‘Whoops’ said the regulator trying to enact Basel 2.5 locally | FT Alphaville

‘Whoops’ said the regulator trying to enact Basel 2.5 locally

Rule-making is a natural response to a financial crisis. There is, of course, also a tendency for the new rules to be more complex than their predecessors. But this evolution has given some regulators pause for thought.

Consider the below a case study, as fuel for debate. It’s an example of when a local regulator managed to fudge the implementation of the edicts from the gnomes of Basel…

This particular nugget of regulation concerns the securitisation of assets held in a bank’s trading book. Such securitisations were, and still can be, especially handy when a bank is trying to rid itself of all that unwelcome mark-to-market volatility that can come from, say, legacy mortgage-backed securities and miscellaneous leftovers from the CDO construction pipeline.

Potential accounting benefits aside, to get regulatory credit for removing assets from one’s balance sheet, a bank must prove that “significant risk transfer” (SRT) of said assets has occurred. In other words, that the bank actually transferred a meaningful amount of the risk arising from the assets to an outside party, or parties.

That’s where Basel 2.5 comes in. It outlines the capital treatment for such deals, putting in place more stringent capital requirements for trading book assets than existed previously. Basel edicts get translated into directives at the EU level, and then it’s over to the local regulations to incorporate the directives into local rules.

But when the UK’s Financial Services Authority did this for Basel 2.5, it seems their first go didn’t quite work as intended. Whoops.

By March 2012, the FSA proposed to fix said boo-boo, as announced in their quarterly consultation (emphasis ours):

4.12 The aim of applying SRT [significant risk transfer] in the trading book is to correct a regulatory failure. Currently, firms can avoid the SRT test by originating securitisations in the trading book, rather than the banking book. This new rule removes this opportunity for regulatory arbitrage by ensuring that the securitisation regime implemented by CRD2 and CRD3 is consistently applied across both the trading and banking books.

4.13 The main benefit from this new rule is to remove the risk that firms do not hold sufficient capital against securitised positions held in the trading book to reflect the risks associated with these positions.

CRD2 and 3 are those European directives that instruct the implementation of Basel2.5, which the FSA needs to put in its local “Handbook”.

In the FSA’s defense, the way CRD3 was drafted makes it sound like the equivalent of an Excel file containing circular references and badly written macros. Cast a quick eye over the following to get a sense of it:

4.7 Under CRD3, it is not clear whether or not SRT applies to securitisations originated in the trading book. More specifically, although Annex I point 16a in the Capital Adequacy Directive (CAD) (Directive 2006/49/EC, as amended by CRD3) cross-refers to Annex IX in the Banking Consolidation Directive (BCD) in relation to the use of capital charges and the supervisory formula method (SFM), it is not clear whether the extent of this cross-reference could be understood to include the remainder of the securitisation rules in Annex IX, specifically the rules on SRT. As a result, an argument could be made that CRD does not require that the SRT requirement be applied to securitisations originated in the trading book. Article 326 of the draft CRD4 Regulation, largely a copy-out of the CRD3 text, is no clearer in this regard…

So when Andy Haldane, executive director for financial stability at the Bank of England, says something like:

Modern finance is complex, perhaps too complex. Regulation of modern finance is complex, almost certainly too complex. That configuration spells trouble. As you do not fight fire with fire, you do not fight complexity with complexity. Because complexity generates uncertainty, not risk, it requires a regulatory response grounded in simplicity, not complexity.

Keep in mind that the above whoopsie wasn’t even Basel III yet. And as Haldane pointed out in the same speech…

Basel I – 30 pages

Basel II – 347 pages

Basel III – 616 pages

As for the costs of the mistake, back to the FSA’s consultation:

4.15 Additionally, this new rule could have a capital impact, where firms must raise more capital in instances where their securitised positions in the trading book do not meet the SRT test. However, our analysis suggests that this impact is likely to be negligible. According to regulatory returns submitted to the FSA in June 2011, only three firms have gained capital relief by originating securitisations in the trading book.9

That’s good, but what was going on before June 2011? Did some banks in the UK (and elsewhere?) manage to hold lower capital buffers at a particularly sensitive time? If so, for how long? We’ll likely never know. Unless, that is, someone on the Treasury Committee wants to pipe up to ask how confident regulators are that they’ll be able to translate CRD IV (the EU directive which Basel III will ride on) accurately into local rules.

Here’s a convenient map through the Basel + CRD jungle, courtesy of Accenture (click to expand):

Fingers crossed.