🤖 AI Summary
This study investigates the on-chain circulation micro-mechanisms of stETH and wstETH—two dominant liquid staking tokens (LSTs) on Ethereum—and uncovers their high-frequency reuse patterns and evolving user behavior within DeFi. To address the lack of granular, address-level empirical analysis, we propose a “micro-velocity” analytical framework—the first to enable large-scale, address-level characterization of LST circulation. Methodologically, we introduce a scalable velocity component decomposition technique that integrates transaction history reconstruction, event log indexing, and smart contract state backtracking to achieve behavior deduplication and causal attribution. We open-source our end-to-end data pipeline and a comprehensive historical dataset. Empirical findings reveal: (1) LST circulation velocity remains persistently high and is highly concentrated among top addresses; and (2) users exhibit a pronounced migration from stETH to wstETH, driven by wstETH’s superior protocol composability, establishing it as the dominant DeFi-native staking vehicle.
📝 Abstract
We introduce a micro-velocity framework for analysing the on-chain circulation of Lidos liquid-staking tokens, stETH, and its wrapped ERC-20 form, wstETH. By reconstructing full transfer and share-based accounting histories, we compute address-level velocities and decompose them into behavioural components. Despite their growing importance, the micro-level monetary dynamics of LSTs remain largely unexplored. Our data reveal persistently high velocity for both tokens, reflecting intensive reuse within DeFi. Yet activity is highly concentrated: a small cohort of large addresses, likely institutional accounts, are responsible for most turnover, while the rest of the users remain largely passive. We also observe a gradual transition in user behavior, characterized by a shift toward wstETH, the non-rebasing variant of stETH. This shift appears to align with DeFi composability trends, as wstETH is more frequently deployed across protocols such as AAVE, Spark, Balancer, and SkyMoney.
To make the study fully reproducible, we release (i) an open-source pipeline that indexes event logs and historical contract state, and (ii) two public datasets containing every Transfer and TransferShares record for stETH and wstETH through 2024-11-08. This is the first large-scale empirical characterisation of liquid-staking token circulation. Our approach offers a scalable template for monitoring staking asset flows and provides new, open-access resources to the research community.