One of yearn.finance’s core tenets is downside mitigated savings accounts. Thus far we have focused on lossless strategies, while the upside is potentially less, the downside is mitigated. This however leaves a lot of desirable opportunities unavailable to vaults.
Lets go through an example; lets assume the current best opportunity for DAI is at 10%, however if we did DAI:ETH we could be earning 20%. This is not a strategy yearn would have considered, since it adds a variable to DAI, namely ETH’s price volatility. If we take 1,000 DAI and convert it to 500 DAI : 500 ETH, and ETH price decreases by 10%, we would have < 1,000 DAI.
What if we could offset any potential price decline in ETH? Our goal would be to mitigate our ETH exposure, while the increased interest minus cost of mitigating such exposure is still greater than the original opportunity. Using the example above 20%-<hedge cost> greater than 10%.
This offset of losses is called hedging your risk. Lets first look at how we would do this with a futures contract, in the example above we bought 500 DAI worth of ETH, for simplicity sake, lets say that’s 1 ETH. We would like to earn yield with the ETH for 4 weeks. We agree with Alice that we will give her 1 ETH in 4 weeks if she gives us 500 DAI. Alice believes the price will go up, so being able to buy ETH for cheaper than market price in 4 weeks is a good deal for Alice. After 4 weeks, the futures contract is settled, 1 ETH is given to Alice, and Alice gives us 500 DAI. We have our 1,000 DAI and Alice has 1 ETH.
A futures market settles the underlying asset (ETH in the above example) at a future determined date. This is one mechanism we could use.
Using the above example, lets say ETH increased to 1,000 DAI. Alice is happy, she made 500 DAI profit, however, if we did not do a futures contract, we could give that 500 DAI profit to our LPs, further increasing their yield. So what if we wanted the option to be able to sell it to Alice in 4 weeks, but we didn’t necessarily want to do it after 4 weeks passed?
For this, we can purchase an “option”, just as the word says, it is the option to execute a contract (like a future). This would work exactly the same as above, however this exposes Alice to a potential downside (since what rational actor would only buy something if it would not make them profit?), so Alice charges a premium, she charges us 10 DAI for the option (right) to be able to sell her 1 ETH in 4 weeks for 500 DAI. If however in 4 weeks the price of DAI is greater than 500, we can simply not execute the contract.
Assuming the price went up, the LPs made more profit, assuming the price stayed the same, LPs made 20%-10 DAI, assuming the price was less then we could execute the option and have 500 DAI.
In the above examples, there is a lot of “micro management”, prices need to be compares continuously and checks on when to execute specific options/futures. What if we could simplify this a bit more? To do so, we can use Options settled/denominated in DAI
Options allow us to offset the “loss” part. So looking at our above example again, lets say the price of ETH decreased to 400 DAI. If our agreement with Alice was a Binary Option settled in DAI, then instead of us giving her the 1 ETH, and she giving us 500 DAI, Alice gives us 100 DAI (that is what her loss would have been). This means the underlying asset is not settled, but instead the profit portion is. So at this point, the pool is 600 DAI : 1 ETH (400 DAI). While the net result is the same, this required a lot less steps. Lets consider that the binary option cost us 10 DAI, if ETH increased, we would have 500 DAI : 1 ETH (600 DAI) = 500+600–10, if the price of ETH decreased and we executed our option, we would have 500 DAI : 1 ETH (400 DAI) + 100 DAI (Alice) = 500 + 400 + 100–10.
This allows us to keep our position neutral, while being able to enjoy optimized yields. An important note here is options pricing. Lets say the option costs 10 DAI for every 1 ETH. That’s 10 DAI over 500, or 7.3% every 4 weeks, so the option pricing here becomes incredibly important.
Next we need to understand option pricing, open interest, and strike variance, but I will cover those in a next post when we discuss the changes/modifications we have been proposing.
