ECB: Launching QE

Published on January 15, 2015
SGH Insight
"Furthermore we believe there is a good chance the ECB may in the end deliver more than the 500 Billion Euro purchase program reported by the press to have been discussed at last week’s interim meeting as the 'maximum' option."
Market Validation
(ECB Introductory remarks 01/22/15) "Under this expanded programme, the combined monthly purchases of public and private sector securities will amount to €60 billion. They are intended to be carried out until end-September 2016 and will in any case be conducted until we see a sustained adjustment in the path of inflation which is consistent with our aim of achieving inflation rates below, but close to, 2% over the medium term."

European Central Bank President Mario Draghi and his key allies on the ECB Executive Board have spent almost a year clearing the path for a politically controversial sovereign bond purchase program in case it is needed as part of the ECB’s unconventional monetary policy tool kit. Well, here we are.

Yet on the eve of a key January 22 monetary policy committee that will include a revision to the already dismal ECB staff quarterly forecasts, and with expectations for QE running high, pockets of resistance to sovereign bond purchases still remain within the Governing Council, led by German Bundesbank President Jens Weidmann.

*** Despite that resistance, we understand ECB officials, including some of the more traditionally hawkish members, are on board and are acutely aware of the need to take further and prompt action. We expect they will announce the launch of a sovereign bond purchase program at their upcoming meeting on January 22, rather than go piecemeal or wait for more data (there may be a small delay to when actual purchases begin). An announcement is also extremely likely to include supranational bond purchases such as EIB bonds and could include the go ahead for a corporate bond purchase program. ***

*** We furthermore believe there is a good chance the ECB may in the end deliver more than the 500 Billion Euro purchase program reported by the press to have been discussed at last week’s interim meeting as the “maximum” option. That larger size could result from, as an example, an option to add or extend to a relatively front-loaded, one year commitment on purchases, or a slightly less front-loaded but longer two-year commitment that adds up to more than 500 billion euros. ***

*** While the one trillion Euro balance sheet expansion “intention” to hit three trillion will help explain, color, and guide the size of the announced program, we do not expect the ECB to explicitly link bond purchases to hitting the three trillion Euro mark (it is after all not a “target” in and of itself). But as a 500 billion Euro program as leaked risks not just disappointing markets but also under-delivering and leaving the ECB “a dollar short” of its objective. And as so much of this exercise hinges on a positive management of inflation expectations, we do not sense that ECB officials will want to jeopardize that with a lower headline figure, or probably even with a 500 billion euro “cap.” ***

*** While not ideal in its signaling, as a nod to “moral hazard” concerns and in exchange for greater consensus, including agreement on a larger size of purchases, most ECB officials appear to expect and be comfortable with some level of compromise on the principle of risk-sharing, which means instructing the National Central Banks to buy and hold a certain amount of bonds on their own balance sheets rather than on the ECB’s. This would add a level of complexity to implementation of the program, but we are told there is some precedent in such a hybrid-purchaser approach both with the ABS program as well as with an earlier version of a Covered Bond Purchase Program. ***

The final contours for a sovereign purchase program have yet to be defined or agreed to, and will not be decided until the meeting itself. But the ECB’s objective is to structure and implement a program such as the one outlined above that while comfortable within both the letter and spirit of the mandate of the central bank, does not unduly compromise the effectiveness of the program for the sake of a unanimity that may not materialize in time for action, if ever at all.

Timing, Composition, and Size

The ECB late last year signed on to the “intent” – some members more willingly than others – to expand its balance sheet by about a trillion Euros to three trillion. It was fairly obvious from day one (hence the dissents) that this could not be achieved through TLTROs, ABS, and Covered Bond purchases. But the subsequent continued plunge in inflation since then has only added to a growing consensus on the urgency and need to embark on sovereign – as well as other – bond purchases in order to expand the balance sheet more rapidly.

A sovereign bond purchase program will almost certainly include supranational bonds as well, such as European investment Bank bonds. We also suspect the ECB will want to announce a readiness to buy corporate bonds, although we are not sure how developed those plans are at this point.

A greater sense of comfort on moral hazard risks and protection of the ECB balance sheet by having some purchases reside on the NCB books, even if ultimately, truth be told, all pea-pods eventually lie in the same bed, is also designed to help afford ECB officials greater latitude for larger purchase amounts.

