Guide

Impermanent loss explained

Impermanent loss (IL) is the hidden tax on decentralized exchange liquidity providers. When you deposit two tokens into an automated market maker (AMM) pool, the contract continuously rebalances your position as prices move — selling the token that rallies and buying the one that falls. If you had simply held the same tokens in your wallet, you would often end up with more value. The gap between “hold” and “LP” is impermanent loss. Swap fees and token incentives can offset it, but in trending markets IL routinely swamps advertised APY. This guide explains why IL happens, how to estimate it at different price ratios, when it becomes permanent, how concentrated liquidity amplifies the risk, and what mitigation strategies actually work — with context for Solana venues like Raydium and Orca. Start with our liquidity pools guide if you need AMM mechanics first.

Why pools rebalance against you

A constant-product AMM maintains x × y = k: as traders buy token A, the pool’s A reserve shrinks and B reserve grows, so the marginal price of A rises. Your LP share is a slice of both reserves. When A doubles in price versus B, the pool has automatically sold A into B along the way — you own less of the winner than you started with.

Economically, LPing a volatile pair is similar to selling rallies and buying dips on a schedule you do not control. That is profitable in a sideways range (you collect fees while prices oscillate) and painful in a one-direction trend (you systematically lighten your best performer). IL is not a bug — it is the price LPs pay for earning swap fees without posting limit orders.

The loss is called “impermanent” because if prices return to your entry ratio, the gap versus holding closes. The moment you withdraw at a different ratio, the loss is real and permanent — hence many experienced LPs prefer the term divergence loss.

The math: IL at common price ratios

For a 50/50 pool, impermanent loss versus holding depends only on how much the price ratio changed since you deposited. If r is the new price divided by the entry price, IL as a fraction of hold value is:

IL = 2 × √r / (1 + r) − 1

Intuition check: at r = 1 (no move), IL = 0%. At r = 2 (one asset doubled), IL ≈ −5.7%. At r = 4 (4× move), IL ≈ −20%. At r = 0.5 (halved), the formula is symmetric — IL is the same magnitude whether the winner is token A or B.

Worked example: you deposit $10,000 — $5,000 ETH and $5,000 USDC when ETH = $3,000. ETH rallies to $6,000 (r = 2):

Holding: 1.667 ETH × $6,000 + $5,000 USDC ≈ $15,000 (+50%).
LP position: pool rebalanced; your slice ≈ $14,142 (+41.4%) before fees.
IL: roughly $858, or ~5.7% versus hold — fees must exceed that to break even.

At r = 4 ($12,000 ETH), hold value hits ~$21,667 while the LP position lags near ~$17,333 — IL near 20%. Meme-coin pairs that 10× in a week can produce IL so large that months of fee income cannot recover it.

Fee income vs impermanent loss breakeven

LPs earn a share of swap fees proportional to pool volume and their fraction of LP supply. Breakeven requires:

Fee income > IL + gas + opportunity cost

Pool dashboards quote APY by extrapolating recent 24-hour fees. That snapshot ignores forward price trends — the main driver of IL. A useful sanity check:

Estimate IL for a plausible price move (e.g. ±50% over your holding period).
Divide by expected days in range; compare daily fee yield to daily IL drag.
Add smart contract, bridge, and token-emission risks — emissions boost APY but dilute and often end abruptly.

High-volume, low-volatility pairs (ETH/USDC, SOL/USDC on deep pools) often clear breakeven because fees are steady and IL grows slowly. Long-tail pairs (new meme / governance token vs SOL) may show 200% APY from emissions while a 3× rally produces 25%+ IL in days. Treat headline yield as marketing until you model IL under stress — the same discipline as position sizing in traditional portfolios.

Pair type changes everything

Not all pools carry equal IL risk:

Stablecoin / stablecoin (USDC/USDT) — prices stay near 1:1; IL is negligible unless a depeg event hits. Fee APY is modest but predictable. See our stablecoins guide for peg mechanics and depeg history.
Correlated assets (stETH/ETH, liquid-staking derivatives) — IL stays small while correlation holds; correlation breaks are tail risks.
Volatile / volatile (SOL/meme, ETH/alt) — maximum IL surface; only rational if you expect range-bound trading or fee tiers are extraordinary.
Single-asset exposure pools — some protocols accept one token and hedge internally; IL risk shifts to the protocol treasury, not away entirely.

Weighted pools (80/20, 60/40) change the IL curve: higher weight on the volatile asset increases directional exposure and alters the hold comparison. Always read the pool’s formula — Curve stableswap, Balancer weights, and Uniswap v3 concentrated ranges each produce different IL shapes.

Concentrated liquidity amplifies IL

Uniswap v3–style concentrated liquidity (and Solana analogs on Orca Whirlpools, Meteora DLMM, Raydium CLMM) lets LPs deposit capital only between chosen price ticks. Capital inside the range behaves like a deeper pool — higher fee APY per dollar — but the position goes 100% into one asset when price exits the range, crystallizing IL-like exposure.

Active LPs must reposition ranges as markets move; each rebalance costs gas and realizes slippage. Passive wide-range positions resemble classic pools; tight ranges around spot are a leveraged bet on low volatility. Many retail LPs discover IL amplification only after a breakout blows through their upper tick.

Rule of thumb: tighter range → higher fee capture while in range → faster conversion to single-asset inventory when price trends → worse outcome versus hold in a breakout.

Mitigation strategies (what works, what does not)

Works in practice:

LP only pairs you would hold anyway, where fee income matches your time horizon.
Prefer deep, high-volume pools on assets with sustained trading demand.
Use stable or correlated pairs when the goal is yield, not directional bet.
Hedge delta with perps or options on the volatile leg — adds cost and basis risk but caps IL on large moves.
Withdraw before known catalysts (unlocks, mainnet launches) if you cannot hedge.

Often oversold:

“IL is always covered by fees” — false in trending markets; check the math.
Chasing emission APY on thin pools — emissions end; IL does not.
Auto-compounding vaults — convenience fee plus smart contract stack risk; IL math is unchanged underneath.

On Solana, low swap fees and fast confirmation make frequent repositioning cheaper than on Ethereum L1, which tempts tighter concentrated ranges — and faster IL if you are wrong. Simulate withdrawals with current pool state before committing size; see DeFi risks for protocol-layer failure modes beyond IL.

Production checklist for liquidity providers

Model IL at ±25%, ±50%, and ±100% price moves before depositing.
Compare projected fee APY (conservative volume haircut) to worst-case IL.
Verify pool contract address, audit status, and admin keys — IL is irrelevant if the pool is drained.
Track entry price ratio; set alerts when divergence exceeds your fee cushion.
For concentrated positions, define exit rules when price nears range bounds.
Size LP exposure as a fraction of portfolio — not your entire stack on one meme pair.
Record withdrawal transactions; IL becomes permanent at exit price.

Key takeaways

Impermanent loss is the opportunity cost of LPing versus holding when the price ratio changes.
At a 2× price move, IL is ~5.7%; at 4×, ~20% — before fees.
Fees offset IL in high-volume, range-bound pools; trending pairs punish passive LPs.
Concentrated liquidity boosts fee yield but accelerates single-asset conversion on breakouts.
Withdrawal at a diverged price makes IL permanent — plan exits, do not hope for mean reversion.