CRD IV: A closer look at items relevant to Nordic issuers

After publishing a revised CRD IV package on 26 March, the European Parliament finally passed the CRD IV package that comprised the Capital Requirements Directive (CRD) together with the Capital Requirement Regulation (CRR) on 16 April.

Florian Eichert
Senior Covered Bond Analyst, Crédit Agricole CIB

While CRD IV is obviously crucial for all banks in Europe, there are some items that are in our view especially relevant for Nordic banks:

• Which LCR category will Nordic covered bonds fall into, especially considering that sovereign debt markets are not big enough to cover LCR asset needs in some Nordic countries?

• How will the specialised covered bond issuers be able to deal with the LCR rules?

Timing of CRD IV implementation
The timing of CRD IV implementation was anything but certain until the last minute. Had the European Parliament not been able to pass the CRD IV package as early as they did, implementation could have easily been pushed well into 2014. Since they were able to agree on the text earlier than some might have feared, though, the CRD IV package will be in force from 1 January 2014.
With regards to covered bonds, only some of the elements — such as the new way to calculate covered bond risk weights or the enlarged list of eligible assets for preferential risk weight treatment — will come into effect on that date, though. Others, such as the LCR, will only be phased in from 2015 onwards, while the European Banking Authority (EBA) has a completely different set of relevant dates:

• The EBA will, for example, have until 31 January 2014 to decide on uniform definitions of high (level 2 in the Basel III language) and extremely high (level 1 in Basel terms) liquidity and credit quality for the LCR. This is when and where covered bonds’ treatment will be decided.

LCR: Level 1 vs 2 assets
We are still waiting for the official version of the CRD IV package document. And strangely enough there is still considerable uncertainty around the level 1 versus level 2 split.
The document that was published on the Parliament website didn’t have a 40% cap for level 2 assets in it. Should the 40% cap on level 2 assets really turn out to have been taken out, the work that is now to be undertaken by the EBA would lose much of its relevance as investors could in fact buy as much of an asset class as they wish irrespective of its categorisation. We are just not sure yet as this seems too good to be true and too material to have escaped everyone’s attention.
Irrespective of this decision, there will be differences between level 1 and 2 assets, though, which will always be an incentive to buy level 1 assets as well:

• Level 1 assets will not be subject to a haircut, whereas level 2 assets will receive a haircut of at least 15%;

• Level 1 assets do not have to be diversified, and there are no size or concentration limits. Level 2 assets, on the other hand, have to be diversified. What exactly this means is not yet clear and probably also in the mandate of the EBA;

• Level 1 assets also do not have to be sold to prove their worth. On the other hand, banks will have to sell a certain portion of their level 2 assets each year to prove that these assets can be liquidated. This obviously means transaction costs.

Where will covered bonds fit in, 1 or 2?
In the opening remarks of the CRD IV package, the text states the objective of the CRR to be for banks to hold diversified liquidity buffers:

• “As ex ante it is not possible to know with certainty which specific assets within each asset class might be subject to shocks ex post, it would be appropriate to promote a diversified and high quality liquidity buffer consisting of different asset categories”

• “When making a uniform definition of liquid assets, at least government bonds and covered bonds traded on transparent markets with an ongoing turnover would be expected to be considered assets of extremely high liquidity and credit quality”

Obviously it sounds great for covered bonds to be named in the same sentence as sovereign debt and be given the chance to move into level 1, which is something that is not possible under the current Basel proposals.
In our view this is, however, something that Danish and Swedish domestic covered bond markets would have the best shot at for now, as they can simply boast the biggest amount of trading related data, which is what the EBA is looking for.
Which bucket covered bonds ultimately end up in is one of the major decisions that has been passed on to the EBA. The technical standards that the EBA is currently drafting have to be presented to the Commission one month after the implementation of the CRR, which means we are talking 31 January 2014.

What would level 1 mean for Swedish covered bonds, for example?
Should, for example, Swedish covered bonds make it into level 1, it would certainly have an impact on investment choices as well as issuing behaviour. Sweden is the only country to date that has already implemented an LCR. Covered bonds are, however, treated as level 2 assets and many Swedish treasuries have also bought non-Swedish government debt from other European countries or the US to fully comply with the ratio.
Should Swedish covered bonds make it into level 1, however, bank treasuries from Sweden would certainly be incentivised to increase the weight of covered bonds as they still offer some 70bp-80bp more than Swedish sovereign debt. Ultimately it would mean that Swedish issuers will cover much of their covered bond related funding needs in Swedish kronor while using euro markets mainly for senior unsecured deals, which are not LCR eligible and thus do not enjoy the same backing as the covered bonds. Structural euro covered bond issuance could very well be reduced to one deal per issuer per year simply to show they are still there.
We would in that case be talking about a structural supply per year of maybe Eu5bn-Eu7bn, which is clearly well below years like 2010 and 2011, when supply from the country was easily in excess of Eu10bn.

LCR and the inflow cap at 75% — there will be covered bond-specific exceptions, which is brilliant news for many Nordic issuers
Previously all LCR inflows were capped at 75%. This represented a big problem for specialised bank covered bond issuers (French, Norwegian, Swedish, some Finnish, and to some extent Irish issuers), or for the Danish pass-through model.
For example, a Norwegian issuer has mainly mortgages on its balance sheet, which are funded via covered bonds. Capping the inflows on the mortgage side at 75% could have meant that to comply with the LCR, issuers would have had to build up a sizeable treasury portfolio, which in many cases could have probably been funded via additional covered bonds but would have altered the cover pool and made it more diverse and harder to analyse. Investors who want to buy mortgage covered bonds are buying them to get exposure to the mortgages and not to sizeable securities portfolios.
Luckily the final CRR version now includes an exception to this inflow cap for, amongst others, the following inflows:

• “Inflows from monies due from borrowers and bond investors related to mortgage lending funded by bonds eligible for treatment as covered bonds”

This should be great news above all for the affected covered bond issuers as their models still work and do not require significant additional securities portfolios just for the sake of complying with the LCR.

Florian Eichert
Senior Covered Bond Analyst
Crédit Agricole CIB