M Shearer, G Rauterberg, and MP Wellman
Proceedings of 4th ACM International Conference on AI in Finance (ICAIF’23), pages 592–600, November 2023.
Abstract
Financial benchmarks estimate market values or reference rates used in a wide variety of contexts, but are often calculated from data generated by parties who have incentives to manipulate these benchmarks. Since the LIBOR scandal in 2011, market participants, scholars, and regulators have scrutinized financial benchmarks and the ability of traders to manipulate them. We study the impact on market welfare of manipulating transaction-based benchmarks in a simulated market environment. Our market consists of a single benchmark manipulator with external holdings dependent on the benchmark, and numerous background traders unaffected by the benchmark. Background traders use standard zero intelligence (ZI) strategies. We explore two types of manipulative trading strategies: manually adjusted ZI, and strategies generated by deep reinforcement learning. We find that manipulation decreases market surplus for the manipulator but increases it (to a lesser degree) for the background traders. It also decreases the quality of market information. Including the benchmark holdings, aggregate profits for the manipulator substantially increase. The negative impacts of manipulation, therefore, fall to the external counterparties to the manipulator’s benchmark holdings, as well as anyone relying on benchmark information for decision making.