of like minting, redeeming fees and things like that. And then all excess collateral value, which is collateral in the system, that was originally put in at higher collateral ratios and that now that FRAX is more algorithmic is no longer needed in the system. So FRAX Share holders actually get that value back in to the FRAX Share distribution.

Bobby Ong (06:30):
Very interesting use case for FRAX Share. How do you guys come out with this fraction algorithmic? Was it you guys or someone else wrote about it and then you guys are the first team to actually implement, execute this idea?

Sam Kazemian (06:41):
I mean, well, first of all, I always like to give credit to people who incremented the field, right? So obviously FRAX, the mechanism is unique, right. The thing that we've come up with it's, I'm proud to say that all of the co-founders and staff had a huge hand in, is this novel market set collateral ratio that adjusts based on the price of the stable coin. Obviously things that, you know, have inspired me, are the original Robert Sam's Seignorage Shares white paper, a lot of the Basis guys and the original Basis white paper, you know, Nader and those guys. And yeah, it's like, it's an amalgamation of a lot of stuff. And in crypto, everything is iterative. Right? And one of the main stories that I like to tell about how FRAX came to be as the whole Tether saga, when they were in a legal dispute with the New York attorney general, it was like about two years ago. And there was a public claim by the New York attorney general that Tether is only $0.74 backed by, you know, actual deposits. Right. And then the rest of it is like not dollars, whatever it is, they stabilize it in whatever way and stuff like that. What was really interesting about that is nothing happened to the Tether price where, you know, when, when that was actually announced publicly, right, cause it was a public lawsuit and stuff. I was looking at it, I was like, "Holy crap, like, Tether is going to go to $0.90 or $0.80 or whatever but it didn't budge, right? And so that was the first thing that I thought of. I was like, "Well, what does that mean?" That basically means that because Tether is so tightly integrated into the crypto community, all of these trading pairs, everyone relies on it, it has extremely high liquidity as money as cash.

Sam Kazemian (08:23):
And you kind of will it into existence. Then, everyone believes this is a dollar, right. And it kind of has this force of monetary premium. Like people actually create value in the sense that this is used as cash. What that meant about Tether, there was at least $0.26, give or take, of monetary premium of this value in there that obviously was not needed to come from the collateral, the dollar deposits, right? And so in that situation, I was thinking like, "How do you create a decentralized transparent and trustless protocol that actually measures monetary premium? Like how would you create something that starts off fully backed?" You know, like people said about Tether and then you can actually test the market's belief. I do think Tether is now fully backed. I'm actually not one of the "Tether truthers" and stuff. I'm pretty sure that if they didn't take care of all of that stuff, they would not be operating.

Sam Kazemian (09:18):
So I do think now they're fully compliant but the idea behind this is that how can we create a completely transparent protocol that, that actually measures the amount of monetary premium of a stable coin? And so that's what led us to trying to just redefine the categorization of stable coins. So as you probably know, before we released FRAX, those stable coin triangles and types where, you know, there's like Fiat backed up here and then there's like over collateralized crypto here and then there was fully algorithmic, right? Like Seigniorage Shares, Basis and stuff like that, right. And everyone would always talk about stuff like that. And so we actually just wanted to redefine the whole paradigm, right. We actually wanted to say "why would you go from, you know, a hundred percent or over to zero, right?" Like why, why is that the dichotomy, right?

Sam Kazemian (10:10):
Why is it hundred, zero, right? Why isn't there something in the middle? And then once we started asking that question, we started saying, "Okay, what is the middle? Is it 50? Is it 60? Is it 70? What's the thing?" And we came to the idea of why don't you let the market decide instead of like trying to have to convince people, "Oh, well, we're going to vote on governance and see if it like de-pegs or re-pegs or whatever. Why don't we just create a really good mechanism, right? We create good game theory that actually just tests the market's belief of what that collateral ratio should be.

