16
Examination of witnesses
Peter Praet and Daniel Gros.
Q101 The Chair: Welcome to this session of the Economic Affairs Committee. We will be hearing from Peter Praet, who is the former chief economist of the European Central Bank, and Daniel Gros, who is distinguished fellow at the Centre for European Policy Studies.
I will begin by asking the first question. Do central banks lack a clear exit strategy from QE?
Peter Praet: Thank you for your invitation. I would not say that. In the ECB we already had experience of an exit in 2018; we decided to stop QE and it worked pretty well. When it fits very well with your reaction function and it is well communicated to the market, exit is not necessarily a problem. The problems come when your targets are not achieved within a reasonable horizon of time and then you go into purchaser bonds for a long period of time, and it never ends. You start accumulating a lot of government bonds on your balance sheet. That is what some people in the market call QE infinity. That is one risk that central banks could face.
The other risk is that inflation surges too fast, which is the other side of the risk and is not impossible, of course. There you get a difficult situation, because when the central bank has to increase rates because inflation increases faster than expected, there is of course tension with government public finances because public finance can get cheap central bank funding. At the same time, the central bank is one of the biggest owners of government bonds, so in a way it risks shooting itself in the foot when it increases the rate. That creates a tricky situation, and we have to see how strong the governance of the central bank will be. The mandate is for stability, so normally the central bank—[Inaudible.]—situation is obviously more complicated.
So there are three scenarios. It works and then the exit is smooth. We in the ECB had another go at this in 2018, and now we are in a new situation with the current one.
The Chair: To what extent do you think that central banks will be able to unwind their QE programmes independently of one another?
Peter Praet: It is a very important question. There are two points. First, you have the flow effect—the volume of bonds that you buy each month, which is easier to do. The other is the stock. What do you do with the huge amount of stock that you have on the balance sheet? On one hand, you want to reuse it, because you have accumulated in Europe something like 30% of government bonds on your balance sheet. For a new shock, if interest rates remain very low you want to build capacity there.
The stock effect is quite important for the market, because it means that if you reduce your stock you re-inject duration in the market and you may have a shock on the interest rate on the term premium, on long-term rates. In the ECB we have an exit on the stock that is very long. We have announced that we will reinvest the stock for a very long period of time, which is fine to counter the markets.
The problem we had in the ECB was that there was a succession of shocks: the global financial crisis, the euro crisis, a bit of a slowdown in the international economy, and then the Covid shock. Each time you increase your balance sheet, the amount of stocks that you have accumulated on your balance sheet increases further. That is a real issue.
The ECB communication today is that one part of the stock—let us say the legacy stock—plus the classical QE will be reinvested for a very long period of time after you have stopped the flows of buying. For the Covid portfolio, which is also very sizeable, it is not clear. It will be reinvested at least until the end of 2023, or more if needed, but that is very unclear. These are huge amounts in the markets, and this is one of the main issues with the balance sheet policy of central banks.
Daniel Gros: As Peter Praet already mentioned, exit could mean exit from the flow of buying or from reducing the stock effects. My view is that central banks do not have a good exit strategy, basically because they are waiting for inflation to increase so that they have an excuse to reduce the buying and then perhaps later also reduce the stock they are holding.
I also believe that the impact of QE has been vastly overestimated. That means that if the central bank bond buying has little effect on the real economy and on inflation, the central bank might have to go on buying for a very long time without really seeing an impact on inflation. That leads to what Peter mentioned, QE infinity.
That is the underlying problem. We do not have a good theory as to why QE should work. Central banks believe that it does, but if the effect is much smaller than they assume, they might be condemned to buy ad infinitum or they must be rescued by luck—namely, an increase in inflation.
Q102 Lord Bridges of Headley: Peter Praet, can I pick up your remarks and ask you to develop them a bit? You have recently said that the relationship between central banks and Governments is in “a sort of honeymoon situation because their interests are aligned”. What happens when those interests begin to diverge? Could you spell out for us in a bit more detail what you think will happen at that point?
Peter Praet: Yes, indeed. As Daniel Gros just said, it is very difficult to assess the impact of QE because it will be related to the context in which you operate. When markets do not function properly, QE is quite important. With the Covid crisis, many central bankers were convinced that having even easier financing conditions would not really stimulate the economy. If you are in a situation of war, people are not going to be influenced very much by lower interest rates or financial conditions in general. That is one of the considerations.
