Experts stunned and confused by APRA’s risk culture pilot project in the financial sector

Another risk expert who spoke on condition of anonymity said the results presented by APRA were unlike anything they had ever seen, with the results of a question designed to weed out inauthentic participants. about 20 times higher than the average.

The regulator declined to respond, but is aware that failure rates were much higher than expected.

Disastrous financial results

APRA has stepped up its efforts to oversee the GCRA practices of regulated institutions following the publication of a capacity review in 2019 that asked the regulator to “focus more intensely” on the subject.

A poor risk culture in the financial services industry has contributed to dire financial results for banks and shareholders, including the account fraud scandal at U.S. lender Wells Fargo and the record $ 1.3 billion settlement of Westpac with AUSTRAC.

The 2021 survey was conducted by consultants for APRA and contained 40 questions on 10 “dimensions of risk culture”. Professor Sheedy, who developed his own model of free risk culture, questioned the model used.

“The problem is that it takes several years to validate a new measurement. All the differences that they point out between the institutions, they don’t know if they are real because they don’t know if the measure means anything, ”she said.

Tender documents reveal that Oliver Wyman was paid $ 361,900 by APRA for a “benchmark investigation” to be completed by June 2021. Oliver Wyman has also been hired by APRA to work on the prudential investigation on the Commonwealth Bank following the AUSTRAC scandal.

Following the publication of a 2019 APRA Capacity Review, author and former ACCC Commissioner Graeme Samuel called the contribution of external consultants to the landmark investigation “a waste of money. money “at a parliamentary hearing.

A risk expert who spoke on condition of anonymity questioned the results, including a graph that appeared to show that companies with lower survey participation rates also had the strongest risk cultures.

“I would even go so far as to say that in my 10 years of surveys … I have never seen this model,” said the expert.

The results of an “attention control question” designed to eliminate inauthentic participants also raised eyebrows. The APRA survey failed 20 percent of participants on this measure while most surveys generate failure rates of only 1 percent.

The APRA pilot survey was sent to 11,600 participants in the 10 insurance companies and received a response rate of 62%. Most respondents did not have any managerial responsibilities that APRA said provided a “unique employee perspective”.

Professor Sheedy said the value of asking entry-level employees to judge a company’s risk framework or providing a board’s risk appetite assessment was questionable.

“I find it hard to see how a typical employee, in a survey, would be able to meaningfully comment on the organization’s risk architecture and the functioning of the board,” she said. .