Ethereum Foundation DevCon 6 - Cryptoeconomics Track, 18 Oct 2022

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Cryptoeconomics Dive: LP Volatility Harvesting Across Yield Rates by Will Villanueva

Description

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.
https://archive.devcon.org/archive/watch/6/cryptoeconomics-dive-lp-volatility-harvesting-across-yield-rates/

This talk furthers the concept of volatility harvesting. Currently, Uniswap and other major dexes see a huge part of their trading volume consist of the result of volatility in the market. Value changes and as a result, trading volume spikes and LPs profit.

When extending yield, which is also quite a volatile concept, to AMMs, volatility harvesting is increased further to not only affect value but also the yield that value creates.

Speaker(s): Will Villanueva
Skill level: Intermediate
Track: Cryptoeconomics
Keywords: DeFi,AMM,Cryptoeconomics

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

Devcon is the Ethereum conference for developers, researchers, thinkers, and makers.
Devcon 6 was held in Bogotá, Colombia on Oct 11 - 14, 2022.
Devcon is organized and presented by the Ethereum Foundation, with the support of our sponsors. To find out more, please visit https://ethereum.foundation/

A

A

Volatility harvesting across yield rates, so um element Finance. We had an event last night. If any of you are there. Thank you for coming. So it was super cool.

A

Super enjoyable, I'll go into what we do in a second, but one of the major things or one of the things that I want to talk about is what is volatility harvesting, so volatility harvesting at its Essence, if you think of amms in the current market, our sort of volatility, harvesting engines- and what does this mean is as spot prices drop, whereas spot prices rise in the space?

A

This gives opportunity for market makers to then go ahead and ARB the market make profit, bringing those spot prices to a certain level and allows LPS to capture fees and get value from those volatility changes. So this is essentially you know what I think of when I think of a volatility harvesting tool. So let's, let's apply this to the yield markets before we do that, let me give you a quick intro into what does element Finance do so element Finance at its core is a yield splitting protocol.

A

So what we do is we take different yield sources in the space. Let's say you have an S to eat position. Let's say it's a urine position. A lot of our things are built on urine and what we do is, if you put a million dollars in we take that position. We split it into two parts, so there's the principle and the interest, so you have a one-year lockup.

A

If you have 10 interest on this position, you have a million dollars in principle and I guess this shows 20, so 20 on a million dollars would be 200k in interest. So at the end of the term, at the end of the year, you can collect both of those uh we use a curve. This was developed mainly by actually yield space. We did an alteration on this curve. um It's called a constant power sum and essentially what we do is during the market.

A

We let people essentially stake the principle that they have that's locked up and what the curve does is it's a time-based curve and it sort of follows this concept. You know where I give you a dollar for 99 cents or 90 cents, so uh this is the concept of the opportunity cost of money. So if I say hey, I'm gonna offer you, you know this million dollars, but you can't touch it for a year. You can't do anything with it. What is it worth for you to have that million dollars?

A

You know I can't put in a savings account. I can't stake it. I can't farm with it I'm losing some type of interest on that right, so I, say: hey I think I could probably get like 15 apy I'll I'll give you 90 of that. So here's a dollar you can't use it for the next year, I'll give you 90 cents for that dollar, and so what this curve essentially does? Is it acts with respect to this time period um between the constant product and the constant sum formula so early on it has price Discovery.

A

You can have the apy change on the principal this million dollars, it's locked up um and later on it sort of converges one-to-one value. If I say hey here's a million dollars tomorrow, you can't use it until tomorrow, you'll probably buy it for a million dollars right versus for a year from now, then you get that opportunity cost. So this is essentially how how that works.

A

So what we saw early on in our platform is these markets actually worked really well, um as people locked up their principal in the amm's, uh people were actually trading these with opportunity cost of money um based on the variable rates that you could see in the market. So, for example, when we had first launched, we saw a curved try, crypto term, where the yield was the fixed apy was at 15 percent and the variable rate was at nine percent in this term.

A

So what ended up happening is the fixed rate. People actually brought the fixed apy to be higher than the variable apy. How does that make sense?

A

um It's hard to grab that it's hard to understand that what happened is we were in a market lull, so the variable rates had all dropped in the space, and people were speculating that they would rise and sure enough they did. They ended up making money off of it and so why? Why is that related to the number value here? So we have.