Today, central banks, and certainly the ECB, basically try to preserve easy financing conditions, which facilitates fiscal policy because fiscal policy can issue at low interest rates in general. With the two fiscal policies together, supported by expansionary monetary policy of a sort of QE, you could say that the central banks empower fiscal policy by what they do.
That is fine as long as the interests are aligned. The acid test comes when inflation goes up. On the one hand, that is good news for the central banks, because that is what they want to achieve, but inflation going up too much is a possibility, as you know. I think the probability is low, but it is a possibility. Then, of course, interest rates will increase. Governments may argue that it does not matter, because they have borrowed the money long term, so they will not be hit very much by an increase in interest rates. However, as you know, the central banks have all that stuff on the assets side of the balance sheet, and if they have to increase rates they will have less revenue because they have to pay on the liability side the money market rates that they decide. Their income will fall and so the money they redistribute to the Government will also fall.
One big central bank of the euro system—the Bundesbank in Germany—has decided this year to risk provision and to avoid distributing profits to the Government. It decided that recently for risk management purposes, I think rightly. That shows how central banks have to prepare to have sufficient provision and the Governments must know that the central banks will decide the inflation rates. At that point there may be a divergent view between the central bank and the fiscal authorities. That is not the case today, but it may come.
Q103 Lord Livingston of Parkhead: I want to come back on the issue of exiting. You obviously talked about the ECB experience. Peter Praet, you said the ECB had exited to some degree, but the oldest and biggest QE provider, the Bank of Japan, has not managed to do so. Is this a bit like an addict in that it takes more and more to have an effect, and they can stop when they want to but they never quite want to? The ECB is probably particularly disciplined because of the separation, but for central banks generally this will be really difficult, particularly if they have economies that are naturally deflationary and we are stuck with this for life.
Peter Praet: You are quite right: this is a worry. There are two ways of seeing the problem. From my experience, we succeeded very much in influencing financial conditions and supporting the economy and inflation. It took time. We had a number of shocks, and you cannot draw conclusions without looking at what happened in the European Union. We had the global financial crisis, but then we had the euro crisis, the slowdown of international trade and the impact on manufacturing and then the Covid shock.
Whatever the reasons for that, the result goes in the direction of what you say. Each time, you accumulate an increasing share of assets, and it is very difficult to know what the markets are today, given the size of the balance sheet of the central banks. How would markets function in the absence of central bank interventions?
Whatever the reasons, there has been a sequence of interventions, I think for good reasons. When I was there we postponed QE until 2015 for good reasons, some of which we are discussing now. It is not so obvious for a central bank to start buying government debt. We tried negative rates, and we tried forward guidance, which, if it works, is a cheap substitute for QE, for managing expectations. But it does not really work; you have to put money on the table at some point.
The big danger is that the markets get used to that, as Daniel said, and at some point it is very difficult to exit because markets start to expect too much from the central bank. With central bank presence in the markets, especially when markets are not functioning properly—when you have disorderly market function—risk managers of private banks start to build the reaction function of the central bank into the risk models, and then you get into a vicious circle whereby the central bank is increasingly taking care of the markets.
Usually central banks try to say, “Look, when there is a big shock in the economy, we have to use QE and non-conventional measures”. Of course, when things normalise you toughen regulation on the banks, on the financial markets in general, and you take care of moral hazard after the shock.
When you have a succession of shocks and intervene every time, like in the last one, the markets start to get addicted to that. It is a real issue. When the Covid crisis is over, we will have to look at how the market functioned. In the US, the central bank had to intervene massively in the most liquid market in the world, which is the Fed bonds. That is incredible. There are lessons to draw, but usually after. I admit that there has been a long period of central bank intervention in markets, and obviously it does not work.
Daniel Gros: The question illustrates exactly what I said earlier. If the impact of QE is second order and you have a naturally deflationary economy, you have to keep going and buying and buying. The one difference between the eurozone and the rest of the world is that in the euro system you cannot buy ad infinitum, because practically speaking you are limited to the 33% of the outstanding stock of any one maturity. That means that you cannot buy more than one-third of the government debt outstanding, and you cannot concentrate your holdings only on the long end because you cannot hold more than 33%, even of the longer-dated bonds, and as you buy more and more you have to go down the maturity ladder until you buy the shorter-trend stocks. For the ECB there is a limit to how much can be bought.
