Impermanent Loss Is the Price of Automated Rebalancing

On May 5, 2021, Uniswap v3 turned impermanent loss from a pool-wide exposure into a range-setting decision: it is the shortfall between an LP’s automatically rebalanced position and the value of simply holding the deposited tokens. The mechanism did not become new that day. Concentrated liquidity made it adjustable—and potentially more severe—because suppliers could commit capital to a narrow price interval rather than the entire curve.

What causes impermanent loss?

Impermanent loss occurs because an automated market maker sells the asset that is rising and accumulates the asset that is falling. In a standard 50/50 constant-product pool, traders move the reserve ratio, while arbitrageurs trade until the pool price again tracks the wider market. That keeps the venue useful without an order-book market maker, but it leaves LPs with a different token mix.

The comparison is always against a specific baseline: holding the original quantities outside the pool. If one token doubles against the other, the classic 50/50 position trails that hold strategy by about 5.7% before fees; a fourfold move produces a 20% shortfall. Those figures assume frictionless arbitrage, no fees and a full-range constant-product position. They are model outputs, not forecasts.

Concentrated liquidity raises the stakes

Uniswap v2 spread every LP’s capital across all possible prices. V3’s contracts let LPs choose a range, improving capital efficiency near the current price but increasing exposure to rebalancing within that band. If price leaves the range, the position becomes entirely one asset and stops earning fees until price returns or the LP moves the range.

That arrangement enables builders to create deeper markets with less capital and lets professional managers express a view on volatility. It also constrains passive users: a narrow range needs monitoring, repositioning costs gas, and changing the range can realize the shortfall. “Impermanent” describes the possibility that relative prices reverse; it does not mean the contracts restore lost value.

Who takes the risk and who gets paid?

The cash flows are straightforward:

• Liquidity providers supply both inventory and the pricing curve, collect trading fees, and bear divergence and smart-contract risk.
• Traders pay pool fees and accept execution prices that worsen as their orders consume available liquidity.
• Arbitrageurs spend capital and transaction fees to correct stale pool prices, keeping the AMM aligned while capturing price differences.
• Protocol stakeholders may receive a share only where the relevant contracts and governance settings activate a protocol fee.

Can fees offset impermanent loss?

Fees can offset it, but the answer depends on the path, not merely the ending price. Volume must pass through an LP’s active range, and the LP’s share of active liquidity determines the revenue. High volume with modest price movement can make the trade attractive; a sharp one-way move can leave the LP holding the weaker asset after collecting too little.

The same baseline discipline applies when comparing withdrawal routes, as our Manta Bridge review shows: cost only makes sense against a defined alternative. For LPs, that alternative is the untouched token basket.

The practical verdict is that impermanent loss is not an accidental penalty; it is the economic cost of continuously quoting both sides of a market. Fee income is compensation, not protection. What remains unknowable in advance is whether future in-range volume will be sufficient for that compensation to exceed the shortfall, gas and any management costs.