The Basel Committee has decided that banks should not be allowed to use internal models to measure capital requirements for operational risk.

What’s happening?

In March 2016, the Basel Committee on Banking Supervision (BCBS) issued a consultation on revising the bank capital framework for operational risks – risks such as fraud, technology failure or misconduct fines. Banks will no longer be allowed to use internal models known as the advanced measurement approach (AMA) to calculate operational risk-weighted assets, which are part of the Basel risk-based capital ratio. The BCBS also issued a revised standardised approach to calculate operational risk-weighted assets, based on a business indicator and the use of historic loss data for each bank.

Why is it happening?

The Basel II framework appeared to grossly underestimate potential operational risk losses. The standardised approach assigned different risk weightings to different business lines, with retail banking assigned the lowest weighting, and investment bank sales and trading activities the highest.

This proved to be an inaccurate assumption. In the UK, banks paid compensation of £22.5bn ($31.8bn) between 2011 and 2015 for mis-selling payment protection insurance to retail customers, according to the Financial Conduct Authority. Moreover, the standardised approach was based on gross revenues over a five-year period. With many investment banks tipping into losses from 2007, their operational risk weights were therefore suppressed for an extended time at precisely the point when conduct losses – for instance, related to the manipulation of interest rate benchmarks – were rising sharply.

The incorrect calibration of the standardised approach also had a knock-on effect on the AMA. Banks were assured of a lower capital charge if they adopted the internal models, to incentivise the extra systems investment required to build these models.

“Supervisory experience with the AMA has been mixed. The inherent complexity of the AMA and the lack of comparability arising from a wide range of internal modelling practices have exacerbated variability in risk-weighted asset calculations, and have eroded confidence in risk-weighted capital ratios,” the BCBS concluded in its consultation.

What do the banks say?

Industry participants planned a meeting for the end of March 2016 to discuss the BCBS proposal. Quantitative analysts were the driving force behind this meeting, with the apparent intention of lobbying Basel to reconsider its abandonment of the AMA. In theory, an improved standardised approach would provide a better reference point for the continuation of internal models as well.

However, operational risk managers and experts accept that there are inherent difficulties modelling a risk that is by its nature more behavioural than statistical. The AMA models were built using internal and external loss data, together with scenario analysis.

“For smaller firms, operational risk modelling is challenging, because these firms will thankfully have nowhere near enough internal loss data, and the available external data mainly comes from larger banks, and so may not be as relevant. This leaves scenarios, which, even with more sophisticated estimation techniques, remain subjective. These issues are also challenges for larger banks – hence this consultation document,” says Michael Grimwade, head of operational risk for the international securities business of Japan’s MUFG.

Will it provide the right incentives?

To maintain a clear risk management incentive within the capital framework, banks can incorporate historic loss data into the proposed new standardised approach. To be eligible to use the historic loss module, banks must have at least 10 years of data deemed adequate by their supervisor, or five years during an as-yet unspecified transition period.

“If the bank is permitted to use its loss data, that is a way to reduce the required capital amount. But there is also an important supervisory element here – in case banks have the incentive not to use loss data because it might increase their capital requirement, supervisors can require the banks to use that data,” says one regulator involved in drafting the consultation.

Simon Ashby, an associate professor at Plymouth Business School who previously worked in operational risk for Lloyds Bank and the UK Financial Services Authority, suggests that regulators will need to promote the message that operational risk is at least as important as the other main Basel risk categories – credit and market risk. While the regulatory rationale for internal models was to provide an incentive to invest in operational risk data governance, the bank motivation was always more likely to revolve around capital optimisation.

“The trouble with operational risk modelling is that there are too many assumptions, too many differences of opinion, so in some ways it is easier to game than the models for credit or market risks. If the time and effort spent on AMA is replaced with more intelligent conversations between supervisors and banks about how they are managing risks, and looking at scenario analysis, then that would be useful,” says Mr Ashby.


All fields are mandatory

The Banker is a service from the Financial Times. The Financial Times Ltd takes your privacy seriously.

Choose how you want us to contact you.

Invites and Offers from The Banker

Receive exclusive personalised event invitations, carefully curated offers and promotions from The Banker

For more information about how we use your data, please refer to our privacy and cookie policies.

Terms and conditions

Request a demonstration to The Banker Database

Join our community

The Banker on Twitter