Impermanent Loss on Solana DEXs: AMM, CLMM & DLMM Explained
What Is Impermanent Loss?
Impermanent loss is the cost of being a liquidity provider. It’s the difference in value between holding tokens in your wallet and depositing them into a liquidity pool. Whenever the price ratio of pooled tokens changes from your deposit ratio, you end up with less value than if you’d done nothing.
Here’s the simplest example: You deposit $500 worth of SOL and $500 of USDC into a SOL/USDC pool — $1,000 total. SOL doubles in price. Your pool position automatically rebalances, selling SOL for USDC as the price rises (that’s how AMMs work — they continuously rebalance). When you withdraw, you get more USDC and less SOL than you deposited.
Your pool position might be worth $1,414. But if you’d just held $500 SOL (now worth $1,000) and $500 USDC in your wallet, you’d have $1,500. The $86 difference is impermanent loss. The pool made you sell SOL on the way up — exactly the opposite of what you wanted.
Why “Impermanent”?
The loss is called impermanent because it only exists as long as prices have diverged. If SOL’s price returns to its original value, the pool rebalances back, and your IL goes to zero. You’re whole again.
But this framing is misleading. In practice, the loss becomes permanent the moment you withdraw while prices have diverged. And prices rarely return to the exact same ratio — crypto trends, it doesn’t oscillate neatly back to entry prices. Most LPs withdraw during divergence, which means most LPs realize their impermanent loss as an actual, permanent loss.
The LP Equation
The core equation every LP should internalize:
Net LP Return = Fee Income – Impermanent Loss
Impermanent loss is the cost. Fee income is the revenue. If fees exceed IL, you profit. If IL exceeds fees, you would have been better off just holding. Every LP decision comes down to this math.
IL on Standard AMM Pools
Standard automated market maker (AMM) pools — like Raydium’s legacy constant-product pools — use the formula x × y = k, where x and y are the quantities of each token and k is a constant. This formula forces the pool to always hold both tokens, rebalancing continuously as prices change.
The IL math on constant-product pools is well-defined and predictable. Here’s what you lose at common price changes:
| Price Change | Impermanent Loss | Value vs. Holding |
|---|---|---|
| ±10% | 0.11% | 99.89% |
| ±25% | 0.60% | 99.40% |
| ±50% | 2.02% | 97.98% |
| ±75% | 3.79% | 96.21% |
| ±100% (2x) | 5.72% | 94.28% |
| ±200% (3x) | 13.40% | 86.60% |
| ±400% (5x) | 25.46% | 74.54% |
A few things jump out from this table:
IL is symmetric. A 50% drop and a 50% rise produce the same IL (2.02%). It doesn’t matter which direction price moves — any divergence costs you.
IL accelerates with larger moves. A 25% move costs only 0.6%, but a 100% move (2x) costs 5.72% — nearly 10x more IL for 4x more price movement. The relationship is non-linear.
Even large moves have bounded IL. A 5x price move — the kind of rally that makes crypto headlines — costs 25.46% in IL on a standard AMM. That’s significant but not catastrophic. Standard AMMs have a built-in dampening effect.
The problem: most standard AMM pools on Solana generate 5–15% APR in fees on major pairs. If SOL moves 50% in a month (which happens regularly), you lose 2% to IL — potentially wiping out a month or two of fee income. On volatile pairs like memecoins, where 5x moves happen in days, IL routinely exceeds fees.
IL on Concentrated Liquidity (CLMM/DLMM)
Concentrated liquidity changed the math dramatically. Instead of providing liquidity across all possible prices (from $0 to infinity), you concentrate it within a specific range. This means your capital is used more efficiently — you earn more fees per dollar — but the tradeoff is amplified impermanent loss.
All the major Solana DEXs now use concentrated liquidity: Orca Whirlpools, Raydium CLMM, and Meteora DLMM. Each implements it slightly differently, but the IL dynamics are fundamentally the same.
How Concentration Amplifies IL
Think of it this way: in a standard AMM, your liquidity is spread across the entire price range. If price moves 10%, only a small slice of your liquidity is actively affected. In a concentrated position with a ±10% range, ALL of your liquidity is active across that range. You’re effectively providing 10x more liquidity in that range with the same capital.
The amplification cuts both ways. If you provide liquidity in a ±5% range around SOL’s current price:
- Fee income: You earn roughly 10–20x more fees per dollar of capital compared to a full-range position (because your capital is concentrated where trading happens).
