Excerpt from speech by European Securities and Markets Authority chair Steven Maijoor, New technologies within and beyond capital markets, AFME/Euromoney Global Innovation Institute in Paris
While there is much hype around technology, we have seen staggering new technologies in recent years across sectors. Capital markets, a crucial engine of the economy, are no exception. As such it is both timely and important that we are here at this event to discuss technological transformation of capital markets.
Regulators face a balancing act. We work to understand and respond to the risks that new technologies and entrants may introduce while at the same time not wanting to stifle innovation by restricting the use of certain technologies. When making this assessment, I think it is important to keep in the back of our minds that common capital markets phenomena like a derivative, a mutual fund, and even a stock exchange once came to life as a financial innovation.
Maijoor specifically identified artificial intelligence for investing, suptech and regtech:
Another way in which technology may affect capital markets and investors is in the form of AI-powered investment and trade execution strategies. AI and machine learning tools are bein used by portfolio managers – especially systematic or ‘quant’ funds – to detect subtle patterns in data to help predict price movements. Their aim is to generate alpha, and to do so they comb through vast datasets from sources as diverse as satellite images and Twitter feeds.
At present the amount of money in AI-based strategies is limited, and so any impact on financial stability are limited too. However, as AI tools become more widely used, we will want to keep monitoring this area. ESMA contributed to a report by the Financial Stability Board, published last November, which noted the scope for new forms of interconnectedness resulting from the use of AI in financial services.
Finally, a very different application of AI outside of the financial sector is that examination boards are reportedly using machine learning algorithms to spot people cheating on exams. As a former academic who had to set exams for my students, I can certainly see the value of such a tool.
Now that I am a regulator, one of the goals of my work is to ensure the integrity of markets. Algorithms can be used to help identify where people may be ‘cheating’ in other ways, such as acting on insider information or other bad conduct. This is an illustration of supervisory technology, or suptech.
Regulators have for example been exploring how best to put in place data analytics and pattern recognition systems to study trading behavior to detect market abuse. While we are still at an early stage in applying tools such as AI-powered surveillance of market conduct, I see significant potential in this area.
The flipside of the suptech coin is regtech: the use of new technology by financial market participants to meet their regulatory obligations such as reporting and risk management. Regtech is for example extensively used to meet the reporting obligations for investment firms under MIFID 2, allowing for more automation in data reporting. Common reporting standards, such as LEI, ISIN and ISO20022
underpin the successful application of RegTech.