APRA plans deeper dive into detailed data

The Australian Prudential Regulation Authority (APRA) plans to enhance its data collection to gain greater insights into the performance of products and emerging risks, the Actuaries Institute summit has heard.

APRA GM Data Analytics and Insights Alison Bliss says the regulator is stepping up its data-driven analysis and will work with the financial services industry as the needle shifts on the amount and quality of data to be collected over coming years.

“We are looking to have those data sets broad and deep and strong enough that we can pose many future questions that will emerge as the industry changes,” Ms Bliss said.

APRA is seeking more information at a “granular level” as the financial services sector is increasingly data driven, she says.

The regulator has introduced the APRA Connect collection system for enterprises lodging information and data and is liaising with the Australian Securities and Investments Commission on data sharing to minimise doubling-up and to ease collection pressures on industry.

Examples of effective data analysis have included an examination of superannuation funds offering My Super products that highlighted areas of underperformance, Ms Bliss says.

“It has been extremely successful for us in actually getting change in industry and getting consolidation and movement in the My Super sector and getting great outcomes for members in those funds,” Ms Bliss said.

“This is the sort of thing that we anticipate doing more of, not just in superannuation but across all industries.”

Ms Bliss says natural language processing (NLP) has also been important in analysing comments made in a survey examination of risk culture within financial services.

“We have been able to add some additional colour and insights to the responses that we have received with those surveys,” she said. “NLP has really been able to help us focus in on additional areas for change.”

APRA is taking account of potential risks that may emerge from data use and technology advancements as it steps up its capabilities, she told the summit.

“Big data is really important, but how we do big data is equally important in terms of ensuring we get good outcomes,” she said.