FS23/1: Synthetic Data Call for Input Feedback Statement [pdf]

The Financial Conduct Authority, FCA, has published a Feedback Statement (FS23/1) concerning the „use of synthetic data to support financial services innovation“. „Synthetic data“ thereby refers to data derived from statistical modelling of „real data“ which replicates – so to speak – real data without the use of confidential information. The statement follows last year’s Call for Input in which the FCA sought views on
– the benefits of synthetic data or data access in general to facilitate financial innovation;
– the scope of usage of such data at this point in time;
– the financial industry’s needs regarding meta and synthetic data;
– possible impediments in the collection or creation of (synthetic) data; and
– current use cases for synthetic data.
In its Feedback Statement, the FCA now outlines the feedback it has received to its „consultation“ and summarizes its next steps in this matter. Some of the key issues raised by respondents are briefly noted below; for further detailed, comprehensive information, please refer to the enclosed Feedback Statement.
– Most respondent unanimously agreed that (synthetic) data „is crucial for financial innovation“.
– Most respondent also agreed that large synthetic data sets may be particularly helpful in the areas of financial fraud identification and prevention, the measurement of ESG performance and ESG reporting, or the conduct of credit scoring, to name a few.
– The greatest potential is seen in the area of anti-money laundering (AML) and the combating of the financing of terrorism (CFT).
– Most respondents, namely 66%, found regulatory issues as well as data security and privacy issues to be the most significant impediments to the collection or sharing of synthetic data. A lack of clarity in related regulations was thereby cited the most.
– 41% of the respondents noted technical issues as key hurdles in this context. Of particular note are thereby data quality issues (lack of standards, format issues) and IT-infrastructure issues.
– A large number of respondents noted testing and the validation of models and solutions and the possible sharing of data as the key benefits of (the use of) synthetic data.
—————
Based on the feedback received, the FCA states that it will set up a Synthetic Data Expert Group which will be responsible for the creation of a „framework for collaboration across industry, regulators, academia and wider civil society on issues related to synthetic data“ and that it will soon host a roundtable with industry, academics, and the Information Commissioner’s Office (ICO) to further investigate the challenges faced by market participants in this area. A corresponding „findings paper“ will be published in due course.