For educational purposes only; not investment advice. Providing liquidity can lose money and may expose you to token, smart-contract, and market risks.
Direct answer
A liquidity pool is a set of assets held by a smart contract and made available under programmed rules. Automated market makers (AMMs) use pools to quote and settle swaps without matching each trade to a specific counterparty. Other DeFi protocols may pool assets for lending, liquidations, collateral, or yield strategies, so “liquidity pool” does not imply one universal formula.
Liquidity providers (LPs) deposit assets and receive an accounting claim on the pool. Depending on the design, that claim may be a fungible LP token or a non-fungible position. An LP normally withdraws its share of the pool’s current assets, not the exact quantities originally deposited.
The pool is not a guaranteed-income account. Fees can compensate LPs for making capital available, but their result also depends on trading activity, changes in the pool’s asset mix, token prices, contract rules, and exit costs.
How it works
In a full-range, two-asset constant-product AMM, reserves x and y follow an invariant based on x * y = k. Before fees, adding deltaX of token X allows the trader to receive:
deltaY = y * deltaX / (x + deltaX)
The trade increases one reserve and decreases the other. Because each additional unit moves farther along the curve, a large order relative to pool depth receives a worse average price. This is price impact. Swap fees reduce the trader’s effective input or output and are commonly credited to LPs, although exact fee allocation depends on the protocol.
Arbitrageurs trade when the pool’s reserve price differs enough from external markets. Their trades tend to realign the price, but they also change LP inventory: LPs end up with less of the asset that appreciated relative to the other and more of the asset that underperformed.
LP ownership is proportional only in fungible, full-range designs. Concentrated-liquidity systems let an LP choose a price interval. Capital can be more efficient while the market is inside that interval, but the position can become one-sided and stop earning swap fees when price moves outside it. Stable-swap and multi-asset pools use different invariants and must be evaluated under their own rules.
Example
Consider a fee-free constant-product pool with 100 ETH and 200,000 USDC, so k = 20,000,000. Its starting reserve price is 2,000 USDC/ETH. A trader adds 20,000 USDC to buy ETH.
The new USDC reserve is 220,000 USDC, so the invariant implies an ETH reserve of 20,000,000 / 220,000 = 90.9091 ETH. The trader receives about 9.0909 ETH, paying an average of about 2,200 USDC/ETH. The post-trade reserve price is about 220,000 / 90.9091 = 2,420 USDC/ETH.
Now consider the LP benchmark. An LP deposits 1 ETH and 2,000 USDC when ETH is worth 2,000 USDC. If ETH later reaches 4,000 USDC and arbitrage restores the simplified pool price, the LP position holds about 0.7071 ETH and 2,828.43 USDC, worth 5,656.85 USDC. Holding the original assets would be worth 6,000 USDC.
The LP therefore trails the matched hold portfolio by about 343.15 USDC, or 5.72%, before fees and costs. This shortfall is impermanent loss. The LP still gained value versus the initial 4,000 USDC, which shows why impermanent loss is a relative-performance measure rather than necessarily a negative account return.
Risks
- Inventory and impermanent-loss risk: arbitrage systematically changes the token mix; fee income may not offset underperformance versus holding.
- Asset and depeg risk: a pool can accumulate the weaker token as traders remove the stronger one. A stable-swap curve can offer low slippage near parity but cannot guarantee that a stablecoin remains redeemable or returns to its peg.
- Smart-contract and control risk: bugs, malicious tokens, upgrade keys, governance actions, emergency controls, compromised interfaces, or bridge failures can cause losses or block withdrawals.
- Concentrated-liquidity risk: a narrow range increases capital efficiency but can become inactive quickly, leave the LP almost entirely in one asset, and require costly management.
- Execution and cost risk: gas, price impact on entry or exit, MEV, transfer taxes, routing, and slippage can reduce realizable returns.
- Fee and incentive risk: displayed APRs are backward-looking or assumption-dependent. Volume can fall, other LPs can dilute fee share, and reward tokens can lose value.
Before depositing, verify the exact contract and token addresses, pool invariant, fee tier and fee recipient, withdrawal conditions, admin or upgrade powers, oracle dependencies, and whether any incentive is separate from swap fees.
Common misconceptions
Myth 1: LP fees are fixed income
Fees depend on volume, fee settings, active liquidity, and the LP’s share. Net performance must also include token-price changes, impermanent loss, gas, and incentives.
Myth 2: A deeper pool eliminates risk
More depth usually reduces price impact for a given trade. It does not remove token failure, contract bugs, depegs, adverse selection, or governance risk.
Myth 3: Stablecoin pools cannot lose money
Stable-swap curves are optimized for assets expected to trade near parity. If one asset depegs or becomes non-redeemable, LPs may be left concentrated in that weaker asset.
Myth 4: Withdrawal returns the original token amounts
An LP redeems the assets represented by the position at withdrawal time. Swaps, arbitrage, fees, range boundaries, and protocol rules can all change those quantities.
Related topics
Sources
- Uniswap v2 Core - Uniswap Labs (accessed: 2026-08-21)
- Uniswap v3 Core - Uniswap Labs (accessed: 2026-08-21)
- StableSwap - efficient mechanism for Stablecoin liquidity - Curve (accessed: 2026-08-21)