Q104 Lord Skidelsky: I am trying to draw together the two sets of comments and I am finding it a little difficult. On the one hand, Peter Praet has talked about a series of interventions necessitated by a series of shocks, which makes it very difficult to unwind the QE. On the other hand, Daniel Gros has said that it has had very little effect. He also said, which I thought was very significant, that there is no good theory of how QE is meant to work.
On the one hand, you have these continuous interventions, because that seems to be what is done. On the other hand, you have no idea how the interventions are meant to improve matters, and no theory about it. Is it simply a placebo, which is designed perhaps to make people feel good? If so, what is one really saying about monetary policy or the state of monetary policy at this point in time? Maybe I could start with you, Daniel, as I am taking up a specific remark you made.
Daniel Gros: I smiled when you used the word “placebo”, because that is also the word I have used in some publications of mine.
The famous joke by Ben Bernanke about QE is that it does not work in theory, and everybody agrees, but it works in practice and that is where the difficulties start. QE is basically the central bank, which is one arm of the Government, buying bonds from the other arm of the Government. It is a left-pocket/right-pocket operation, which in a perfectly working market has zero impact. But markets are not totally perfect, are not full arbitrage opportunities and so on, and people have a preferred habitat, so there might be some effects. How large are they? It is difficult to estimate. I would say that the quieter and more developed the markets are, the lower the impact. It must be a second order in the long run.
Quantifying it has been exceedingly difficult. You might have seen the recent paper by some economists showing that economists from central banks have in general found QE to have a strong impact on inflation, whereas independent economists have not. There is a certain bias in the available evidence. I repeat that, in an emergency, when markets are in upheaval, central bank bond buying can be very important. Secondly, I do not think that interest rates can be lowered by a significant amount—let us say, more than 50 basis points—by bond buying by the central bank.
Peter Praet: I agree with Daniel. If markets functioned perfectly, QE would have no impact. There are a lot of frictions in the markets, and that is where the discussion starts.
There are basically three channels, as you know. There is a portfolio rebalancing effect, because of the central bank’s extra duration. I remember when we tried to calibrate how many bonds you should buy. It is not obvious; is it €500 billion, €800 billion? How much do you buy from the market? We estimate how much we need to take from the stock of bonds in the market and how much central bank liquidity to give in exchange. What sort of impact will that have?
Then you have the expectation effect, because that is a strong signal from the central bank of future short-term rates when you intervene in markets. The expectation channel, which I think is key—and I think Daniel would agree—is very difficult to manage and to assess, but it is a key channel.
The third is liquidity. It is the flow effect when the central bank intervenes at some point in time. It can have a very big impact at that point in time and avoid, for example, non-linear adverse multiple equilibrium effects. When markets start to panic and the central bank intervenes, you can stop the panic and calm it down. It depends on the context.
What I do not like usually, and where I agree with Daniel, although I think the impact is much bigger than he says, is that you cannot just use elasticities: you put in €100 billion of QE and you get X basis points in long-term rates, and that will stimulate growth by a certain percentage over some period of time. It is context-dependent and it is an art.
The bottom line is that QE had a big impact on financing conditions in general and at some point it supported growth and inflation.
Q105 Viscount Chandos: What are the differences between the ECB’s QE policy and that of the Bank of England? In particular, do you see those differences having led to any tangible difference in stimulating the real economy?
Daniel Gros, you have flagged your scepticism about the impact in terms of growth, whereas, Mr Praet, you said in your speech in Frankfurt in 2018 that you estimated that the ECB’s QE had added nearly 2% to both GDP and inflation in the period between 2016 and 2020. How do you think that compares with the Bank of England’s success?
Peter Praet: From the estimates I have seen—Daniel is right to say that one has to take these estimates with caution—I remain convinced that QE both in the UK and in the ECB has led to good results in growth and inflation. It is very difficult to prove; the counterfactual is very difficult. But there are a number of very detailed granular stories. I headed the economics department, where we did a lot of research into the impact of QE. I can tell you that it works quite independently. We saw some of the results.
What strikes me the most when you compare the Bank of England with the ECB policy is that the ECB has embedded QE in very sophisticated guidance. Look at the guidance on interest rates. Interest rates are going to stay low or even lower for an extended period of time. There is a whole bit of guidance linked to what we call state contingency: you keep the rates low or even lower until you see inflation in the eyes, when the results are there. That is in the ECB’s September 2019 guidance.
