FSA puts pressure on banks to adopt slotting in place of ‘ropey’ risk models

Regulator tells banks to start using slotting’s ‘simple buckets’ to weigh up their property loan risk

The Financial Services Authority expects UK banks to be using the slotting method of risk weighting for their commercial real estate loans by this summer.

Andrew Bailey, head of the FSA’s prudential business unit, told MPs at the Treasury Select Committee this month: “We are withdrawing the modelling concessions for many banks in the area of commercial property because the models were so ropey that they should never have been there in the first place.

Bailey for web
Bailey: bank risk models “were so ropey that they should never have been there”

“We’re moving them to a non-model approach called slotting. It’s saying: ‘Just put your loans into simple buckets and use simple weights against them. Don’t try to do pyrotechnics, because we know what happens then.'”

The move underlines the FSA’s scepticism about the reliability of the mathematical models banks use to calculate how much regulatory capital they need to hold, and how they are using those models.

In 2011 the FSA asked banks to calculate the risk weights for an identical, hypothetical portfolio of sovereign, bank and corporate debt.

The results showed massive variations – the risk weightings used implied the most prudent banks would be holding twice a much capital as the most aggressive banks (see below).

Slotting is therefore likely to significantly increase the amount of capital banks will have to set aside to cover their property lending, although the impact on each bank will vary depending on their loanbook.

The UK property industry and lending banks have been fighting a rearguard action against slotting, arguing that introducing it now will raise the cost of borrowing and push down property prices.

However, the FSA’s view is that slotting has been under discussion for nearly two years and the end game is in sight.

Banks’ models ‘not sound or reliable’, says BoE director

At the 15 January Treasury Select Committee meeting, Bank of England head of financial stability Andy Haldane spelt out why regulators don’t trust banks’ capital calculations.

 “The results of the FSA’s hypothetical portfolio exercises make clear that for a common exposure, banks can hold very different amounts of capital – and in some cases that means a number of banks will   be holding too little,” Haldane said.

Haldane: “These models are not a sound basis for capital regulation”

“If you look at the average risk weight banks have assigned to assets in the past 20 years, you’d be astonished to see how far they fell over the period.

“One interpretation is that assets on banks’ balance sheets are a lot, lot safer than they were 20 years ago. But that rather flies in the face of the evidence of the past five or so years, and of the evidence the FPC put in the Financial Stability Review from November.

“Many of these models are not just complex and lack transparency, but more fundamentally they lack any robustness. They are not a sound or reliable basis for the setting of capital regulation.

“That problem is being increasingly recognised, not just by regulators but by investors in banks – and banks themselves. Many of the loudest complaints about those models have come from banks.

“So it is time for a drains-up. The regulatory community made a significant error in the 1990s in the reliance it placed on deeply complex and deeply fragile models. That balance needs to be redrawn.”