Asset encumbrance: discussions gaining momentum?

Asset encumbrance is one of those topics that everyone talks about but where discussions hardly ever go beyond very shallow, general statements.

In recent weeks there has been a lot of talk — especially from the Nordic countries, but CRD also mentions the issue. Below we sum up the status quo and give a few thoughts as to where things could go.

A role in risk weights under CRD IV?

Some members of the European Parliament have long been very critical of the increasingly important role covered bonds play and if they had their way we would already be looking at a pan-European covered bond issuance limit. Fortunately they have not succeeded so far and proposals for a limit didn’t find their way into the CRD IV text.

The more general topic of asset encumbrance has still found its way in there, albeit in a much less concrete and threatening way. Article 478 states:

• “The Commission shall, by 31 December 2014 and after consulting the EBA, report to the Parliament and the Council, … , on whether the risk weights laid down in Article 124 … are adequate for all the instruments that qualify for these treatments and whether the criteria in Article 124 are appropriate.”

• “The report and the proposals … shall have regard to: (c) the extent to which covered bond issuance by a credit institution impacts on the credit risk to which other creditors of the issuing institution are exposed.”

There is thus a chance that, going forward, CRD IV could link covered bond risk weights to asset encumbrance levels. It is at this point still fairly vague, though, and we’re definitely not looking at implementation in the coming two years.

Case-by-case consensus grows in Norway

There has been a lot of talk about asset encumbrance in Norway since last autumn. The discussion went from imposing hard limits on covered bond issuance to increasing risk weights for mortgages to 35% to imposing additional qualitative limits on which mortgages can be transferred to covered bond companies, thus limiting future issuance rather than imposing a hard cap.

The Ministry of Finance had last December asked the Norwegian central bank and the FSA (Finanstilsynet) to state their case and both have replied in public letters:

• Norges Bank: Limits should be considered on funding mortgages via covered bonds, but these limits should take into account any other measures that are currently being undertaken to reduce the overheating in Norwegian mortgage markets and protect the Norwegian deposit protection fund. These could include bail-ins of senior unsecured debt, higher fees payable into the deposit protection fund, as well as in general better disclosure. Norges also highlighted that there are significant differences between individual Norwegian banks’ usage of covered bonds.

•  Norwegian FSA: Is thinking about qualitative rules on shifting loans to covered bond entities, but oversight should be done on a case-by-case basis. Banks should draw up their own guidelines for the responsible use of assets as collateral and should run stress tests. The FSA will analyse these plans on a bank-by-bank basis.

Florian Eichert
Senior Covered Bond Analyst, Crédit Agricole CIB

The prevailing opinion thus seems to be that there will be no hard issuance cap for now. Banks are merely asked to be more transparent and prudent about using covered bonds as a funding tool. The Norwegian authorities will pay close attention to this and look at each case individually, taking into account the bank’s overall situation. At the same time, higher risk weights for mortgages as well as additional qualitative restrictions (like for example lower maximum LTV limits for mortgage eligibility) are supposed to calm down the mortgage market.

In fact this approach isn’t all that new to covered bond markets. Both the UK and the Dutch authorities take a similar case-by-case approach. The Dutch FSA stress tests covered bond programmes as well. And both the UK and Dutch authorities can impose issuance limits with the actual level differing from institution to institution based on overall capital levels and the quality of unencumbered assets.

Our take

The discussion about asset encumbrance will stay with us for the foreseeable future. By the sheer nature of things, unfortunately, covered bonds will always be an integral part of these talks as they are the most obvious and transparent source of encumbrance.

It won’t come as a surprise to anyone if we say that the strong focus on covered bonds in this regards is overdone as they are by no means the only source of encumbrance.

• If, for example, the share of central bank funding of the Portuguese banking system is around 10%, as was the case for most of 2012, it is probably a fair guess to say that depending on the assets used and haircuts applied somewhere between 10% and 15% of Portuguese banks’ balance sheets are encumbered for central bank purposes alone.

• Looking at end-2012 figures for covered bonds, the roughly Eu34bn of outstanding Portuguese covered bonds and Eu46bn of cover pool assets make up around 6% of banking system assets when looking at covered bond volumes and 8% when looking at the cover pool data — clearly less than central bank-related asset encumbrance.

In our view the main motivation behind the focus on asset encumbrance by regulators and politicians is to protect deposit protection funds (DPF), which in case of a bank default would probably be the single biggest senior unsecured creditor of a defaulted bank. Wholesale senior unsecured investors are hardly at the centre of attention here; rather the opposite if one thinks about bail-in discussions.

There are two ways to better protect taxpayers’ money — lower the probability of a default (PD) and/or lower the loss given default (LGD). Asset encumbrance clearly affects the latter since the more assets there are that are encumbered, the fewer there are that are left for the DPF. Focussing on those two components, we can say the following:

• Covered bonds actually help to lower the PD, as they provide stable long term funding. They are thus also helping to protect the DPF and ultimately taxpayers’ money.

• On LGD, you can achieve a better position for the DPF by either limiting encumbrance or by increasing the buffer of those parties that are taking a loss before it’s the DPF’s turn.

As mentioned, covered bonds reduce the PD of an institution. And while collateralised funding clearly increases the LGD, measures such as increasing capital levels, introducing write-downs of sub debt and ultimately all the way up to senior unsecured debt increase the buffer that sits below the DPF and counters some of the increased risk coming from encumbrance.

In our view, the strong focus on asset encumbrance is politically fashionable at the moment but overdone. Collateralised funding in general and covered bonds in particular add stability while there are already measures being undertaken that limit the LGD in a resolution scenario by increasing the buffer that sits below the DPF.

Covered bonds in particular have been put in the spotlight as they are the most obvious source of encumbrance there is and the product’s success has certainly been a thorn in the side of many outside the market. However, we don’t think that we will see any hard issuance limits at a European level anytime soon. The CRD text is merely talking about potentially adjusting risk weights at some point after 2014.

There might, of course, be measures taken at a national level before that. However, the most advanced countries in this regards are Norway and Sweden and if we end up having a Norwegian style regime where there is more transparency on asset encumbrance (and not just that coming from covered bonds) coupled with case-by-case decision making about the use of covered bonds by individual banks that takes into account specific circumstances of a given issuer rather than having a “one size fits all” issuance cap, I think we can actually live with that quite well.

Something that could, however, add quite a bit of heat to the discussion and harden the stance on the part of regulators to the disadvantage of covered bond issuers is if we continue to see more and more new asset types being used as collateral for covered bonds.

Florian Eichert

Senior Covered Bond Analyst

Crédit Agricole CIB