- IL: If price exits your range, you suffer roughly 10–20x more IL per dollar of capital compared to a full-range position.
What Happens When Price Exits Your Range
This is the critical mechanic that catches new CLMM LPs off guard. When SOL’s price moves above your range, the pool has sold all your SOL for USDC — you hold 100% USDC. When price moves below your range, you hold 100% SOL (the pool bought SOL with all your USDC as price fell).
Either way, you’re holding 100% of the token that moved in the wrong direction. And your position earns zero fees while out of range — you get the worst of both worlds: maximum IL and no fee income.
Consider a concrete scenario: You LP SOL/USDC with a ±5% range at a $150 SOL price (range: $142.50–$157.50). SOL drops to $130.
Your position is now 100% SOL, worth roughly $130 per SOL instead of $150. On a $10,000 position, you might have $8,667 in SOL. If you’d simply held your original 50/50 split ($5,000 SOL + $5,000 USDC), you’d have $4,333 SOL + $5,000 USDC = $9,333. Your IL on this concentrated position: roughly $667, or 6.7%. On a full-range AMM, the same price move (13.3% drop) would cause only about 0.2% IL. Concentration amplified IL by over 30x.
Meteora DLMM: A Different Approach
Meteora’s Dynamic Liquidity Market Maker (DLMM) uses discrete price bins instead of continuous price curves. Each bin represents a specific price point, and you can distribute liquidity across bins in custom shapes — uniform, concentrated around current price, or skewed in one direction.
The key difference: Meteora’s dynamic fees adjust automatically based on volatility. When markets are volatile (when IL is highest), fees increase to partially compensate. During calm markets (when IL is low), fees normalize. This doesn’t eliminate IL — the underlying math is the same — but it helps the fee side of the equation keep up with the IL side during volatile periods.
In practice, Meteora DLMM positions tend to slightly outperform equivalent Orca and Raydium CLMM positions during volatile markets because of this fee adjustment. During stable markets, performance is similar across platforms.
When Do Fees Beat IL?
The fundamental question: can you earn more in fees than you lose to IL? The answer depends on three factors — volume, range width, and price movement.
The Break-Even Calculation
For any LP position, you can estimate the break-even point:
Daily fee income = (your liquidity ÷ total pool liquidity) × daily volume × fee tier
Daily IL cost = depends on how much price moves in a day
If SOL/USDC does $50M daily volume on Orca at a 0.30% fee tier, and the pool has $10M liquidity, total daily fees are $150,000. If you provide $10,000 (0.1% of pool liquidity), you earn $150/day. At that rate, you can sustain up to ~5.5% monthly IL and still break even.
But volume fluctuates. During bull markets, SOL/USDC volume might hit $200M daily. During quiet weeks, it might drop to $10M. Your fee income moves proportionally, but IL during quiet-to-volatile transitions can spike.
High APY Displays Are Misleading
When you see a pool advertising 200% APY on a Solana DEX, read the fine print:
- Is the yield from fees or token emissions? Many pools incentivize LPs with protocol token emissions (RAY, ORCA, MNDE). These emissions dilute over time, and the token itself can lose value. Fee yield is organic and sustainable; emission yield is a subsidy that eventually ends.
- Is the APY calculated on a 24-hour snapshot? A pool that did high volume in one day gets an annualized APY that looks enormous. Tomorrow’s volume might be 10% of that.
- Does the APY account for IL? Almost never. The displayed APY is gross fee income. Your actual return is fees minus IL. A pool showing 50% APY might deliver 10% actual returns after IL — or negative returns if price moves sharply.
Correlated Pairs: Where IL Is Minimal
Pairs where both tokens move together have minimal IL because the price ratio doesn’t diverge much. On Solana, the best examples:
JitoSOL/SOL: Both tokens track SOL’s price. JitoSOL slowly appreciates against SOL by the staking yield (~7% annually), so the price ratio drifts by less than 0.02% daily. IL on this pair is negligible — fractions of a percent annually. Fee income almost always exceeds IL.
USDC/USDT: Both track $1.00. The ratio rarely deviates more than 0.1%. IL is essentially zero. Fee income is low (tight spreads, small fee tiers), but it’s nearly pure profit.
These are the pairs where passive LPing genuinely works. For everything else, you need to be more active.