Bobby Ong (10:46):
I think it's a very interesting story on how you sort of Tether, thought Tether wasn't fully backed to a dollar. I thought it was a big issue in the crypto as well. I've been telling a lot of people myself, Tether's kind of the biggest risk crypto, we don't know if it's fully backed but if it falls for whatever reason. I hope it doesn't fall by if it does fall, then it's gonna be a bad day for crypto. Thankfully I think, I mean, Paolo from Bitfinex said that they all fully comply and registered in US these days. And I think they have like [inaudible] these days but there's still this aura, there's this halo effect where this is so much negative news from Tether but I mean, we don't know. I hope, I hope, I hope it stays, but I was very worried when it started to fall in time, but now I can see, everybody seems to agree to that Tether is $1. Sort of like kind of Fiat money kind of system, like if it really goes down at $0.75, people still believe it's worth a dollar. And it's interesting how you kind of think that, I mean, the market seems to have like extremes, right.

Bobby Ong (11:36):
So, I mean, I was doing a slide for Coin Gecko annual report two days ago, where we categorize the various stable coins between fully Fiat collateral like USDC, USDT to like, overcollateralized stable coins, like DAI, XUSB and then to the extreme algorithmic stablecoins like Ample and Basis. So why isn't there something in the middle and FRAX falls right in the middle of that. Yeah, so I thought that was very interesting. I was personally very fascinated with this innovation going around and thought that the way that it's maintained is in fact was quite an elegant solution for this collateral ratio that is determined by the market. Let's talk a little bit more algorithmic stablecoins, right? So I'm sure when you work on FRAX, they are competitors out there. I guess the earliest one being Basis, the original project did not launch, but two years ago, some anonymous guys launched Basis Cash.

Bobby Ong (12:22):
And, uh, sometime late last year, September, if I remember correctly Empty Set Dollar launched. And then before that, I think earlier 2019, there was Ampleforth launching, early 2020, I think so, Ampleforth. So where the supply increases and decreases. Single token Ampleforth. So which seemed that these algorithmic stable coins, like it aims to achieve a $1 stable coin pack. But like in reality, it becomes sort of a money game played by whales, where the price fluctuate wildly between anywhere between $0.25 to $2, $3 or so on. I'm just curious to hear from your point of view, right, what do you see a problem with this sort of stable coins, the Ampleforth sort of stable coins and the algorithmic kind, like Basis, ESD?

Sam Kazemian (13:06):
Yeah. So first of all, I'm a pretty big fan of algorithmic stable coins. Just the general concept. I like to say the important thing about them is they're trustless mediums of exchange. So actually one thing I'd really like, uh, the DeFi space to maybe stop calling them necessarily stablecoins because Tether and USDC and the Fiat coins have kind of co-opted that name, right. And there's going to be a lot of regulation around stablecoins, some good, some, you know, difficult to adhere to. But I like the fact that the real value proposition of algorithmic stablecoins, right, are that they're trustless mediums of exchange. And now there's a bunch of types. I'm, I'm pretty big fan of studying each of the different mechanisms. I think just overall, the one thing that is important to think about with FRAX compared to the other ones, is that what we've shown at least so far is that if you combine a fractional algorithmic model, the actual amount of stability in the algorithmic portion of the market cap actually becomes much more stable, which is actually one of those situations where the whole thing is more valuable than its individual pieces added together.

Sam Kazemian (14:21):
For example, right now there's a little bit over a hundred million FRAX in circulation at a 84% collateral ratio. That means that there's around 16 million FRAX, that is not backed, right? It's algorithmically stabilized. FRAX is almost exactly a dollar right now, as we're talking. If you ran this idea in your head of what, if there was a Basis structure or like an Ampleforth structure fork or something. And that had a $16 million market cap, which is quite small, right? But it was 16 million in market cap. Do you think it would be this stable? Do you think that would be as stable around a dollar as FRAX is now we have data and the real market is no, absolutely not, right. Not even close, right. And so the fact that now there is actually like a algorithmic supply of a stable coin, 16 million, which is small, but it's sizable, now.

Sam Kazemian (15:19):
You can't just say, "Oh, this is just, you know, wrapped USDC." Cause it's not, there's 16 million of unbacked FRAX and that's as stable as the collateralized portion. I think that's really important to talk about. And that's something that we've discovered and, you know, we've so far we've shown is quite interesting and promising. So the other designs kind of lack that, right? And so one of the first things we're starting to see for example is ESD V2 has this partial, I think they call it partial reserve model and it takes a lot of, you know, inspiration and interesting ideas and stuff from FRAX and other colateralized coins DOT, Maker and stuff like that, right? And it tries to test, you know, different ratios. I think it's, as far as I understand, it's set by their governance, which isn't very reflexive in my opinion, but they'll probably make it more reflective as they continue to do that.