What is different from the UK is that the ECB also links the whole package of measures. QE is linked to the guidance on interest rates. For example, the ECB says, “We’re going to continue to buy €20 billion per month and we’ll stop buying shortly before interest rates increase, but interest rates will increase only when inflation is achieved”. So there is a sort of sequence between QE and interest rates, and the two are very much linked. We used to say that they were chained together. We thought, and it is still ECB policy, that that would bring the maximum in terms of guiding markets’ expectations of future interest rate policies and the whole maturity curve would go down. That is one difference.
The other big difference is Covid, and Daniel may expand on this. There are now two QEs in Europe, which worries me, not in the present circumstances, which are exceptional, but over time. One of the QEs, the one I described, which is very much linked to the guidance on interest rates, has not changed since September 2019. The other is the pandemic emergency purchase programme, which is a different sort of animal, because there the intention is not only to compress the risk-free yield curve, but to ensure that easy financing conditions are spread in all the different jurisdictions.
The ECB now also looks at the weighted average of the sovereign yield curves of the different member states, which is quite new. As Daniel mentioned, the limits on how much you buy in different jurisdictions does not apply under that programme. The problem for the ECB, of course, is the exit from QE when the pandemic period finishes. What will the transition be to a more normal QE? The UK does not have that sort of problem.
This is one of the main challenges of QE nowadays for the ECB: how do you organise the transition from a very flexible QE across countries where one central bank tries to have easy conditions in all the different states of the monetary union, which works pretty well? You can see that the spreads did not fail very much, so you could say that this intervention worked pretty well. The question is how you exit from that. When things normalise you could expect that to be relatively easy. It will not necessarily be the case, of course.
That is one of the biggest challenges and a big difference compared with the UK. The main difference is that the QE is embedded in complex guidance in the euro system.
Daniel Gros: There are fundamental differences between the way QE is executed or can be executed in the eurozone and in the UK, and they depend basically on the fiscal framework, which is very different.
Let me start with the limit on holdings. Peter mentioned that within the PEPP, the pandemic emergency purchase programme, the ECB has said that it will not stick totally to this limit, but in the medium to long run these limits hold. That is the first difference, because the Bank of England might be able to buy all the long-term debt of the British Government and therefore extract much more duration than the ECB can ever do, because the ECB can only buy one-third of the entire spectrum. So the duration extraction is limited.
The second point, which is more important, is that I misspoke. I said that the ECB can buy. In the euro system, as set up, it is not the ECB that is buying government bonds. Most of the buying is done by the national central banks themselves, and—that is key—on their own risk. When we read that the ECB is buying hundreds of billions of Italian Government debt, that is wrong. It is the Banca d’Italia that is buying, and it is buying at its own risk. That comes back to the left-pocket/right-pocket nature, which is at the federal level in the UK and in the United States but is at the national level in the euro area.
This has one key implication, because part of the aim of the QE in the euro area was also, as Peter mentioned, to compress risk spreads. Most of the academic estimates of the impact of the PSB highlights the impact it had not so much on the interest rate, but on Italian and Spanish risk spreads.
That is where the question arises: can you really expect some permanent impact on these risk spreads when all the buying is done by the national central banks at their own risk? There is no risk transfer involved. That is why I would say that the QE, as practised and as possible within the current fiscal set-up of the euro area, can only be of limited importance.
Periods of big market turbulence are different. March 2020 was different, but during periods when the markets work pretty well, when people have time to think, “What are the Italian Government worth over the next 10 years?”, the fact that the Banca d’Italia is buying its own government bonds will not have a big impact.
Q106 Lord Fox: We might come back to this when Lord King is asking his questions, because I still think there is some clarity needed.
Going back a question, the co-dependency of the market and central banks in terms of this huge central bank balance sheet was alluded to by Mr Praet. What risk does this hold?
Expanding on that, one of our previous witnesses talked about the decoupling of what he called the real economy and the market economy as a result of this process. Do you see this as one of the risks? If not, what are the risks of this co-dependency?
Peter Praet: There are two questions there. One fundamental question is that central banks have intervened massively in the bond market for a long time, since 2008, and it is very difficult to imagine how the bond market would function without the intervention of central banks. This is a fundamental question in trying to explain why. It may be bad luck, because we had a sequence of shocks, but that is the reality today.
