The UK regulators’ Digital Regulatory Reporting initiative could dramatically reduce reporting costs. Such digital technologies will require significant investment that financial firms need to prepare for. By Alexander McGill and Henry Williams, financial services experts at PA Consulting.
Over the past decade, the Financial Conduct Authority (FCA) and Bank of England (BoE) have launched pioneering initiatives to foster market competition and improve regulatory processes, partly through the use of digitisation.
One of these is Digital Regulatory Reporting (DRR) — a project to explore the use of emerging technologies to digitise and automate costly regulatory reporting. The progress of this project was recently reviewed by PA Consulting to further shape the initiative’s strategy going into 2021.
The report concludes that the scale of the challenge to make DRR successful is huge — not least the leap needed to bring promising technical solutions into full market implementation. Nonetheless, early estimates show that a full-scale implementation of DRR across the UK could reduce industry costs by as much as 40%.
Regulatory reporting is one of the most challenging and costly elements of compliance for financial institutions. The complexity, scale and cost of reporting has increased in line with the growing regulatory rulebook over the past decade. Much of this is attributed to the fact that regulatory reporting remains a heavily manual process, reliant on human input, interpretation and assurance. This also includes the challenging task of interpreting regulatory rules and converting them into data items to generate the required reports.
The resulting process is expensive and can be prone to error. As outlined in the BoE’s Future of Finance report, regulatory reporting was estimated to cost the UK industry £2.0–4.5bn per year. In parallel, regulators are dealing with the repeated inaccuracies in the reports that are provided. For example, one in five markets in financial instruments directive (MIFID) II reports have been estimated to be inaccurate and the Prudential Regulation Authority (PRA) has recently issued heavy fines and commissioned reviews into errors in reporting.
Ambitious vision and semantic technologies
DRR is looking to use innovative technologies to approach the problem from a radical new perspective, to achieve a fully automated regulatory reporting mechanism.
To deliver on this vision, the initiative has focused on exploring how regulators could provide rules in a format which can be read, understood and executed by firms’ internal systems in order to increase efficiency and reduce errors in reporting.
Three technical solutions were considered for digitising reporting, with two quickly discounted; firstly, the translation of regulatory policy into code, where the inherent ambiguity in regulation and the human effort needed would be too expensive. Secondly, the use of Natural Language Processing, while a promising technology, could not deliver accurate results reliably.
However, a third option has shown real promise. This is to use semantic technologies as an interim step in between natural language regulation and the code required to execute these rules on firms’ data. Semantic technologies provide a way for computers to understand both the context and meaning of data by linking it together in structures.
One advantage of semantic technologies is that they can be written by non-technical policy writers and automatically converted into a range of coding languages, providing a low-cost and accurate way of writing machine-executable regulation.
Focusing on data standards
There are many elements of DRR which need further exploration before a technical solution can be pursued. As PA recommended, this should focus on establishing a common industry data model and data standards rather than just reporting automation. Having high-quality data and minimum standards across the industry will lay the foundations for a fundamental shift in how regulatory data is collected and analysed, and support any future industry-wide initiatives.
Preparing for DRR
There are several activities industry participants should be doing now to prepare for DRR’s arrival:
1. Developing a regtech strategy
DRR will digitise and automate the end-to-end regulatory compliance process, and this will touch on every aspect of compliance. This includes redrafting the regulatory handbook in a machine-readable format, aligning the entire industry to common data model and standards and accessing data using APIs. This requires a major change in underlying technology to transform regulatory compliance. Industry participants should develop a regulatory technology (regtech) strategy for reporting to tackle this issue.
2. Establish a regulatory reporting data model and maintain strong data governance
For many firms, one of the greatest regulatory reporting challenges is organising and structuring their operational data into a format they can submit to the regulator.
As discussed in PA’s review, the development of a common industry regulatory reporting data model is top of UK regulators’ agendas and there could even be an ambition to develop a global regulatory reporting data model.
Data models which are easy to maintain with standards that are inclusive for the whole industry will be favoured by regulators, so industry participants should keep this in mind when developing the data model.
3. Collaborate closely with industry peers and regulators to address the challenge
DRR will require a significant amount of investment from the industry and its vision requires a revolution in regulatory compliance. It is also clear that the regulators want DRR to be co-created by industry participants and third-party vendors that support this ecosystem.
So far, DRR has directly involved nine leading UK banks in the pilot phase. However, the PRA and FCA are keen to expand industry participation. The industry has a one-off opportunity to shape the future of regulatory reporting and organisations should be looking to get involved.