A

This system called yield token compounding which essentially lets you go through and leverage your exposure to variable interest to go long on variable interest and it's this concept where you meant these principles, these yield tokens. You sell your principal for a fixed apy percentage, and you do this repeatedly until you own a stronger exposure to the yield.

A

So this is like I. Don't want to go into too much depth here, but if the variable apy is at 20- and let's say the fixed apy is at 10, I have a 10 spread if I, basically re-hypothecate and do this I deposit, I, sell, I deposit, I, sell I deposit I, sell I. Do this six or seven times you know, I can 7x that to around 70.

A

So this causes an upward price action in the fixed rates in the market, and so, if I believe, the variable rates are going to go significantly higher I might be willing to let the fixed rate go higher than what the current variable rates are. So we saw some other interesting volatility behaviors um at one point there was I actually did a talk about this at another conference there was issues with mem I. Think a lot of you remember that there's a whole Scandal there.

A

We have a lot of those, sometimes in our industry and the fixed APR, like you know, popped up to 130 percent, and then we had another case where we had a wbtc vault, where it dropped completely to zero percent. Actually, um and during this time we saw a lot of volatility movement, the price action. Basically, the the fixed apis went up and down. They went to 50 to 100 down to 20 back up to 70..

A

It moved all around all around the place, um and so this is sort of you know an example of the volatility in action, eventually it normalized to around 10 percent. So uh what do we notice in the fixed rate? Market? Is it generally sometimes, except for some of these exceptions? It tends to follow the variable rate ends up being somewhat similar, um the apy to what the variable rate is in the market, sometimes a little less, sometimes a little bit higher. They tend to track each other they're correlated.

A

This is what we've seen from the data that we've gotten in our current V1, and this is where you know I've done. A talk on this before is essentially these fixed rate markets that we do they're not really just fixed rate markets, they're they're, yield markets, we're essentially creating this value capture mechanism for yield markets, and so um what is like element Finance like and and these systems and fixed rate systems as a whole. You can sort of see them as it's more than a fixed rate system.

A

It's a highly liquid Arbitrage Market that lets you sort of play and create profit, as yields shift within the space, so as you'll Spike to tan they dropped. Two I can create profit off of this in a similar way that, if eth spikes up and then drops I can create profit mechanisms off of this. It also allows for really cool things like Leverage going long variable being you know negative on the you know, future variable interest, like I, believe this one's going to go to zero percent.

A

This this Farm shot up in value, so sort of the the dream for me and it's not just element, it's sort of the fixed rate space which is akin to you know some some similarities to zero coupon bonds. Other things like that is that it ends up becoming this intermediate layer for yields across lending and borrowing platforms, and also like not just blending and borrowing platforms but yield platforms as a whole. So real, quick, uh I'm gonna go into like one example. This is going to be I.

A

I can't go into a lot of the depth, but I want to sort of paint a picture. So I did a Twitter thread. You can check check it out on Twitter, it's pinned um where I actually coded I went for a week in my room and coded a bunch of simulations, and this simulation particularly was on something called Fiat Dao and what they did.

A

It was akin to maker where they had a one percent stability set fee and that one percent stability fee is basically your borrow rate right and what they did is they took principal tokens as collateral, so I could take a principal or a fixed rate. Let's these principal tokens are these fixed rate.

A

Tokens right so I could buy a fixed rate, use as collateral borrow Fiat swap that to die, buy more fixed rate, die I could basically leverage into the fixed rate side, and so, if the borrow is at one percent, apy or let's say makers, stability, if you want to think in relation to maker or other systems, is that one percent apy um and uh essentially what what we're able to do is if the fixed rate's at three percent, you can leverage into that fixed rate until it basically converges to that one percent value um which is really cool, and what does this also allow for so there's um this concept of fixed rate, borrowing mechanism or adapter, someone there's a few people who are working on this on the element platform uh actually currently, and essentially what this is is you could take these uh these instruments that we have and you can plug them into compound and Ave and existing systems?