Sam Kazemian (16:15):
Basis, I'm actually a big fan. I actually farmed it. And in early days I was pretty active in the early community. I still check in, I like actually the Basis Cash design. One of the things about it is that I think that just with no collateral at all, it's difficult to regain the peg when it's at a certain price, like it's like at $0.80, right. And then the discount on the bonds, which how it's calculated is the price of the stable coin squared, right? If in that situation, it doesn't make sense to buy the bonds. You get this situation where the price gets stuck at under a dollar, right? Cause like, let's say you have the price at 80, right. And the bond discount is like, you know, it's like $0.60 or whatever, if no one buys, if at this time bonds at like $ 0.64 is not a good deal, no one will buy it.

Sam Kazemian (17:09):
Right? And then like the price will just, it'll just be flat at like 80 or under the peg. And I think that that's one of the main issues is that's just happening, right. That these coins are getting like stuck at 80 or is it getting stuck at like 77 or they're getting stuck at 92, right. And there's no reflexive components. What's interesting is I think they're also, I follow a lot of these stablecoins and I think they're also having governance proposals and stuff for having like part of their expansion go to like a treasury or something like that, or some kind of collateral, which I like to think we inspired that or something like that because we showed that it's kind of, you know, it actually creates more value than the actual collateral value, right. There's like more stable value in FRAX right now than the actual collateral value, right. There's only 84 million USDC but there's like a hundred million FRAX and they're all at a dollar. So I think that's the main thing is that some kind of system where there's outside exogenous value that actually is able to be part of the monetary policy of these protocols.

Bobby Ong (18:17)
Interesting how you brought these up. So I've been monitoring during this collateral ratio, right. For the past few weeks, since I started learning about FRAX. And it started out, I think when I first started noticing it was probably 94 or maybe around 95% or so collateral ratio. And over the past couple of weeks, it has gone down to around 84% as you mentioned. I don't have a mental model of what would be the stable state collateral ratio. I keep asking myself every day, what would be like equilibrium collateral ratio and my gut just feels that I think the market has bought around 40 or 50% collateral ratio. But I don't know and would like to hear it from you, like, do you have any idea on, what do you think is a stable state collateral ratio? What do you think the market can absorb I mean, we saw Tether, you mentioned that it can easily absorb 25%, right? So what do you think? Can the market absorb 20% collateral ratio for FRAX, for example? What do you think?

Sam Kazemian (19:08):
Yeah, I mean, I get asked that a lot and the other people in the community always talk about the collateral ratio, but the reason we designed it exactly like this is when we're originally thinking about it, we're like, "Oh geez. Like, we don't want to like, have any kind of real say, we want the market to decide." What's funny is then we came up with this interesting system, but then people will still ask us, right. They're like, "Hey, when do you think?" But the whole point is I honestly have no idea, right. And then I try to actually temper a lot of expectations because at first people told us, "Hey, this is really cool, but we're just worried the collateral would just get stuck at a hundred percent." Like it would just, it would always just get hard and perfectly hard and stuff like that.

Sam Kazemian (19:50):
And it would just get stuck like at 99, 98, a hundred or something. That's clearly not happening, right. Then people said, "Oh, well maybe it'll never go below, you know, 90 or 85 or something." Now we're under 85. I have no idea how long we'll stay at like 80 something, right. Maybe next week we'll be at, you know, 70 something, maybe it'll re-collateralize back up to close to 90. I actually don't know if that's part of, kind of the whole thing at the beginning. I was like, I have no idea how this system will react. And so one of the things that I like to make sure the community and everyone knows is like, we could be at 80 something percent for like a month. We could be at it for like two months, right. And if we're supposed to just let this deterministic system, you know, work as it's intended.

Sam Kazemian (20:38):
Um, and then in like another two weeks, we could go from 80 to like 65. The whole point is as long as the stable coin actually stays around the peg and there's slight demand increase, right, it's like a slight, a higher amount that the minting is continuing to go up and stuff. It'll continue to de-collateralize. It'll continue to go more and more algorithmic, but how quick, or if it'll, you know, if it'll find some steady state for a while and then it'll go down and find some steady state, I think that's actually how it's going to be. Cause that's how it's looking like right now. But I don't know. I have no idea. Now, one thing I will say is like, we're thinking of updating the collateral ratio PID controller design to something that's slightly more encompassing of like the realities of the system.

