Deconstructing LP Payoffs: Fee Accrual vs. Inventory Drag in AMM Mechanics

Liquidity provision on Uniswap is often described as fee yield. At Haven Capital Advisors, we approach it as a dynamic inventory and execution process.

Uniswap is an automated market maker, or AMM: instead of matching buyers and sellers through an order book, it allows traders to swap against pooled assets priced by a programmatic curve. Liquidity providers supply those assets and earn a share of swap fees, but their inventory changes as prices move.

That trade-off became more pronounced with concentrated liquidity. Uniswap v2 distributes liquidity across the full price curve. Uniswap v3 lets a provider concentrate capital within a chosen price range, increasing its active presence near the current market price while the range holds.

The trade-off is straightforward: once price moves outside that range, the position is held entirely in one asset and no longer earns fees until price returns or the range is changed.

For capital allocators, the question is not simply what fees a pool generates. It is whether those fees justify the resulting inventory exposure, time out of range, and the costs of managing the position.

This article sets out a framework for assessing LP positions: measuring fees against inventory divergence, range inactivity, and execution costs relative to a defined benchmark.

The Core Analytical Framework: Deconstructing Net Return

The net performance of an LP position relative to a holding benchmark can be formalized through a straightforward decomposition:

Net LP Return = Realized Fee Accrual (γ) - Path-Dependent Divergence Loss (ΔIL) - Friction & Execution Costs

Where:

• Realized Fee Accrual (γ): The gross swap fees accumulated by the position as trading volume crosses the pool's price curve.
• Divergence Loss (ΔIL): The opportunity cost and inventory drag incurred as the pool's internal price ratio diverges from the initial entry price.
• Friction & Execution Costs: The operational drag resulting from network gas fees during position creation, rebalancing, and withdrawal, alongside execution slippage and maximal extractable value (MEV) leakage.

Understanding how these three variables interact requires examining the transition from constant product curves to concentrated liquidity mechanics.

From Constant Product (x · y = k) to Concentrated Liquidity

In a constant-product AMM such as Uniswap v2, liquidity is distributed across the full theoretical price curve—from zero to infinity—according to the invariant x · y = k, where x and y represent the pool’s token reserves. [1]

That design makes liquidity continuously available across all prices. It also means that a meaningful portion of an LP’s capital sits far from the current market price, where it contributes less to near-term trading depth.

Uniswap v3 introduced concentrated liquidity. Rather than supplying liquidity across every possible price, an LP can select the price range in which its capital will be active. [2][3]

Capital efficiency

For a fixed amount of capital, concentrating liquidity closer to the current market price can provide more active trading depth around that price than spreading the same capital across the entire curve. In this context, capital efficiency describes where in the price curve capital is placed and how much is available to support trades while price remains within the selected range.

This is not to say that this method must produce superior returns. Concentration of liquidity ranges alone does not establish higher fee income, superior risk-adjusted outcomes, or the value of active management after costs.

The trade-off

A narrower range places more capital near the current price but covers fewer possible price outcomes. If price moves outside that interval, the position becomes concentrated in one asset and stops earning swap fees until price returns or the LP changes the range. A broader range covers more price outcomes, but spreads the same capital more thinly at any given price, thus earning less in fees. [2][3]

For an allocator, the relevant question is therefore not whether concentration is inherently better, but whether the selected range, expected active time, inventory behavior, and operating costs are appropriate for the market being evaluated.

Uniswap v2: Full distribution

Capital placement

Across the full theoretical price curve (0 to ∞).

Mechanical effect

Capital is available across every price level, but the same notional values may be less concentrated near the current trading area.

Uniswap v3: Targeted distribution

Capital placement

Inside a selected price interval [P_lower, P_upper].

Mechanical effect

More of the same value is concentrated near the current trading area, while the selected range remains active.

The Dynamics of Inventory Drag and Divergence Loss

Divergence loss is an inherent feature of constant product pricing. When the external reference price of an asset changes, arbitrageurs execute trades against the AMM pool until its internal quote mirrors the broader market. In doing so, the AMM automatically sells the appreciating asset and accumulates the depreciating one.

