Why Impermanent Loss Occurs

"Impermanent loss" is one of those terms that sounds reassuring until you examine the mechanics. The word "impermanent" implies it eventually goes away — and sometimes it does. But the name is more optimistic than the reality warrants, and it obscures what's actually happening to your assets.

Most explanations start with the math, which makes the concept harder to grasp than it needs to be. The mechanism is straightforward once you understand what an automated market maker actually does with your assets when you deposit them.

How AMMs Rebalance Your Position

When you deposit tokens into an AMM liquidity pool — say ETH and USDC on Uniswap — you're not simply parking assets somewhere for safekeeping. You're entering a pricing system that continuously adjusts your position according to a formula.

The dominant formula in DeFi is the constant product: x × y = k, where x and y represent the pool's holdings of each token, and k is a constant that never changes. Every trade through the pool adjusts the ratio of x to y to maintain that constant. This is how the AMM prices trades without a central order book — traders push the ratio, and the formula quotes the price.

Here's what that means for a liquidity provider. As ETH becomes more valuable relative to USDC, traders buy ETH from the pool and sell USDC into it. The pool ends up with less ETH and more USDC. Your proportional share of the pool follows those same ratios — you now hold less ETH and more USDC than when you deposited.

That rebalancing is the source of impermanent loss.

The Numbers

A concrete example makes it clear. Suppose ETH is $2,000 when you deposit. You put in 1 ETH and 2,000 USDC — equal value on each side, $4,000 total.

ETH then doubles to $4,000. Arbitrageurs push the pool price to match the market rate, continuously rebalancing the pool as they trade. Once ETH reaches $4,000 in the pool, your share looks like approximately 0.707 ETH and 2,828 USDC — worth around $5,656.

If you'd held that same 1 ETH and 2,000 USDC outside the pool? $6,000.

The $344 difference is impermanent loss. Not a fee. Not a security exploit. Not a bug. It's a direct consequence of the constant product formula rebalancing your position as prices diverge.

The "impermanent" label applies to one specific scenario: if ETH drops back to exactly $2,000, the pool rebalances toward your original ratio and the divergence disappears. Loss gone — and you collected trading fees the whole time. That's the case where the name earns its meaning.

If you exit while ETH is at $4,000, the loss is realized and permanent. The trading fees you earned are an offset, but the IL is locked in. Whether fees outweigh the IL depends on trading volume, how long you held the position, and how far the price moved — all variables that differ significantly by pool.

When It's Worse, When It's Not

Impermanent loss scales with price divergence. A 2× price move produces roughly 5.7% IL. A 5× move produces around 25% IL. The wider the spread between entry price ratio and exit price ratio, the larger the loss relative to holding outright.

This is why asset pairing matters more than most liquidity providers initially realize. Stablecoin/stablecoin pools — USDC/USDT, DAI/USDC — have near-zero impermanent loss because the assets barely deviate from each other by design. Correlated pairs like stETH/ETH diverge minimally because stETH tracks ETH's price closely by construction.

The pairs with the highest IL exposure are volatile, uncorrelated tokens. Holding a position in a newer token against ETH during a sharp price run will generate significant IL — often more than the trading fees collected. This isn't a risk that goes away with better execution. It's structural.

Concentrated Liquidity Changes the Calculus

Uniswap v3, launched in 2021, introduced concentrated liquidity: instead of spreading capital across all possible prices, LPs choose a custom price range. Within that range, capital is more efficient — you earn more fees per dollar deployed because you're competing against less distributed liquidity. Outside the range, the position earns nothing and sits entirely in one asset.

The tradeoff is sharper IL exposure. A tight range near the current price earns high fees when the price stays in-range, but the rebalancing that would have occurred across a wide range gets compressed into a narrow window. When price moves through a tight range, the IL can be significant relative to the fees earned.

Active management emerged as a response. Third-party protocols — Gamma, Arrakis, and others — help liquidity providers maintain in-range positions automatically, attempting to capture fee income while managing the rebalancing cost. The tooling has matured considerably since v3 launched.

There's been some experimentation with protocol-level IL insurance too. Bancor v3 tried this — the mechanism worked in principle, but the insurance pool became stressed when token prices fell sharply and redemption demand exceeded reserves. It's an open engineering problem, not a solved one.

What Would Confirm or Break This

Confirmation signals: Active management protocols demonstrating consistent fee-over-IL surplus across full market cycles. Continued adoption of stable-pair AMM curves (Curve Finance, Balancer stable pools) for correlated assets — these use different formulas optimized to suppress IL for pairs that shouldn't diverge much.

Invalidation signals: An AMM design that eliminates the IL mechanism while preserving capital efficiency and price discovery — nobody has built this. Intent-based architectures like CoW Protocol and UniswapX reduce on-chain rebalancing by matching orders off-chain before AMM fallback. If these absorb dominant DEX volume, the economics of AMM liquidity provision shift materially.

Timing

Now: Concentrated liquidity is the default for major token pairs. If you're providing liquidity today, understanding your price range, IL exposure at various price levels, and expected fee capture rate is load-bearing information — not optional context.

Next: Active management tooling is maturing. The protocols that automate range management are worth understanding as the design space develops over the next 12–18 months.

Later: If intent-based execution takes large DEX market share, the fee economics of AMM liquidity provision change. Still a multi-year question with no clear resolution timeline.

What This Post Covers and What It Doesn't

This explains the mechanism behind impermanent loss — why it occurs structurally, when it becomes permanent, and how concentrated liquidity changes the exposure profile. It doesn't constitute a recommendation to provide or avoid liquidity in any protocol or pair.

Whether trading fees compensate for IL in a specific pool depends on variables — volatility, volume, range width, time in pool — that shift continuously. That analysis belongs in live tracking, not a static explanation.

The mechanism is the mechanism. Knowing it doesn't make it disappear.