Lending pools are decentralized applications which allow mutually untrusted users to lend and borrow crypto-assets. These applications feature complex, highly parametric incentive mechanisms to equilibrate the loan market. This complexity makes the behaviour of lending pools difficult to understand and to predict: indeed, ineffective incentives and attacks could potentially lead to emergent unwanted behaviours. Reasoning about lending pools is made even harder by the lack of executable models of their behaviour: to precisely understand how users interact with lending pools, eventually one has to inspect their implementations, where the incentive mechanisms are intertwined with low-level implementation details. Further, the variety of existing implementations makes it difficult to distill the common aspects of lending pools. We systematize the existing knowledge about lending pools, leveraging a new formal model of interactions with users, which reflects the archetypal features of mainstream implementations. This enables us to prove some general properties of lending pools, and to precisely describe vulnerabilities and attacks. We also discuss the role of lending pools in the broader context of decentralized finance and identify relevant research challenges.
|Title of host publication||Financial Cryptography and Data Security|
|Editors||Matthew Bernhard, Andrea Bracciali, Lewis Gudgeon, Thomas Haines, Ariah Klages-Mundt, Shin'ichiro Matsuo, Daniel Perez, Massimiliano Sala, Sam Werner|
|Publication status||Published - 2021|
|Event||2nd Workshop on Coordination of Decentralized Finance - Virtual event|
Duration: 5 Mar 2021 → 5 Mar 2021
|Conference||2nd Workshop on Coordination of Decentralized Finance|
|Period||05/03/2021 → 05/03/2021|
|Series||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
Bibliographical noteFunding Information:
Massimo Bartoletti is partially supported by Conv. Fondazione di Sardegna & Atenei Sardi project F74I19000900007 ADAM. James Hsin-yu Chiang is supported by the PhD School of DTU Compute. Alberto Lluch Lafuente is partially supported by the EU H2020-SU-ICT-03-2018 Project No. 830929 CyberSec4Europe (cybersec4europe.eu).
Acknowledgements. Massimo Bartoletti is partially supported by Conv. Fondazione
© 2021, International Financial Cryptography Association.