For a standard constant product pool, the divergence loss ratio relative to holding the underlying assets 50/50 is defined by:

IL(k_p) = (2 · √k_p) / (1 + k_p) - 1

Where k_p = P_current / P_entry represents the relative price ratio change. For example, a 2x price shift results in an unhedged divergence loss of ~5.7% relative to holding. [1]

Concentrated liquidity changes the LP’s exposure to price movement within a chosen interval. Tighter ranges can increase capital efficiency while active, but they also make out-of-range inactivity and range-management decisions more consequential. [2][3]

Once price moves outside a position’s selected range, the position becomes entirely concentrated in one asset and stops earning swap fees until price returns to the range or the LP changes the position. Which asset is held depends on price direction and token ordering. [2][3]

The final outcome still depends on the price path, the starting and ending inventory, trading volume, and all implementation costs.

Market structure and adverse selection. Swap volume alone does not determine LP outcomes. During rapid price moves, arbitrage helps align AMM prices with external markets, but it can also impose costs on passive liquidity when pool prices lag external reference prices. Fees may compensate for some of that exposure, but the relationship depends on the market, fee tier, volatility, and implementation.

Options Payoff Analogy and Its Structural Limits

Financial analysts may compare concentrated positions to short volatility strategies, such as writing covered calls or cash-secured puts. While this comparison offers intuitive value—both collect cash flow in exchange for taking on asymmetric directional risk—the analogy has critical structural limits:

• No Upfront Option Premium: An options writer collects a fixed, upfront premium at trade inception. An LP collects no upfront premium; fee income accumulates variably over time, dependent on realized trade volume and tick-range structure. The structure may be viewed, however, as simply a variable streaming premium, which offsets some limitation.
• No Fixed Expiration: Standard options contracts have defined expiration dates where payoffs settle. LP positions are open-ended perpetual commitments until manually closed or rebalanced.
• Forced Continuous Rebalancing: An options writer decides what to hedge and executes their hedges at the market. An LP's inventory is determined by, and rotates continuously along the price curve—selling the appreciating asset, accumulating the depreciating one—at pool quotes against arbitrageurs. This execution drag (loss vs rebalance, LVR) is the structural cost fees must overcome, and it cannot be paused while the position is open.

Operational Frictions: Gas Drag, Slippage, and Rebalancing Costs

LPs can reposition concentrated-liquidity ranges to respond to inventory exposure or out-of-range inactivity. That process is not friction or cost-free.

Repositioning requirements. Adjusting a position will usually require withdrawing liquidity, exchanging inventory, and launching a new range. The specific process and cost depend on the strategy, transaction design, and market conditions.

Realized divergence effects. Repositioning can crystallize divergence loss relative to a holding benchmark—rebalancing resets the inventory ladder, so in a trend the position realizes a shortfall in the amount of the outperforming asset, rather than allowing price retracements to recapture the losses. The economic effect depends on the position, price path, and the trades that affected it.

Gas, slippage, and adverse execution. Rebalancing will face gas costs as well as potential slippage and adverse swap execution costs opposing arbitrageurs (LVR). The significance of those costs depends on transaction design, liquidity conditions, and network conditions, and how well the LP pre-structures inventory for fast operations.

Conclusion

The analytical framework begins with a clear benchmark. From there, an LP position’s net result can be separated into fees earned, inventory divergence, time out of range, and execution costs.

For a full-range v2 position, the main question is whether fee income offsets the inventory effects of continuous rebalancing. For a concentrated v3 position, range selection and repositioning add further variables: capital can be more concentrated near the active market price, but can also become concentrated in one asset and stop earning fees when price leaves the selected range.

For capital allocators, this turns liquidity provision from a yield screen into an underwriting decision. The question is not which pool displays the highest fee APY. It is what inventory exposure, operational burden, and implementation risk the strategy requires—and whether the resulting net return justifies them relative to simpler alternatives.

That is also how a liquidity manager should be evaluated: by the benchmarks it sets, the inventory and range risks it accepts, the controls it applies, and the clarity of its net-performance reporting.

At Haven, this framework informs how we evaluate pools, set and monitor ranges, manage protocol concentration, and assess whether repositioning is justified after costs.

Fee income is an important input but it is not the whole investment case.

Haven Capital Advisors, LLC is a North Carolina-registered investment adviser. Registration as an investment adviser does not imply a certain level of skill or training. This material is provided for informational and educational purposes only. It does not constitute investment, legal, or tax advice, and it is not an offer to sell or a solicitation of an offer to buy any security or investment product. Digital asset strategies involve substantial risk, including the risk of total loss of capital. Liquidity provision exposes capital to price divergence, execution and network costs, and protocol risk, and any income generated is variable and not guaranteed. This material discusses legal and regulatory concepts for informational purposes only; it does not constitute legal analysis and should not be relied upon as such. The views expressed are as of the date of publication and are subject to change.