The last shock was a repeat. You will remember the discussion that was often had that banks are much safer. They have much more capital than before, so what will the next crisis be? That was before the Covid. The question was how the central bank would react if the asset managers or the money market funds, the non-bank financial institutions, started to sell? Would the central banks intervene or not? Now we know the answer: the central banks have intervened massively in these markets, as you see in the US with the T-bonds market. This is a very uneasy situation.
One has to look carefully at regulation. The central banks cannot always come to the rescue when something does not work in America. Of course Covid was a big shock. You can understand to some extent why the central bank would intervene. It is very difficult to swallow the fact that the central bank had to intervene massively in the US bonds market, so you have to look deeper into the system at what sort of measures you have to take. It will take some time, but it is a necessary exercise.
There is another point to make about the relationship between government and the central bank. As you know, Governments issue long-term debts. Part of that is bought by the central bank, and the central bank issues central bank money, so there is base money creation in the system. When things change, interest rates have to increase, so the first question is whether the central banks will be inhibited in doing what is necessary, given their mandate, or whether they have a conflict of interest because they are the biggest owners of public debt? Governments should know, of course, that the central banks may have a problem of profitability, depending on the interest rate increase later on. In the UK with the Bank of England, you have the P&L problems; the problems of risk related to QE are outside the Bank of England. As you know, these risks are taken by the Treasury. That is a fundamental difference with the ECB, because the risks are taken directly by the central banks.
As Daniel correctly said, what concerns the purchasers of national sovereign debt is that it is done on the balance sheet off the national central banks. By the way, I have been very much in favour of that policy these years for different reasons. But the national central banks should buy and take the risk on themselves. That, to some extent, would send a signal to the national treasuries that there is some circularity there and that they should be careful about the sustainability of public debt. The UK’s situation is quite different, because the P&L issues are taken in a transparent way by this special vehicle.
Daniel Gros: One short comment. The statement that a large participant in the market would somehow change prices is not compatible with the view that the market is efficient. There are enough hedge funds and other people around who think, “What is this government bond worth” and they will price it accordingly, independently of how much is held by the central bank. Therefore I think that the impact of large central bank holdings on the government debt market is overrated.
Q107 Lord Haskel: Mr Praet, you told us about the differences between the Bank of England and the ECB. The decision of the ECB to extend its asset purchase programme sends out certain signals to the market. The Fed and the Bank of England also read these signals. What impact does it have on their policies?
Peter Praet: That is a very important point. You know that market participants tend to look at central bank balance sheets on an aggregated basis as well. They look not only at their own central bank but at the balance sheet of all central banks, basically because central banks, by these purchases, want to influence risk behaviour, and they do in practice, whatever the mechanism.
When the ECB started its QE programme in 2016, there were spillover effects on the rest of the world, because you buy bonds from non-residents—basically they were the first sellers of these bonds—and in exchange they get cash. What do they do with the cash? They may buy US equities and US dollars. That would have an impact on the exchange rate and on financial conditions even in the United States, so there would be a spillover.
That is always the case when a central bank changes its monetary policy, an important central bank being the Fed. Personally I do not see big differences with interest rate interventions historically. The context is different, of course, but for me it is always the same story. You change your monetary policy compared to the others, so there will be impacts on financial conditions in other countries. One of the channels is the exchange rate.
Daniel Gros: I would interpret the signalling effect a little bit differently. My view is that if a central bank announces that it does QE or increases its purchases, it sends out the signal that the economy may be even more deflationary than people had thought. The central banks are pretty good observers with inside knowledge. Therefore they will lead markets to reassess their expectations of future interest rates. Since the euro area is pretty much integrated in the rest of the world, any downward assessment of the euro area will have a similar impact on the rest of the world—a muted impact, but still. That could mean that in this way it could be transmitted from one central bank to another.
Lord Haskel: Can the Bank of England learn any lessons from the European Central Bank?
Peter Praet: No, I do not think so. It has a good framework. When I compare the communication of the two central banks, what strikes me—and I understand why—when you look at the communication of the Bank of England compared to the ECB is that it is much simpler. As we know, it is a very complex situation, and the very complex environment of the euro area. I would not really draw lessons and try to use more complexity in the framework of the Bank of England. I do not think it is necessary.