Sam Kazemian (21:27):
And by that, I mean, right now it just takes a look at the Oracle price. Literally just the FRAX Oracle price above a dollar, right. And it's only gone one or two cents above a dollar, as you said. The other thing we want to look at is like, what about looking at the market cap of the FXS token or the liquidity of it on chain. Because you can do that on chain entirely, you know, because you can do that entirely on chain now. And that's important, right? Because FRAX being at a $1.01 when it has 500 million market cap, which we're seeming like we're on our way to, hopefully right, versus it only having $5 million market cap is totally different, right? Those are totally different supplies that require much more robust algorithmic stabilization in the FXS liquidity and market cap, right, to stabilize. So one of the main things we're thinking of doing is updating it a little bit so it's not completely only price based because the price doesn't necessarily get the data of market cap or liquidity, right.

Sam Kazemian (22:34):
And we'll propose some obvious like, governance stuff. We have a open. People are putting proposals up there. But we'll be talking about it in the community. There's actually a lot of people were talking about it. Actually, I wrote on the whiteboard here, uh, I don't think you can really see it, but we're listing all of the different things that you could design the PID controller around. Like you could take in the FRAX market cap, the FXS market cap, we could actually take in time slash memory of how long it's above a dollar, right? Like if it's over a dollar for 10 minutes, it's different than if it's over a dollar for the whole one hour window or something, right. We're actually talking about that earlier today. So there's a lot of improvements, incremental improvements, just like any good protocol that we're working on. But overall, just past three weeks, things have looked quite good.

Bobby Ong (23:29):
Interesting ideas, processing all this new information that I'm hearing from you and kind of makes sense of that. I think earlier you mentioned about the demand for FRAX and FXS, the collateral ratio will go down incrementally as function of the FRAX demand. When there's more demand, the price will most likely stay at one dollar or $0.10 in this example, this collateral ratio. What is driving the demand for FRAX at this point in time and it's a stablecoin, right? I mean, you could own Tether, USDC, XUSD, any other sort of stable coin. Why FRAX? And I guess that's the first question. The second question is what can you do with FRAX? How do you plan to increase adoption of FRAX in the ecosystem positively?

Sam Kazemian (24:10):
Yeah. And that's like the most important thing, right? Because at the end of the day, if this is supposed to be trustless medium of exchange or money, it's gotta be accepted in places for lending, for trading, for purchases and things like that, which I think actually comes, purchases comes towards the end. But we actually have been integrated in quite a number of places just in the three weeks that we've been up. We recently passed CREAM, governance vote. We're going to be on there for lending and borrowing. Andre today, if you guys saw, released the permissionless Curve Pools, which is really cool and the inaugural pool event that he launched, where the Algo pools starting with FRAX and then the other ones are Basis and Mith cash. So that's really exciting. So now we have Curve pools for FRAX. We're getting on the Sushi Onsen menu, which is like their exclusive, you know, for Sushi rewards and an extra liquidity, you'll soon be able to add liquidity and earn FXS as well as Sushi.

Sam Kazemian (25:09):
That's pretty good. The goal is to basically, two things. One is the whole point of algorithmic stable coins over existing ones is for it to be a trustless medium of exchange with no custodial risk of, you know, the dollars or having blacklists and things like that. The other thing is you have to actually have enough stability where people actually will hold FRAX instead of Tether or USDC, right. And I think that conversation in the algo space is now getting to the point where it's like, okay, how long do those other projects and stuff take to figure out to be as really tight around the peg for people to actually think about holding them or how, or are they just games, right? Are they just stabilization games, which are fun but they're not in the same space as other stable coins. We think FRAX actually has already in the few weeks that it's been out proven that it can likely stay close enough to the peg that it's actually useful.

Sam Kazemian (26:10):
It can stand a dollar, $0.99, dollar one if it's expanding and it could actually be useful. So the goal here is to get it accepted in every place that USDC and Tether is, have it be as stable as them virtually and then remove the custodial risk and other things so that if people really highly prefer using a fully autonomous decentralized solution, they can do that. And so what that actually allows us to do is, for example, implement privacy tech and things like that. In fact, one of the things we're looking at is, for example, what if we incentivize something similar to a Tornado Cash style mixer, right? What's actually interesting is since minting and redeeming FRAX is kind of like a pool, right? You put in collateral and you get out FRAX, right? Or you put in FRAX and you get out collateral and FXS, right?

