Uniswap v3 launched concentrated liquidity in May 2021 with a compelling pitch: instead of spreading capital across an infinite price curve, LPs could focus their liquidity within a defined range and earn far more fees per dollar deployed. Capital efficiency improved by orders of magnitude on paper. In practice, a significant portion of retail LPs have been quietly losing money relative to simply holding the underlying assets.
The mechanism is clean in theory. An LP providing liquidity to an ETH/USDC pool between $2,800 and $3,200 earns fees only when ETH trades within that band. Outside it, their position sits idle - fully converted to one asset at the range boundary, earning nothing while the market moves. The tighter the range, the higher the fee yield when in-range, and the higher the impermanent loss (IL) exposure when the price exits.
What makes this genuinely tricky is that IL in concentrated positions isn’t just larger - it’s asymmetric in ways passive holders don’t intuitively model. When ETH drops from $3,000 to $2,600 and an LP’s range was $2,800–$3,200, that LP exits the range fully denominated in ETH just as ETH is falling. They’ve bought the dip automatically, without deciding to, and now hold a depreciating asset with no fee income to offset it. Rebalancing into a new range means realising that loss and paying gas.
The Fee Math Doesn’t Always Save You
Fee income can offset IL - that’s the argument LPs make for staying in tight ranges on volatile pairs. But the fee APR figures displayed on most front-ends are calculated from recent volume and don’t account for out-of-range time. A position that’s in-range for 40% of a two-week period earns 40% of the displayed APR, not the headline number. On ETH/USDC with significant daily volatility, that’s not an edge case.
Professional market makers using automated rebalancing strategies - Arrakis, Gamma Strategies, and similar protocols - can manage this more systematically. They’re adjusting ranges algorithmically, hedging delta exposure off-chain, and absorbing gas costs across larger capital bases. Retail LPs competing in the same pools are doing none of that.
Who It Actually Works For
Concentrated liquidity is well-suited to stable or correlated pairs - USDC/USDT, wstETH/ETH - where price rarely leaves a narrow band and IL stays minimal. The fee yields on those pairs are lower, but the position behaves more like a savings account than a volatility trade.
For volatile pairs, it’s closer to writing a covered call without the premium explicitly priced in. The fees are the premium. Whether that premium is sufficient depends on realised volatility over the position’s life - which nobody knows in advance.
The honest version of concentrated liquidity isn’t that it democratised sophisticated market-making. It’s that it gave retail LPs access to a strategy that sophisticated players execute better, in markets where sophistication increasingly determines who captures the spread.