How AI and Big Data Can Help Banks Adapt to a New Accounting Standard
One of the big learnings from the 2008 financial crisis was the urgent need for a change to accounting standards for banks.
It was widely acknowledged that the old, backward-looking approach to measuring loan losses had contributed to the collapse.
But the shift to a new, forward-looking system is not without its challenges, as many banks have neither the data collection and analysis capabilities nor the tools to accurately predict future losses.
In this paper from the University of Toronto’s Rotman School of Management, the author Scott Liao explains the new loan loss accounting models, and discusses how Big Data and AI can help banks calculate more reliable forecasts.
Scott Liao
Professor of Accounting
Rotman School of Management
University of Toronto
Highlights
Challenging banks’ capabilities
Banks may not have existing models, expertise, or information systems to collect the data needed to forecast forward-looking credit losses.
New technologies can bridge the gap
Big Data analytics are relatively easy to implement and can help to accurately forecast bank expected credit losses. Machine learning can also be used to predict borrower default risk and macroeconomic indicators.
Embracing new approaches
Although they require more data and technology literacy, these new approaches offer enhanced computational power and flexibility to arrive at better estimates of lifetime expected losses.