The data shows a contradiction that few are willing to face. Hyperliquid’s RWA perpetual open interest hit $3.6 billion in July, surpassing Bitcoin. Yet the protocol’s quarterly revenue dropped 43% from its peak, and buybacks—the core of HYPE’s deflationary thesis—were nearly halved. This is not a healthy divergence. It is a structural warning.
Context: The HIP-3 Mechanism and Its Discontents
Hyperliquid’s HIP-3 proposal introduced a permissionless market deployment system. Anyone staking 500,000 HYPE (approximately $28 million at current prices) can launch a perpetual market and keep 50% of the trading fees. The remaining 50% flows to the protocol’s Assistance Fund, which buys back and burns HYPE. This is a radical departure from the industry norm. Synthetix, for example, caps external builder fees at 30%.
Kain Warwick, founder of Synthetix and Infinex, has publicly stated that the 50% split is unsustainable. His argument is straightforward: the platform controls the fee structure. Hyperliquid can unilaterally cut the builder’s share at any time, or even absorb the markets entirely. The builders, in turn, have no contractual guarantee. They are dependent on a protocol that can change the rules.
Warwick has been through this exact game. His perspective carries weight. But the market has not yet priced in the full implication of his critique. The question is not whether the fee split will change—it is when, and how the transition will affect the HYPE token.
Core: The On-Chain Evidence Chain
Let me lay out the numbers. I have tracked this data since Q3 2025.
- Total protocol revenue has fallen from $357 million in Q3 2025 to $202 million in Q2 2026. That is a 43% decline.
- Buybacks dropped from $290 million to $149 million over the same period. A 49% reduction.
- HYPE price is at $57.66, down 24.8% from its $76.67 high. The deflation narrative is losing steam.
Yet the volume has not declined. Warwick noted that trading activity remains strong—the fees are simply flowing to different parties. The HIP-3 builders, particularly trade.xyz, now capture a significant share of the revenue. trade.xyz alone accounts for over 90% of all HIP-3 open interest. This concentration is both a growth engine and a systemic risk.
The RWA perpetual market has grown from 2% of Hyperliquid’s total volume to 50% in less than a quarter. That is a rapid structural shift. But the revenue that the protocol retains from this growth is shrinking because the 50% fee split channels half of it to external builders.
Here is the core paradox: more volume, less protocol revenue. The bull market euphoria around RWA derivatives masks a deteriorating unit economics for HYPE holders.
I have seen this pattern before. In 2020, during DeFi Summer, I analyzed Uniswap V2 liquidity pools and identified a similar disconnect between volume growth and LP profitability. The market initially ignored the warning signs, but eventually the arbitrage opportunities corrected. The same dynamic is playing out here.
Contrarian: The 50% Split May Not Be the Real Problem
Conventional wisdom says that reducing the builder fee share will boost protocol revenue. That is mathematically true, but it ignores the behavioral response. If builders lose incentive, they may withdraw liquidity. The 90% concentration from trade.xyz means that any change could trigger a sudden collapse in HIP-3 volume. The protocol could absorb those markets, but that would require operational capacity and might deter future builders.
Warwick’s argument assumes that builders will accept lower fees because they have no alternative. Hyperliquid is the “mothership” for RWA perpetuals—no other platform offers the same user base and liquidity depth. But the 500,000 HYPE staking requirement creates a significant locked position. Builders are already committed. They are not leaving easily.
This is where the correlation-versus-causation trap matters. The 43% revenue decline is not entirely caused by the fee split. It also reflects a broader market slowdown in crypto-native trading volumes. Hyperliquid’s core crypto perpetuals have seen lower activity. The RWA segment is the growth area, but it is still early. The fee split is a contributing factor, not the sole cause.
Moreover, the 50% split might be a temporary growth hack. Hyperliquid is buying market share in the RWA derivatives space. If they successfully establish dominance, they can later adjust the split downward. The builders, having built their infrastructure on Hyperliquid, may stay even with a lower share. This is a classic platform strategy.
Takeaway: The Next Signal to Watch
The next quarterly report will be critical. If protocol revenue continues to decline while RWA volume grows, the pressure to adjust the fee split will intensify. A formal HIP-4 proposal to reduce the builder share to 30% or 35% would be a bullish catalyst for HYPE, as it would directly increase the buyback flow. But the market reaction will depend on whether builders stay or leave.
Ledgers do not lie, only the narrative does. The current narrative is that Hyperliquid is the undisputed leader in RWA perpetuals. The hidden truth is that the token holders are not fully capturing the value of that growth. Survival is the ultimate alpha in a bear. If you are long HYPE, you are betting that the protocol will eventually rebalance the incentives. The data suggests that adjustment is coming. The question is whether it will be smooth or disruptive.
Trust the math, ignore the hype. The math says the buyback engine is losing fuel. The hype says the RWA volume is unstoppable. One of these is wrong.