Bread and buffers

Ironing out the pro-cyclical effects of regulatory capital and accounting standards has become a focal point for regulators and politicians across the globe. Talk has centred on an overhaul of the incurred loss model in favour of expected loss or dynamic provisioning, along with the introduction of capital buffers, leverage ratios and a new stressed value-at-risk measure. Regulators are also pushing banks to consider probabilities of default (PD) through-the-cycle, instead of point-in-time.

While some practitioners have warned about the pro-cyclical nature of Basel II for some time, the financial crisis has created a sudden urgency to tackle the issue. It has become painfully clear any regulation that requires risk weightings to be set based on credit quality will lead to higher capital levels in a downturn, potentially aggravating any recession by causing banks to cut back on lending.

Putting aside extra capital in good times means banks could dip into a pool of funds during any recession. Likewise, moving away from point-in-time modelling of default probabilities would mean banks consider the risk of a borrower through an entire cycle. In other words, the internal rating and capital assigned to any entity should not spike suddenly due to changes in economic conditions.

However, the piecemeal alterations by supervisors create potential contradictions within regulatory capital rules. For 10 years, regulators have moved towards making regulatory capital rules more risk sensitive, and to eliminate the gap between the risk models banks use internally and those used for regulatory purposes. Central to this is the concept of the use test. In obtaining validation for their risk models for regulatory capital purposes, banks must show these systems are truly employed for internal risk management.

The problem is, the likely addition of capital buffers means overall regulatory capital levels will no longer be as risk-sensitive. Some bankers would argue the behaviour of borrowers and PDs are inherently pro-cyclical: borrowers are more likely to default in a downturn. As a result, internal risk calculations should always measure this, arguably by considering PDs at each specific point in time. While the foundations of the Basel II framework will be retained, the proposed bolting on of a capital buffer and measuring of PDs through-the-cycle means there is likely to be a split between what banks are doing internally and what they are doing for regulators.

Capital buffers should strengthen the financial system, and hopefully avoid some of the capital headaches of the recent crisis, supervisors believe. But there are potential implications for regulatory use tests. After striving for a decade to eradicate the gap between internal risk capital and regulatory capital, this gap could now open again.

Nick Sawyer, Editor.