I have been away from the purely banking risk area for a little while, so I had dropped my regular visits to the BIS website. Having headed over there for some stats recently I noticed how much they had beefed up their coverage.

A few interesting things have also appeared.

Deposit Insurance

The CP on deposit insurance that they issued in March has been updated into a full set of recommendations – the paper is here. Regular readers would know my opinion on deposit insurance, but in general the principles in this paper are sound. A worthwhile note is on page one:

The introduction or the reform of a deposit insurance system can be more successful when a country’s banking system is healthy and its institutional environment is sound. In order to be credible, and to avoid distortions that may result in moral hazard, a deposit insurance system needs to be part of a well-constructed financial system safety net, properly designed and well implemented. A financial safety net usually includes prudential regulation and supervision, a lender of last resort and deposit insurance. The distribution of powers and responsibilities between the financial safety-net participants is a matter of public policy choice and individual country circumstances.

Advice to our PM and Treasurer – do not do this in a hurry and in a poorly thought out manner. Nuff said.

Basel II Changes

Pillar I

The latest tweaks to the main Basel framework seem to be trying to do a little stable-door shutting on securitisation – with a fair bit of language added around understanding the risks and a few risk weights being increased.

Pillar II

The language of the amendments to Pillar II is also interesting – this places a large amount of emphasis on improved risk management at the banks, including reporting structures, concentration management and, in particular, the off-balance sheet stuff.

One recommendation in particular is interesting – the separation of the CRO’s office out of the business lines to report directly to the CEO and Board (pillar II amendments – 19).


This one is also good (pillar II, para 40):

A bank should conduct analyses of the underlying risks when investing in the structured products and must not solely rely on the external credit ratings assigned to securitisation exposures by the CRAs [credit ratings agencies]. A bank should be aware that external ratings are a useful starting point for credit analysis, but are no substitute for full and proper understanding of the underlying risk, especially where ratings for certain asset classes have a short history or have been shown to be volatile. Moreover, a bank also should conduct credit analysis of the securitisation exposure at acquisition and on an ongoing basis. It should also have in place the necessary quantitative tools, valuation models and stress tests of sufficient sophistication to reliably assess all relevant risks.

I would agree – but it should not take a regulator to tell a bank this. I wonder who is going to pay for all the extra analysts – or will banks just walk away from this market?
The whole document is worth a close read and I should get around to it soon. From my scan, though, I cannot see it having a significant impact in Australia. APRA was already doing most of this – but the changes to securitisation may cause some heartache amongst those institutions that rely on them, such as many credit unions and other smaller ADIs. I would be looking to the specialist mortgage providers and the bigger banks to benefit from these.


As expected, as well, the BCBS has boosted its look at liquidity – with almost as many mentions in the 41 pages of this as in the 400 of the Accord itself. Given much of the problems (like those at Northern Rock) at least started as liquidity issues this is not surprising. The language looks mostly drawn from the relevant Sound Practices document, but this incorporates it into the Accord itself. Again – not much to see here, as much of this is already in the relevant Australian liquidity regulation, and APRA is well across this.

Stress Testing

Brand new is a whole two pages on stress testing. As I highlighted nearly three years ago in one of the perennial favourite blog posts here – the original accord had little to say on the subject beyond telling you that it had to be done. The two pages here go some of the way to addressing that gap, but not nearly all the way. They are obviously leaving the wording fairly general here and leaving it up to national regulators still.

Pillar III

Much of this is focussed (as can be expected) on improving disclosure around the areas that have caused problems recently. Almost all of the changes to Pillar III are about increasing disclosure of securitisation exposures – with some consequential changes to the CRM stuff. None of this should cause any banks to sweat too much – these sorts of numbers have been demanded of the banks by the regulators for a while. It is just publishing the disclosures rather than sending them to the regulators.

Wrap Up

All of these comments, though, are fairly preliminary. I will need to go through this in a bit more detail and give it some thought. If I have missed anything, feel free to point it ou

t in comments.