A

And what you do is you essentially create a hedge Market? You transform them into fixed rate, borrow markets. So let's say I have a compound borrow position, um let's say I'm borrowing to I from compound. Let's say we hit, like you, know very low utility rate, and in this case the apy to borrow goes low. Let's say it goes to half a percent. This is great I want to lock that in you know, so what you can do with sort of the system that you know we have.

A

Is you take out that borrow and you hedge, by going directly into the yield? So these are the yield tokens I talked about earlier and what you do is, let's say in this case I'm saying: okay uh lending is at three percent. Whatever borrow is that one percent apy? If suddenly the borrow, goes up to 10 apy The Lending side also is going to spike up to around 10 or 12, and so because I hedged. My borrow position with this yield exposure on the lending position.

A

That means my borrow position, gets pretty stable, I'm hedged against Stark drops or rises in the interest rate. This.

B

Is really cool.

C

A

We can basically turn it in any platform. Lending platform doesn't matter what chain into a fixed rate borrow system this can be built on top. This can be an adapter. So what does this mean? Is if you sort of take a step back, you can see that these mechanisms, these fixed rate and variable rate in these yield markets that we have what they do. Is they create this convergence layer for defy rates for Lending Market rates?

A

You use it to take out, borrow positions. You can create essentially fixed rate and variable positions on the lending side, the side where you get apy and when you go cross-platform, if I basically do a fixed rate, borrow on platform a um and then I essentially use that to sort of go into and purchase the principal tokens on platform B that are certain value. You end up having this sort of liquid layer in between that brings all those rates to convergence and brings them together.

A

It's really cool I wish I had a better diagram for this, but I'm sort of starting to introduce this topic and and playing with this, and this works across different l2s l1s. um It's it's really really powerful, actually, um and so. We've already started batching, um basically activities on Aztec for ethereum, and so what I sort of see is this world where we can sort of batch a lot of the activities from different um different layers, different chains all on one chain, and you can sort of interact and Arbitrage?

A

Basically, the borrow the lending rates, the yield rates within that one chain that sort of works as a centerpiece for.

C

A

So I want to talk for a second about you know amm, so amms I think are there's a lot of downsides to them. um So you know we had unique V3 with concentrated liquidity model. um You can kind of see. Amms also are sort of a free straddle option for market makers. A lot of the value that's captured off amm's uh goes to market makers not to the people, providing the liquidity not to the lpers an organic way. This is why a lot of people are doing research on cool ways.

A

How do you bring Mev into the actual, um the actual amm? How do you bring transaction ordering and validator activities into the actual amm, because sort of the most altruistic position in the markets is being an LP, especially in an organic market where you don't have emissions things like that, and they take on a lot of risk, so spot prices? Can you know, go up down through the roof?

A

They a lot of times can see less gains than if they had held one position or um you know worse gains on the other Spectrum, and so there there's definitely work that needs to be done on amms, but they're really really good for yield prices and yield tokens and and principal tokens and everything that's involved there and the reason.

A

Why is because if I have uh you know fixed rate usdc and we call these principal tokens in our platform and that's at 10 apy over a year, that means it's going for 90 cents day by day that 90 cent value converges to one dollar. You don't really experience any impermanent loss, it's sort of more like a stable swap um the fees are, are you know pretty cool as a percentage of that yield that people uh secure and swap with?

A

And you know that that's really interesting, and so one of one of the issues we've had in V1 and what we've been doing um is we have this liquidity fragmentation? It's like I have this six month term, this fixed rate term that I'm interested in, but it's three months through and now I need to switch to another six month term, or um you know, as an LP I need to. It makes sense for me to pull out my position and go into another one.

A

These are some of the weaknesses, so we did a bunch of simulations and Analysis in our current fee markets, uh capturing volatility in the space um which is super interesting. um This is fairly complex. The main thing that I'll share uh the biggest learnings is, as you see, uh yields rise.

A

um That is the highest one of the highest value capture mechanisms for LPS. As you see a yield rate drop. This is actually also if you're active, extremely profitable. So what happens if a yield rate drops? So in this case we have a situation where we're going from 10 to 2.5 apy and in 2.5 months it drops to 2.5 apy, um and this is a six month term, so that drop essentially brings that principal token right. If 90 cents on a dollar would be 10.

