High-Frequency Trading and Market Quality: Evidence from Account-Level Futures Data

Date Author Names Title

Date	Author Names	Title
July 2023	Coughlan, John, Orlov, Alexei	High-Frequency Trading and Market Quality: Evidence from Account-Level Futures Data
Journal of Futures Markets, 43, 1126– 1160. https://doi.org/10.1002/fut.22404 We use rich regulatory data on intraday transactions and end-of-day positions of traders to examine how participation of high-frequency traders (HFTs) affects market quality. Panel estimation evidence shows that greater participation by HFTs is strongly associated with improvements in market quality, whereas higher rates of aggressive, directional trading produce an adverse, partially offsetting effect. While futures contracts are sensitive to market uncertainty, as measured by the VIX, they are even more sensitive to own price volatility. We take advantage of the 2015 change in CME's daily settlement methodology for agricultural commodities to address potential endogeneity using a fixed-effects difference-in-difference setup. Our results are robust to relying on alternative estimation techniques, using overly conservative (clustered) standard errors, modeling various forms of cross-sectional and temporal dependence, as well as studying each market separately.

July 2023

Coughlan, John, Orlov, Alexei

Journal of Futures Markets, 43, 1126– 1160. https://doi.org/10.1002/fut.22404

We use rich regulatory data on intraday transactions and end-of-day positions of traders to examine how participation of high-frequency traders (HFTs) affects market quality.
Panel estimation evidence shows that greater participation by HFTs is strongly associated with improvements in market quality, whereas higher rates of aggressive, directional trading produce an adverse, partially offsetting effect.
While futures contracts are sensitive to market uncertainty, as measured by the VIX, they are even more sensitive to own price volatility.
We take advantage of the 2015 change in CME's daily settlement methodology for agricultural commodities to address potential endogeneity using a fixed-effects difference-in-difference setup.
Our results are robust to relying on alternative estimation techniques, using overly conservative (clustered) standard errors, modeling various forms of cross-sectional and temporal dependence, as well as studying each market separately.