Q108 Lord King of Lothbury: Good afternoon, Peter and Daniel. Some central banks have been talking openly about practising yield curve control. One or two of your comments this afternoon, Peter, and certainly Christine Lagarde’s recent comments, suggested that the ECB was indeed practising yield curve control. Would you accept that as a valid description? If so, do you not worry that the independence of the ECB in practice is jeopardised if people come to think that the policy is one of putting a cap on long-term interest rates?
Peter Praet: I fully follow your point, Lord King. The ECB says that it is not doing yield curve control. It is very clear in its communication. At the same time, it wants to ensure favourable financing conditions during the pandemic period, so it is relatively vague and it brings confusion in the market. Is it doing it or not? It is true that when you do QE and you calibrate QE, you try to see how much you should buy in the markets, and how you communicate by signalling to reach a certain impact on long-term rates. You do it for a period of time, and it can be calendar-based, state-contingent or whatever, and then you look at the reaction of the markets and decide to recalibrate and to recommunicate or not.
When you do yield curve control, you fix the price, which is not easy to do and the quantities have to adjust. When Mario Draghi came with the “whatever it takes”, he did not have to spend any QE money; it was just that markets believed it. That is in the good state of the nature when you can influence expectations in some special circumstances. If markets do not believe you, it can cost you a huge amount of money. It would be totally incompatible with the treaty of the euro area, because the central bank would fix the price and adjust the quantities—whatever it takes. It is a very risky game.
Japan does it, but the situation is quite different for the euro area. Japan is not comparable to what happens in the euro area. There is a difficulty with yield curve control; this is why the ECB also comes with financing conditions, which is a vague concept. In my time, pre-Covid, we looked at trying to influence the risk-free curve. Now the ECB wants to ensure easy financing conditions everywhere in other jurisdictions, and it has succeeded in doing that. What it did is amazing and very bold. The question, of course, is what you do after when things start to normalise, because all the policy that is being done now is for the pandemic period. The transition post-pandemic will be tough.
It depends if things normalise, the economy is better, and inflation goes up a bit. That is a good scenario, of course. It may be much bumpier than that. I fully agree with your point. First, it said that it was not doing yield curve controls and it protested, but at the same time the markets tended to believe that it was leaning in that direction because it wanted to ensure easy financing conditions. It pulled back by saying, “No, that’s not really what we mean”. That communication is a little difficult for the time being. Admittedly the situation is not easy, though, so I have a lot of sympathy with what it is doing.
Daniel Gros: I think the ECB would like to be able to do yield curve control, but it cannot. One reason is: what long-term interest rate? It is the average of the euro area. Then you have the risk spreads for one half of it. The second point, which I made earlier, is that the ECB as a euro system is limited in the amount of bonds it can buy; it is limited basically to one-third of the overall outstanding.
If you put that together with my own opinion that the QE has very limited impact, this means that you would be in a very dangerous situation in which you would have to buy big amounts, but you cannot. At the same time you would have to be held responsible for risk spreads within the euro area, which are largely outside your control. If you get a euro-sceptic Government in Italy, what will you do? Therefore, I agree with Peter that the ECB should make it clear that it does not want to do yield curve control.
Q109 Lord Fox: In your first answer, Mr Praet, you talked about it being easier to exit when you reach your targets. In your fourth answer I think you started to talk about what those targets might look like. They all looked incredibly nuanced to me. In your view—and I would also like Mr Gros’s view—what are the targets of the ECB and the Bank of England? How will we know we are there?
Peter Praet: The communication on QE. I am not talking about the exceptional pandemic QE—
Lord Fox: I am, to be honest, because that is where we are now.
Peter Praet: In a way it is easier for the pandemic, because QE will stop when the governing council assesses that the pandemic crisis is over, which is very vague.
Lord Fox: That is extraordinarily vague.
Peter Praet: That worries me personally. You start thinking about the transition to the post-pandemic environment, and I am sure it is doing that. But when you go to the post-pandemic environment, you go back to the other QE, which stops before you increase interest rates, and interest rates stay low until you have seen the inflation outlook converge to a level sufficiently close to but below 2% within the projection horizon. The convergence has to be consistently reflected in underlying inflation dynamics.
I was not there when this guidance was drafted. It is a bit complicated, but basically the idea is: “It’s an open-ended QE. I will do QE as long as needed, and when I see inflation I will stop QE”. This is indeed what we call an open-ended QE, and it raises one of the risks we mentioned before: if, for any reason, monetary policy does not work, you end up with something that is very difficult to exit because you committed to do something and the results do not show up. It is what people call QE infinity.
