Today I’m going to dive deep into Element, the leading rate speculation platform and then compare it to its peers. Data from July 2nd.
There is no token for this one (yet) but rate speculation platforms are a new Defi primitive that you might want to be aware of. The basic idea is to split your yield bearing asset into 2 parts, a principal token that can be redeemed for what you put in at expiry and a yield token that can be redeemed for all the profit on your yield bearing asset.
Element’s starting yield bearing asset is LUSD3CRV-f which is a Curve pool LP token for Liquities LUSD token. This token earns fees from people trading but critically there is also a Yearn vault that is currently earning 9.54% by farming and selling CRV.
When you mint using element you get 2 tokens:
ePyvCurveLUSD-27DEC21: Element Principal yearn vault Liquid USD expiring Dec 27, 2021. This is the P token.
eYyvCurveLUSD-27DEC21: Element Yield yearn vault Liquid USD expiring Dec 27, 2021. This is the Y token.
The yv portion of that token name indicates that the LUSD3CRV-f token you provide is deposited into the yearn vault. That’s where it gets the yield the Y token accrues.
So let’s dig into some use cases using numbers on their dashboard.
- Fixed Rate Lending
Naturally, since the P token is just dead capital until expiry people are going to sell it. This leads to a price on the principal asset that is strictly less than the principal you can claim with it but which should be 1:1 at expiry time. Right now the P token is valued at 0.9722. They do the math for you regarding price and duration assuming it will be 1.0 at expiry and provide an APR for those buying the P token: 7.87%. I’ll note that Blockfi just cut their rates to 7.5%, so if you want to pull some capital out from there and buy some P token here you can let Blockfi know you thought that was uncool by voting with your wallet.
In the simple case you just hold this until December then claim for fun and profit. However, if you’re hungry for more gains you can also monitor the rate over time and trade in and out of this asset between now and expiry. You’re going to eat some trading fees but just last night the rate was as low as 5.87% while I was looking at it so that’s quite a bit of variation.
The other side of this is the Y token. Basically this is just a time cost of money thing. If you think the Y token will be redeemable for some amount at expiry, then you clearly have some opinion about the average rate between now and expiry. Let’s say for the sake of argument you believe the market rate of 7.87% is fair. You wouldn’t buy a token for $1 that is only redeemable for $1 at the end of the year. Clearly you’d rather have your money earning interest for you than locked up doing nothing. So, to buy the Y token you not only need to believe it will be redeemable for $X, you need to be able to purchase it for $X-plausible interest.
- Rate speculation
Whomever is minting and selling P tokens right now is probably doing so because they believe they can get more than 7.87% on the asset between now and expiry. They are long the rate. When they sell they have their LUSD3CRV-f token back less 2.78% of their capital and they still have their Y token which we’ll talk about more in a second. The simple thing to do would just be to deposit the remaining capital to the yearn vault and earn 9.54%. If the rate holds you walk away with some profit. But this wouldn’t be a good post without some amount of lunacy so here we go…
First I will note the very shallow liquidity of the Y token pool at $20,557 simply isn’t going to let me trade $10k LUSD3CRV-f for Y tokens.. That makes some sense because the people minting the P tokens are generally long the rate so they are likely holding their Y tokens so we don’t expect the liquidity pool to be very deep. Therefore if we want more Y tokens we’re probably going to have to get them by minting… a lot. Where there is recursion in Defi there is a use case for flash loans. So if you really believe the rates are going to be higher than 7.87%… you want to turn $10k LUSD3CRV-f into y tokens.
- The P token price is 0.9722.
- Borrow 10k/(1-.9722)~=$360k stablecoin as a flash loan.
- mint LUSD3CRV-f worth ~ 360k
- Mint $360k*.9722=~350k P token.
- Sell the P tokens and your original $10k LUSD3CRF-f and repay the loan.
- Keep Y tokens minted from 360k in principal until expiry.
- Redeem for fun and profit.
For simplicity let’s say the rates averaged 10% over the next ~6 months. The Y tokens you redeem be worth ~18k for your original 10k investment giving you ~160% yield (80% yield over 6 months). Nice. That’s how we do it with Defi! I’m clearly ignoring some factors such as trading fees, slippage, etc. It’s an illustration of the concept.
- LP
Yo dawg, I heard you liked speculating. Instead of speculating on the rates you can speculate on the speculators speculating on the rates by being an LP in the pools for the above tokens. This not only makes some profit on the LUSD3CRV-f tokens from Curve but also people trading the P and Y tokens over time. There should be steady trading volume on the P token pool over time because the value should go from .9722 to 1 over 6 months.
If you’re in the LUSD3CRV-f/ePyvCurveLUSD-27DEC21 (god I love Defi) pool then you stand to make trading fees on this correction and profit from rate volatility and speculation in the market from people like 2) above.
Unfortunately, it does sacrifice the rates from the yearn vault on half your capital sitting in the AMM because for some reason the AMM uses LUSD3CRV-f and not yvCurveLUSD. I asked about this on their discord. You can read their reply [here](https://discord.com/channels/754739461707006013/826211846541148211/860407531187470346).
I’m even less convinced on LPing the Y pool but they write an article on this that might convince you more than me [here](https://medium.com/element-finance/the-automated-fixed-rate-rebalancer-1bf78010dce9).
