September ECB: 6-month QE extension justified, but risk of disappointment

This post previews this week’s ECB policy meeting. The main points are:

  • In July Draghi stressed that the ECB was open to all options on further easing, were watching emerging data closely and would use flexibility to meet their €80bn a month purchase target (they only managed €60.5bn in August).
  • Several developments indicate that ending QE purchases in March 2017 would be premature, so there’s a good case for an extension of the €80bn a month QE (PSPP) programme by at least six months (i.e. to September 2017): (i) actual inflation remains disappointingly low, with underlying price pressures continuing to lack a convincing upward trend; (ii) weaker survey-based inflation expections and professional forecasts; (iii)  inflation breakevens remaining worryingly low (see Chart); (iv) signs of slowing near term growth momentum, replacing initial limited apparent Brexit spillovers; (v) the ECB  staff growth and inflation forecasts suffering limited downgrades from the adverse impacts of the Brexit vote.
  • Overall the large stakes and asymmetric risks mean that the ECB should ideally remain in pro-active risk management mode. So there’s around a 60% chance of such a QE extension being announced this week. But given that it seems increasingly priced the market impacts would likely be limited – small bond rallies and limited EUR downside (EURUSD potentially below 1.11) – depending on technical details and Draghi’s tone. Theres less than a 10% chance of the more radically dovish option of cementing the de facto open-ended nature of the programme
  • But several factors could delay the ECB announcing a QE extension until December (40% chance) including: (i) they may prefer a “wait and analyse” approach given the relative proximity of the previoius easing measures (i.e. its “too soon”); (ii) the relatively-small size of the latest setback – the outturns aren’t yet significantly-different to the June forecasts and the September forecasts could be boosted by greater ECB optimism about the efficacy of easing meaures; (iii) headline HICP inflation will likely rise near-term on more positive energy price base effects; (iv) ECB political barriers, most obviously from Germany; (v) extending QE exacerbates prospective bond scarcity problems (see chart) but ECB probably won’t be ready to immediately implement required PSPP technical parameter tweaks to address that (they’ve thus-far rebutted scarcity suggestions).
  • Such an ECB disappointment would likely lead to an unhelpful bond market sell-off and a limited EUR bounce (EURUSD towards 1.14), especially if accompanied by continued stone-walling on bond scarcity. That said, Draghi would likely send dovish signals e.g. announcing expert groups to examine future options.
  • The ECB faces the issue that the least contentious PSPP paramater tweaks are also least likely to resolve the bond scarcity issue. Relaxing the 33% issue limit for non-CAC bonds is the most straightforward change (50% chance) but wouldn’t impact scarcity issues. Relaxing the 33% issuer limit is less likely (around 30% chance) but would at least facilitate greater Portuguese and Irish bond purchases (but not Bunds). Relaxing the prohibition on purchasing bond yielding less than the -0.4% ECB depo rate could substantially ease scarcity effects, but it raises tricky issues so there’s less than a 20% near-term chance. Abandoning the capital key determination of asset purchases would usefully target stimulus where it’s most needed and sidestep bond scarcity problems. But there’s less than a 5% chance of it being announced this week given steep political resistance and the current macro situation.
  • It’s also extremely unlikely near-term that the ECB will tweak the TLTRO II parameters, expand the range of elibible assets in the QE basket (e.g. to include bank debt) or further cut the rate (although the risks are skewed towards rate cuts in December).
  • But Draghi seems very likely to try and up the pressure on governments to pull their weight via fiscal support and structural reforms, given growing concerns of ECB policy  approaching its limits and potentially having adverse side-effects (see Coeure’s Jackson Hole speech).
  • ECB ations will struggle to have much traction on the EUR – recent moves have only been weakly correlated with interest rate moves (especially EURUSD) – in an environment of continued market scepticism about central bank actions (see here) and continued huge EA current account surplus.


Small downgrades to already-weak ECB growth and inflaton forecasts motivates PSPP extension

The June 23rd Brexit vote has likely transformed the pre-referendum chance of ECB growth and inflation forecasts being revised up for the first time in several of years into instead the risks being of small forecast downgrades. I agree with the consensus that the 2017 growth forecast, put at 1.7% y/y in June, is most vulnerable to a cut (several pp).  There’s obviously considerable uncertainty about the scale of the adverse impact on the UK, taking account of the MPC’s sledgehammer response (see here, here and here). But the risks are skewed to negative spillovers onto EA growth prosects – principally via weaker external trade but also via potential investment downgrades on greater uncertainty (recall that EA growth has recently been more supported by domestic demand).

Moreover, the initial picture of limited Brexit spillovers onto EA activity indicators has recently become less positive as EA data have disappointed (see Chart) with, for example,  the composite PMI falling to it’s lowest since January 2015 including similarly weak new orders and signs that the period of sustained job creation could be cooling.  This picture of steady but sluggish growth likely falls short of ECB aspirations: last weekend Yves Mersch commented that “the pace of recovery in the euro area remains unsatisfactory”. Surprisingly, the nascent slowdown appears most pronounced in Gemany, with services PMI falling to a three-year low (with weak new business) alongside a plunge in the Ifo index (see chart) and weaker ZEW indices and July indistrial production falling sharply.