Lord Fox: To Infinity and Beyond, I think, is the title.
Peter Praet: To prepare for the post-Covid environment, the ECB is engaged in a strategy review. That is a good opportunity to rewrite the whole guidance framework and to reintegrate the two QE programmes in an orderly way. It will not be easy, but I think it is feasible.
Lord Fox: Mr Gros, you have a less clear view on the effectiveness of QE or you have a clear view on the ineffectiveness of QE in the first place. Presumably targets will never be reached, because it is not doing anything.
Daniel Gros: Yes, it is very likely, and then you have the typical situation of an irresistible force meeting an immovable object. The irresistible force is QE as long as inflation has not reached 2% and the immovable object is the limitation of the treaty, which you can get around for a while but not fundamentally. That is why the ECB either has to change its talk or has to continue QE but buys different stuff—private sector bonds, private sector equity. There it does not face that many limitations. If you go outside the realm of buying government bonds, yes, the ECB could go on for ever, but for government bonds it has its limits and it risks getting into a very difficult situation.
Q110 Baroness Kingsmill: The discussion that has preceded my question has basically suggested that QE is very effective in times of crisis, but we have become accustomed to QE, or rather perhaps Governments have become accustomed to QE, and it has become much less effective. Perhaps the current situation with the turbulence in the bond market at the moment is symptomatic of this, in the sense that yields are rising rapidly while returns are consequentially dropping.
Does this mean that the central banks are running out of options? The bond markets and the financial markets generally are very sophisticated now, and they can anticipate what is going to take place and so perhaps pre-empt some of the issues that the central bank might otherwise have undertaken. Things like growth rates and economic earnings growth and so on are not being taken sufficiently into account by the central bank perhaps.
Peter Praet: They are good points that you are making, and they are difficult questions. Think about the spillover from the US. You will have another discussion about the US policy. The central banks basically control the short term, the money market rates, and then by forward guidance they can have some influence without any QE two or three years later. That is the horizon. After that, they can say that rates will remain low for an even longer period, but markets will not believe it, so central banks that buy assets extract duration to show that they mean it.
They can also intervene heavily for some months in the markets just to show that they lead the show and they can do it. This has limits, of course, and the fundamentals will determine the direction. What you see in the United States is that the support of domestic aggregate demand is so strong because of fiscal policy that the long-term rates go up despite the communication from the central bank. The short end is very well controlled, but you do not know the long end. This is a question.
Then you get the spillovers to Europe. I must say that there is room to decouple the European long-term rates from the US. We did it in 2013 when there was the taper tantrum in the US. It is a mix of fundamentals. Markets assess that the fundamentals are different, but they also look a little bit at relative monetary policies.
All in all, we succeeded in the ECB to decouple the long-term European rates, the risk-free curve, from the US curve for a number of years—some people say by luck, other people say it was because of the guidance, or whatever. It is a mix of both, probably.
In the situation today, I would say basically the same. The ECB has expressed its intention to intervene in markets on a short-term basis to avoid the unwarranted tightening of financing conditions that would come from the long end of the bond market. That is the sort of strategy that can work in the short term. You just want to be present in the market that is punishing some of the speculators, but you cannot just lean against fundamentals.
The ECB’s assessment, which is wide, is that fundamentals are quite different from the US: “If there is a spillover from long-term rates in the US to Europe, we think it’s unwarranted and we’re going to go in the markets and try to hit you at some point”. Market participants will pull back for a while and see how the fundamentals will evolve. It looks a little bit like foreign exchange interventions, but here we know the budget. They have a quantity there, about €1 trillion, and they say they may use them more on a quantity basis.
I do not think it changes the fundamentals very much. It can help if this phenomenon does not take on too much dynamism. When markets gain momentum it is very difficult to stop. Intervening at some point in the market allows you, if you do it well, to calm down a bit and to gain time. It does not change anything fundamentally, but it can be useful. It is not easy to do, I agree. Thank you for your question.
Daniel Gros: I have two short comments. The recent increase in interest rates was not in disorderly market conditions. That is totally different from last year, when there were risk threats and increases and people were no longer able to do arbitrage with the markets. Here we just have a reassessment of the economic outlook for the United States. Expectations about firms’ profits matter, of course, because they have an impact on investment, demand and so on.
