The recent scandal involving TD Securities LLC[1] has served as a stark reminder of the dark side of high-frequency trading. While this particular case has garnered significant attention, it is far from an isolated incident. Spoofing, a form of market manipulation that involves placing and canceling large orders to deceive other market participants, has become a pervasive problem in today’s highly automated financial markets. From Wall Street to emerging markets, this insidious practice is eroding investor confidence and undermining the integrity of our financial system.

The Dodd-Frank Act of 2010 was the first legislation to specifically mention spoofing, and it triggered enforcement activity that increased over the following years. Current regulation is founded on the principle of prohibition and prosecution. However in practice, illegal market manipulation is often assessed and punished case by case. Consequently, market-regulation enforcement is a matter for the administrative authorities and the criminal courts, both of which have some discretion. The effectiveness of enforcement actions is regularly proven. Landmark cases include – in addition to the recent case involving TD Securities – United States V. Coscia, which involved a trader in commodity futures markets who was sentenced to three years in prison, a civil fine of more than $25 million, and a $12 million disgorgement penalty. In another case, Chicago-futures-market trader Igor Oystacher and his firm were ordered to pay $2.5 million in civil fines and placed on probation for three years. In 2020, J.P. Morgan was ordered to pay $920 million for spoofing involving Treasury futures. This is the largest fine ever imposed by the Commodity Futures Trading Commission (CFTC).

But along with these successful prosecutions have come other challenges and an increasing belief that these cases are just the tip of the iceberg. In a new article, we provide three explanations for why :

• Spoofing is hard to detect: There are so many legitimately canceled orders that regulators have to process millions of data records daily to detect manipulation and to ascertain that an order was not bona fide.
• Spoofing is hard to prove: Sanctions are constrained by the burden of proof in a high-frequency and algorithmic framework, which implies using thealgorithms itself as evidence of intent. Moreover, trading supervision is mainly focused on detecting suspicious and abnormal price changes – but price impacts are assessed in a world where HFT manipulators often trade insignificant quantities within a low-latency recurrent strategy.
• Spoofing is hard to discourage: As spoofing is frequent and diffuse, it does not formally or systematically affect price accuracy or the spread in real time.

If prosecution is the public weapon, market design could be the private one. The underlying principle is to make the cost of engaging in spoofing activities exceed any potential gains. While regulators can increase fines to deter such behavior, improvements in market design, such as enhanced surveillance systems and reduced latency, can also significantly diminish the profitability of spoofing.

Yet, both regulators and exchanges are acutely aware that even minor tweaks to a market infrastructure (e.g., decimalization, order priority, latency, or transparency) can trigger significant ripple effects.

With this in mind, how can we reduce the profitability of deceptive practices without compromising market quality or the interests of other participants? Given the risks associated with trial-and-error approaches, regulators and market operators need a more systematic alternative. Our research offers a novel solution by conducting in-depth simulations of various market scenarios.

Using sophisticated software[2] to replicate a virtual financial market, we have analyzed the impact of different market designs on the profitability of spoofing. This methodology, known as agent-based modeling, is widely used in economics and management research and more specifically in financial market microstructure[3]. By conducting simulations, we can compare before and after scenarios. For instance, we analyzed the profitability of spoofing strategies before and after the introduction of a random delay in market-order execution, which is the most direct and targeted approach to mitigate high-frequency spoofing tactics that rely on rapid order cancellation and execution. Our findings show that this subtle alteration in the market design can effectively reduce the attractiveness of spoofing without compromising overall market quality.

Of course, these results are not generally applicable, nor is that the goal of this methodology. Its strength lies in its ability to provide a unique, context-specific view of market dynamics, from individual trader behavior to broader price movements. Our study offers a valuable contribution to the ongoing debate about how to effectively address manipulative practices and foster the growth of market designs (like IEX) that promote market fairness and integrity.

ENDNOTES

[1] https://www.sec.gov/newsroom/press-releases/2024-160

[2] https://artificialmarket.univ-lille.fr/

[3] See for instance, the seminal works of Santa Fe Institute: Palmer, R. G., Arthur, W. B., Holland, J. H., LeBaron, B., & Tayler, P. (1994). Artificial economic life: a simple model of a stockmarket. Physica D: Nonlinear Phenomena, 75(1-3), 264-274.

This post comes to us from Daniel Ladley at the University of Leicester School of Business in the UK, Nathalie Oriol at the University of Côte d’Azur – GREDEG – CNRS in France, and Iryna Veryzhenko at LIRSA-CNAM in France. It is based on their recent article, “High-Frequency Spoofing, Market Fairness and Regulation,” available here.