Best Pairs for Minimizing IL on Solana
| Pair Type | Example | IL Risk | Fee Income | Net Expected Return |
|---|---|---|---|---|
| Stable-stable | USDC/USDT | Very low | Low (tight spreads) | Slightly positive — reliable but modest (3–8% APR) |
| LST-SOL | JitoSOL/SOL | Low | Moderate | Usually positive — best risk-adjusted LP on Solana (5–12% APR) |
| Major-stable | SOL/USDC | Medium-high | High (deep volume) | Variable — depends entirely on SOL volatility over your LP period |
| Major-major | SOL/ETH | Medium | Moderate | Usually slightly negative — correlated but not enough to offset IL |
| Volatile-volatile | BONK/SOL | Very high | High but inconsistent | Usually negative for passive LPs — only profitable with active management |
| Memecoin-stable | WIF/USDC | Extreme | Very high short-term | Heavily negative — extreme IL dominates any fee income within days |
The pattern is clear: the more correlated the pair, the better the risk-adjusted LP return. The more volatile and uncorrelated, the more likely IL destroys your position.
Common IL Mistakes
Chasing High APY Without Understanding It
The most common mistake. A pool displays 500% APY. A new LP deposits. The APY is based on one day of high volume from a memecoin launch. Volume drops 90% the next day, the token price crashes 80%, and the LP withdraws with 40% less value than they deposited. The “500% APY” was real for that one day — but the IL was real forever.
Using Narrow CLMM Ranges Without Active Management
Narrow ranges are tempting because they show higher projected APY. But they require constant attention. If you set a ±2% range and check your position once a week, price will almost certainly exit your range. You’ll sit earning zero fees while accumulating maximum IL. Narrow CLMM ranges are a full-time trading strategy, not a passive yield play.
Not Accounting for IL When Calculating “Profit”
Many LPs track their claimed fees and think they’re profitable. They earned $500 in fees over a month — great. But they don’t calculate that their position lost $800 in value compared to simply holding. They’re down $300 but think they’re up $500. Always compare your position value to a “just held” baseline.
LPing Memecoins
Memecoin prices can move 90%+ in a single day. On a standard AMM, a 90% drop causes approximately 20% IL. On a concentrated range, it can cause 40–60% IL. Meanwhile, fee income on memecoins is inconsistent — volume spikes during hype and collapses afterward. By the time you realize the fees aren’t covering IL, the damage is done.
If you LP memecoins, treat it as a directional bet that you’re comfortable losing, not as a yield strategy.
How To Manage IL as a Solana LP
Stick to Correlated Pairs for Passive LP
If you want to deposit liquidity and not think about it for weeks, only LP correlated pairs: JitoSOL/SOL, mSOL/SOL, USDC/USDT. These generate modest returns (5–12% APR) but with minimal IL risk. This is the only LP strategy that consistently works for passive participants.
Use Wider Ranges on Volatile Pairs
If you LP SOL/USDC on a CLMM, resist the urge to set a narrow range for higher APY. A ±20% range earns fewer fees than ±5% but survives normal market volatility without going out of range. For most LPs, staying in range and earning moderate fees beats earning high fees for two days and then sitting out of range for a week.
Rebalance Ranges When Price Moves
Active CLMM management means adjusting your range as price moves. If SOL moves from $150 to $170 and exits your $140–$160 range, you withdraw liquidity, rebalance your token ratio, and create a new position centered around $170.
This works but has costs: each rebalance incurs transaction fees and locks in any IL from the previous position. You’re also realizing a taxable event each time you withdraw and redeposit. Kamino offers auto-managed vaults that handle rebalancing for you — they charge a performance fee (typically 10% of earned fees) but remove the manual overhead.
Track Real P&L Including IL
Use portfolio tracking tools (Step Finance, Sonar Watch, or protocol-native analytics) that calculate your actual P&L including IL. The number you care about is: “How much is my position worth now, compared to what I’d have if I never LP’d?” If that number is negative after fees, you’re losing money. Stop looking at APY dashboards and start looking at total position value.
Know Your Break-Even
Before entering any LP position, do the math: “If SOL moves X% in a month, I lose Y% to IL. My expected fee income is Z% per month. Is Z greater than Y at the price moves I consider likely?”
If you can’t answer this question, you don’t have enough information to LP profitably. And if the answer is “fees only beat IL if SOL moves less than 10%,” ask yourself how realistic that is in crypto. Be honest with yourself — the market doesn’t owe you yield just because you provided liquidity.
Impermanent loss is not a bug — it’s a fundamental property of how AMMs work. Every swap that happens in your pool is the market taking value from your position. Fees are the compensation for that. Whether the compensation is adequate depends on the pair, the range, the volume, and — most importantly — whether you’ve done the math.