Offsetting the negative Brexit impacts and recenrt data, however, is likely greater ECB optimism about the efficacy of their existing easing meaures. The June ECB forecasts were conservative in their assumptions of the impacts of the measures annouced in March, given their unusual nature.  But further analysis and the actual implementation of all the measures seems likely to swing towards greater assumed impacts.  Ewald Nowotny recently argued that monetary policy has proved “more potent” than previously thought.  Similarly, the Bundesbank head of projections argued that “We may well underestimate the impact of non-standard measures…we are in a learning process.”

But overall a weaker EA growth forecast means that the output gap will be closed more slowly, which will translate into weaker inflation dynamics.  Indeed, a recent ECB staff paper argued that the EA output gap could be up to 6% of GDP.  While that conclusion was data-driven by inflation’s persistent weakness, rather than by identifying fundamental factors driving the output gap, it could perculate into the lower ECB inflation forecasts. But overall the risks are that the June HICP forecasts could be edged down margially – and that’s worrrying given that in June inflation was expected to peak at only 1.6% y/y in 2018. And such prospective continued inflation undershoots render ending QE purchases in March 2017 premature.  Don’t forget that recent ECB introductory statements have for months stressed that the QE programme is “intended to run until the end of March 2017, or beyond, if necessary, and in any case until the Governing Council sees a sustained adjustment in the path of inflation consistent with its inflation aim“. And the July minutes (accounts) indicated that the ECB is concerned about the lack of underlying price pressures.

Further near-term inflation (and inflation expectation) disappointments

Unfortunately, several developments indicate that’s not yet happened despite all the ECB’s previous monetary easings: (i) Headline HICP flat-lined at only 0.2% y/y in August, while core HICP disappointingly ticked down to only 0.8% (see Chart); (ii) the PMI prices balances report worrying signs – “Inflationary pressures are also cooling amid intense competition” – with services output charges declining for eleven consecutive months and input prices rising at their weakest rate in five months; (iii) The European Commission surveys of household and corporate inflation expectations shows some signs of fragility (see Chart); (iv) Professional forecasters have also become less optimistic about inflation prospects (see Chart) – and Dragh has previously paid attention such SPF movements; (v) Most provocatively, 5y5y inflation breakevens have recently been declining again (to less than 1.3% compared to around 1.7% at start-2o16, see Chart) despite more positive oil price trends.



Hence an extension of the €80bn a month QE (PSPP) programme by at least six months (to September 2017, from the current March 2017 end-point) seems appropriate.  And there seems to be a good chance (around 60% probability of it) being annouced this week.  While there are some arguments for a delay until October or November (see below), the large stakes and asymmetric risks increase the chances of the ECB remaining in pro-active risk management mode. Recall that in July Draghi argued that “…given prevailing uncertainties, the Governing Council will continue to monitor economic and financial developments closely.”

ECB could nevertheless be tempted to delay announcing the almost-inevitable QE extension

Despite those worries, there are several potential push-backs against an immediate extension of the the PSPP programme: leading to around a 40% chance that the QE extension announcement could be delayed (until October or December).

First, the thus-far limited nature of recent negative developments: big picture the activity and inflation outturns aren’t dramatically different from the June ECB forecasts. Second, headline HICP inflation looks set to be boosted by positive energy price base effects over the next few months.  Third, the relative proximity of the latest ECB easing measures, which given lages will only now be impacting.

Fourth, the negative side-effects of easing measures show signs of rising up the ECB agenda.  Benoit Coeure’s Jackson Hole speech argued that “extended use of unconventional measures could come with rising side effects, for instance on financial stability, financial intermediation and international spillovers. Thus far, the benefits of such measures have clearly outweighed their costs, but we cannot rule out a situation where the side effects are such that the negative consequences prevail.” Similarly, Yves Mersch also recently warned that using “extreme policy measures with unacceptable side effects” to shore up the Eurozone’s weak economy could undermine trust in the single currency. Of course, divisions within the Governing Council are far from unprecedented and the more hawkish GC members have previously been outvoted.

Finally, and probably most importantly, extending QE exacerbates prospective bond scarcity problems, which raises policy sequencing issue.  Specifically, ECB probably won’t be ready to immediately implement the required PSPP technical parameter tweaks to address them.  After all, they’ve thus-far rebutted suggestions of scarcity issues. In July Draghi argued that they had the flexibility to side-step such issues: “I think in worrying about the coming months, whether we’ll actually be able to fulfill this objective, proper attention should be given to evidence we’ve given in past few months and the ability to exploit flexibility.” But that sounds over-optimistic given investment bank estimates indicates that the Bundesbank will be unable to continue buying the €10bn a  month in Bunds substantially beyond the March 2017, or possibly earlier (see FT Report).  Indeed, the chart below illustrates that the share of bunds yielding less than the -0.4% ECB Deposit rate has increased as the Brexit-driven MPC easing and mixed US data gave fresh impetus to the global bond market rally.