Sam Kazemian (27:06):
It actually is already kind of in the structure. We would just have to change how the transaction worked to include some ZK proofs and things like that. And you can have the advantages of fully anonymous stable asset, which would be really cool. So we are able to do that. Obviously I don't think USDC, and Tether, and stuff are going to allow that, in fact, there's probably an argument or thing to be thought about that as more stable coin regulation comes out, if USDC and Tether will be required to, for example, blacklist major privacy smart contracts, such as like Tornado Cash or the other solutions that are coming out. I don't know. But, I would say that it's possible. It's almost, I would say it's about 50-50. And all of those things create just a unique value proposition for something like FRAX, like an algorithmic stable coin, that's truly, you know, decentralized and completely permissionless.

Bobby Ong (28:08):
Let's talk about this risk of USDC or USDT blacklist certain addresses, because you said we could blacklist the Tornado Cash contract. Let's run through a scenario. So there was a hack in one of the centralized exchanges, where let's say a million USDC was stolen for example. And this million USDC is deposited into FRAX and then FRAX is minted. And then two days later USDC decides to for example, blacklist, I don't know how do you do it, but blacklist this USDC. Now FRAX will be impacted with it because it's supposedly 86% collateralized, but it could go down to immediately 50% because of this blacklisting of USDC. I don't know how it works because it's sort of held in a pool. So it becomes quite a big risk as well if USDC decides single-handedly to blacklist the entire pool address, then everything's kind of in trouble, right? So I'm sure you get a thought about this. So want to hear your thoughts. What do you guys think of this risk?

Sam Kazemian (29:00):
Yeah, so definitely. First of all, we decided to launch with a stable coin, so like USDC as the collateral to start, so people can really understand and wrap their head around the whole fractional algorithmic mechanism. In theory, you can actually build this exact same mechanism with a volatile crypto, like ETH and BTC, obviously like the collateral ratio would change as volatile as the value of the volatile crypto goes up and down. Um, and there's actual ways to make it a little bit better but we wanna think about that for like FRAX V2 but as long as V1 is still out, basically you're totally right. Right. The collateral has custodial risk while like the actual stable coin is algorithmic and permissionless. So basically we've just put it one level down, that we put the custodial risk one step removed.

Sam Kazemian (29:52):
I think obviously as FRAX gets more and more algorithmic, the risk of it actually goes down, right. And it becomes much less part of the risk of the whole system and becomes less systemic but it is currently a risk three weeks in, it's possible. I think that we really do recognize that and so for V2, people can basically be confident that there's going to be other collateral types. There's definitely going to be like, I'll actually give some stuff to think about, like one of the things that's interesting that we're exploring, not finalize, we don't really know if this is what we'll release or anything, but we were thinking of having FRAX be backed by $1 of LP token value. So that's, that's something that's kind of interesting. That's in that same thought of as Fi Protocol, if you guys have heard, they're kind of have this similar fractional stuff with a different type of mechanism.

Sam Kazemian (30:48):
So for example, if you have a FRAX-ETH LP token and we say that you can mint a $1 of FRAX, you mint one FRAX for $1 value of the FRAX-ETH LP token, that's actually interesting, right? Because it's backed by $1 of liquidity. So, and then that liquidity is backed up with ETH but then the liquidity token is not as volatile as ETH, right. It's actually softened volatility because it's against the stable coin itself, right. So that kind of stuff is very interesting. It's, I think that's where stable coin design is headed. I wouldn't be surprised if other stable coins started looking at that. In fact, it's kind of similar to stable credit, right? If you, if you think of Andre's stable credit, Andre's a brilliant guy. Probably one of, if not the biggest galaxy brain in DeFi, right. And, uh, I think he's really onto something in terms of the stable credit design.

