NN IP: Draghi tries to achieve a delicate balance

The key issue for the ECB remains that it has to strike a fine balance between two things. On the one hand, the ECB wants to acknowledge the improvement in the growth outlook because this will help to cement more positive private sector nominal growth expectations. On the other hand, the ECB wants to prevent financial conditions from tightening in response to the improved growth outlook by hammering home the message that its policy response to this will be vastly underwhelming. Having said that, the ECB cannot completely refrain from reacting to a changing economic environment (which is essentially what the BoJ is doing!) because a statement to this effect would not be credible. After all, if the ECB maintains the current pace of QE for another year it will run into hard scarcity constraints. These constraints can in principle be lifted but the cost of doing so is substantial in terms of implications for market functioning and political or even legal challenges. Hence, the ECB’s game plan is to substitute between two instruments: QE and rate forward guidance. QE will be tapered towards zero but at the same time rate guidance will be enhanced in the direction of “lower for longer” if this should prove to be necessary. Only there is a slight additional problem with this game plan, because the ECB has explicitly linked the pace of asset purchases to progress towards its inflation objective. Hence, if there is no progress whatsoever on the latter the ECB cannot taper without losing credibility.

All in all there are thus enough ingredients for markets to get carried away in their thinking about the future monetary policy stance. In other words, expectations about future monetary policy are pretty much footloose because there are several instruments, many of which are new so markets cannot benchmark them to the past. The ECB has its own way of dealing with this. To convey the message that the policy response will be vastly underwhelming, Draghi uses a set of key words. According to the ECB, confidence in the recovery (i.e. the first part of the balancing act) requires “patience, prudence and persistence” (the second part of the balancing act).

Patience refers to the fact that it will take time for inflation to converge towards the target. This convergence requires that the ECB is persistent in its efforts to keep overall financial conditions at an accommodative level. Prudence then refers to the need to make any adjustment the policy parameters in a very gradual way so as to reduce the risk that the market will over-interpret the policy action taken and make an unwarrantedly big adjustment to its expectation of the future policy stance. Back in June the market got carried away in bond space as yields jumped after Draghi’s infamous Sintra speech. Since then the market has been getting carried away big time in FX space as the euro appreciation which was ignited in Sintra developed into an appreciation bandwagon effect which in the end caused yields to fall from their “Sintra” peak as markets started to anticipate an easier policy stance in response to the appreciation. All this goes to show that prudence is a lot easier in theory than it is in practice, which is something the Fed also learned the hard way over the past years.

Lessons from the first two mini exit steps

Before we discuss the outcome of the press conference it is useful to review the first two small exit steps the ECB has taken.

The first step was the decision to “slow and stretch” purchases in December 2016 which was accompanied by dovish foam on the runway in the form of an explicit QE easing bias. The rationale for this step was the reduction in deflation risks on the back of a better growth momentum which should ensure a continued tightening of the output gap.

This rationale was also behind the second step which was the removal of the easing bias on rate forward guidance in June. Once again there was some dovish foam on the runway in the form of a statement that policy normalization was not under discussion and the QE was sufficiently flexible to deal with scarcity constraints. The latter can be seen as an attempt to keep the QE easing bias alive. The next step will obviously be some kind of tapering. The rationale will no longer be the diminution of deflation risks, because these have all but disappeared now in the eyes of the ECB (even though the risk of inflation expectations settling at a below-target level is still alive). The likely rationale this time will probably be the fact that core inflation has moved sustainably above 1% and is expected to hit the target over the medium term. Core inflation has indeed risen above 1% but it may be a bit early to label this as a sustainable movement.

That is one reason why the ECB has postponed its QE decision to October and may even wait until December. Another reason for waiting stems from the fact that the exchange rate is currently the main source of sleepless nights for the ECB Council. The ECB probably believes that some appreciation is warranted by the fundamentals (improved growth momentum relative to the rest of DM as well as declining political risks) but the exchange rate is clearly in danger of overshooting. Still, the nice thing about an exchange rate is that there are two sides to it and the ECB may get lucky if a reduction in US political risks and/or a more hawkish Fed repricing reverse general dollar weakness somewhat. All this is a big “if” though, so the ECB cannot bank on it. The question then arises what they can do to deal with euro strength. The first port of call is verbal intervention (which is what Draghi did during the press conference) but this can easily backfire if the ECB is unable to back it up with policy action. As for the latter the options are pretty limited. Because the exchange rate is the relative price of two monies it tends to react more to short-term rate differentials and there is little scope for reducing EMU short-term rates further. Meanwhile, QE tends to have most of its effect on the term premium on longer-dated bonds, because of which it should be less effective in influencing the exchange rate. However, QE can also be used as a signal to cement the “lower for even longer” message on rates stronger. This is indeed the most fruitful route for the ECB because rate guidance is explicitly linked to the end of net asset purchases.

