BoE Governor Outlines Work on Using Technology for Bank Supervision

Mr. Carney mentioned that, last year, he announced that Huw van Steenis would lead a review of the future of the UK financial system, including recommendations for how BoE should respond. In two months, Huw will publish his conclusions and BoE will announce a number of concrete steps to create an environment for a more resilient, effective, and efficient financial system. 

He emphasized that big data is opening up new opportunities for more competitive, platform-based finance of small and medium enterprises. Search and social media data are supplementing traditional metrics to unlock finance for smaller enterprises whose assets are increasingly intangible. He also highlighted that BoE is building a platform for fintech innovation, which considers how general purpose technologies, including advanced analytics such as artificial intelligence, can increase the resilience of the financial system. The financial sector is investing heavily in the Cloud, machine learning, and artificial intelligence, said Mr. Carney. Banking is already the second biggest global spender on artificial intelligence systems (after retail) and is expected to invest a further USD 10 billion on artificial intelligence by 2020. Solutions enabled through artificial intelligence are increasingly important in fraud detection as well as automated threat intelligence and prevention. There is also significant potential in credit assessments, wholesale loan underwriting, and trading. Such advanced analytics are also likely to lead to changes to the way BoE conducts supervision.

Mr. Carney added that PRA promotes safety and soundness based on forward-looking, judgment-based supervision, in which it identifies the key risks facing firms and sets supervisory strategies to mitigate them. As a process, it can be broken down into three simple steps: rule-setting and reporting; analysis and monitoring; and setting and communicating a supervisory strategy to mitigate identified risks. Each of these aspects of supervision is amenable to automation, machine learning, or artificial intelligence to some extent.

With regard to rule-setting and reporting, he said: "At over 638,000 words, the PRA Rule Book is longer than War and Peace. It is also somewhat less interesting and infinitely more complex. We are currently using advanced analytics to understand the complexity and interconnectedness of the PRA rulebook, to identify ways to simplify our rules, and to make compliance with them easier for firms." He added that BoE is also exploring ways to make reporting more efficient and effective. To this end, it is running a Digital Regulatory Reporting pilot, with FCA, on machine-readable reporting requirements, which firms’ systems could interpret and ultimately automate regulatory data collection. These initiatives create the potential to unlock the power of artificial intelligence to improve the quality supervision by BoE. He concluded that, "By adapting our hard and soft infrastructure, the Bank of England will help create the conditions for such innovation to flourish to promote the good of all the people of the United Kingdom."

Related Link: Speech

Keywords: Europe, UK, Banking, Insurance, Regtech, Suptech, Big Data, Digital Regulatory Reporting, Machine-Readable Regulations, BoE