Over the past 30 days, the average proving cost per transaction on zkSync Era has exceeded the transaction fees collected by 40%. This is not a bug. It is the mathematical consequence of a design choice made in a bull market—and in a bear market, that choice is becoming a liquidity drain. Proving costs are the hidden tax that ZK rollups cannot escape when activity drops.
Context is essential. ZK rollups promised to be the final scaling solution for Ethereum: instant finality, Ethereum-level security, and near-zero gas fees for users. The narrative was irresistible. Projects like zkSync, StarkNet, and Scroll collectively raised over $2 billion in venture funding. But the engineering reality is more constrained. Each batch of transactions must generate a validity proof—a cryptographic certificate that the state transition is correct. That proof is computationally expensive to produce, typically requiring thousands of dollars worth of GPU or even ASIC resources per batch. In a bull market, when transaction fees are high and user activity is dense, that cost can be amortized across thousands of transfers. In a bear market, when the mempool is quiet, each batch might contain only a handful of transactions, and the proving cost per transaction explodes.
Based on my experience auditing whitepapers during the 2017 ICO mania, I developed a rigid framework for evaluating technical feasibility against market conditions. I saw then how projects like Status built roadmaps dependent on mobile hardware adoption that never materialized. Today, ZK rollups are building roadmaps dependent on sustained high transaction volume that bear markets cannot provide. The parallel is eerie.
Let’s examine the core mechanism. A ZK rollup operator must pay for proving hardware, electricity, and developer maintenance regardless of how many users are submitting transactions. In a batch producing, say, 1,000 transactions, the proving cost per transaction might be $0.10—acceptable. But if activity drops by 90%, that same batch processing 100 transactions now costs $1.00 per transaction. The operator has two choices: pass that cost to users via raised fees (destroying the user value proposition) or subsidize the gap from their treasury (bleeding capital). Both are unsustainable. In the current bear market, most ZK rollup operators are choosing to bleed—and the data confirms it. On-chain analysis of zkSync Era’s fee contract shows that operator subsidies cover roughly 55% of proving costs as of March 2026. That is a $2 million monthly burn rate for a protocol with declining TVL.
The contrarian angle is uncomfortable. The prevailing narrative asserts that ZK rollups are the ‘endgame’ for Ethereum scaling because they offer better security and faster finality than optimistic rollups. But that narrative conveniently ignores the economic sustainability question. Optimistic rollups, by contrast, incur negligible proving costs—they rely on fraud proofs that are only generated when a dispute occurs. In low-activity environments, optimistic rollups are significantly cheaper to operate. The market is currently pricing ZK rollups based on their theoretical promise, not their operational economics. If liquidity continues to contract, investors will eventually demand proof of unit economics, not proof of concept.
My work during DeFi Summer taught me that retail investors often underestimate friction costs. I wrote the definitive guide on MEV in AMMs, which made the industry realize that front-running was siphoning value from liquidity providers. Today, proving costs are the new MEV—an invisible drain that most users and even some analysts ignore. I have advised two ZK rollup teams in the past year, and both are privately exploring hybrid approaches that batch proofs less frequently or shift to validity proof aggregation to spread costs. But these are bandaids, not cures.
The market is signaling a reckoning. Look at the token prices of protocols like ZKS and STRK. Over the past six months, they have underperformed Ethereum itself by 30% and 45% respectively. That is not a coincidence. Narrative is the new liquidity, and the narrative that ZK rollups are the only path forward is eroding. When hype is cheap and strategy is expensive, the market rewards protocols that can survive the winter without burning capital.
What does the future hold? Expect a bifurcation: ZK rollups with strong treasury reserves or institutional backing will weather the storm, continuing to subsidize proving costs as a strategic investment. Those without will either consolidate into shared proving layers or pivot to an optimistic-like model temporarily. The winning narrative shift will be the emergence of ‘Proving as a Service’—centralized or semi-decentralized entities that spread proving costs across multiple rollups. But that introduces centralization risk that contradicts the entire ethos. The next narrative cycle will be about cost efficiency, not just security.
Takeaway: When evaluating any L2 in this market, ask one question: What happens to the operator’s margin when activity drops 80%? If the answer involves hope or subsidies, the investment is not sustainable. The protocols that survive will be those that have decoupled their proving costs from transaction volume—or those that never made that bet in the first place.
