Why Uniswap V2 Pools Still Exist: When to Use Legacy Liquidity vs. V3 Concentrated Positions

A liquidity provider reviewing their portfolio might ask why Uniswap V3 exists if it offers superior capital efficiency. The answer is more practical than theoretical: V2 and V3 pools coexist because they serve different risk profiles, market conditions, and operational constraints. V3’s concentrated liquidity mechanism can deliver up to 4000x capital efficiency compared to V2’s uniform distribution, but that concentration comes with management overhead, impermanent loss exposure, and the need for active monitoring. For some pairs and some liquidity providers, V2’s simpler model remains the better choice despite its lower efficiency metrics.

Understanding when to deploy capital in each version requires clarity on what capital efficiency actually means and what it costs. An LP who achieves 10x efficiency but experiences severe impermanent loss may have worse returns than one who accepted 1x efficiency but faced a narrower price range and stable conditions. The two versions are not a hierarchy where one is universally superior. They are distinct tools that produce different outcomes under different circumstances, and the choice between them shapes risk exposure, gas costs, and realized returns.

The capital efficiency paradox: why V3 efficiency does not always equal returns

Uniswap V3’s concentrated liquidity feature allows an LP to provide liquidity within a specific price range rather than across the entire 0 to infinity spectrum. This concentration multiplies the capital used in a narrower band, producing up to 4000x efficiency depending on the range width. A 1% width around the current price—a very tight concentration—can theoretically amplify a capital unit by 4000. That sounds like an unconditional improvement, and in controlled backtests it often appears that way. Real market conditions are less cooperative.

The capital efficiency gain assumes that the price stays within the chosen range. When it does, an LP earns fees on capital deployed efficiently. When the price moves outside the range, the position becomes entirely composed of one asset—often the less desirable one. At that point, the LP has become a hodler of the less-concentrated asset without earning fees, while watching the price potentially move further away. This is impermanent loss in its sharpest form. A V2 position, by contrast, automatically rebalances across the full range. It never stops earning fees and never becomes 100% of one asset, though it earns fees on less efficiently deployed capital.

The practical question is whether fee income from tighter concentration exceeds the cost of impermanent loss and position maintenance. For a highly volatile pair, such as an altcoin/ETH or ETH/USDC during market stress, the math often breaks in favor of V2. The LP keeps earning fees even if the price swings, and the simpler mechanics mean less attention required. For a stable pair, such as USDC/USDT or a similar synthetic stablecoin pair where price movement is measured in basis points, V3 concentration makes sense because the probability of moving outside the range is low and the fee income on capital is high.

A common mistake is to treat “higher efficiency” as “more money.” An LP who deploys $10,000 in a 0.5% V3 range on an altcoin pair and has the price move 2% in one day may earn excellent fees but suffer 5% impermanent loss, resulting in a net loss despite high capital efficiency. An LP who deployed the same $10,000 in V2 and accepted the lower efficiency may have earned modest fees but lost nothing to price movement because the position remained balanced. The V2 LP lost opportunity cost—they could have had more capital working—but they did not lose capital itself.

Fee tiers and liquidity depth: why V2 remains attractive for some pairs

Uniswap V3 introduced multiple fee tiers: 0.01%, 0.05%, 0.30%, 1.00%, and 10.00%. This granularity allows pairs to establish different markets. A stablecoin pair might use the 0.01% tier because price movement is minimal and traders are extremely price-sensitive. An exotic pair might use 1% because liquidity is sparse and the risk is higher. The fee structure also changes LP incentives: a 0.01% tier generates very tight margins unless capital is extraordinarily efficient, while a 1% tier can sustain less efficient but more stable LP returns.

V2 pools have a fixed 0.30% fee, period. This simplicity is both a limitation and a feature. For many pairs, 0.30% is exactly right: not so high that slippage becomes a barrier, not so low that LPs struggle to justify the capital. For volatile pairs where impermanent loss is already a drag, the fee layer is relatively thin; the 0.30% earned may not offset price movement. For stable pairs, 0.30% may be higher than necessary, and traders might prefer a lower fee if available. Yet because the fee is uniform, V2 pools often contain a critical mass of liquidity for mainstream pairs, particularly on Ethereum mainnet. That liquidity depth itself becomes a feature: deep liquidity means tighter spreads and lower slippage for traders, which in turn attracts more trading volume and feeds more fee income to LPs.

