Providing liquidity: AMMs, impermanent loss and whether the fees cover it
Impermanent loss is neither impermanent nor a loss in the way the name suggests, and misunderstanding it is why so many liquidity providers underperform simply holding.
How an automated market maker prices things
A traditional exchange matches buyers with sellers. An AMM has no counterparty: it holds a pool of two assets and prices them with a formula. The most common is the constant product rule, where the quantity of one asset multiplied by the quantity of the other must stay the same after every trade.
Buy asset A from the pool and its quantity falls while asset B rises. To keep the product constant, the price of A must rise. That is the entire mechanism, and it means the pool automatically quotes a price and automatically moves it as people trade.
You become a liquidity provider by depositing both assets in the correct ratio. You receive a share of every trading fee, proportional to your share of the pool.
Impermanent loss, worked through with numbers
The pool always rebalances toward whichever asset is falling. You end up holding more of the loser and less of the winner. That is not a bug, it is the mechanism doing its job, and it has a cost.
The size scales with how far the ratio moves. A 25 percent divergence costs roughly 0.6 percent, doubling costs about 5.7 percent, a fivefold move costs about 25 percent, and a tenfold move costs over 40 percent.
Concentrated liquidity, and why it made things harder
Later AMM designs let you provide liquidity only within a chosen price range rather than across all possible prices. Within that range your capital does far more work, so you earn substantially more fees for the same money.
- The advantage is real. Capital efficiency can be many times higher, and fee income rises correspondingly.
- The cost is that price can leave your range. Then you are entirely in one asset, earning nothing, holding the one that fell.
- It converts a passive position into an active one. Ranges need managing and rebalancing, which costs gas and attention.
- Impermanent loss is amplified within the range, because your exposure is concentrated exactly where the price is moving.
- Studies of real positions have repeatedly found that a large share of concentrated liquidity providers underperform simply holding the two assets. Narrow ranges are a trading strategy, not a yield product.
When providing liquidity actually makes sense
How to evaluate a pool honestly
- Find the pool's actual fee income over the last thirty days, not the projected annual rate.
- Divide by the total value locked to get a real realised fee yield.
- Look at how far the pair diverged over the same period and calculate the impermanent loss that implied.
- Compare the two. If fees did not exceed the loss, the pool did not pay for the last month.
- Check what share of the advertised yield is fees and what share is token emissions, because emissions are dilution paid to you in a falling asset.
- Then compare the whole thing against simply holding the two assets, which is the benchmark that matters.
BEFORE YOU MOVE ON
Common questions
What is impermanent loss?
The gap between the value of a liquidity position and simply holding the two assets. The pool rebalances toward the falling asset, so you end up with more of the loser. A doubling in one asset costs about 5.7 percent, a tenfold move over 40 percent.
Is providing liquidity profitable?
Only when accumulated fees exceed impermanent loss. That is reliable on stablecoin and correlated pairs, and frequently false on volatile pairs where the advertised yield is quoted and the loss is not.
Is concentrated liquidity better?
It earns far more fees per unit of capital while the price stays in range, and it amplifies impermanent loss and requires active management. Studies of real positions have found many concentrated providers underperform holding.
Risk warning: crypto is highly volatile and largely unregulated. You can lose everything you put in. Nothing here is financial, investment or tax advice.