Abstract
We present a novel way for cryptocurrency miners to manipulate the effective interest-rate on loans or deposits they make on decentralized finance (DeFi) platforms by manipulating difficulty-adjustment algorithms (DAAs) and changing the block-rate. This presents a new class of strategic manipulations available to miners. These manipulations allow miners to stretch and squeeze the time between consecutive blocks. We analyze these manipulations both analytically and empirically, and show that a 25% miner can stretch the time between consecutive blocks by up to 54% in Ethereum and 33% in Bitcoin, and squeeze it by up to 9% in Ethereum. Ethereum is particularly vulnerable, and even relatively small miners can seriously affect the block-rate. An interesting application of these manipulations is to create an artificial interest-rate gap between loans taken from one DeFi platform which accrues interest according to block height (such as Compound [66]) and deposited in some other platform that does so according to elapsed time (like a bank, or other DeFi platforms such as Aave [99]). Hence, stretching and squeezing the block-rate can decrease the interest paid on DeFi loans relative to external financial platforms. The profit made from this interest-rate gap provides a large incentive for miners to deviate. For example, a 25% Ethereum miner using our manipulations can increase mining profits by up to 35%, even after taking potential losses into consideration, such as less block-rewards. Our analysis of these manipulations and their mitigations has broad implications with regards to commonly-used cryptocurrency mechanisms and paradigms, such as Ethereum's difficulty-adjustment algorithm and reward schemes, with Ethereum's handling of uncle blocks being particularly manipulable. Interestingly, Bitcoin's mechanism is more resistant Ethereum's, owing to its larger incentives and a more resilient DAA.
Original language | English |
---|---|
Title of host publication | EC 2022 - Proceedings of the 23rd ACM Conference on Economics and Computation |
Publisher | Association for Computing Machinery, Inc |
Pages | 65-88 |
Number of pages | 24 |
ISBN (Electronic) | 9781450391504 |
DOIs | |
State | Published - 12 Jul 2022 |
Event | 23rd ACM Conference on Economics and Computation, EC 2022 - Boulder, United States Duration: 11 Jul 2022 → 15 Jul 2022 |
Publication series
Name | EC 2022 - Proceedings of the 23rd ACM Conference on Economics and Computation |
---|
Conference
Conference | 23rd ACM Conference on Economics and Computation, EC 2022 |
---|---|
Country/Territory | United States |
City | Boulder |
Period | 11/07/22 → 15/07/22 |
Bibliographical note
Publisher Copyright:© 2022 ACM.
Keywords
- blockchain
- cryptocurrency
- decentralized finance
- proof of work