ECB signals prudence towards exit

With all this in mind we can say the following about the ECB’s September meeting:

There was no change to any of the policy parameters this month. In particular, the easing bias on QE was maintained and the forward guidance on rates was not changed. The latter contains the element which says that rates will remain at present levels until “well past” the end of QE. Over the past few months there has been a lot of speculation that the ECB may remove these two words so as to be able to hike the depo rate very soon or even before the end of QE. Our view has been for a while that there will be no change in any of the policy rates until QE is done and our confidence in this has only been strengthened by the fact that Draghi chose to emphasise this during the press conference. For all practical purposes this closes the debate on sequencing: QE will end first and only then will rates rise at some point. This still leaves the door open for a removal of “well past” but this will only happen if euro appreciation pressure abates significantly.

The latter was clearly very much in focus for the ECB. The statement identified the euro appreciation as a source of uncertainty that needs to be monitored. Also, FX was explicitly mentioned as a downside risk to growth. On top of this, the recent appreciation was the main driver behind the downward revision to the inflation forecast by 0.1pp in 2018 and 2019 to 1.3% and 1.6% respectively. Finally, Draghi explicitly linked future policy setting to the trajectory of the euro as he stated that the “calibration of our policy instruments beyond the end of the year” will depend on both “the expected path of inflation and the financial conditions needed for a sustained return of inflation rates towards levels that are below but close to 2%”. Of course, both the inflation forecast and the level of financial conditions are very much influenced by the euro exchange rate. In this respect Draghi labelled the exchange rate at being “very important”.

Despite all this, the euro appreciated substantially during the press conference while Bund yields fell. These moves have been partially reversed. The fall in Bund yields is probably driven by the fact that the market believes that a stronger euro will lead to a more dovish exit. The latter will mostly imply that the ECB will “slow by less or stretch for longer” when it decides on the future course of QE in the fall. Nevertheless, it is unclear how much effect this will have on the euro since exchange rates are mostly driven by short-term rate differentials. In this respect, market pricing for the first rate hike has already been pushed out to 2019. Hence, one explanation for the euro appreciation is that the market believes that the ECB does not have much ammunition to fight it. Of course, this may be a big invitation for the bandwagon to continue. More fundamentally, we view the euro exchange rate as the main victim of the ECB balancing act between confirming the better growth outlook and signalling an underwhelming policy response to it. Back in 2014 the Fed more or less faced the same issue. At the risk of repetition, this balancing act is a pretty difficult thing to get exactly right so expectations can easily embark on an explosive path. And of course, exchange rate expectations are less anchored than expectations governing other asset prices even under normal circumstances. Besides that, exchange rate expectations also depend on what “the other side of the equation” is doing and the ECB has no control over that. This explains why central bankers tend to shy away from strong words about exchange rates, especially when they cannot be backed up by action.

The ECB will thus announce its QE game plan in the autumn, most likely so in October, even though some details may have to wait until December. The latter details probably pertain to the parameters governing the program which may well need to be tweaked to avoid running in to hard scarcity constraints. Draghi once again emphasized there is some room here. Most of this room will probably be found in the capital key, even though Draghi emphasized that any deviation will be temporary. One possibility is to allow countries with scarcity constraints (Germany) to postpone purchases until more bonds become available. In the case of Germany, a fiscal expansion after the election may help somewhat in this respect. Alternatively, the eligible universe could be increased by including more supra-nationals, or German quasi-public and state/local issuers.

Our base case for the ECB is now that QE will be extended by six months at EUR 40 billion per month in 2018. After that we expect tapering towards zero by the end of that year. This will probably happen in the form of another “slow and stretch” step which keeps the option of extending QE into 2019 open. In this respect we also expect the QE easing bias to be maintained until well into 2018. Essentially this a cheap option as it describes the reaction function in case of an adverse shock. Furthermore, we expect rate guidance to be maintained in its current form, including “well past” until the middle of next year as well. Whether or not “well past” is dropped in some form after that will depend very much on the euro exchange rate. The risks to this view reside on the dovish side. In particular, we see a significant risk that QE will be extended into 2019 at a very low pace to enhance the credibility of rate remaining “lower for even longer”. A first rate hike will thus come in Q2’19 at the earliest and, once again, only if the euro is well-behaved.

Emerging markets: easing FX regulation in China

With most emerging countries having reported their foreign exchange reserves position for August, we have enough information to estimate total EM capital flows for last month. We come to a small net inflow of USD 3 billion, substantially below the USD 16 billion of July, but in line with what we have been seeing so far in 2017. Remarkable are the strong inflows in the riskier countries with a sizeable carry and the outflows in East Asia. This has been the picture throughout the year.

The most important aspect of the EM flow data is that Chinese outflows remain modest and manageable. The Chinese authorities have even seen room to let the CNY appreciate a bit relative to its trade-weighted basket. 3% in the last two months is not a big deal, but it is another confirmation that the Chinese government has regained control over its macro-economy. Last Friday, it announced that it will ease FX regulation by eliminating the reserve requirement that banks had for FX derivative sales. In 2014 and 2015, a strong demand for FX forwards was seen as speculation and a key cause of capital outflows. This is why the now abolished FX reserve requirement was introduced.

The recent appreciation of the CNY and the easing of FX regulation should mean that the renminbi risk is more balanced now. This observation is important, as a more balanced CNY helps creating a favourable environment for EM currencies overall. Lower perceived China risk keeps the window open for EM exchange rates to appreciate more.