Yves Mersch: Interview with Reuters

Interview with Mr Yves Mersch, Member of the Executive Board of the European Central Bank, conducted by Mr Balazs Koranyi on 14 March 2018 and published on 16 March 2018.

The views expressed in this speech are those of the speaker and not the view of the BIS.

Central bank speech  | 
22 March 2018
PDF version
 |  4 pages

Mario Draghi took a cautious view on inflation, arguing that economic slack may be bigger than earlier thought. What is the implication of this unexpected slack?

The policies we have undertaken, especially since 2014, have been quite successful in getting us closer to our inflation aim. But there are uncertainties about the process of inflation, given factors like globalisation and technological changes. Mario Draghi raised important questions, which have been discussed among central bankers: what is the non-accelerating inflation rate of unemployment (NAIRU), what is the output gap and how big is the slack? He was exploring the possibility that the degree of slack might be bigger than what you find in classical assessments.

We are continuously surprised by the response of the labour market to increasing aggregate demand of a growing economy because rising demand has been leading to continuous positive supply reactions in the labour market.

ECB staff estimates that the euro area output gap is closed, and there is already a positive output gap in Germany. But there is little inflation. Maybe there is another explanation. Perhaps we are getting a supply of labour that is pushing the frontier out. In Germany we know that migration from Central and Eastern European countries with relatively comparable skills is making an impact. Without such a labour supply reaction, you would have had wage inflation about 1 percentage point higher, and as a consequence, higher inflation.

Would that mean that the output gap is not yet closed?

That is an open question. We are all surprised by the reaction of the labour market to better economic conditions. The increasing participation of women and older people is certainly making an impact. In term of hours worked, we have not yet recovered the level of 2007. But in terms of employment, we have. It seems there is a trend for people to work less or part time.

Now we are reaching a floor for unemployment but it's not so clear if we're going to see wage increases and Mario Draghi was opening the conversation on the possibility that there may be more slack in the economy. That still needs to be confirmed but we already have strong evidence of a strong labour supply reaction.

The survey of professional forecasters indicates that unemployment will fall to 7.5 percent in 5 years. It's not clear if that's the new NAIRU. Economists have been surprised by the fact that the NAIRU has been continuously revised downwards. The supply of labour has increased more than expected, so there were reservoirs of labour. It could be migration or longer working hours.

It is clear that if you believe that the degree of slack is higher, then the process of convergence to below, but close to, 2% over the medium term would be drawn out.

We should not draw implications for policy immediately but it shows you the richness of the discussion at a stage where the economy is booming and we don't see very strong wage increases and little pass through to prices.

However, we still see improvements in the path of inflation, so the bottom line is not bad at all. It means that the economy may have unexploited potential capacity.

If slack is bigger, wouldn't it mean that the path would be shallower?

Well, other things being equal, it would be shallower.

Another explanation for low wage growth is that there is some backward looking behaviour in the wage-setting mechanism and in inflation expectations. For the time being I take an open view.

Monetary policy also plays a role. Our asset purchases programme has a minimum horizon of end September, or beyond, if necessary, and in any case until we see a sustained adjustment in inflation. It is true that when we communicate about what happens beyond September, we want to have confirmation of data. Things are going in the right direction but there are a number of uncertainties. So the message is patience, persistence and prudence.

When do you need to communicate what happens beyond September?

It can't be on the very last day, but I argue that it shouldn't be too early, either. There are not too many Governing Council meetings between now and then.

Can you see a scenario that you use the June projections to set up the decision and then take the actual decision in July?

The projection exercise is always important. There are four per year and there will be one in June. It's true that projections in general are useful to guide and structure the discussion. But I would caution against concluding that there is a pattern. We should not put too much weight on the projections.

You mentioned earlier that market expects the first rate hike between the first and second quarters of next year. How comfortable with those expectations?

The expectation implicit in the forward curve currently fluctuates around April and May. Market analysts' surveys indicate expectations of a slightly delayed lift-off.

I don't want to directly answer this question. But we try to judge if market expectations reflect fundamentals and if market expectations fit with our views. It's a two way street: we learn from markets and markets learn from us.

Markets have a certain expectation about the end of the net asset purchase programme (APP). We try to see if this fits with the narrowing between the inflation objective and market indicators of inflation expectations. For example, I look at the gap between the objective of below, but close to, 2% and the 5-year/5-year forward swaps-derived inflation rate, which is now at 1.7 percent. So the gap is narrowing, which is why markets don't expect us to continue the APP much longer.

