Groundhog Day

The ECB buys covered bonds — again. It’s like changing your password on your computer. You leave the letters identical and increase the number at the end by one to keep it simple — in this case, to 3.

Mario Draghi announced an ABS and covered bond purchase programme at the European Central Bank press conference last Thursday. He didn’t explicitly mention a number, but Bloomberg reported a combined figure of Eu500bn over the next three years.

The official details will be outlined after the next governing council meeting on 2 October. We don’t, therefore, know the overall volume of the programme, the breakdown between ABS and covered bonds, nor the duration of CBPP3 or its focus.

The initial reaction from the market was quite dramatic. Spreads across the board have tightened by up to 40bp in the periphery (Monte 2024s, for example) and up to 10bp in core sectors. The first issuers have also been able to reap the initial benefits of the programme, with Caffil and CFF pricing deals through swaps on the back of massive order books, and Aareal pricing a three year at mid-swaps minus 10bp.

Focus is on credit easing

During the press conference Mario Draghi spoke about credit easing and trying to incentivise banks to lend to the real economy. He also spoke about having expertise in both markets as the Eurosystem already holds significant volumes of securitisations and covered bonds in its collateral system. The volume of covered bonds pledged with the ECB at the end of Q1 was Eu378bn while ABS stood at Eu307bn.

According to ECB data, the overall covered bond market’s size stood at Eu2.6tr at the end of 2013. The volume for Eurozone countries came in at Eu1.6tr. This includes everything from big benchmarks to small private placements to retained deals. A Eu500bn figure looks enormous even if it will cover future issuance in the next few years (which will be dented by the TLTROs, by the way).

Having said that, the ECB will ultimately struggle to find sufficient bond volumes in the open market in our view to get anywhere close to these figures, particularly if we think of the last programme, which closed at Eu16bn after nine months.

Whatever happened to the first two programmes and what could they buy?

In addition to the ECB’s normal portfolio purchases, they have conducted two purchase programmes in recent years.

The first purchase programme, between July 2009 and June 2010, was a big success with the full Eu60bn programme money having been spent.

The second programme, from January 2011 to October 2011, however, already faced similar problems to those the ECB will face this time around — there wasn’t enough issuance and secondary markets had dried out. Of the planned Eu40bn, only Eu16bn was spent and the programme was closed prematurely.

Even though the ECB offered bonds in repo transactions, it didn’t outright sell any in the aftermath of the programmes but is holding them to maturity. From both programmes combined, the ECB still holds Eu46bn, which is around 60% of the initially purchased amount.

Anything the ECB can actually buy?

First, the overall market the ECB could be considering (euro-denominated covered bonds from banks from within the Eurozone) was around Eu1.55tr at the end of 2013. Around Eu0.46tr are backed by public sector assets, though, while the mortgage backed ones amount to Eu1.09tr. In the past the ECB had always preferred mortgage backed covered bonds to stay further away from indirectly funding the public sector.

Second, the ECB could look at retained issuance. The numbers from Ireland, Italy, Portugal and Spain amount to around an estimated Eu200bn. The ECB will have to do a lot of convincing, though, and/or be extremely aggressive in the prices it is willing to pay these banks. Getting large volumes of retained bonds will be hard and will only happen at much tighter spreads, or it will require additional pressure by the ECB (higher haircuts yet again on retained deals?). After all, the alternative is to simply continue using the bonds as repo collateral for the TLTROs.

Third, the primary market seems to be the main target for the central bank if it wants to stick to the more traditional covered bond types. However, post the ECB base rate drop by 10bp to 0.05%, it has at the same time also lowered the TLTRO funding cost for banks, making it even more attractive. While the ECB has, however, lowered its base rate, yields in the market have collapsed as well, especially at the short end of the curve.

When factoring in TLTRO all-in costs based on banks using sovereign debt as collateral, Pfandbrief issuers can actually beat the TLTRO by issuing in the market at the very short end out to two years. And when considering the fact that, for example, specialised mortgage lenders in Germany don’t have substantial unencumbered securities portfolios and would pledge, for example, commercial mortgage loans at much higher haircuts, market funding beats TLTRO money out to three or even four years. Deutsche Pfandbriefbank’s recent three year deal was issued with exactly this rationale behind it.

How large could the covered bond share be in the overall programme?

A key question investors have been asking is what the ECB could actually buy in the covered bond space. If the main focus of the ECB is on freeing up banks’ balance sheets, as Draghi said during the press conference, covered bonds should only play a minor role in the grand scheme of things.

During CBPP2, for example, the ECB ended up buying around 9% of the eligible primary issuance. Overall, the central bank managed to buy Eu6bn in primary markets as well as Eu10bn in secondary markets during the 10 months the programme was active. And, given the state of secondary covered bond markets, we’d be surprised if they manage to get a similar secondary market buying figure during CBPP3.

Bottom line

Every way we look at it, the ECB will struggle to find sufficient volumes in the covered bond market to satisfy its hunger for credit easing, even if the main focus of the programme ends up being on ABS markets. Consequently, the central bank will continue to squeeze the market to bits with spread differences between different issuers, countries and broadly speaking different levels of risk further coming down.

In our view this is part of the bigger plan, however. The ECB is looking to offer banks potential capital relief through the ABS purchases, on the one hand, and make sure their funding levels are as cheap as possible, on the other hand. At the shorter end up to four years we have the normal MROs at 0.05% and the TLTRO at 0.15%. And since the central bank doesn’t really operate at the long end of the curve it is using the squeezed covered bond market and CBPP3 to anchor banks’ funding levels at the long end of the curve at very low levels. Especially with the stress test and AQR results looming and the ECB taking over the role of a banking supervisor, it seems that the ECB wants to add an effective ceiling to any potential volatility should there be any surprises in the results. Starting in October fits that argument timing-wise quite well.