Sam Kazemian (31:47):
Now it's not a algorithmic token, right. He's basically building this on AMM curves that are essentially, it's actually almost like a better Maker instead of it being over collateralized with, you know, pooled assets, it's over collateralized with LP tokens. So that's a big improvement. That's actually really cool. With FRAX, one of the things we're thinking of is what if it's just fractionally collateralized with LP tokens? So that's where I think a lot of the algorithmic research is headed and we want to still be at the top of that. And so that's one of our answers to how we remove the custodial risk and the centralization risk of the USDC.

Bobby Ong (32:31):
Do you see, with the next version of FRAX, do you think the volatile assets would be like Bitcoin, WBTC, ETH or would it most likely be the LP tokens coming in? [inaudible]

Sam Kazemian (32:43):
I think it'll be a really collaborative process with the community, because first of all, we don't want to put out any of these things with just ourselves, because there's a bunch of really smart people in the FRAX community that we actually really respect and they've come up with really good ideas. So one of the things we might do actually is that we might actually release different collateral pools but with very small ceilings. Right. And so basically right now with USDC, that's just one. So there can't really be a ceiling, right. But for example, we can release a Bitcoin or ETH collateral option but only a maximum of 3 million FRAX can be backed by that. Right. And so that's small enough where it can easily be algorithmically stabilized if something happens and those that collateral dumps, you know, 40% or something like that. But, uh, it's big enough where you could see the demand for it, how it actually functions in the wild. So we might actually launch a few and very small and just let it actually build up with the community and see what people think and increase the ceiling through governance and things like that. So I think, definitely expect a lot of like, good and different collateral types coming up.

Bobby Ong (34:01):
So it's sort of like in the future, in V2 it will be like FRAX, currently every FRAX is backed with USDC but it could be a future where there will be several pools in FRAX, where big portion is backed by USDC, a smaller portion be backed by BTC, ETH and some LP tokens. Would it be sort of right to say it looks something like that?

Sam Kazemian (34:20):
Yeah. And so we could see how those, and there would be small enough where it would not be a systemic risk because we don't want to actually just throw it in there and then see this whole system that's probably going to be really big by then potentially, you know, implode or lose a peg but it would be big enough for us to actually be able to see what the demand is and how they perform. So, yeah, I think it a future where there's different pool collateral other than just USDC. Again, we launched it with USDC so that it would be super simple. So people can like wrap their head around these fractional algorithmic idea that no one's built. And so people can really understand it, see if they like it to be a part of the community. It's easier to communicate with people.

Sam Kazemian (35:02):
And then, the other stuff you could do for example, is some people in the community have suggested, interest bearing collateral, right? So there's no reason why, you can't back it with $1 worth of interest bearing collateral. And as that accrues interest, so like for example, Yearn tokens, right? Like yDAI or something. And as that accrues interest, the interest goes back to FXS pool, right? And so that's actually very elegant. We don't even have to decide to, we don't even have to create our own strategies or FRAX vaults or anything. We actually just outsource that to the really smart guys at Yearn and all of them and all of them. And then, uh, we do get the interest for it, right. And then that goes back to FXs holders, which is great.

Bobby Ong (35:48):
I see a future where like, if you guys play your cards right. I mean, you guys can pretty much take over as the de-facto a decentralized stable coin. I mean, it was supposed to be played by DAI but there's some controversy, I mean it's a bit hard as a lot of DeFi project on DAI but like, I mean, there's some difficulty in growing DAI after some point. But in terms of like when you introduce BTC and if you see it kind of like pretty much resembling, maybe it will be fractional, I mean, for Maker it will be like 150% backed by, like for most stable coins but for you guys, would it be the same, like because the collateral value of the underlaying assets very volatile, right?

Sam Kazemian (36:27):
Yeah. So that's why we haven't really decided exactly how to launch those kinds of things, especially because we want to have a very coherent vision for them. So for example, one of the things we've said is if we do launch, you know, volatile, just raw volatile collateral, like ETH or WBTC or something, we would likely launch a bond token with it. So that the whole protocol becomes a three token design, debt token, so the bond, you know, the shares, which is what proves all of the excess remaining value once bonds are repaid and then, you know, stable coin. That would actually help recover any collateral value that's unexpectedly lost, right? The bond token would help cushion that. So one of the things we're thinking of is if we do launch multi-collateral, would it require to make the design work well?

