Fee level dominates revenue, skew has minor effect in Lightning routing
The claim
The authors assert that setting the fee level is far more important than fee skew for maximizing Lightning routing revenue
Published as: “Trim: Fee Level Matters More Than Skew in Lightning Routing”
So what
Operators can boost revenue by focusing on a single fee level per channel and only add a skew when costly refills are frequent, simplifying fee-management tools.
The five-minute version
The paper finds that setting the fee level captures most of a Lightning node’s revenue, with a well-chosen flat fee within 1.6% of the optimum on median channels, while adding a skew only changes gross revenue by a few percent. Evidence comes from a 2026 network snapshot (5,507 nodes, 27,063 channels) with synthetic payments; no fee rule beats the revenue ceiling set by demand and replenishment, and tuned tools are at most a few percent above the uniform-level frontier. The authors note that the results rely on synthetic traffic and a simple sender model.
What they found 3
- A well-chosen flat fee is near-optimal on exact channel instances.
“On 144 exactly solved channels a well-chosen flat fee is nearly optimal, within 1.6% at the median (at most 9.6%)”
- No fee rule beats the revenue ceiling set by demand and replenishment.
“no fee rule, flat or balance-dependent, beats the ceiling that demand and replenishment set.”
- The skew changes gross revenue by at most a few percent.
“The skew changes gross revenue by at most a few percent, as Proposition 4.4 allows here; it pays only where refills are frequent.”
What undercuts it the paper’s own results
- The skew, while secondary, can still add up to a few percent of revenue in some settings, contradicting a strict interpretation that only the level matters.
Stated limits 4
- Traffic: payments are synthetic, acceptance is exponential and demand is stationary.
- The peak’s location reflects the simulated senders; the ordering holds across settings.
- Routing: senders take the cheapest single path, blind to balances except through mission control.
- Scaled capacities put network success (0.70-0.75 at deployed fees) below the 99.7% a large node reports.
What’s new here 4
- Derives a two-layer fee model (level + skew) for a Lightning channel as a two-sided inventory.
- Proves a revenue ceiling that no fee rule can exceed given demand and replenishment.
- Shows that the optimal flat fee (level) is near-optimal across a wide range of realistic channel instances.
- Provides a practical rule (Trim) for choosing a uniform level and optionally adding a skew only where refills are frequent and costly.
Artifacts & methods 5
- Trim modeltheory
- Exact channel solver (relative value iteration)system
- Network simulation with synthetic paymentsevaluation
- Reimplemented deployed tools (charge-lnd, feeadjuster, LNDg)system
- Synthetic payment generatordataset