Covered bonds do exactly what they say on the tin

On September 28 the only publicly outstanding Greek benchmark covered bond matured. A day earlier the last outstanding transaction issued by the now defunct US mortgage lender, Washington Mutual, was also repaid on schedule.

The €1.5bn Greek deal was issued in September 2009 but three years later, when the Greek sovereign debt crisis was at its nadir, the issuer conducted a cash tender offer in which it bought back €643m notes at 70 cents. This represented a considerable 15 cent premium to where the bonds had been trading in the secondary market and was about twice as high as where Greek sovereign bonds were trading.

In the end sovereign private sector investors ended up incurring great losses but in the case of the covered bond they received full and timely repayment. Both the US and Greek deals should support confidence in the covered bond market, especially when it comes to weaker banks from weakest jurisdictions.

The positive outcome for Greek covered bond investors also calls into question the wisdom of sovereign rating ceilings which for the most part are capped at B- in the case of Greece.

The Greek country ceiling is driven by the imposition of bank transfer restrictions since July 2015, which represents a de facto imposition of capital controls.  These controls means it is illegal to take out €100,000 from a Greek bank and put it into a German bank, for example.  And, as long as these capital controls remain in place rating agencies are likely to maintain their approach to sovereign rating ceilings.

However this approach is at odds with the timely redemption and coupon payments of NBG’s covered bonds. It is evident that in this case the capital controls had no influence whatsoever. The covered bond was allowed to make payments to the paying agent under the covered bonds issued. Despite this, rating agency methodologies assume a sovereign default is affectively akin to a covered bond default. 

Technically the Greek government defaulted and yet it didn’t force any private sector banks into insolvency. A failure of the government does not therefore necessarily constitute a failure of the covered bond. This was also the case for Icelandic government debt and covered bonds, though they were issued in a different currency.

Washington Mutual was a completely different beast as the bank failed and defaulted on its senior debt. The concept of transferring the cover pool along with the overcollateralization to a solvent bank has happened on many occasions in Germany. But in this case the transfer to JP Morgan occurred in a country where covered bonds were not systemically important.

Under such circumstances, where there is less history and use of covered bonds, the rating agencies widely assume that the government is less likely to step in and sort of the muddle and they factor this into their assumptions. However, perhaps this is too cautious — covered bonds could scarcely be less important to US banks, and yet the transfer went without a hitch.

But overall there is another even more important story. Looking through the failures of covered bond issuing banks, 20 mainly European institutions went down, of which about half have defaulted on their senior debt. 

However, none have defaulted on their covered bonds. 

The list includes rehabilitated (*) issuers such as SNS Bank, Bank of Ireland and Bankia which have all returned or are poised to return to the covered bond market, as well other institutions like Bradford & Bingley and Washington Mutual, which no longer exist.

(*) Ibercaja conducted a non-compulsory debt buy-back in 2012 and therefore cannot be considered a rehabilitated issuer as was originally stated in this article.