The starting point for the discussion was the reason for a covered bond at all given the funding situation of the banks both in terms of high levels of deposits and, in many cases, access to cheap intra-group liquidity. The need to reduce dependence on parental liquidity and short term and potentially volatile deposits in the interest of financial stability was emphasised. A point made from the audience was that the relatively healthy loan to deposit ratio at banks in the region was partly due to the paucity of other local fixed income investment opportunities and that covered bonds should also be recognised as a healthy development for the buyside of the market.
The roundtable then went on to discuss the rationale for covered bonds to be developed on a pan-Baltic basis. The relatively small size of the mortgage market in each of the three countries was felt to be particularly problematic for covered bonds due to investors in that market having a strong preference for larger, liquid bonds.
Representatives of all three countries briefly summarised the development of the laws in their country so far. Estonia is the most advanced with a covered bond law recently having been passed by parliament. It was acknowledged that further work was needed to ensure integration of the law with the Latvian and Lithuanian laws, to reflect the final outcome of the Covered Bond Directive and to develop certain secondary regulations. In Lithuania a law has been drafted and is currently subject to review with the intention for its submission to parliament after the summer recess. A similar time frame is being targeted in Latvia even though the law there is still being drafted. Clearly there is considerable goodwill from all stakeholders to ensure that the Latvian law progresses rapidly to “catch up” with the process in the other countries.
What is a pan-Baltic framework?
One of the key topics discussed was what a “pan-Baltic” covered bond framework actually means.
As a member of the project team explained, the objective was to have three separate covered bond laws — one in each country — but to make them as similar as is practically possible given national specificities. This was necessary as banks in the region operate with different business models — some are based in one country with branches in the other two, some operate as full subsidiaries in all three countries and some only operate in one of the three countries. For all possible permutations a prospective issuer must be able to use a covered bond law in the country in which they are supervised. Thus covered bond laws must exist in all three countries, but should be similar wherever possible and should be able to easily interact with assets in the other countries — this needs legal changes in both the country of the issuer and the country of the assets.
A lively discussion followed on the topic of whether this was necessary given existing EU rules on enforcement of creditor rights granted in another EU member state’s covered bond law. As a lawyer from the working team pointed out, there is a big difference between having the right to enforce and the practicality of actually doing this, in particular given the sensitivity and complexity that always accompanies changes to insolvency or bank resolution rules in a country.
One speaker pointed out that the design of the framework can best be understood by looking at the alternative frameworks that the project team had rejected. In response to a question from the audience it was pointed out that it would be impossible for a law in one country to make the necessary changes to the other countries’ laws. And the alternative of a bank issuing separate covered bonds in each country and pooling them in a combined covered bond — as is contemplated in the Covered Bond Directive — was quickly ruled out given the far higher cost implications in a smaller mortgage market.
Choice of model
A further topic of discussion was the choice of model within each country. Before the pan-Baltic framework had even been contemplated Estonia had adopted a “ring fence” approach for cover pools while Lithuania had decided, in common with most new covered bond frameworks, to retain the assets in a separate legal entity owned by the issuer. Latvia, after detailed consideration, has chosen to adopt the Lithuanian model. The choice of model was “not for fun”, as one speaker put it, but was based on what was actually possible within each country’s legal framework.
As the project team emphasised, and investors and rating agency confirmed, the choice of model was of more importance to structurers than it was to investors or to the eventual rating of the bonds — the objectives of the pan-Baltic framework are not compromised by the need to adopt different models in the three countries. A researcher with an investment bank confirmed that topics such as over-collateralisation, the quality of the assets and the risk protection for investors were all far more important.
Success?
This led to a discussion of what a successful pan-Baltic covered bond framework would look like. The objective is that, despite different starting points in terms of legal background, all three laws should facilitate a seamless transfer of assets cross-border and should offer both investors and issuers similar outcomes. Representatives of rating agencies and investors confirmed that such an outcome would be constitute a successful outcome for pan-Baltic covered bonds.