All that matters for the market, which is natural when people think that the US is expanding much more and the euro area is lagging behind. But it cannot lag behind for ever. Part of the reassessment should also concern the euro area, so I do not think there is any reason for the ECB to be too concerned. People know that the euro area economy is much more sluggish, but perhaps it is a little less sluggish than people thought earlier.
Baroness Kingsmill: What should the central bank be doing?
Daniel Gros: Nothing.
Baroness Kingsmill: Nothing? Just carry on as is?
Daniel Gros: At this particular juncture.
Q111 Baroness Kramer: Is the central bank mandate a narrow mandate or a broader mandate? Are we getting a broader mandate, whether covertly or overtly, and should it be reined back?
Daniel Gros: I would keep the mandate as narrow, as the bank is independent.
The Chair: You can expand on that if you want.
Daniel Gros: The more independent a central bank is, the narrower its mandate has to be. Translation: the ECB’s mandate is totally independent, so it had better stick to a very narrow mandate. In other countries where the mandate can be arranged or changed by political decisions more easily, you can have a wider mandate or a wider interpretation of the mandate.
Baroness Kramer: Mr Praet, would you say that where the mandate expands independence disappears?
Peter Praet: This is the first time I fully agree with Daniel on this. I would add only that in the ECB’s governing council deliberations it should devote more space to the longer-term consequences. I think it is doing that now in the present environment, but it should have more discussion about the side-effects of policies—it has communicated a lot on proportionality—and how financial stability considerations intervene in the decision-making on monetary policy in the short term and the longer term.
It has a little too much on inflation, and we know that price stability is not enough. It does not mean that we have to change the mandate—I fully agree with Daniel—but it is a bit short in saying, “We’ll look only at price stability”. In its deliberations, it also has to explain how the other considerations intervene in the decision-making process. Naturally, you could say that in the short term that you have to intervene, but in the longer term it may create financial stability risks. You then have to explain how this trade-off has been managed. I fully agree with Daniel that independence goes together with a rather narrow mandate.
Baroness Kramer: I am curious. How do you distinguish between a blurring of the boundary, between fiscal and monetary and simple co-ordination?
Peter Praet: There is no quid pro quo in principle. That means that if you do your fiscal expansion, I can guarantee you that interest rates will remain low for a long period of time. That commitment is conditional on inflation price stability—that objective. Where it becomes difficult, and I follow you fully, is when the communication starts to go in the direction of saying, “I will ensure that financing conditions remain for a very long period of time”. Yes, maybe, but it is conditional on what happens with inflation. Treasuries should know that it is conditional, and the risk management at the Treasury should take that into account.
It is not a blanket guarantee that says, “Don’t worry. The rates will remain low for a long period of time, whatever happens”. What worries me a little bit, to be frank, is that if inflation goes up one day and the central bank increases the rates, I can see the Minister of Finance saying, “You promised you wouldn’t change your interest rates. Now you increase your rates and you get me in trouble”.
One has to communicate continuously that this guidance about future short-term rates is conditioned on inflation and that inflation may increase one day. They may succeed even more than they hope now. Who knows? That is a question of risk management.
The difficult situation today is that monetary policy during the Covid shock is not the most efficient tools to stimulate the economy. You need fiscal. The zero lower bound means that you have to buy government bonds. That creates a complementarity, but it does not mean what we call a quid pro quo, where you say, “You do this and then I do that”. I do this, but it depends on inflation, my objective.
That is not always understood in communication, and certainly not, in my view, to the general public. If the day comes when interest rates have to go up, the Treasury will say, “No, you promised not to increase the rates for a long period of time”. Then I will have to raise taxes to do something. Then it goes back to the citizens, and the citizens say, “What is the central bank doing?”
Communication from the central banks must be very clear. Now we will see, because ex ante you do not know. You only know after whether the system was sufficiently strong in terms of independence. You never know ex ante. You have to see ex post. It is difficult to tell today.
Daniel Gros: In the eurozone, when the lines between monetary and fiscal policy are blurred, the Germans go to court.
Baroness Kingsmill: That is one answer.
Daniel Gros: But they do not succeed. They never succeed.
The Chair: On that note, I thank Peter Praet and Daniel Gros. It has been a very interesting session, and I thank you for your answers, which have been very helpful to the committee. We are going to move straight on to Stephen Cecchetti next. You are very welcome to stay and listen to the session if you wish to do so.