We expect the ECB to announce a set amount of monthly purchases for a period of time, which could be a two-year commitment rather than one, and which will by definition imply a total program purchase size amount. They could, and these numbers are our own speculation, for example, announce a 30-35 billion euro per month bond purchase plan for two years, which would total 720 – 840 billion Euros. Those numbers could even be bumped slightly higher. Alternatively, the ECB could announce a one-year commitment to buy, for example, 40 billion Euros per month, for an annual total of just under 500 billion, with a commitment to revisiting and extending purchases for another year at the end of the term if needed. We think the ECB is leaning towards a two-year commitment.

The ECB is also considering the option of open-ended purchases akin to the Federal Reserve’s QE3, but ECB officials appear unlikely to be ready to go so far as to leave purchases completely open-ended and tie a program entirely to hitting a single specific target, be it a balance sheet target or the 2% inflation mandate. They note the Fed’s difficulties both in coming up with its initial target “threshold” (which as a reminder was tweaked down the road) and in communicating its eventual exit from the open-ended QE3 program regime.

But in order to pre-empt any unwarranted or misplaced questioning of its ultimate commitment, we expect the ECB will communicate that it can assesses, down the road, if its program is (or has been) sufficient in meeting its commitment to hitting its ultimate inflation mandate, and continue with more if need be. That could be inferred to mean the three trillion Euro balance sheet size is not in and of itself the ultimate objective or even potentially the upper limit to its balance sheet size – its ultimate target is getting on course to hit its “2% or slightly below” inflation mandate.

Guidelines and “Modalities”

We do not expect the ECB to publicly provide a total cap on the ultimate potential size of a QE program, but there is nevertheless a great deal of sensitivity to the notion of “owning” any individual country’s bond market. We suspect the ECB may use an internal and hypothetical guideline perhaps on the order of magnitude of approximately 30% of each market as a soft limit for its purchases, if not explicitly announced. Yesterday’s European Court of Justice Advocate General opinion was clear in arguing there is no requirement in its view whatsoever to make such guidelines public and explicit.

For context, and as precedent, the Federal Reserve also established (public) guidelines that the Fed’s SOMA purchases should not exceed 35% of any individual issuance. While that ceiling was eventually broken a few times, it provided more than enough room for the Fed to achieve its monetary goals. With a European sovereign bond market outstanding size of around nine trillion Euros, even a trillion or more of sovereign purchases would fall well short of, for example, a 30%, which would mean in the aggregate three trillion, guideline limit for “owning” any single sovereign market.

The fact that issuance is so much lower in Europe than in the US means that a proportional QE response there will have a greater market impact than if simply weighed and compared to the US by GDP. And the ECB will end up buying a good deal of new issuance, but in the secondary market. That is not a problem – for an extreme precedent, the Bank of Japan for its part has had no qualms in buying the equivalent of the entire monthly new issuance of JGBs.

The ECB can and is likely to buy bonds across the curve. Draghi limited the OMT bond purchase program designed to calm the collapsing peripheral markets to buying only 1-3 year bonds in order to alleviate any lingering concern that its purchases could constitute selective monetary financing of any given country. But a QE program is so clear from a mandate perspective in constituting monetary, and not fiscal policy, that there is no need to limit purchases to the short end to prove or reinforce that point.

Likewise the ECB can and will most probably buy bonds trading at negative rates. While negative rates are a complicating factor, the BOJ as an example has continued to buy bonds even at negative interest rates, and ECB officials note to us that if there is a small carry cost loss to the ECB in doing that, so be it – their objective is monetary policy stimulus and not a positive P+L.

The ECB may limit its purchases to investment grade bonds, which currently includes every Eurozone country except Cyprus and Greece. Portugal is rated BB by S+P, which puts it one step below investment grade, but it is still investment grade according to DBRS. We suspect the NCBs could have the option of buying lower rated bonds on their balance sheets, which could provide a backstop to Greece, as long as of course it remains within the Eurosystem and in good standing.

