image missing on blob storage!!!


10 November 2014
Richard Kemmish

Food, culture and history are all traditional measures of the rich diversity of our European continent. But in my opinion their importance is wildly overstated and in terminal decline. European opinion on Irish pubs, Mozart, Tapas bars and the first world war are now more or less totally convergent from Helsinki to Lisbon.

In contrast, there is immense (and in some cases growing) divergence in the single most important building block of the financial wellbeing of most Europeans: housing. The mortgage market relative to GDP in Holland is four times larger than that in culturally similar Austria. The percentage of people who rent rather than own their own homes in Germany is twenty times that in Romania.

Such fundamental differences in the role of the housing market obviously has many consequences including both covered bond regulation, the single supervisory mechanism and the dividing line between them (a topic I will return to in a later post). For the mean time though, what do these differences say about the credit worthiness of the bonds themselves?

The practice in some countries of capping the exposure of the cover pool to certain asset classes, buy-to-let mortgages for example, is predicated on a view on the riskiness of that asset class. Buy-to-let mortgages in a country with a high home ownership percentage – such as the UK - are marginal to the mainstream of housing finance, therefore more likely to be risky. That ‘buy-to-let is riskier than owner occupied’ is taken as a given reflects the fact that the credit market is broadly defined in places where that is true. In a country with a low home ownership percentage, where even the wealthy rent, the credit is very different.

Differing risk also applies to the LTV ratio. Countries with higher home ownership and ‘anglo saxon’ style mortgages (that is, fully amortising over the borrower’s expected working life) are the norm in rating agency models and are predicated on the need to minimise state pension provision (as housing costs are eliminated from the household budget when the mortgage is paid off, usually roughly at retirement). But what of those countries where interest only mortgages are normal as borrowers save for their pensions in a separate vehicle? Economically the same but with very different credit implications for the SPV owning the mortgage (but not the pension) and for the rating agency models.

Then there is the macro-prudential regulation of the issuers. The regulatory risk weights of mortgages vary enormously across Europe (and bear no correlation to the riskiness of the same assets as calculated by the rating agencies), could the agencies and regulator’s different perceptions be explained by the extent that the housing market diverges from the agency model base case?

Or encumbrance limits, there is a correlation between the encumbrance of the banking system and the size of the residential mortgage market. Or the probability of systemic support? Or attitudes towards non-residential assets in covered bond legislation?

Personally I love the diversity of mortgage markets in Europe and I’m only scratching the surface of its implications.

Event, Articles and Videos that might interest you

No related events !

Ring fencing

23 February 2018 | Richard Kemmish

Is the ringfencing of UK banks a good thing for covered bond creditors? Yes, but with some fairly significant ‘buts’.

Tags: Covered Bond

How many covered bonds next year?

23 February 2018 |

Predictions of next year’s covered bond issuance volumes, never particularly accurate, aren’t even considered important now. We might be wrong to dismiss them totally.

Tags: Covered Bond