Suptech tools for prudential supervision and their use during the pandemic

The Covid-19 pandemic has prompted authorities to rely on virtual inspections, including the increased use of suptech tools to support supervisory risk assessments. This paper takes stock of 71 discrete suptech tools used for prudential supervision in 20 jurisdictions and explores the benefits, risks and implementation challenges.

We find that more than half of the 71 suptech tools assess mainly qualitative data, underscoring the importance of analysing textual information in prudential supervision. The remaining tools are split between those that analyse mainly quantitative data and others that scrutinise both quantitative and qualitative data. Despite these variations, all tools aim to extract deeper supervisory insights or to improve supervisory efficiency. 

Limited data science skills of supervisors, data quality issues that underpin suptech models and settling on an appropriate calibration of suptech parameters hamper broader adoption of suptech tools. As more tools become operational, a critical consideration is to ensure that the tools' outputs support, rather than replace, supervisory judgment. In this context, a comprehensive suptech strategy - that addresses many of these challenges – becomes indispensable, particularly as more supervisory activities migrate to a virtual setting.  

JEL classification: C45, C88, C89, G20, G38, O31, O32.

Keywords: suptech, prudential supervision, data analytics, innovation, AI, artificial intelligence, ML, machine learning, NLP, natural language processing.