Market prices seem to be coherent with the narrowing of the gap between market-based inflation expectations and our objective. It's a good sign. There is a convergence between market expectations and our intended end date, with the optionality.

When approaching the end of our net asset purchases, our forward guidance on policy rates will have to be further specified and calibrated as appropriate to keep inflation on a sustained adjustment path towards our inflation objective.

So what happens afterwards?

Our resilience criterion for inflation is not always understood well. It means that for us to stop net asset purchases, we have to be sure that the financial conditions resulting from that announcement will not compromise the adjustment path of inflation.

This path is by itself conditional on easy financial conditions. And financial conditions are the result of three instruments: net purchases, re-investments and our forward guidance on the short-term policy rates. Today we say that our policy rates will stay at their present levels well past the horizon of our net asset purchases. This indication is an important conditioning variable - together with net purchases and re-investments - behind the convergence path that we project today. As we move forward in time, as is natural, the relative importance of the three policy tools will change, and the main tool for shaping the policy stance will become the path of our key interest rates and the forward guidance about their likely evolution.

Today, by and large, markets quantify the "well past" interval as 'up to next spring'. But at some point that 'well past' indication will not be sufficient any longer to give the public enough information about the most likely outlook for our policy.

Once you stop net asset purchases the signaling aspect of the APP disappears and you therefore have to be much more precise about the future path of the short term rates. We will then have to re-specify our forward guidance on interest rates. Most likely, we will calibrate it as appropriate for inflation to remain on the sustained adjustment path to our inflation aim. So, as in the past, we will ensure that monetary policy controls the short end of the yield curve.

We will proceed at a gradual pace, or a measured pace as Mario Draghi said, in a way that is most appropriate for inflation convergence - taking into account continued uncertainty about the size of the output gap and the responsiveness of wages to slack.

You said you would have a discussion about revising your guidance in early 2018. Should you revise the guidance at the same time as you decide what happens beyond September?

Indeed, early this year, we have removed the easing bias on our APP. Looking forward, I wouldn't say there is a date or a deadline. The fact that the two of us are talking about this now is already very telling. However, I would not revise the guidance too early, because that could send wrong signals about the end of our net asset purchases, which in any case are intended to run until the Governing Council sees a sustained adjustment in the path of inflation.

When the first rate hike comes, will you only raise the deposit rate until the interest rate corridor is symmetrical again?

I think it is premature to discuss this question. What counts in the markets is the level and the expected path of the very short term rate at which banks borrow from each other. Because of sizeable excess liquidity expected to prevail over the coming years, the deposit facility rate is the policy rate that can steer those interbank conditions most effectively.

Benoît Cœuré argued that given your oversized balance sheet, you can retreat from asset buys without risking an unwarranted rise in term premia. Do you agree with this argument?

The float - the share of government bonds held by the private sector - has decreased with our purchases. The marginal intervention you have when the float is smaller has the potential to have some bigger effects on prices. But it is a relatively short term consideration. The main argument, and I think it is widely shared, is indeed that the stock becomes very important.

But we need to consider that, as the stock ages automatically, there is a 'duration leak', if you do nothing about reinvestments. That's where the reinvestment comes in. And we say we will reinvest for an extended period of time. It's another element that in the future will have to be clarified.

Obviously, reinvesting in short term debt has not the same impact on duration as investing in longer term debt. The Fed tries to be relatively market neutral. That's something we may wish to clarify in the future.

Clearly, we will remain an important participant in the bond market. But we do not dominate the market. Markets will continue to play a very strong role. That means our actions are very much scrutinized.

The influence we have on market prices reflect the credibility of our policy, of the central bank and the inflation objective.

Is tapering at the end of the programme necessary?

Let me start with what I read from markets: markets expect us to avoid cliff effects. I think it is a good illustration of the interaction I mentioned before. It's again a question about being coherent with what you say. When you say you have a policy of prudence, because of a number of uncertainties, what you will do at the end of the program must be coherent with that sort of communication. So people can judge what is coherent with a notion of prudence and could conclude that we will gradually bring net asset purchases to an end once the Governing Council sees a sustained adjustment path. I think it is well understood by the market.