On-chain Peak Shaving of Ethereum Gas Fees

Abstract

Abstract While blockchain technology is expected to reduce transaction costs, network congestion execution costs remain understudied in operations management. Analyzing 62,142 Ethereum transactions from seven firms (January–March 2026), we study "on-chain peak shaving"—scheduling transactions toward low-congestion windows to reduce gas fee exposure. Gas fees peak at 10 AM ET, driven by speculative-arbitrage rather than operational activity. Firm scheduling responses vary: only three firms transact off-peak, while four transact during peak windows due to deadlines or governance cycles. This heterogeneity is driven by transaction deferrability and gas intensity. We formalize these into an On-Chain Scheduling Matrix mapping firms to four regimes that predict fee savings and residual cost floors. Theoretically, we extend Transaction Cost Economics to account for time-varying execution costs from congestion externalities, classifying gas fees as execution costs in timing but maladaptation costs in origin. Ultimately, managing gas fees requires systematic operational planning akin to energy procurement.