Leave it to technocrats?
Those who believe that governance by technocrat will solve Italy's ills should think again. The IASB is currently working on a three-bucket approach for financial asset impairment. The idea is that newly originated or purchased loans - the model must work for both - are allocated by an entity to one of three buckets. And in very general terms, assets will move from one bucket to another in order to reflect deteriorating credit quality and credit losses. This is the board's third stab at developing an impairment model since 2009.
Another issue is whether the allocation to each bucket is absolute or relative. The board has favoured a relative approach under which an entity allocates loans of differing credit quality to bucket one. It would move loans across the buckets to reflect deteriorating credit quality and credit losses. The banks hate this approach, citing its cost and complexity implications, including the need to develop costly tracking systems to monitor credit quality.
By way of contrast, under an absolute or credit-quality approach, all loans in bucket one would feature an allowance for a future amount of expected losses - a downside where the borrower might have no default history.
But more gripping than the technical details are the findings of the board's outreach activities with banks. Remember, standard setters are in a bind: they need the people who bear some of the responsibility for the mess we are in to help them develop the alarm bell for the next crisis. Of course, standard setters simply scrap the amortised cost model, fair value everything, and leave the banks to argue with analysts over the ‘more relevant' amortized cost numbers - basically what purchasers of sovereign debt are doing right now.
The intrigue all starts with a non-public meeting. Apparently, the board holds ‘three-way-dialogue' meetings. Typical attendees have been IASB representatives, banking regulators, and the Institute of International Finance. If the meetings take place in the US, the SEC and the US FASB attend. According to IASB member John Smith, one preparer said during the latest meeting: "If we had day-one loss, [we] would not be able to participate in higher-risk lending." This was, Smith added: "Because if they take a loss it will affect their capital and the hit was enough to preclude them from engaging in that kind of business."
Sound familiar? HSBC's Russell Picot, during a 10 September 2009 classification and measurement roundtable, warned that standard setters had better be careful what they develop, least we end up without any solvent banks: "Pragmatically, I can't imagine that the world could afford [the US approach]. If you look around the world's financial system and you look at the disclosed numbers for customer loans at amortised cost and fair value, they are very significantly under water."
Following on from the three-way-dialogue meeting, Smith said he had received a call from "one of the representatives there who is personally more in favour of an absolute model, where you would recognise day-one losses and he seemed to confirm in that conversation that perhaps some of the operational concerns of the relative method have been overplayed."
Smith added that "one of the regulators" who attended the meeting had accused the banks of sending the standard setters mixed messages, with the regulator supposedly saying: "You guys are sending these poor standard setters on a wild goose chase, because on the one hand you're telling them it's operationally difficult, on the other hand you're telling them that it's not and they don't know who to believe."
The IASB's financial instruments specialist, Sue Lloyd, confirmed this was the case, adding that the board has received a paper from "four of the large banks in Europe". In the paper, the banks confirmed that they do not support an absolute approach. Moreover, they disapprove of the day-one loss effect as a result of assets moving into bucket two from bucket one.
The IASB director added that although the banks might well face operational challenges in applying the three-bucket approach, "we do have a situation where when we talk to different parts of the banks we are getting different messages."