PERPS ATLASPERPETUAL MARKETS INTELLIGENCE
RESEARCH PREVIEW 01
Home / Research / Fee Buyback Mechanisms Compared
← Research library
TOKEN ECONOMICS · 22 SEP 2026

How six perp DEXs turn fees into token mechanics

A grounded, source-by-source comparison of documented fee-to-token mechanisms — not a ranking of which is best.

Why "buyback" alone tells you little

Several protocols in our coverage document a mechanism that converts trading fees into token-related benefits. The mechanisms differ in what they touch (buy, burn, stake reward), what triggers them (fixed rule vs. discretionary), and whether a realized dollar figure is independently verifiable. Read the source before comparing headline percentages.

Documented, rule-based allocations

Aster’s June 2026 documentation allocates 99% of daily platform fees to automated ASTER purchases distributed to veASTER stakers, with separate biweekly reserve burns until a 3-billion-token cap — the burn and the staker distribution are documented as distinct flows, not double-counted. Avantis’ foundation dashboard describes a live 30% of daily fees allocated to buybacks, with higher rates gated behind milestones. Hyperliquid’s fee documentation describes automatic conversion of Assistance Fund fees into HYPE, which is then burned — a supply-reduction mechanism, not a direct cash payment to holders.

Discretionary or partially documented allocations

Paradex’s DIME documentation permits net fees to fund buybacks and/or growth at the protocol’s discretion — not an unconditional payout rate. edgeX’s tokenomics dashboard reports buybacks and burns occurring, but does not disclose the exact recurring formula in the sources reviewed. GRVT’s token introduction describes membership benefits (fee discounts, margin benefits) without a documented conversion of net revenue into a cash-like holder distribution.

What this means for a valuation model

A documented, fixed-rate mechanism (Aster, Avantis) is closer to a modellable holder flow than a discretionary one (Paradex) or an undisclosed formula (edgeX). None of this determines whether the resulting per-token value is attractive — that depends on the realized dollar amount, the diluted supply it is spread across, and a defensible comparison multiple, none of which this article supplies. See each project’s valuation section for the specific, current blocker.

Referenced profiles

Further reading & sources

By the PerpsAtlas research desk · Educational analysis · Published 22 September 2026.

Compare the protocols →