The pitch for DeFi yield farming sounds straightforward: deposit two assets into a liquidity pool on Uniswap or Curve, earn a share of trading fees from every swap through the pool, and withdraw your assets plus accumulated yield whenever you choose. Many new DeFi users take this at face value — and discover only at withdrawal that they have less than they started with, despite holding positions that appreciated in price. The gap between their actual return and what a simple "buy and hold" strategy would have returned is impermanent loss.
Impermanent loss is not a scam, a bug, or a fee you missed. It is a structural, mathematically inevitable consequence of how automated market makers work. Once you understand the mechanism, you can evaluate any liquidity pool opportunity accurately — and avoid the mistake of providing liquidity to a volatile pool with insufficient fee income to cover the IL cost.
The Mechanism: Why AMMs Create Impermanent Loss
The dominant AMM design (Uniswap V2 and most forks) maintains a constant product formula: x × y = k. The pool holds equal dollar values of two assets; when price changes, arbitrageurs restore the ratio to match external market prices. This is the mechanism that keeps AMM prices accurate — but it's also the source of IL.
Walk through a concrete example:
You deposit into an ETH/USDC pool when ETH = $2,000. You deposit 1 ETH and 2,000 USDC — $4,000 total. Assume the pool now has 10 ETH and 20,000 USDC total (k = 200,000). Your share: 10%.
ETH price doubles to $4,000 on external markets. Arbitrageurs will buy ETH from the pool until it's priced at $4,000 there too. The constant product formula determines the new pool composition: if ETH price = USDC/ETH = 4,000, then for the pool to price ETH at 4,000 USDC/ETH, the pool must hold approximately 7.07 ETH and 28,280 USDC (7.07 × 28,280 ≈ 200,000 = k).
Your 10% share is now worth: 0.707 ETH + 2,828 USDC. At $4,000/ETH, your ETH is worth $2,828. Total position value: $2,828 + $2,828 = $5,656.
If you had simply held 1 ETH and 2,000 USDC instead of depositing: 1 ETH = $4,000 + $2,000 USDC = $6,000 total.
The LP position is worth $5,656. The hold position would be worth $6,000. The difference — $344, or about 5.7% — is the impermanent loss. You earned trading fees, but your underlying position underperformed simply holding by 5.7% of capital.
The IL Table: Know Your Exposure Before You Deposit
These are the precise impermanent loss percentages at various price change multiples for a standard 50/50 two-asset AMM pool:
| Price Change (one asset) | IL vs. Holding | In Dollars (on $10,000 position) |
|---|---|---|
| ±25% (1.25× or 0.8×) | -0.6% | -$60 |
| ±50% (1.5× or 0.67×) | -2.0% | -$200 |
| ±100% (2× or 0.5×) | -5.7% | -$570 |
| ±200% (3× or 0.33×) | -13.4% | -$1,340 |
| ±400% (5× or 0.2×) | -25.5% | -$2,550 |
| ±900% (10× or 0.1×) | -42.5% | -$4,250 |
These are the losses relative to holding — not necessarily nominal losses. If a 2× price move causes 5.7% IL but the pool earned 15% in trading fees over the same period, you still outperformed holding by 9.3%. The question is always: are the fees earned sufficient to exceed the IL cost?
Why "Impermanent" Often Becomes Permanent
IL is called impermanent because it fully reverses if price returns to the original ratio at the time of deposit. If ETH returns from $4,000 to $2,000, the pool returns to its original composition and the IL disappears completely.
The problem: in crypto, prices rarely return to exactly where they were on the relevant timescale. If ETH goes from $2,000 → $8,000 during your liquidity provision period, the IL is 25.5% relative to holding. When you withdraw (because you've decided ETH is overvalued), that loss is realised. If ETH then drops back to $2,000 six months after your withdrawal, the IL reversal happens — but you're no longer in the pool to benefit from it. The "impermanence" is a mathematical property of the position, not a guarantee of recovery for any individual LP.
Which Pool Types Carry the Highest IL Risk
Volatile/volatile uncorrelated pairs (highest risk): ETH/SOL, BTC/LINK, or any two altcoins with low price correlation. Both assets can move independently in large magnitudes, and the combined divergence creates maximum IL. During a period where one asset 5×s while the other is flat, IL approaches 25% vs. holding.
Volatile/stable pairs (moderate risk): ETH/USDC, BTC/USDT. The stable doesn't move, so all IL comes from the volatile asset. In a bull market where ETH goes 5×, IL is ~25% vs. holding. The question: would you have held ETH or USDC in your hold scenario? If you were planning to hold both anyway, the relevant comparison changes.
Correlated volatile pairs (lower risk): stETH/ETH, WBTC/ETH (more correlated than ETH/USDC). These pairs tend to move together, reducing price ratio divergence and therefore IL. The stETH/ETH pool on Curve has historically had minimal IL because stETH and ETH trade very close to parity.
Stable/stable pairs (minimal risk): USDC/USDT, USDC/DAI. Stablecoins maintain ~$1 peg; ratio rarely diverges more than 0.5%; IL is negligible. These pools earn lower fees but are appropriate for stablecoin yield with minimal IL exposure.
Evaluating Whether a Pool is Worth It
Before depositing into any LP position, estimate:
- How much price divergence do you expect between the two assets over your intended holding period?
- What is the pool's average APY from trading fees (and any additional liquidity mining rewards)?
- Does the fee APY exceed your IL estimate with enough margin to justify the smart contract and position management risk?
Use APY.Vision, Daily DeFi, or Uniswap's own analytics to check historical APY for the specific pool. Compare to the IL table above. If a pool pays 8% APY but you expect one asset to 3× relative to the other (13.4% IL), providing liquidity will underperform holding by 5.4% per year — before fees and gas costs.
Liquidity mining rewards (additional tokens paid to incentivise LPs) can dramatically change the calculus — 50%+ APY in mining rewards can easily cover IL. But mining reward tokens are often themselves volatile and declining in price, which erodes the real value of the yield. Always convert mining APY to USD terms using current token prices and consider the likelihood of reward sustainability.
Minimising Impermanent Loss
Stick to correlated or stable pairs where IL is structurally limited. The stETH/ETH pool, USDC/USDT pools, and similar "low divergence" pools offer yield with minimal IL exposure.
Use Concentrated Liquidity (Uniswap V3) with active management: V3 allows providing liquidity in a narrow price range, dramatically increasing fee income per dollar of capital. However, when price moves outside the range, the position converts to 100% of the underperforming asset — amplified IL. Active management (repositioning ranges when price drifts) is required. Protocols like Arrakis Finance and Gamma automate this.
Provide liquidity during ranging/low-volatility periods: IL accumulates fastest during strong trends. During sideways, low-volatility markets, fee income accrues while IL remains minimal — the ideal LP environment.
Use IL protection protocols: Bancor V3 historically offered IL protection after 100 days of providing liquidity. Check current availability and terms as these features vary by protocol version.
Summary
Impermanent loss is a structural, mathematical consequence of AMM design — not a risk that can be eliminated entirely in volatile asset pairs. A 2× price divergence causes ~5.7% IL; a 5× divergence causes ~25% IL relative to simply holding. IL is "impermanent" only in theory; in practice it often becomes permanent at withdrawal. Evaluate every LP opportunity by comparing expected fee income to estimated IL — only provide liquidity when fees clearly justify the IL exposure. Focus on correlated, stable, or low-divergence pairs for lower-risk yield generation in DeFi.
0 Comments
Leave a Comment
Your email won't be published. After submitting, you'll receive a quick verification email — click the link to publish your comment.