Banks are struggling to deal with the regimes of stress testing imposed since the global financial crisis of 2008, a survey has revealed.
Financial regulators around the world are using stress testing to mitigate risk as they impose higher levels of capital adequacy.
Only one in four banks are mature in their ability to conduct stress tests, while 40% are stuck in the early stages of stress test development, according to the survey from SAS and Longitude Research – which was highlighted at the SAS Premier Business Leadership Series conference in Las Vegas.
SAS CEO Jim Goodnight said in the past few years regulations have been introduced to banking which require risk computation to do stress testing. This has seen an increase in sales of SAS's high-performance analytics, which helps banks test systems such as logistic regression – a model used to calculate probability to default on loans.
"That used to take 15 or 16 hours, and can now be done in 15 minutes," he said. "That means that your modelling people, who are a scarce resource, become more productive.”
For more on data management in banking
The survey of more than 100 senior banking officials in Europe and the US assessed the effects of stress testing.
Respondents signalled data issues (35%) and skills shortages (32%) as the biggest problems. Data aggregation and consolidation were cited as especially difficult to do well.
Editorial director at Longitude Research James Watson said while stress testing is now a fact of life for banks, there are a significant number that still have work to do.
“This is especially true when it comes to the next step ahead – using stress testing as a tool to improve the wider management of the business,” he said.
Only 24% of banks acknowledged making any changes to their strategic decisions as a result of stress testing.
Global product marketing manager for risk management systems at SAS Brooke Upton said stress testing should not just be viewed as a compliance exercise.
“It’s time to move stress testing from the dreaded manual process to a more automated business-focused opportunity, where high-performance analysis of the data is also used to make better risk decisions,” she said.