Of course, CSPP is a prime example of an ECB innovation being announced but precise technical details following later.  So this isn’t sequencing issue isn’t necessarily a show-stopper for a QE extension announcement this week.


The previously-mentioned fact that the ECB only managed to buy €60.5bn of bonds in August, rather than the €80bn target, seems partly seasonal (i.e. quiet summer markets).  But given that a similar picture could well hold in December, September-November will have to be bumper months of purchases if the market isn’t going to start questioning ECB policy.

PSPP Parameter tweaks: Easiest options unfortunately aren’t the most effective

To facilitate, or complement, the likely 6-month QE extension there’s a range of options available to the ECB in tweaking the paremeters of the PSPP (QE programme).

The primary aim of the 33% issue limit is to prevent the ECB from obtaining “a blocking minority for the purposes of collective action clauses”. ECB doesn’t want to place itself in a position of blocking a bond restructuring. Easing the limit could cause distortions in the market, reducing the chances of it being implemented.  Nevertheless, relaxing the 33% issue limit for non-CAC bonds (to say 50%) is the most straightforward PSPP parameter tweak this week (around a 60% chance of happening).  But it wouldn’t resolve scarcity issues.

Relaxing the 33% issuer limit is a little less likely.  The ECB views this limit asa means to safeguard market functioning and price formation as well as to mitigate the risk of the ECB becoming a dominant creditor of euro area governments.” But it’s nevertheless arguably somewhat arbitrary – it constrasts with the 70% limit the ECB set in its corporate bond purchases – and hence a candidate for relaxation. Such a relaxation would particuarly benefit Irish and Portuguese bonds – given that the ECB purchases under the SMP/ANFA schemes mean that the 33% limit is already close, necessitating slower ECB purchases. Indeed, recent ECB purchases of Portuguese government bonds have only been around half the level prescribed by the Capital Key. But it won’t resolve the difficulty in buying the quota of Germany bonds (plus the Nertherlands and Finland).

Relaxing the prohibition on purchasing bond yielding less than the -0.4% ECB depo rate would help ease scarcity effects, with the extent depending on how far below the Deposit rate purchases are allowed.  But it seems very unlikely near-term given the tricky issues it raises – there’s probably less than a 20% chance of it being announced this week. First, the ECB would be esentially locking in a loss-making position into it’s purchase programme: that’s not great for the ECB balance sheet and could potentially be interpreted as monetary financing.  Second, it calls into question the ECB deposit rate supposedly representing the floor for (appropriate maturity/risk) yields: it looks internally inconsistent for the ECB to push yields below it’s own floor.

Abandoning the capital key determination of asset purchases would very usefully target stimulus where it’s most needed and sidestep bond scarcity problems. In effect it would shift QE purchases away from German bunds and towards the periphery, especially Italy given the large size of the market but also towards French OATs for similar reasons hence likely tightening spreads.  But it faces the steepest political resistance, due to concerns about money financing and a backlash from Northern (German) electorates.  The ECB probably wouldn’t want to stoke such political issues.  But luckily the current EA situation appears far from the adverse scenario where it becomes the end game. So there’s a very low probability of (less than 5% chance) of it being implemented this week.  Indeed, given the political-economy considerations Draghi could again push back against questions about the CG discussing the capital key, which the market could take as mildly hawkish.  But equally,  Draghi could adopt a more neutral academic approach – adopting the adage “never say never”, should the EA economy continue persistently underperforming.

ECB ations could struggle to gain much traction on the EUR

Overall, Thursday’s ECB meeting seems unlikely to provoke sharp asset price moves. But this may be particularly the case for the FX market: EURUSD could potentiall fall to around 1.11 on a dovish ECB (annouce 6-month QE extension and loosening issue/issuer limits plus hints at looking at other PSPP parameters) but rise towards 1.14 on a more hawkish Draghi (no QE extension announced, albeit with continued dovish words).  Of those, the risks are slightly skewed towards the ECB disappointing.

But it’s notable that since Q1 2016 the €’s movements have been largely disconnected from EA-foreign relative interest rates, having previously tracked them fairly closely.  Within that, EURUSD has been substantially stronger than the 1.07 level implied EA-US rate differentials (see first Chart below) while EURJPY has been substantially weaker (second chart below). Interestingly, EURGBP has continued to closely track EA-US rate differentials despite the Brexit shock: I noted this tendency, with UK rate moves being the main driver, here.


This disconnect seems to reflect a combination of factors, including: (i) ECB policy focus moving away from weakening the € and towards stimulating domestic demand via credit easing measures; (ii) continued FX market scepticism about central bank policies – mainly about Fed’s rate hikes (see here) and the BoJ’s continued hints at doing more (the 21 September BoJ meeting is likely to be more FX market imoactful than Thursday’s ECB meeting); (iii) the continued huge EA current account surplus







One thought on “September ECB: 6-month QE extension justified, but risk of disappointment

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s