And that folks, is how you use these rate speculation platforms.
4) Comparison
So with that as a baseline, let’s have a look at the other competitors in this arena. They all work similarly to Element with the idea being deposit a token they can generate yield with and then they mint some form of Principal and Yield token pair which you can use to speculate on rates in various ways.
Element gets yield from Yearn finance which is cool because yearn has an affiliate fee program where they give partners some of their 20% fee for bringing liquidity to their platform. These affiliate fees are what funds Element’s treasury/development. Generally this is yield you wouldn’t otherwise have access to if you used Yearn directly so it’s nothing lost and is an all around good solution. Otherwise, Element has no fees, and no discrete plans to add them which is unique (though they could add one to their Balancer pool if they chose).
Element’s P + Y token setup is the most straightforward. Their Y and P tokens are going to be valued at the time cost of money. That’s how the zero-coupon bond mechanism works on the P token. It’s how it offers the fixed rate lending use case.
As I poke fun at above, the pools have ridiculous symbol names. It’s important to know for comparison that half the LP for Element is a non-interest bearing asset. I’ll come back to this when discussing capital efficiency tradeoffs.
Element is unique in that it deliberately offers no early redemption process whereby you could claim the underlying value of the tokens before expiry. As with all one way bridges this creates a price ceiling on the P and T tokens but no price floor. As to why they did this:
We do not support early redemptions, one reason is to reinforce time as a variable and not to penalize users (typical early redemptions punish users in trad markets).
I don’t have the data to back that up, but there ya go.
So, let’s compare this to Pendle. Pendle’s Y token isn’t directly redeemable for yield. Instead your yield for each pool is redeemable on their dashboard given the duration of time that you have held the Y token. Think of it like farming, you hold/stake a token, you get paid every block, and can claim anytime you wish. Naturally, since the Y token isn’t redeemable for anything, it’s value decays to 0 at expiry. This makes the Y token resemble something closer to an options platform which may be what they were going for.
The downside is this adds gas accounting to the ERC-20 token itself making each send/trade more expensive and of course this means they needed to make a custom AMM because a normal Liquidity Pool where one of the assets is purposefully going to zero is going to be rather devoid of liquidity after the first round of people get burned. So, it creates more transactions, each transaction is more expensive, and they had to write a custom AMM which has added more smart contract risk. In addition to this Pendle takes 3% of the yield and 0.05% of the volume on the AMM for themselves. The upside is you get a Pendle token you can farm and flip unlike Element which has none. In the sake of fairness, here’s what they said:
Yes you are correct that to be able to do this, both our contract system and the AMM will need to be more complex. However, there are certain benefits. One of those is the simplicity in the definition of YT and flexibility in how you can use it. YT directly represents the right to receive yield from the underlying yield bearing assets over the time period that you hold it, guaranteed on-chain by its exact mechanism. As such, people can hold YT for any period of time that they wish to receive the yield from, and sell it whenever they like. Its value is solely derived from the future yield from this point to the expiry, without any kind of past yield baked in it.
Next up is Tempus. Remember how I mentioned that those Element LP’s were forfeiting yearn yield on half their balance and I was all emo about it? Well, Tempus fixes that! Instead of having separate B/Y and B/P pools you have a single Y/P pool. Obviously this improves the capital efficiency for LP’s and is designed to attract more capital. Why wouldn’t you LP at that point? P + Y should equal B because of arbitrage and early redemptions so rather than depositing in yearn with B directly, deposit into Tempus which deposits in yearn for you and you still get your yearn yield plus the trading fees on the liquidity pool! This draws people into the system who aren’t even interested in rate speculation which is a great design.
Now, the downsides. Tempus is going to take some of the yield (starting at 0 but planned for governance in the future). A larger problem is Tempus is planning to directly farm from protocols instead of using Yearn. In theory they could just copy whatever Yearn does and not have to give up 20% of the harvest but we both know it’s never that simple with smart contracts. The result is probably going to be less yield on the base asset. For example, let’s say they have an aUSDC pool. Are they going to farm and sell AAVE for you? If so, that’s a lot of responsibility they are accepting onto themselves to both maximize yield and run a rate speculation platform. If not, the yield just won’t be competitive, LP fees or not, compared to other professional liquidity pools with competitive strategies. I personally think they should change that plan and just go with an alUSD3CRV-f pool to start. They kept mentioning Lido so they’re probably thinking of the stETH Curve LP to start.
Last in line is APWine which I’ve been able to gather fewer details about. What I do have though is incredible so hold onto your butts. What if I get tired of withdrawing, redeeming, depositing, LPing, etc every few months as we roll from expiry to expiry and I just want to earn my comically oversized yields in peace? APWine has you covered! I really think UMA should pay attention and call these guys. Basically instead of having 1 pool per expiry eg: LUSD3CRV-f/ePyvCurveLUSD-27DEC21 they’ll have one pool per base asset with a revolving door of underlying tokens in that pool. I await more details!
Which system I use is of course going to be wherever I get the highest, safest, yields. We’re all just liquidity locusts around here. We’re looking to park our capital where it’s treated best. That being said, I think the Tempus design is the most exciting while I think Element has the lead in terms of exposure/liquidity.