A

If it drops to five percent, then that's 95 cents on a dollar right. Five, five cent discount- um so essentially it's worth more and what you can do is it drops. Is you just sell the principal token so someone who locked in a 10 fixed rate apy in 2.5 months when it dropped to 2.5 percent, they sell the principal tokens.

A

They got 4.2 percent return in 2.5 months because they sold it, which equates to 20 APR. So if you're going into position- and you see you know what I believe this like rate's going to drop- it makes sense to buy the fixed rate side, because once the rate drops, then you can flip that you can sell that um and then you can basically get an early Redemption on your apy.

A

You get fulfilled quicker on the other case, if the apy drop or Rises, um you get a higher exposure to the fixed rate apy, so you get more more yield exposure, it's also profitable, Endeavor and mechanism, and so what? What does this lead to this sort of leads to products that can be built on top active strategies? Vaults really interesting things um even Bots that market makers can do to really capture a lot of this value.

A

The value is absurd that can be caught, especially once you're doing borrowing lending markets once you're hedging on those once you have printable tokens as collateral once you have these markets like running truly smoothly, there's a lot of ways to sort of make profit off of changing yield rates. So it's it's fascinating.

A

So another thing we saw is it's sort of unideal for an LP, because, with these like terms, we used to have were essentially uh you in element. We have like usually six-month terms, and we do a new six-month term. But what happens is like here's a simulation, the fees drop off as this term ends. It doesn't really make sense for me as the lp to stay the full term. uh It makes more sense for me to pull out.

A

Why do the fees drop out because, as time goes on, the value converges, so you have less of a differential, so the fees are less and also people are less likely to trade on, like one month left of apy um Johnny Ray, my co-founder. We're both uh Eve to researchers, um but he actually recently came up with a new model. um We're calling this hyperdrive and it introduces no more terms.

A

um Lps is Perpetual positions and essentially this new amm that we've been researching and we'll be releasing a lot of data on the simulations on it here soon lets you basically underwrite someone to take out a fixed rate term um on whatever time they want. So they can say six months three months and it's a brand new term. You don't have to worry about these terms. Going halfway into a term I have one month left you underwrite that position immediately, and this is really good for LPS too.

A

This is really good for users if I want to take out a loan from Ave I want a one month, three month six month, I don't want this weird halfway position thing for an LP: it's a Perpetual position. I get I, get average exposure to. um Basically all these different term lengths.

A

um It's it's a really good situation for them, they're able to capture and Garner more fees, and this also allows for just better systems to be built on top Simplicity um and better, better Vault, better value capture mechanisms, better ways to go long and variable yield on the market, better ways to be negative on variable, yield um and sort of play with these these markets and and have fun. So uh this is LP volatility, harvesting, cross, yield rates, I'll, uh try and release a lot more uh simulations and data here soon on Twitter follow us.

A

This is really cool. We are doing some like new and groundbreaking research here and there's a lot of profit mechanisms so like pay attention is all I have to say. Thank you. Thank.

B

You we have a little minute for some Korean, a anybody have a question.

C

Okay, so earlier this year there was a project called defrost Finance on Avalanche and they did um Leverage The Yield farming, um but they had a lot of trouble with keeping the liquidity in order to do that. So in your example, where you had that three to seven leverage on the dies, how do you maintain that liquidity to keep that so people actually want to trade? The other side on you.

A

Yeah so great question, um so we actually saw like I think it was through the bull market, something like 400 million dollars of trades on our platform, um so we actually saw a really active uh activity on yield token compounding or increasing your exposure on the variable side. um I think. Maybe some of your question is how do you match that if you have leverage to the other side, because that's cell, but you match that on the purchase side?

A

That's why things like uh like having these fixed rates is collateral right being able to leverage into buying the fixed rates is important. Liquidity is another thing like liquidities down significantly in the market, including our platform, uh I, sort of think we build sustainable, strong products. Things like hyperdrive. That makes sense that are a significant leap on what exists in existing tradify like this stuff doesn't exist. It's it's crazy, cool um and I.

A

Think it's just a matter of time of sort of building and garnering um garnering that that space I've been looking at a lot of things too. Looking at real world assets like some some other things that can also play in as yield sources, I think staking derivative is a really good yield sources. People are going to do those regardless Mev is an interesting one, etc, etc. So.

B

No more questions, so thank you will cool. Thank you.