Suboptimality in DeFi

Aviv Yaish; Maya Dotan; Kaihua Qin; Aviv Zohar; Arthur Gervais

Paper 2023/892

Suboptimality in DeFi

Aviv Yaish

, Hebrew University of Jerusalem

Maya Dotan

, Hebrew University of Jerusalem

Kaihua Qin

, Imperial College London

Aviv Zohar

, Hebrew University of Jerusalem

Arthur Gervais

, University College London

Abstract

The decentralized finance (DeFi) ecosystem has proven to be popular in facilitating financial operations, such as token exchange and lending. The public availability of DeFi platforms’ code, together with real-time data on all user interactions with them, has given rise to complex tools that find and seize profit opportunities on behalf of users. In this work, we show that both users and the aforementioned tools sometimes act suboptimally. In specific instances which we examine, their profits can be increased by more than 100%, with the highest amount of missed revenue by a suboptimal action reaching 428.14ETH ($517K). To reach these findings, we examine core DeFi primitives which are responsible for a daily volume of over 100 million USD in Ethereum alone: (1) lending and borrowing funds, (2) using flashswaps to close arbitrage opportunities between decentralized exchanges (DEXs), (3) liquidation of insolvent loans using flashswaps. The profit which can be made from each primitive is then cast as an optimization problem that can be solved. We show that missed opportunities to make a profit are noticed by others, and are sometimes followed by back-running transactions which extract profits using similar actions. By analyzing these events, we find that some transactions are circumstantially tied to specific miners, and hypothesize they use their knowledge of private orderflow for a profit. Essentially, this is an instance of miner-extractable value (MEV) “in action”.

Note: Accepted to the 4th Annual CBER Conference.

Metadata

Available format(s): PDF
Category: Applications
Publication info: Preprint.
Keywords: Cryptocurrency Blockchain DeFi Miner Extractable Value
Contact author(s): aviv yaish @ mail huji ac il
mayadotan @ mail huji ac il
kaihua qin @ imperial ac uk
avivz @ cs huji ac il
arthur @ gervais cc
History: 2024-04-04: revised; 2023-06-09: received; See all versions
Short URL: https://ia.cr/2023/892
License: CC BY

BibTeX

@misc{cryptoeprint:2023/892,
      author = {Aviv Yaish and Maya Dotan and Kaihua Qin and Aviv Zohar and Arthur Gervais},
      title = {Suboptimality in DeFi},
      howpublished = {Cryptology ePrint Archive, Paper 2023/892},
      year = {2023},
      note = {\url{https://eprint.iacr.org/2023/892}},
      url = {https://eprint.iacr.org/2023/892}
}