Algorithmic Trading Regulation: The Frameworks for Human Supervision and Direct Market Interventions

Abstract

This paper examines the regulation of algorithmic trading in the capital markets and focuses on the human supervision and director market interventions. We compare the regulation in the UK, the EU and the US to find a common basis and additional regulatory techniques. In Part I, we examine the requirements for the internal risk management process a firm has to conduct before an algorithm can be implemented on the market in each of the three jurisdictions. In Part II, we examine the direct market intervention methods and focus on circuit breakers and assess their effectiveness. In Part III, we investigate the liability issue in the algorithmic trading space, including the liability of trading firms and venues for breaches of regulatory requirements, compensation claims of individual investors, and the liability of the regulators. The possible contributions of this paper are to fill the gap of human supervision and direct market interventions in algorithmic trading and build a novel framework for machine learning regulations in finance.