Bill Thornhill, writing in GlobalCapital, has argued that ESNs are not about improving funding for banks but about ring-fencing the concept of covered bonds. I respectfully disagree.
As a starting assumption, it sees reasonable to say that covered bonds – let's call them ‘traditional covered bonds’ – will always be a better funding option than European Secured Notes. Investors will demand less spread, rating agencies will demand less over-collateralisation. Given that there are €5 trillion of residential mortgages in Europe that aren’t in cover pools, the banking industry can issue a whole load of much more efficient traditional covered bonds before it has to trouble itself with ESNs.
Whilst that might be true in aggregate, it ain’t necessarily so at the level of individual banks or even, individual countries. Let’s look at some of the times that it might not be true.
Firstly, I hate to say it but, some covered bond regimes are less than perfect. It might be easier to upgrade a shoddy covered bond law than to write a new ESN one, but it might also be impossible with too many vested interests, sunk costs and bonds outstanding. For example, those regimes with high barriers to entry are supported by those who have already paid that barrier – and they are the loudest voices in the debate – but opposed by smaller banks for whom those costs are a very real concern. ESNs could be differently structured to open up secured funding to a wider issuer base.
Then there is the problem of mortgages. It’s a commonplace in most of Europe that house prices are excessive. The only debate is whether they are sustainably so. A house price crash could have three effects that could promote the development of ESNs: it could drastically reduce the collateral available for traditional covered bond programmes; it could hit the health of the banking system forcing issuers to favour secured over unsecured issuers and it could increase investor preferences for secured debt that isn’t as correlated to house price indices as the rest of their portfolio.
Even without a house price crash, mortgages can overnight become poor quality collateral by misguided (but popular) governmental policies. The introduction of non-recourse (datio in solutum) rules or the redenomination of foreign currency mortgages at off market rates are just two recent examples. It isn’t difficult to imagine others.
Then there are factors at the level of the individual issuer. Not every bank has large residential mortgage portfolios, some of them have constraints – IT or legal for example – on pledging them, some of them have already pledged everything they have to covered bonds used as repo collateral.
But the ultimate proof of my point is empirical. As Bill rightly points out the spread compression that we have seen has reduced the potential cost saving from ESNs. At the same time the elevated levels of liquidity in the banking system have reduced the need for such funding.
And yet, even in this unpropitious environment for ESNs, we have seen loans to SMEs used as collateral for secured funding – most famously by Commerzbank, but also in securitisations that are used as repo collateral (thus becoming de facto dual recourse secured debt), in central bank facilities and in Turkish covered bonds.
Imagine how much more of this type of funding we will see in less benign funding environments going forward. Not tomorrow, admittedly, but legislation should look further than that, to funding conditions in 5 or 10 years.
If the ESN market is going to develop anyway it seems sensible to put in place safeguards now to help the market develop in a harmonised and effective way.