Constant Product vs. Concentrated Liquidity: Choosing the Right AMM Design

Constant Product vs. Concentrated Liquidity: Choosing the Right AMM Design Nov, 18 2026

You deposited your ETH and USDC into a liquidity pool, expecting passive income. Instead, you watched your balance shrink as the price of ETH swung wildly, only to realize that the fees you earned barely covered the loss from selling low and buying high. This isn't bad luck; it’s a feature of how Automated Market Makers (AMMs) are designed. For years, the standard was simple: spread your money across every possible price. Then came a new approach: concentrate your money where the action is. But does concentration actually make you more money, or just complicate your life?

Quick Summary

  • Constant Product AMMs (like Uniswap v2) offer simplicity and always-on liquidity but suffer from low capital efficiency.
  • Concentrated Liquidity AMMs (like Uniswap v3) boost capital efficiency by up to 4,000x in narrow ranges but require active management and carry higher risk.
  • Impermanent Loss is not eliminated by concentration; it often accelerates because positions convert fully to one asset when prices move out of range.
  • Use CP-AMMs for long-tail tokens and passive strategies where you can’t monitor prices daily.
  • Use CLAMMs for stable pairs, major assets, and if you have tools or time to rebalance positions frequently.

The Constant Product Model: Simple but Inefficient

Think of a Constant Product AMM as a bucket with two types of water-say, Ethereum and Dollar tokens. The rule is strict: the amount of Ethereum multiplied by the amount of Dollars must always equal a fixed number, k. If someone buys Ethereum, they remove some ETH and add some USD. To keep k constant, the price of ETH rises automatically. This is the x·y=k invariant that powered Uniswap v2.

The beauty of this design is its robustness. It works at any price, from $1 to $1 million. You don’t need to predict where the market will go. However, this flexibility comes at a cost. Most of your deposited capital sits idle at prices the market rarely visits. If you provide liquidity for ETH/USDC, half your funds might be supporting trades at $5,000 ETH while the current price is $3,000. That capital is doing nothing. It’s paying no fees. It’s just sitting there, exposed to volatility but contributing little to execution depth near the current price.

For traders, this means higher slippage on large trades because the liquidity is thin near the active price. For providers, it means you’re effectively over-capitalized for the actual trading activity occurring in the pool. It’s like stocking a grocery store with 10,000 bananas when you only sell 100 a day. The inventory is safe, but the return on investment is diluted.

Colorful geometric band representing concentrated liquidity density

Enter Concentrated Liquidity: Precision Over Simplicity

Uniswap v3 changed the game by letting you choose where your money works. Instead of spreading liquidity from zero to infinity, you pick a specific price range, say $2,900 to $3,100. Your capital is compressed into this band. Inside this range, the pool still follows the constant product formula, but locally. Outside this range, your position holds only one token.

This compression creates massive capital efficiency. Uniswap’s whitepaper claimed up to 4,000x efficiency compared to v2 when concentrating in a tight 0.1% range. Why? Because all your dollars are fighting for the same trades happening right now. A trader swapping $10,000 worth of ETH hits a wall of liquidity instead of slipping through a thin market. For professional market makers, this is a dream. You earn significantly more fees per dollar deployed because your liquidity is actually being used.

But here is the catch: precision requires maintenance. If the price of ETH drops to $2,800, your position is now entirely in USDC. You stopped earning fees the moment the price crossed your lower boundary. You didn’t lose your principal value in terms of token count, but you missed out on the upside recovery if you didn’t rebalance. And rebalancing costs gas fees and potentially realizes losses.

Stylized scales tipping to show impermanent loss risk

The Impermanent Loss Reality Check

Everyone talks about impermanent loss (IL) in DeFi, but few explain how AMM design changes its impact. IL is simply the difference between holding your tokens and providing them to a pool. In a CP-AMM, IL grows gradually as the price diverges. You always hold a mix of both assets. Even if the price doubles, you still hold some of the original asset.

In a CLAMM, the dynamics are sharper. If you set a narrow range and the price moves out of it, your position becomes 100% one asset. You have essentially sold everything on the way down (or bought everything on the way up). When the price returns, you haven’t bought back in unless you manually reposition. Academic studies, such as those by Loesch et al., suggest that while concentrated liquidity can generate higher fee revenue, it also exposes LPs to faster and more severe divergence losses during volatile periods. The