Leeds Building Society has announced it is to use an artificial intelligence (AI) data engine to estimate missing Energy performance Certificate (EPC) information and more accurately evaluate the emissions of its mortgage portfolio.

The society has selected 4-Xtra Technologies, a software firm formed at Leeds University, for its AI-based extreme event forecasting and synthetic data generation engine.

Leeds revealed that the relationship follows a successful three-month proof of concept, when 4-Xtra demonstrated it could provide the missing data to a level that would satisfy the society’s requirements.

With the UK committed to achieving net zero by 2050, regulators now expect financial institutions to factor ESG considerations into their risk processes and decision-making. The Prudential Regulatory Authority has also stressed the need for financial institutions to address capability gaps in identifying, measuring and managing climate risks.

EPCs, which measure the energy efficiency of buildings, are a valuable source of data for the risk processes related to mortgage applications. They are also key for calculating mortgage portfolio carbon emissions to satisfy climate change reporting requirements. Not all properties have a valid EPC, which means financial institutions have to estimate the missing data.

Chief risk officer at Leeds Building Society, Andy Mellor, commented: “We want to demonstrate how we are effectively incorporating climate risk considerations into our risk management processes and credit decisions. Missing EPC data was a particular obstacle in meeting these requirements for our mortgage portfolio.

“The successful PoC undertaken with 4-Xtra Technologies and its SaaS solution has demonstrated that we can now instantly estimate data for any gaps, which in turn improves our credit decisioning processes and enables us to report our mortgage portfolio emissions more accurately.”

Built upon academic work, 4-Xtra’s extreme event forecasting engine and complementary synthetic data generator can be used for a variety of use cases in financial services, with data outputs delivered via web-based interfaces or via APIs for ready integration with a firm’s own systems.

4-Xtram CEO, David Potter, added: “We’re delighted to support Leeds Building Society in factoring in ESG considerations to its risk processes.

“Our SaaS solution can accurately predict missing EPC data, while also providing clients with enhanced data analytics capabilities to manipulate the large data sets involved. We are working on providing further enhancements to our solution, to include other physical risks while also actively developing other financial services solutions using our extreme event forecasting and synthetic data generation technology.”

Share Story: