Regulatory Data Management: Managing data collections for effective supervision

The amount of data we create every day in the digital era is hard to fathom. In 2018, more than 2.5 quintillion bytes of data were created every day. At the beginning of 2020, the amount of data in the world was estimated to be 44 zettabytes - and one zettabyte is 1,000,000,000,000,000,000,000 bytes.  

That data contains valuable information; but storing, processing, and analysing that data can be a huge challenge. 

What is Regulatory Data Management?  

In the financial services sector, regulatory data management refers to how financial regulators who oversee banks and other financial institutions like the pensions and insurance sectors process the data they provide. In most countries, the central bank is responsible for this.  

Financial institutions must provide huge reams of data to their central bank to show that they comply with financial regulations. These include Basel III, IFRS 9, MIFID II, EMIR, and FATCA. The goals of collecting this data include mitigating risk, performing stress tests, and examining analytics.  

Before the 2008 financial crisis, regulation was seen as ‘light-touch’, but those days are now gone. Today, surveillance is much more intense. Regulators demand much more data from financial institutions, at much more frequent intervals, with the aim of preventing another crisis.  

Initially, financial regulators dealt with this huge volume of data by hiring new staff to monitor, process and analyse it. But the volume of data continues to increase exponentially while the number of staff a regulator can hire is limited. As such, regulators are increasingly turning to technology to help with their regulatory data management.  

How can data modelling support Regulatory Data Management?

As regulatory requirements have become more complex, and financial institutions have to file more frequently, the way regulators collect data has evolved. Increasingly, the focus has moved from aggregate data to granular data, from complex forms to a no-forms approach, and to data modelling.  

When data is managed manually, the regulator creates a new requirement and it's up to each institution to interpret it and figure out what it means for them. Then, they find out what IT or human resources might be needed to meet it. For larger retail firms, this costs about $450,000 annually. The process is bloated with institutions overwhelmed by the scale of requirements, and regulators unable to effectively analyse the data. 

Data modelling - a representation of all the data and the relationship between different datasets - can help address this.  

Maturity assessments often find that organisations that are not using data modelling are mired in issues including siloed data, duplication of data, inconsistent and hard to analyse data, and a lack of ownership. To solve these problems, data is categorised by department or project on the platform and data owners are identified. The data is then published with stakeholder buy-in. 

Regulators can play an important role in this modernisation by providing clearer data models, more machine-readable data models and rules, as well as more automated means of submission and interaction of data.  

Establishing a new model for Regulatory Data Management 

A six-step process can be used by regulators who want to move from a scenario where data is convoluted and confusing to one where data modelling supports automation: 

• Vision and mandate: Getting buy-in from all stakeholders 
• Processes, standards and best practices: Ensuring good design and embedding the solution into the IT architecture 
• Community: Consultation with stakeholders including other regulators 
• Technology: Build and test the infrastructure needed including interoperability 
• Resources and skills: Team, training, and capability models 
• Governance: Steering group, change management and communications 

This evolution in terms of an operating model works in parallel with new methods of data collection. The human workforce can then spend their time on more valuable tasks like working on policy, managing trends, and data analysis, instead of manually processing data.  

Working with financial regulators across the world, we have developed best practices around Regulatory Data Management, and the technology to match. Our regulatory data management products and service represent decades of our domain knowledge and experience matched together with encompassing processes, tools and training.