image missing on blob storage!!!

A worrying obsession with small details

23 October 2014
Richard Kemmish

Growing up in Birmingham in the 1980s my education involved learning a lot of things that are important in a career making cars (and nothing about financial markets). But just as you never forget the firing sequence of a straight-6 cylinder block (1-5-3-6-2-4, since you ask) you never lose that obsessive interest in the minutiae that make machines work.
Which could explain my unhealthy obsession with a detail of covered bond structuring that makes normal people (defined, for these purposes, as non-Brummies) suddenly want to change the subject: the Asset Cover Test.  

The ACT, was invented in 2003 by Allen and Overy and by my estimate is now embedded in 45% of covered bond programmes. People’s reluctance to engage with the details is understandable, it’s description typically stretches to about five pages of gratuitously complicated legal drafting.

What it does in a nut shell is to determine the number of assets you need in the cover pool. It says that the collateralisation should take into account both the quality of the assets and a cushion for potential structural problems, both of which are re-calculated every time you run the test. The alternatives are to rely on a hard-coded value in the covered bond law or covenant, this is obviously a blunt tool. Or to leave the amount of collateral up to the regulator’s discretion – as the latest iteration of the Pfandbrief law does. This probably works as well as the ACT in those countries where the regulator can be trusted to protect covered bond holders, but...

Part of the ACT’s (unnecessary) complexity is that it runs both the ‘how much do you need?’ test and a separate ‘LTV cap’ test to replicate those caps used in ‘traditional’ covered bond jurisdictions. Strictly speaking if the ‘how much do you need?’ test works, the LTV caps are redundant from a credit perspective, they are only in there to make the bonds imitate traditional covered bonds and therefore conform to EU law. 

If you don’t have the time to understand the whole ACT, then understand the calculation of just one term of it: the Asset Percentage. Adjustments, modifications and floors aside, the AP is the overcollateralization ratio that you need (or at least the reciprocal of the ratio) and typically refers to the value that the rating agencies have calculated as being the minimum to achieve the then highest achievable rating under their methodologies.

To over-simplify, the AP is a commitment on the part of the issuer to do – like Mario - whatever it takes to maintain the rating. That’s why it is so powerful.

But here is the problem, the spread of the ACT is now so prevalent that investors expect to see it. There is a risk that the concept gets diluted and ‘asset percentage’ used to refer to either fixed values (X%) or numbers that aren’t defined as strongly as ‘whatever it takes’.

EBA, by all means standardise pool reporting or asset definitions but please, please set a common understanding of what an ACT is, it’s the single most important component in the covered bond credit engine.

Event, Articles and Videos that might interest you

No related events !

Euromoney and the ECBC are heading to Vancouver

15 January 2018 |

Later this year, we head to Vancouver for an event focused on the North American covered bond market.

Singapore’s 4% isn’t Canada’s 4%

15 January 2018 |

On a superficial level, Canada and Singapore don’t have a lot in common. But in covered bond land we look beyond the superficial (usually) and can see some remarkable market, regulatory and structural similarities.

Tags: Covered Bond