The liquidity depth question is crucial. A V3 0.01% tier on USDC/USDT might have substantially less total liquidity than a V2 USDC/USDT pool because LPs know that 0.01% fees do not compensate for the effort of managing concentrated ranges. Conversely, a V3 1.00% tier on an experimental token pair might have more liquidity than a V2 version of the same pair because LPs are willing to concentrate capital in exchange for higher fees. An LP choosing between versions cannot optimize capital efficiency in isolation; they must consider where other LPs have already deployed their capital and what the resulting liquidity distribution means for their own position.

Position management and operational overhead in V3

A V2 position is passive once created. An LP deposits a pair of assets in the ratio that maintains the constant product formula (x*y=k), and the pool rebalances automatically as trades occur. The LP earns fees proportional to their share of the pool. That is the complete list of decisions: how much capital, which pair, and when to withdraw. Monitoring is optional.

A V3 position requires decisions before creation and ongoing attention afterward. The LP must choose the price range—decisions on both lower and upper bounds. This requires a view on where the price is likely to go and for how long. If the view is wrong, the position becomes inactive and stops earning fees. The LP must also choose between active management (rebalancing the range as the price moves) or passive management (accepting that the position will eventually be out of range). Active management means paying gas fees to periodically adjust the range, harvest fees, and rebalance. Passive management means accepting concentrated exposure to one asset when the price moves far enough.

For a mainstream pair on a low-fee chain such as Arbitrum or Optimism, gas costs for rebalancing a V3 position might be $10–50 per rebalance depending on network congestion. For a volatile pair, rebalancing might be necessary weekly or even daily. That $50 per rebalance compounds quickly and becomes a real drag on returns, especially for smaller LPs. A V2 position avoids this overhead entirely. The LP pays gas once to deposit and once to withdraw. The protocol handles all rebalancing automatically.

For institutional LPs or strategies running on mainnet Ethereum where a single gas transaction can cost $100–500, the overhead calculation shifts. In that regime, the question becomes whether the superior capital efficiency of V3 justifies the transaction costs and the management complexity. For a large position, the answer is often yes. For a retail LP with $5,000–50,000 deployed, V2 or a passively managed V3 position might be more appropriate than active rebalancing.

Uniswap V3 concentrated liquidity in practice: when tight ranges work

A USDC/USDT pair trading near price parity is an ideal V3 use case. USDC and USDT are synthetic stablecoins; their exchange rate should be 1:1 or very close to it. Price movements are usually in basis points. An LP might establish a V3 position with a range from $0.999 to $1.001—a 0.2% band around parity. If the price stays in that range (which it usually does), the capital in that narrow band earns fees on nearly 100% of the trading volume. The same capital in V2 would earn fees on the same volume but would be “wasted” on price ranges from $0 to $0.999 and $1.001 to infinity, where no trades occur. The concentration directly improves returns without adding real risk because the price rarely moves that much.

Similarly, a pair like ETH/USDC might support a tight V3 range if the LP has strong conviction about a near-term price. If ETH is trading at $2,500 and the LP expects consolidation between $2,400 and $2,600, they might establish a V3 position with that range. If price stays in the range, they earn concentrated fees. If price breaks the range, they have impermanent loss, but they also have a signal to close the position or adjust. The key is that the range is intentional and tied to a view on price behavior, not arbitrary.

A volatile altcoin pair presents the opposite scenario. If an asset is trading at $10 but could reasonably move to $8 or $12 within a month, a V3 range of $9.50 to $10.50 is too tight. The probability of being out of range is high, and when that happens, the concentrated position becomes a pure directional bet rather than a liquidity-earning position. A V2 pool, or a much wider V3 range, becomes more appropriate because it continues to earn fees regardless of price movement.

Flash swaps, arbitrage, and the ecosystem around each version

Uniswap V2 and V3 have slightly different supporting ecosystems due to how they integrate with external protocols. V2’s simpler mechanics made it the first target for aggregators, flash swaps, and complex arbitrage strategies. Flash swaps on V2 allow a user to borrow tokens from the pool, use them elsewhere, and repay them with a 0.30% fee—all within a single transaction. This pattern enabled MEV searchers and algorithmic traders to develop sophisticated strategies around V2 pools.

V3 supports the same flash swap pattern but with more precision. Because V3 allows multiple fee tiers on the same pair, a searcher can execute more granular arbitrage between, say, a 0.05% USDC/USDT V3 pair and a 0.30% USDC/USDT V3 pair, or between a V2 USDC/USDT pool and a V3 pool. This expanded design space has attracted institutional market makers and quant strategies to V3. However, the added complexity means that V2 remains simpler for basic use cases and remains the reference standard for many analyses and tools.

