How I Learned to Stop Worrying and Love Impermanent Loss
Photo from Stanley Kubrick's How I Learned to Stop Worrying and Love the Bomb
There has been a lot of FUD about impermanent loss (including from myself). And rightly so, because impermanent loss:
- is confusing
- has difficult math
- and no one wants "loss".
In this article I'm going to share my current view of what impermanent loss (IL) is and a different way to look at it which I think makes it less scary. In fact, you may (like me) see IL as a good thing!
What is Impermanent Loss?
Impermanent loss was first explained by this post on Medium by Pintail to describe what happens on Uniswap when people deposit into a liquidity pool. Basically since you always need to deposit 50% of each side of the pool, you will gain coins on one side and lose tokens on the other with each price change.
Let's look at this with my favorite farm, CUB-BNB, during a time when BNB nearly doubled in USD value. At this time, we would expect IL to be pretty bad.
Here we see the big move that BNB has made in 3 weeks, from $316 to $578. We calculate that in CUBs (because IL is calculated in the divergence between the two prices) 78.395 CUBs to 162.104 CUBs, or a bit over a 2X move.
Looking at the number of tokens (and seeing that the stake has not moved so no tokens have been bought or sold), we see that the IL caused us to lose 1.476 BNB (worth $853) and gain 169.784 CUB (worth $606), which is a net loss of $247, or 8%.
Here's the classic chart for IL. I'm actually surprised that I calculate a loss of 8% because it should be 5.7% according to this. But it doesn't really matter for my point to be valid.
And if text is easier to understand:
a 2x price change results in a 5.7% loss relative to HODL
a 3x price change results in a 13.4% loss relative to HODL
a 4x price change results in a 20.0% loss relative to HODL
So we have a loss of between 6-8%.
Now let's compare what type of interest we would have earned in that amount of time. Rather than calculate it, I can simply look at today's APR and remember that it was better in the past. So right now we have an APR of 262.36% which gives this Return on Investment:
This assumes compounding daily, and so the daily interest rate is 0.72%. If we put that into any program that calculates interest like this one, then we see that the interest gained in 21 days from the initial $3,050 is $495.93. Subtracting the IL then leaves $249. So IL is taking away some of my gains, but I'm still coming out ahead over anything else available.
Conclusion: A clear win for farming.
What about the other farms?
You could do a similar calculation with CUB-BUSD and you will find even less IL, because:
- Just like with CUB-BNB pool, the value of CUB relative to BUSD also went down. So you would have more CUB than before due to IL (the loss would apply to the BUSD).
- The APR on the CUB-BUSD cool ranges from a little to a lot more than the BNB pool, and so you would have gained more there.
But I feel pretty confident that the 2X gain in BNB made up for the lower APR compared to CUB-BUSD, even without doing the math.
And what about if CUB went up relative to the other token? Well, you would see the same exact situation but reversed, so again 50% of your APR profits would be lost by IL.
And if CUB went up even faster? Well, first of all remember that your token value will always be larger than IL, so we're comparing a big win with a bigger win. And yes, if CUB doubled overnight, you would almost certainly have been better in that moment out of the LP.
But how do you know when that time is? Sure, it will moon SOON but when is soon? How much CUB are you not gaining because you fear that CUB will moon?
It's all perspective
Another perspective is whether you want more CUB or BNB. In the example, I lost BNB. If CUB had outperformed BNB, I would have lost CUB.
Whether you think this is good or bad is a bit of a Rorschach test: are you more bullish about CUB or BNB? If you think CUB will go up higher than BNB in the future, then or course you want more CUB.
If you see a pair of bears, you might be bullish on CUB
And if you don't know which will go up, then regardless of what happens, you will be happy. Because as the CUB price goes up (relative to BNB) you will be selling CUB high and buying BNB low. When the opposite happens, you will be selling BNB high and buying CUB low.
This is why I have most of my deposits in CUB-BNB: I want to have both of them in my wallet. And while I think CUB is going to go up more than BNB, I really don't know. And in this short term where BNB has been outperforming CUB, the IL is in fact giving me extra CUBs.
Impermanent loss is an illusion
From leofinance Discord chat
Even better, impermanent loss automatically rebalances your portfolio. It continually and automatically sells high and buys low, which is exactly what you want. And if you realize you actually don't know where the price will go, then there's nothing at all to worry about. You don't need to wait until the prices have the same relative value as when you put it in.
Sure there is a cost for this automatic rebalancing portfolio (this is the IL), but I'm ok with that. I would never have had so much BNB in my portfolio as I would if I didn't have that farm. And so for that, even if I lose a significant portion of my gains due to IL, I'm sure I will come out ahead.
What do you think? Does impermanent loss still scare you?
Not financial advice / DYOR :)
Posted Using LeoFinance Beta