A Material Sociology of High-Frequency Trading
Our project – ‘A Material Sociology of High-Frequency Trading’ – is funded by the UK Economic and Social Research Council (£411,398, January 2018-December 2021). You can view a video of Donald MacKenzie talking about this research, or read the project's working papers. Prospective project interviewees can find more information on the project's commitments to anonymity and its respect for commercial confidentiality on our Participant Information Sheet.
The research team is led by Professor Donald MacKenzie, FBA, FRSE, who has held a personal chair in Sociology at the University of Edinburgh since 1992. Others in the team are Dr Iain Hardie (Senior Lecturer in Politics and International Relations, University of Edinburgh), and Dr Juan Pablo Pardo-Guerra (Honorary Fellow of the University of Edinburgh and Assistant Professor of Sociology, University of California at San Diego). The goal of our research is academic articles and eventually a book on the topic.
The project has two main goals:
- Understanding how today’s automated trading (especially so-called ‘high-frequency trading’ or HFT) has developed and is currently conducted. For example, what was involved in transitioning from open-outcry trading floors to automated trading? How does a succesful trading firm organise the relationship between IT functions (software, hardware, networks) and traders? What do the research processes underpinning automated trading involve: automation of succesful human strategies? Novel human ideas that are then back-tested? Machine learning (or more traditional linear regression techniques)? To what extent does the research process and its technological implementation differ between generic clusters of strategies, such as market-making and liquidity-taking strategies? To what extent are the particularities of different asset classes important in shaping strategies?
- Understanding the interaction between automated trading and market structure (in equities, futures, foreign exchange, sovereign bonds, listed option, interest-rate swaps and cryptocurrencies), both in the US and Europe. For example, HFT firms have often found themselves in the situation of 'challengers' in markets that have features that make HFT difficult. Do they simply live with those difficulties, or does HFT lead to changes in market structure? What role if any does regulation play in the outcome of processes such as this? What factors explain radically different outcomes such as the considerable presence of HFT in the trading of Treasurys and its marginal role in the trading of Gilts and Eurozone sovereign bonds?
The primary way in which we will be collecting data is by interviewing people who work in high-frequency trading or for the electronic trading venues in which it takes place, as well as those who have other roles in those markets (e.g. as broker-dealers), who regulate trading and who supply trading firms with software, hardware and communication links. We will also take part in events such as industry meetings and training courses.
We will fully respect the anonymity and confidentiality of participants in the study, as well as the commercial confidentiality of the firms for which they work (when, for example, we speak to people in automated trading firms, we are careful never to probe for ‘secret sauce’). Those who take part in this study are guaranteed that nothing will ever be attributed to them or to their organisations in anything we write or say, unless they give us their explicit permission to do so. For more information, see our Participant Information Sheet.