Paper Funded by OFR Grant Examines High-Frequency Trading and Large Datasets
Published: May 5, 2015
A recent paper funded by the OFR through its joint grant program with the National Science Foundation (NSF) offers insights into the impact on the financial system of trading activity at the nanosecond level. The paper, the first produced through this OFR-NSF partnership, also contributes to developing technologies for working with large datasets, an important area of research for the OFR, and fosters a better understanding of market liquidity.
The paper, “Tick Size Constraints, High-Frequency Trading, and Liquidity,” by University of Illinois at Urbana-Champaign researcher Mao Ye and his coauthor Chen Yao, examines high-frequency trading and tick size, or price increments for trades. Because there is a uniform tick size — one cent — lower-priced stocks have a higher relative tick size than higher-priced stocks. The authors explore whether this larger relative tick size “constrains price competition and generates higher revenue for liquidity provision.”
To test the hypothesis, the authors use NASDAQ limit-order book snapshots and trades for 117 stocks every minute from 9:30 a.m. to 4:00 p.m. The authors demonstrate that traders identified by NASDAQ as high-frequency traders (HFTs) provide liquidity more often for stocks with higher relative tick sizes. A larger relative tick size leads HFTs and non-HFTs to quote the same price, allowing the speed advantage of HFTs to trump price considerations. As relative tick size decreases, price becomes more of an advantage over speed.
The authors also take a novel approach to identify the effects of relative tick size by looking at leveraged Exchange Traded Funds (ETF) that track the same index. The authors use leveraged ETFs that have undergone splits as the test group and those that track the same index but have not recently split as the control. They show that after a split, which reduces the price of the ETF and raises the relative tick size, liquidity providers previously able to compete on price could no longer do so and HFTs are able to dominate, based on speed.
This paper, which reflects the views of the authors and not the views of the OFR or the Treasury Department, contributes to a better understanding of HFTs, tick size, liquidity, and working with large datasets. It is an example of the type of research that the OFR seeks to support as part of its statutory mandate to conduct, coordinate, and sponsor research.
The innovative and computational nature of the research directly matches the characteristics of the OFR-NSF grants program. For more information about the program, including how to apply, please see the Dear Colleague Letter on the OFR website.
The OFR is also partnering with the NSF on a second grants program, called Critical Techniques and Technologies for Advancing Foundations and Applications of Big Data Science & Engineering (BIGDATA). The BIGDATA program seeks novel approaches in computer science, statistics, computational science, and mathematics, along with innovative applications. The OFR and NSF are seeking proposals that apply big data techniques to financial stability research. The deadline for full proposals is May 20, 2015.
Greg Feldberg is Acting Deputy Director for Research and Analysis at the Office of Financial Research