US finalises LCR rules, another EC leak

The completion of the Liquidity Coverage Ratio (LCR) rules among various countries is proceeding, with the US releasing its final policy. On the euro front, another document was leaked, with most parameters remaining the same.

Covered bonds not a part of US rules

To begin with the US, the Federal Reserve Board, the Federal Deposit Insurance Corporation, and the Office of the Comptroller of the Currency presented their final version of their LCR rules on 3 September. The rules are broadly in line with the Basel ones when it comes to maximum exposure to Level 2A and 2B assets. However, there are two fundamental differences:

• Phase-in happens faster, as US banks have to comply with the LCR to 100% by 1 January 2017 — so a full two years earlier than foreseen in the Basel rules.
• The scope of eligible assets for the LCR will be tighter than in the Basel rules and certainly tighter than what is currently being discussed in Europe. Covered bonds and private label securitisation are not part of the US rules. Not even muni bonds are — regulators, however, mentioned that they’re at the moment working on a proposal for future consideration

The rules will be applicable to all banking organisations with total consolidated assets in excess of $250bn (Eu193bn) or total consolidated on-balance sheet foreign exposure of $10bn or more, as well as any subsidiary depository institutions with $10bn or more of total consolidated assets (further described as “covered institutions”), while all banks have to comply with it in Europe.

However, covered bonds will not be recognised as HQLA in any of the three LCR categories. As far as we are aware, the US is the only region where covered bonds are not eligible as HQLA for the LCR. In contrast, Basel rules allow AA- and better rated covered bonds to be in level 2A. Countries outside Europe such as Australia or Canada, have taken this approach. In Europe, we are even looking at an extra category (1B) that covered bonds can fall into.

Covered bonds had never been on the agenda as potentially LCR-eligible assets in the US. The agencies argue that the domestic covered bond market is not highly developed and therefore covered bonds don’t meet the standards they set in terms of liquidity.

Europe: another month, another leaked LCR document…

Another LCR document of the European Commission was leaked at the beginning of September. The treatment of covered bonds in this new version contains many familiar items and the overall treatment of covered bonds has not changed compared to previous versions. But there have been some changes in two areas that are rather important. Below we want to highlight the timeframe again, and then go into as much detail as we can at this point on the similarities and differences between the prior document and this one.

Timeline

While the US has this week finalised its LCR rules, we are still waiting on the final version in Europe. The timeline still foresees the final version by end-September, while the LCR is supposed to be implemented from October 2015. The latest document seems to confirm this. So while the US is sticking to the Basel starting date of January 2015, Europe is delaying it by 10 months.

What are the similarities between the last two documents?

Covered bonds still seem to be part of Levels 1B and 2A. Also, the haircuts seem to have remained the same, at 7% for Level 1B and 15% for Level 2A. Transparency requirements from 129(7) CRR are also retained in the current document. Overall, the broad parameters are unchanged, which is good news for covered bonds.

What are the differences?

The differences between the two documents relate to two new additions in the July document:

• Requirement for cover pools to be composed of “homogenous assets”.
• Ban on exposure to institutions.

Both items are nowhere to be seen in the newest document and it is good that they are gone.

The homogenous asset requirement is something the EBA already mentioned in its covered bond report and during the covered bond hearing earlier this year in London. It wanted to ensure investors know what they buy, but more importantly can have a high level of certainty that the pool they have bought today does not change dramatically over time (from residential to commercial mortgages, for example).

We sympathise with this thinking and certainly prefer straightforward and easy to understand cover pools, but implementing hard limits can cause quite some problems for issuers even if they are trying to do all they can to the benefit of investors. Think of liquid asset requirements, for example, in a cover pool with only one remaining bond outstanding. The moment the maturity is less than six months away, issuers have to hold the outstanding bond notional in liquid assets. So even if they leave all of the mortgages in the pool, they will have to top up the pool with a lot of cash and other liquid assets, reducing the mortgage share dramatically.

More importantly, however, the ban on exposure to institutions is gone as well. Covered bond issuers do hold liquidity with other banks, typically in the same institution for the universal bank covered bond model or their sponsor banks in case of countries such as France or the UK. And European regulation allows them to do this as exposure to institutions is explicitly allowed up to 15% of the outstanding covered bonds’ notional. Article 129 1c states:

exposures to institutions that qualify for the credit quality step 1 as set out in this Chapter. The total exposure of this kind shall not exceed 15 % of the nominal amount of outstanding covered bonds of the issuing institution. Exposures to institutions in the Union with a maturity not exceeding 100 days shall not be comprised by the step 1 requirement but those institutions shall as a minimum qualify for credit quality step 2 as set out in this Chapter

Allowing exposure to institutions in one piece of legislation while banning it in a related one would have been rather strange. The current LCR draft still mentions exposure to institutions but harmonises the requirements. Banks can hold exposure to institutions as long as this exposure complies with 129 1c. As this rule has been in force for a long time already we do not see any problem here.

What could still happen?

We are still some time away from the end of September and a few weeks can be a long time in Brussels. We thus would not rule out that yet another document is leaked in the coming weeks and, just to be on the safe side, we have reserved some extra space in our Excel file containing the table above.

According to The Covered Bond Report, for example, there is recognition among the European Commission that an appropriate alternative to external ratings should be explored. What that means exactly, which alternatives we could be talking about, whether it would apply to 1B and 2A and could thus allow covered bonds with certain quality characteristics to get up to 1B irrespective of their external ratings, we do not know. It does sound like a complete U-turn versus the June document, though, where not only external covered bond ratings but also issuer ratings were proposed as minimum standards.

Bottom line for us is that LCR regulation is and will remain very supportive of covered bonds. Bank treasuries will thus continue to have a big incentive to look at the market to cover their liquidity needs. And as sovereign debt yields start moving into negative interest rate territory further and further out the curve, 20bp-25bp covered-sovereign spreads become ever more attractive. The big problem will rather be whether they can source any paper. After all, they will be competing with a rather strong-minded and determined Italian for the next three years.