Fra BNP Paribas:
The ECB seems to be pacing a more patient exit than the Fed did. Compare the “taper torpor” post-ECB announcements with the “taper tantrum” Mr Bernanke unleashed in 2013.
One reason is that the ECB seems to be guided more by actual inflation, whereas the Fed seems driven more by growth and unemployment. The ECB puts a bigger weight on FX.
The ECB is tightening later in the cycle and is more flexible, allowing inflation expectations to rise whereas the Fed has diminished them. The risk is the ECB getting behind the curve.
One cannot help but think that if Mario Draghi was on the FOMC, he would probably be
voting the same way as Charlie Evans and Neel Kashkari.
We want to compare how the Fed and ECB have gone about reducing a monetary accommodation, to see what lessons can be learned. We believe that so far in the tightening process, the ECB is making a better fist of it than the Fed at the same stage. This may well be because the Fed started first, and the ECB has been able to learn lessons. For example, the taper tantrum was a very instructive episode and has clearly encouraged the ECB’s exit from QE to be a gentler, slower process. The ECB made many mistakes earlier: the premature rate hikes in 2011; delaying QE for too long; and the speed of balance sheet reduction when the LTROs ran off.
The first lesson is what the exercise is called. Mr Draghi refuses to call scaling back the rate of purchases tapering, whereas Mr Bernanke explicitly referred to it as that from the beginning. Tapering implies a reduction to zero and the ECB is steering clear of implying a definite end date – that is, its programme is left open-ended. We view this as a very important difference.
This is because QE’s end gives a very important signal. Everyone knows that so long as the economy remains broadly on track, the next logical monetary step after QE has ended is a rate hike. While the time gap between the last bond bought and the first rate hike is uncertain and data-dependent, we would argue it has to be at least six months.
Mr Draghi has gone out of his way to try to avoid markets pricing in hikes too soon, first by fudging the end-date for QE and second saying that the inaugural rate hike would not come until “well past” the end of QE. Going further and being explicit, until the last press conference, about the possibility of QE being stepped up again or extended yet again underlines the message to markets – don’t price in higher rates too soon.
The language accompanying ECB’s latest decision to scale back the pace of purchases was decidedly dovish – Mr Draghi avoided suggesting the recalibration signalled a re-orientation of the monetary policy stance, as appeared to occur with the 2013 US taper tantrum. In fact, Mr Draghi portrayed it as an open-ended recalibration. If QE were to end in September 2018, the first rate hike would likely not be before March 2019. If QE were extended further, it would be even later in the year. Mr Draghi’s eight-year term as President expires on 31 October 2019, and he cannot be reappointed.
The ECB guidance can be interpreted as suggesting there is a decent chance that the inaugural rate hike will be left to Mr Draghi’s successor. At least, his successor will have to handle the bulk of the tightening process. The first reduction in the ECB’s rate of purchase came in March 2017 (EUR 80bn a month to EUR 60bn), announced in December 2016. If QE does end in September 2018, tapering will have taken 18 months (and it might be longer), whereas the Fed’s took eight.
This signals a much greater reluctance to hike rates than the Fed showed – because in our view the first hike can only come at the earliest six months after the end of QE. In the Fed’s case it was 14 months afterwards. Extending QE, even at a much-reduced rate, is the sincerest form of forward rate guidance as the ECB made clear with a very firm line on sequencing. The ECB’s reluctance to be explicit about the timing of rate hikes is a big contrast with the Fed. The FOMC was talking about rate hikes before it announced tapering.
In August 2011, the FOMC already began talking about lift-off, saying that conditions would likely warrant keeping the federal funds rate target near zero at least through mid-2013.
In the January 2012 Summary of Economic Projections (SEP), most of the Committee wanted to be raising rates by 2014, with some wanting to start in 2013 or even 2012 (see Chart 1). Based on figures interpolated from OECD annual estimates, we reckon the output gap was 3¾% of GDP at the start of 2012, whereas the unemployment rate the previous month was 8.5% (the central tendency of the January SEP was to think equilibrium unemployment was 5.2% to 6.0%).
No wonder taper talk unleashed a tantrum – tapering would open the door to the
higher rates that the Committee had been hankering after, when the economy was a long way off full employment. The Fed was too explicit, in our view, about the path it thought appropriate for rates when there was so much uncertainty after a massive financial crisis. It pretended to have more knowledge than it had.
This was especially true about two unobservable variables – the equilibrium interest rate (r*) and the equilibrium level of unemployment (NAIRU). In January 2012 the Fed indicated it thought the equilibrium unemployment rate was 5.6%, now the FOMC reckons it is 4.6%. The median equilibrium interest rate was estimated at 4.25% in the January 2012 SEP; in the most recent SEP 2.75%. By having too high an estimate of the terminal rate as well as too high a NAIRU, the Fed gave a too-hawkish a message.
The fact that it has consistently had to revise down its estimates of r* and NAIRU tells us how uncertain these concepts are and that the Fed has remained too hawkish in successive iterations of its messaging. This is very important because at the zero lower bound, communication constitutes a large chunk of the available toolkit for central banks. Too hawkish messaging holds down inflation expectations.
The Fed’s guidance has always suggested it would take some time to return to ‘normal’, but ‘normal” was its base case for the longer run, for growth, inflation, unemployment rates and the balance sheet… Given that evidence from financial crises in the past, especially with balancesheet deleveraging involved, hoping for normalization was probably premature. It takes a decade or more to recover from such a crisis, not half that time, which is when Fed rhetoric shifted to indicate rate hikes on the horizon.
Mr Draghi has steered clear of indicating what he reckons the equilibrium interest rate is and has not published, as former Chair Yellen has, different rules for deciding where the policy rate should be. Neither has the ECB suggested what the timescale is for achieving what it regards as ‘neutral’ rates. While the ECB clearly subscribes to the Phillips curve model, judged by its references to the likelihood of inflation rising as unemployment rate falls and shape of its forecasts, it has not tried to pin the jelly on the wall by being definite about where NAIRU, or r*, may be.
It may have working assumptions on all these things underlying its forecasts and communication, but it kept mum on where they may be even though this makes it less transparent than the Fed. There is a case, when it refers to some of these ‘parameters’ as
having shifted that it should at least specify broad ranges for where they might be. If the ECB is going to say it looks like the NAIRU has shifted down, for example, it is legitimate to ask “from what to what?”
Fed communication about future rate moves has been based on a model of inflation, the Phillips curve that has proved in practice to be fragile and unreliable. In contrast, while the ECB has not moved rates so far, its communication suggests they will be more driven
by inflation outturns than by the predictions of models. In other words, it looks as if the Fed has been too theoretical in its setting of rates and too-reliant on models that give a big weight to prerecession relationships.
The ECB appears to be adopting a simpler evidence-based approach –where’s inflation today? That same approach is the one that is advocated in the FOMC by the most dovish members. One cannot help but think that if Mario Draghi was on the FOMC, he would probably be voting the same way as Charlie Evans and Neel Kashkari. There’s not been that feeling about the Fed process – QE quickly became a down escalator that the Fed did not want to get off, even though taper targets were set meeting to meeting. The Fed
appears to like set timetables, as is suggested by 17 successive 25bp hikes in the 2004-2006 hiking cycle.
