The ESN debate has focussed on loans to SMEs at the expense of infrastructure funding. Dual recourse infrastructure bonds are viable, they just aren’t that similar to traditional covered bonds.
Everyone pictures ESNs as being as similar as possible to traditional covered bonds. As with the migration of securitisations from residential mortgages to car loans, to credit cards it was realised that tools that mitigate credit for one granular type of asset can be applied to another, even if it is worse in terms of any given asset’s default probability.
When I was asked by the European Commission to consider ESNs in the context of loans to SMEs and loans to infrastructure projects it was clear to me that the first asset class is similar in terms of its granularity but the second was more similar in terms of its credit behaviour. Actual defaults on infrastructure projects are, according to rating agency data, far lower than actual defaults on SMEs and not much greater than losses on residential mortgages.
But the vast majority of people that I spoke to whilst conducting my study thought that maybe SME ESNs would work, but definitely project finance ESNSs wouldn’t. The EBA were certainly of that opinion in their study on the same topic.
They are of course correct if, and only if, you take a fairly narrow definition of the covered bond model; the current definition. If you had applied the definition of covered bonds twenty years ago – before the Directive, before the EBA’s best practice principles, before the supervisory and structural convergence that we have seen between markets – project finance ESNs would have been more palatable to the average covered bond banker.
Why do I say that?
I think that the industry developments have been driven by the granular residential mortgages that have driven its growth, particularly in the newer jurisdictions. The convergence has not been driven by commercial mortgages - which could be argued to look a bit like project finance loans, given their frequent lack of granularity and line-by-line due diligence needs. Convergence has certainly not been driven by the public sector or ship loan covered bonds which share these characteristics.
But if you strip the idea of covered bonds back a generation we get to a supervised, dual recourse instrument backed by credit worthy assets that are important to wider society. That definition is a lot more compatible with the idea of project finance loans.
As covered bond markets have converged, they have improved; as their issuers have been downgraded, the rating uplift has increased. Hence the relative stability in the universe of covered bond issuers.
The lumpier assets, in particular commercial mortgages are also still able to reach top credit ratings but this is only tested in the older, more established markets with strong legal frameworks, such as Germany.
Which fact underlines one of the objections to project finance ESNs – that they wouldn’t be able to get to AAA in many cases. Quite possibly not but do they need to? No, from an investor perspective, obviously. But less obviously, no from a market identity perspective. It is still a covered bond if it can only get a four notch upgrade.
And I don’t say that the directive’s structure, the EBAs rules or national supervisors mindsets are necessarily appropriate for this asset class. But I don’t think that any of those are necessary for the concept to work.
Another thing that I discovered, chatting to project finance experts as part of the study, was that a lot of the lending to European infrastructure projects comes from non-European banks. For that reason alone it seems appropriate to consider project finance ESNs more carefully before dismissing them as too different to be viable.
Event, Articles and Videos that might interest you
No related events !