© The Financial Times Ltd 2015 FT and 'Financial Times' are trademarks of The Financial Times Ltd.
April 13, 2012 5:35 pm
It is crunch time for risk managers and modelling experts at banks and regulators round the world, as they prepare for the 2013 start date of the global “Basel III” set of standards to make big banks safer.
Aimed at preventing a repeat of the 2008 financial crisis, the Basel deal, signed by 27 financial centres and followed by most of the world, broadly seeks to force banks to increase both the quality and quantity of capital they hold against unexpected losses.
Because the requirements are expressed as a percentage – banks must have capital equal to 7 per cent of their risk-weighted assets (RWA) – the figure for each institution depends heavily on the way it measures and calculates risk.
For instance, a bank with a balance sheet of top quality corporate and government bonds must hold far less capital than one loaded with high risk unsecured loans.
But few of the calculations are as simple as that.
Under the last big banking regulation reform, known as Basel II, banks could either adopt a simple methodology that applies relatively high risk weights to broad categories of assets, or they could invest in internal models that predict the probability of default and, if that happens, the “loss given default” of each loan or security.
Since 2007, banks that use such modelling have been generally rewarded with lower aggregate risk weights because regulators wanted to encourage them to spend money on better risk management.
Now that Basel III is starting to be phased in, more and more banks are adopting the models – known as the “internal ratings based” (IRB) approach – and are tinkering with their models in hopes of trimming their RWA and their overall capital needs.
Concern is growing that this “risk weight optimisation” may lead some banks to opt for models that understate the risks.
That in turn would give them a short-term advantage over more cautious peers and leave them dangerously exposed to the next big economic downturn.
The biggest US banks, that are only just now switching to the previous Basel II accord, have long contended their European competitors were understating their risk. Some European bankers, most notably Alfredo Sáenz, chief executive of Santander, have recently added their voices to the criticism.
Jon Pain, head of financial services risk consulting at KPMG, says: “An analysis of the ratio of RWA to the total assets for European banks shows wide discrepancies.
“Some of this can be explained by different business models and accounting rules, but the remaining difference appears to be down to local application of EU directives.”
He adds: “This is not good for consistency and comparison. Until the politicians agree a common approach to RWAs, the same economic position in two countries can result in starkly different numbers. This makes comparison difficult and creates uncertainty.”
Some experts say the concerns about banks tinkering with their RWA calculations are overblown.
They believe the big reductions in RWA that some banks have achieved stem partly from decisions to sell off highly risky parts of their portfolios and partly from efforts to improve models based on recent experience.
“Banks are tidying up the calculations, not manipulating them,” says Patricia Jackson, head of prudential advisory at Ernst & Young. “When banks rushed to get over the finish line for Basel II IRB in late 2007, there were various data problems that had to be dealt with in a range of ways, particularly with added conservatism in models ... As the IRB has run on for longer it is possible to improve estimates of probability of default and loss given default.”
She adds: “Banks are also reviewing whether some portfolios are profitable and getting out of some activities.”
Partly as a result of the concerns, both the Basel Committee on Banking Supervision, which sets the global rules, and the European Banking Authority, the pan-EU regulator, are looking closely at how banks measure risk with an eye to harmonising the methods and the rules.
Among other things, they are asking banks to apply their models to several sample portfolios to see if they can identify the causes of the big differences across institutions and countries.
“This is not an area where there are easy answers or short cuts,” says Clifford Smout at Deloitte’s Centre for Regulatory Strategy for Europe, the Middle East and Africa. “It’s a lot of work to explain apparent differences, and regulators are preparing to do this globally.”
Please don't cut articles from FT.com and redistribute by email or post to the web.