Sam Kazemian (37:19):
Would it require a debt token or bond token of sorts? One of the things about that is, I don't think the current bond designs like ESD coupons and Print Basis bonds work as well as probably originally intended, as evidenced by ESD V2 is moving to a different bond design or Coupons as they call it. And then I think the Basis guys are improving their bonds. I've seen some Basis improvement proposals that have said to slightly tweak the bond designs. I don't think there's a bond design that has proven to work, which is just plainly sad, right? Like none of these things have proven to work at scale and at the, you know, useful amounts. So until we feel confident that like there's a good bond design, we won't really release FRAX bond, but we're thinking about it. So that's, that's one thing in order to launch volatile collateral, we probably need a bond, which is emitted when there's not enough collateral. Like when the collateral ratio is not being able to be held to retract the supply of FRAX in the open market so that it moves in lock step with whatever collateral issue you need. A lot of mechanism design, a lot of things to really stimulate and explore.

Bobby Ong (38:35):
Yeah. I was reading your documentation and one of the thing that are sort of "coming soon" is the FRAX bond and so I thought like, I would like to ask you more on FRAX bond. Seems like a lot of things are still being in development, so there's no finalized design yet. I mean, I'm just going to ask you, like, do you think like it will come out sometime this year? I mean mid this year or something like that. And also like, how do you see the design being different from ESD Coupon and Basis Point? Do you see this FRAX bond having an expiration date or not having an expiration date; it goes on to perpetuity. That's the main difference between ESD Coupon and Basis bonds, right?

Sam Kazemian (39:11):
Yeah. And so, first of all, we don't really like to release the exact same thing as any other protocols, because A, we think that FRAX is so unique that if we release something, it's got to be in the image of what's, you know, symbiotic within the protocol. So our bonds would probably be anchored to the actual collateral ratio. So for example, if we had a bond, I think what we would probably do as a starting point is it would only be admitted if the collateral value is dropping compared to the actual target collateral ratio. So let's say there's 90 million collateral in the system because it's a 90% collateral ratio and a hundred million FRAX in circulation. If the value of the collateral drops to 80 million, right now there's a $10 million gap of collateral that's empty. That's when we would actually emits our bond.

Sam Kazemian (40:07):
I think that's the correct way to do it. And the discount rate, whatever it ends up being, would be dependent on this remaining, you know, this gap, right? This gap of value that the system actually needs to recover either by retracting the supply of FRAX or by having the collateral eventually, you know, appreciate and value into that, right? So that's the bond design we're thinking of. We don't like expiry bonds cause it just hasn't seemed like that has been a success as evidenced by the fact that ESD itself is moving off of them, in V2. We like the auto converting bonds, ones that actually just convert to the stable coin at maturity. The main issue with those bonds is fungibility, right? Because if you, for example, have a one month bond, right? And it automatically converts after a month, regardless of whether the protocol is expanding or not, right?

Sam Kazemian (41:08):
The issue is you have to continue to issue bonds that mature on the same exact date. Because if you, you know, the next week you issue bonds that mature at a new date. Now you have two non-fungible bonds with each other. You have the bonds that mature on the first date and then you have the bonds and mature on the second date. And then you can't have high liquidity for those. You can't have a one Uniswap pool, that's trading bonds against a stable coin. You can't ever have exchanges list your bond tokens because they're not the same tokens. They're always maturing at different dates. So like they can't actually list them under the same ticker. And so you have a huge, huge liquidity issue because they're not fungible, right? And so that's one of the main challenges that we're researching is if we do release a bond, it's got to be highly fungible, highly liquid, extremely unique and specialized to, you know, FRAX's fractional algorithmic design and allow keeping up the whole system at the collateral ratio. that's currently being targeted very robustly. So those are the three main characteristics that we're looking.

Bobby Ong (42:15):
I start to see how, what you mean now with the need to launch FRAX bond before you can support volatile assets because of how the underlying collateral value changes and how the bonds can help stabilize these things. I mean, now it's simple. You guys are only back it to USDC and USDC has shown historically that has been quite stable to a dollar. I mean, it could be a case of, I mean, I don't think it would happened but USDC goes down and then the value goes down $0.70 then like, you have issue with FRAX bond, then that may be a bit troubling. But yeah, I see the point of where is FRAX bond coming. It's very interesting the points and how you mentioned that it should be a tradeable token, if you look at Coupon and Basis bonds, those are not tradable and they are technically non-fungible because every one of them has a different epoch.