Importantly, a uniswap trader or LP does not need to choose between the versions for each transaction. The Uniswap router and aggregators abstract away the version difference: a user can send a swap request and receive quotes from both V2 and V3 pools, with the best price automatically selected. From the end user’s perspective, Uniswap is a unified platform. The version distinction matters most to liquidity providers and to those building on top of the protocol.

Governance, fee capture, and long-term incentives on V2 versus V3

The Uniswap DAO governs both V2 and V3, but the governance structures create different incentive alignments. V2 was deployed with governance disabled; the 0.30% fee entirely funds LP rewards. This simplicity is both a strength and a limitation: there is no mechanism to capture protocol fees for treasury or governance rewards. V3 introduced a fee switch that allows governance to capture a portion of fees (up to 50%) for the DAO treasury. This fee capture has not yet been activated, but the capability exists.

From an LP perspective, this means V3 positions carry the potential risk that governance will vote to divert some of the fee income to the DAO. A V2 LP has no such risk; the 0.30% stays with LPs entirely. For long-term LPs, this is a material consideration. If the DAO votes to capture 50% of V3 fees, a V3 0.30% tier would effectively become 0.15% to LPs—equivalent to V2 but with the added complexity of concentration. That scenario has not occurred, but the possibility exists in the governance structure.

Conversely, governance could allocate UNI token incentives to V3 positions to encourage migration. The Uniswap DAO has experimented with liquidity mining on specific V3 positions to boost capital deployment in priority pairs. A V2 position does not qualify for these incentive programs because the governance mechanism is not available on V2. An LP chasing governance incentives needs to use V3.

Choosing the right version for different LP strategies

An LP managing a stablecoin market-making strategy should strongly consider V3 with tight ranges. The risk profile is low, the fee tier can be optimized downward without hurting returns, and capital efficiency directly translates to higher yields. A 0.01% V3 USDC/USDT position with a 0.1% range can generate 50%+ annual yield if deployed efficiently, while a V2 position would generate far less because the capital would be wasted on out-of-the-money ranges.

An LP in a volatile altcoin pair should default to V2 or accept that V3 requires active management. A passive V3 position in a volatile market will regularly move out of range and stop earning fees. That is a real cost. The LP can mitigate it by choosing a much wider range—say, $8 to $12 on a $10 asset—but at that width, the capital efficiency gain diminishes rapidly. They might as well use V2 and save the operational overhead.

An LP with small capital ($1,000–10,000) should factor gas costs heavily. On Ethereum mainnet, creating a V3 position might cost $100–300 in gas. If the position is managed actively, each rebalance adds another $100–300. A V2 position costs $100–300 to create and then requires no further action. For small capital, V2 often has better realized returns because the gas drag is proportional to the capital and hits smaller positions harder.

An LP with large capital ($500,000+) and a professional setup should evaluate V3 concentrated liquidity seriously. At scale, active management becomes economical because the fee income is large enough to justify the operational cost. The capital efficiency gains compound across multiple positions. Institutional LPs running market-making operations on Uniswap typically use V3 because the precision, fee tiers, and integration with risk management systems matter at that scale.

The practical reality: V2 and V3 will coexist indefinitely

Uniswap’s design allows both V2 and V3 to exist simultaneously, and there is no migration pressure. V2 pools do not need to be retired or deprecated. The Uniswap DAO does not benefit from eliminating V2; in fact, many traders and LPs rely on V2’s simplicity. Removing V2 would break existing positions and strategies, creating unnecessary friction. Instead, the protocol supports both versions, and market forces determine how capital flows between them.

As of 2024, V3 has captured the majority of TVL (total value locked) on Uniswap across most pairs, particularly on Layer 2 networks where gas is cheaper and the active management overhead is less of a barrier. V2 still holds significant liquidity on Ethereum mainnet for major pairs, partly due to historical path dependence and partly because V2 genuinely makes sense for those LPs. On other blockchains like BNB Chain and Polygon, V2 pools sometimes have more total liquidity than V3 for the same pair, reflecting the regional preferences of LPs and the different cost structures of those networks.

The coexistence also reflects a deeper truth about market design: efficiency is not the only variable that matters. Simplicity, cost, predictability, and the ability to passive-invest without constant monitoring have real value. A V2 LP who earns 12% annually with minimal effort may be better off than a V3 LP who attempts 30% annually but experiences volatility, impermanent loss, and operational stress. The choice depends on the LP’s capital size, time horizon, risk appetite, and willingness to engage with the protocol actively.