NCB Balance Sheets, Risk Sharing, and Greece

To date the ECB has been able to accept bonds of lower than investment grade quality (meaning Greek) as collateral so long as the country is in a Troika program. Before the European Court of Justice ruling today, the Greek issue could easily have been punted as the ECB would not have been able to buy Greek bonds until the review was underway and until it was completed. By then (end of February and possibly later if extended) there would be some clarity on the Greek government and its policies.

But the European Court of Justice Advocate’s opinion yesterday on the OMT case has introduced a new wrinkle to that dynamic. Specifically, the ECJ advocate pointed out the inherent conflict in the ECB sitting on the Troika that sets fiscal guidelines for borrowers and at the same time is buying bonds according to whether a country is adhering to the guidelines the ECB itself has helped set, evaluates, and monitors.

The most direct consequence of the ECJ case will be for the ECB to remove itself from active participation in future Troika programs, something it has wanted to do for some time already anyway. In the interim, the use of the criterion for whether a country is or is not in a program to determine whether or not the ECB buys its bonds will need to be addressed.

The ECB is hell-bent on moving forward with QE without getting twisted up over what to do with the tiny Greek market. Limiting purchases to investment grade and keeping a portion of sovereign bond purchases on National Central Bank balance sheets can square that circle for now.

Anything short of full mutualization of purchases on the ECB balance sheet is clearly not the most ideal signal to markets, but there is also precedent to the hybrid-style approach and so, especially in context of a massive purchase program, sharing purchases with the NCBs need not lead directly to howls of the end of the Eurozone unity as we know it. It is nevertheless as good as guaranteed that certain London tabloids and US cable TV stations will have a field day with exactly that.

The current ABS program has both a “risk sharing” and NCB component, and we are told so did one of the earlier versions of the covered bond purchases. Bringing in the NCBs nevertheless introduces a new complication in that if their purchases are left to be conducted on a voluntary basis, the ECB may be faced with the unwelcome prospect of the Spanish and Italian central banks jumping at bond purchases while the Dutch and Germans decide to take a pass. So NCB participation in purchases must be with the advance broad agreement of the NCBs and come in the form of explicit directions from the ECB to the NCBs to purchase a certain portion of bonds on their own.

In addition to helping gain greater consensus in addressing some Governing Council member concerns over the moral hazard issue, keeping bonds on NCB balance sheets will also make ECB officials that much more comfortable in granting “pari passu” creditor status to the bonds that it does buy in the markets. This is a key consideration in avoiding the “crowding out” and “shoving down” effect of a QE program on private sector bondholders.

In reality the ECB creditor status, as well as the NCB’s, is more complicated than that, in that the ECB specifically has vowed never, ever to agree to a haircut on its bond holdings. The ECJ appears to be very friendly to the ECB in this regard, with the Advocate welcoming the ECB’s acceptance of pari passu status in the case of OMT purchases, acknowledging that it makes the ECB’s intervention tool more effective, and discounting the impact or likelihood of an actual loss of any real significance or magnitude from those purchases.

Indeed, the ECJ opinion states that ECB hypothetical purchases of bonds issued by a solvent sovereign entity, even if it has temporary funding or “transmission mechanism” impairments helps with that sovereign’s liquidity, making it all the more creditworthy. As long as the ECB is not in the game of guaranteeing such purchases from a specific country (or guaranteeing that they will hold them on their books forever, but rather retains the option to turn around and sell them at any point), then these OMT purchases do not breach monetization prohibitions. The parallels from that to a QE program are clear.

Indeed there is an argument being made that having a country’s own central bank own its bonds provides a disincentive to defaulting or restructuring that would not exist under joint, mutualized, or foreign ownership. A case in point in the extreme is the Bank of Japan, where the BOJ and domestic Japanese banks hold the vast bulk of Japanese sovereign debt, so what purpose would a restructuring serve there? Closer to home, and more pointedly for the ECB, if the Bank of Italy and domestic Italian banks were to continue to hold the bulk of Italy’s debts, what benefit could there possibly be to the government in reneging?

And so we do not believe that in the end a compromise on burden sharing is all that negative, especially if it allows for greater consensus and a more aggressive, larger bond purchase program. Of course from a signaling perspective, ideally the ECB would put all those bonds on its own balance sheet. We would be pleasantly surprised if with all the moving parts for the ECB to negotiate that were to happen. But even if not, so be it.

Back to list