Sam Kazemian (43:00):
Yeah. So actually that's true of ESD. The current Basis bonds are, I believe, are fungible the original Basis whitepaper with the first in, first out queue of bonds. I think that was not very well thought out because those wouldn't be fungible, right? Cause they have a order that, that they're redeemed but the current Basis Cash bonds, I think since they all get paid, like pro-rata, if they put them in the treasury, I think that they are fully fungible, which is good. The problem with those bonds that the reason we don't want to use that type of structure is those don't really seem like debts. Those just seem like a call option on the next expansion, right. They just seem like, you take out like a call option at like a discount for, right, the price of stable coin squared on the next expansion.

Sam Kazemian (43:52):
And I think that's what their problem is, is that that's why sometimes Basis Cash seems to get stuck under the curve, right, under the peg because sometimes it's not actually profitable or a good deal to buy call options on the next expansion at the current price that the bond pricing curve is offering you. So people just don't buy it, right? And then you have a problem, right. Cause you need to retract because you're at $0.80. Not enough people are buying it because they think it's the bad deal. And then it's like a ball rolling down a curve and then it like slowly just like, stops here. Cause there's, that's exactly where there's no force, right. There's no retraction or expansion. It just stuck, right. And so that's the issue with those. And so we obviously don't want to release something like that, unless we think that there is a better mechanism than that.

Bobby Ong (44:44):
Yeah. That's a very good analogy of how you explain the issue of Basis bond. I've been observing these for the past couple of weeks as well. I mean, it seems to be quite stuck. I mean, all this algorthmic stable coins, ESD, DSD, Basis Cash, they all kind of like stuck below a dollar at this point in time, different issues with each of them, but like interesting points. Yeah. I think sort of, asked most of my questions on FRAX. I think it's been interesting. Now before I end this podcast, maybe is there anything that I should have asked that I have not already asked or anything that you want to say about FRAX?

Sam Kazemian (45:16):
No, I mean, this is actually really fun. The only thing I guess I say is we're really focused on continuing to lead stable coin, especially algorithmic stable coin research and features. This is like a passion project of mine. I kind of see all of us, me, Jason and the back there, Travis and the team, and staff, as this is pretty much our passion, The same way for example, low-level L1 tech is to Vitalik and then a lot of the hardcore ETH guys. So we're super interested in collaborating. I actually, I love the other algo stable coins projects and communities. It might sound like I was a little tough on them but it's actually just to actually give constructive criticism. I actually love to help out in the Basis Cash community. The ESD community has some of the smartest people, Andrew, Mike, Lewis, Scott, Wil Price. All of those guys that are really, really smart guys. So I really think that this is kind of the next big area in DeFi and the stuff that I've been talking about before it was this hot, right. FRAX was started in 2019. So we're just super excited to be working on this area. So that's something anyone that's listening really likes to do hit us up. We're really available. We literally just work on this stuff nonstop. I'm available on Twitter, Telegram and super active in the FRAX Telegram group and community so.

Bobby Ong (46:42):
Thank you very much. Thank you very much. It's been a great pleasure talking to you and learning about FRAX. I thought I knew a lot about FRAX but talking to you and just like, I just learned so much more about FRAX today. Thanks a lot for taking the time to come on the CoinGecko podcast.

Sam Kazemian (46:55):
Of course. Take care, man.

Bobby Ong (46:57):
All right. Take care.

Bobby Ong(46:59):
All right, that wraps up the show. Thank you for listening to the CoinGecko podcast with Bobby. If you like our show and want to know more, check out or please leave us a review on iTunes. If you have any feedback, do drop us an email at Join us for more next week. See ya.

Disclaimer (47:30):
This podcast is provided as part of the overall information on cryptocurrency contained on our website is where your general information only and does not howsoever constitute any endorsement, financial or investment advice nor any solicitation or offer of securities or other financial instruments, coin gecko, and the podcast presenter makes no warranties implied or express of any kind in relation to this podcast, including without limitation the accuracy and updatedness of its content. All opinions and recommendations there in the podcast are based on the personal opinion of the presenter. Please conduct your own research and procure professional advice. Should you at your own risk, decide to howsoever invest or trade in relation to the content contained in the podcast.