Non spreadsheet risks

06 Jul 2015 | Richard Kemmish


For me one of the dominant themes of the Global Borrowers and Investors Forum that has just finished was a growing awareness of what one panellist described as non-spreadhsheet risks to the financial system (in general) and your investment (in particular). Non-spreadsheet risks can probably best be defined as non-quantifiable events and it seems odd how little attention they get given their track record.
 
One of my favourite stories about ‘spreadsheet risk’ in the securitisation market came from a rating agency. Their exceptionally elaborate spreadsheet model (that had of course been signed off by the rocket scientists) assumed that the maturing mortgages could be refinanced. The spread for this refinancing – and hence the proceeds for a given pool of mortgages – was modelled based on the last ten years of the securitisation market which had more or less conformed to a normal distribution. The AAA scenario stress test was a 2 standard deviation move in this variable: 10 basis points ‘to be on the safe side’.

When spreads widened by 400 basis points some bright statistician calculated that, assuming that refinancing spreads were normally distributed, this represented a 1 in 1.5 billion year risk. Perhaps that normal distribution assumption was flawed.

Could it be that this same look-at-the-spreadsheet-only mentality underscores some of the incredulity that some market participants display when a regulator adopts a conservative stance towards certain products?

Should we be worried about the growing use of spreadsheets in the covered bond market? In particular the growth of sophisticated models by regulators, an increasing number of whom feel the urge to second guess the work of the rating agencies? Even in Germany, where you would expect this sort of approach least, the ability of the Bafin to set issuer by issuer minimum collateral levels presumably encourages some issuer by issuer spreadsheet modelling?

I’m not so concerned about it for two reasons. Firstly it is an add on. In all of the established countries many ‘non-spreadsheet’ safeguards have been put in place – such as a requirement for a catch-all sign-off by senior management. Spreadsheets are not the first line of defence. Countries without a history of supervisory oversight need to avoid an over-reliance on quantativie models, particularly if they are jurisdictions with a long tradition of ‘quant-heavy’ securitisations.
 
Secondly, simply, the dynamic nature of the pool. Non-spreadsheet risks can often be addressed easily in the context of a revolving pool. For example, the forced redenomination of Swiss franc denominated mortgages at off market rates in some central European countries would never have been predicted by hitting the F9 key. But it could be addressed in covered bond pools. It couldn’t have been in static pools.

But a healthy level of scepticism towards the outputs of a spreadsheet is not the same as a proper understanding of what the non-spreadsheet risks might be. If the regulators and the rating agencies are busy quantifying the quantifiable, the rest of us can concentrate on being more imaginative about what might actually go wrong.


Go back to the Blog Homepage

Contact the author at covblog@euromoneyplc.com

Any views or opinions expressed in this blog are those of the writer, Richard Kemmish, and not those of Euromoney Conferences. The opinions expressed are done so in the spirit of stimulating open debate. This blog does not constitute investment advice. Links, sources and information published are subject to change and may not be accurate or valid over time. All comments, presentations and questions on this blog are the sole responsibility of the individual who makes them. Individuals are strongly advised to familiarise themselves with their own corporate, regulatory and institutional guidelines.