Banks suspected of tweaking risk measure

The EU’s banking watchdog has found that half the variation in banks’ risk-weighted assets cannot be explained by objective factors such as portfolio and regulatory differences.

The study by the European Banking Authority, which is charged with writing rules for banks across the 27-nation bloc, may heighten suspicions that some institutions are tweaking the way they measure risk to cut capital requirements.

EBA looked specifically at RWA versus total assets in the banking books of 89 banks in 16 countries and found that some banks were using risk models that required them to hold 70 per cent less capital than their peers.

The watchdog found that about 50 per cent of the difference was because of objective factors, including different asset mixes, national regulatory differences and the extent to which each bank used its own models versus a simpler “standardised” approach that assigns risk weights by category.

The gap is important because a bank’s capital ratio – the most critical measure of financial strength – uses RWAs as its denominator and a bank that understates its risk will appear stronger than it actually is.

“Greater disclosure could go some way to appease the concerns raised by investors and market analysts,” said Andrea Enria, EBA chairman.

“But this is not enough. The remaining dispersion is significant and calls for further investigations and possibly policy solutions. We are strongly committed to deepening this analysis to ensure RWAs are seen and understood as reliable and consistent,” he said.

The study comes as investors, regulators and some bankers have questioned whether banks can be trusted to use their own risk models. The UK’s top bank supervisor recently said bank models for commercial property lending were “ropey” and several other regulators are calling for capital requirements to be based on total assets, rather than RWA, to prevent cheating.

Banking Weekly podcast

The banking team discuss Barclays’ cost-cutting strategy plan, RBS’ Libor settlement and small banks rounding on regulations

The EBA will now do more detailed studies of the way banks measure the riskiness of residential mortgages and lending to small and medium-sized businesses to see if the variation is justified. It also plans to send hypothetical portfolios of sovereign and top-rated corporate debt to different banks to test how each one measures risk. It may also try to impose a consistency on the way banks decide whether a loan is in default, a key input to the models.

The 70 per cent variation discovered by the EBA is significantly smaller than the 700 per cent differences uncovered by the Basel Committee on Banking Supervision in a global study that focused on the bank trading books. But credit risk accounts for 77 per cent of the risk-weighted assets of EU banks, compared with 10 per cent for market risk, so the impact of any variation in the banking book is magnified significantly.

Dan Davies, an analyst at BNP Exane, said that the EBA results suggested that regulators would taking a more judicious approach. “The more you look at the variances the less murky they look,“ he said.