The tests that will decide covered bonds’ eligibility for liquidity buffers under CRD IV are still very much up in the air, a panel discussion at an ECBC plenary in Munich last week made clear, with qualitative measures firmly in the mix of potential eligibility criteria.

Liquidity requirements in the form of a Liquidity Coverage Ratio (LCR) under the CRD IV regulatory package are a key focus for the covered bond industry, which in a draft ECBC position paper on the subject argues that the asset class should be considered to feature “extremely high liquidity and credit quality” and therefore count towards banks’ liquidity buffers without any restrictions (as a Level 1 liquid asset).

A requirement that qualitative criteria be taken into account when determining eligibility for bank liquidity buffers was not in the European Commission’s (EC’s) CRD IV proposal, but introduced by the European Parliament’s Economic & Monetary Affairs Committee (ECON) in its version of CRD IV – a move welcomed by the ECBC.

ECBC plenary, Munich

Providing an update on the latest developments concerning CRD IV for a panel on the subject at the ECBC plenary last Wednesday (12 September), Fritz Engelhard, German head of strategy at Barclays, noted that ECON put forward more than 1,600 amendments to the Commission’s proposal, which were dealt with in a three hour meeting – equating to a rate of 6.5 seconds per amendment. The legislative file has entered an open-ended Trialogue process, he added, with a decision expected by the end of October.

The official timetable is for the prudential requirements set out by the CRD IV package – a directive (CRD) and regulation (CRR) – to come into effect 1 January 2013, with a technical report due mid-2013.

Under the EC’s CRD IV proposal the European Banking Authority (EBA) has the mandate to come up with the criteria by which an asset class’s eligibility for liquidity buffers will be tested, but CRD IV texts from the other bodies involved in the EU legislative process provide for the European Central Bank and the European Securities & Markets Authority (ESMA) to be involved.

Lars Overby, policy expert at the EBA, said that the EBA is coordinating with other bodies. He resisted other panellists’ efforts to glean information about the criteria that may end up being chosen as determinants of an asset class’s eligibility for liquidity buffers, noting that the EBA is carrying out an impact assessment of Basel III implementation and that it is “important for us to look at the data and understand the impact of the LCR”.

In relation to the quantitative factors that the EBA is charged to consider, he said that data availability is not as bleak as it appears from the covered bond market’s perspective, and emphasised that the EBA is focussed on collecting data and forming an overview, and that it also wants to take into account qualitative measures.

“It is too early to say anything about concrete proposals,” he said. “The statistical evidence is very useful but there are many other aspects to take into account.”

Quantitative criteria that the EBA is due to take into account in proposing how liquid assets should be determined include turnover ratios and other trading information.

Richard Kemmish, head of covered bond origination at Credit Suisse, said that while comprehensive data is not available, empirical evidence of bond trading “matters less than we think” because data can be manipulated, and that the inclusion of qualitative measures in the factors that are to be reviewed was recognition of the limits of empirical data.

Florian Eichert, senior covered bond analyst at Crédit Agricole, noted that a minimum rating requirement of AA- no longer appears in the latest compromise CRD IV proposal by the Danish Presidency, and asked what other qualitative criteria could be chosen as tests of liquidity buffer eligibility.

Suggestions put forward by Barclays’ Engelhard included the level of transparency – although he said he does not observe much differentiation between good and poor transparency – and the use of “intermediate steps of the rating process” rather than final ratings.

The EBA’s Overby said that covered bonds’ eligibility for repo with central banks is an important factor, but that collateral frameworks differ across national central banks and that it is not clear this is “the correct solution”.

Danes fight potential LCR ‘absurdity’

Danish financial institutions have been especially concerned that their covered bonds may not be designated Level 1 assets, which in contrast to Level 2 assets are not subject to caps or haircuts.

Kim Laustsen, chief analyst at Nykredit Realkredit, said that Danish covered bonds are as or more liquid than government bonds, noting that secondary market trading is very transparent and continued during the financial crisis, with bid-offer spreads widening less than those on government bonds, and that the pass-through system in Denmark creates liquidity.

Given a small stock of government bonds the eligibility of Danish covered bonds as Level 1 assets is crucial to Danish banks’ ability to comply with the LCR, said Laustsen.

“We would consider it absurd if they were not included,” he said.

Overby at the EBA noted that the Basel III framework foresees alternative ways of meeting liquidity buffer requirements for banks in jurisdictions with insufficient liquid assets, including the option that central banks offer credit lines, as is the case in Australia, and that Level 2 assets still count towards liquidity buffers, albeit with haircuts and a cap on their share of total liquid assets.

In Australia the Reserve Bank will establish a committed liquidity facility (CLF) that authorised-deposit taking institutions (ADIs) can tap for liquidity in exchange for a 15bp fee and pledged collateral. The Australian financial supervisory authority, APRA, will recognise the central bank’s commitments to ADIs for the purposes of compliance with the Basel III liquidity standards.

But such routes to fulfilling LCR requirements are not satisfactory in the Danes’ eyes, with Laustsen pointing out that they come at a cost and are therefore arguably not a fair solution for jurisdictions with too small a stock of sovereign debt for the purposes of complying with liquidity buffer requirements.