Treasury Committee
Oral evidence: Bank of England Inflation Reports, HC 596
Tuesday 21 November 2017
Ordered by the House of Commons to be published on 21 November 2017.
Members present: Nicky Morgan (Chair); Rushanara Ali; Charlie Elphicke; Stephen Hammond; Stewart Hosie; Mr Alister Jack; Alison McGovern; Catherine McKinnell; Kit Malthouse; John Mann; Wes Streeting.
Questions 1 – 102
Witnesses
[I]: Sir Jon Cunliffe, Deputy Governor, Financial Stability, Bank of England; Ian McCafferty, Member of the Monetary Policy Committee, Bank of England; Dr Gertjan Vlieghe, Member of the Monetary Policy Committee, Bank of England; Michael Saunders, Member of the Monetary Policy Committee, Bank of England.
Written evidence from witnesses:
– Sir Jon Cunliffe’s annual report to the Committee
– Ian McCafferty’s annual report to the Committee
– Dr Gertjan Vlieghe’s annual report to the Committee
– Michael Saunders’ annual report to the Committee
Witnesses: Sir Jon Cunliffe, Ian McCafferty, Dr Gertjan Vlieghe and Michael Saunders.
Q1 Chair: Good morning, gentlemen. Thank you very much indeed for appearing before the Treasury Select Committee this morning. It is very good to see you all here, and we are looking forward to hearing your evidence.
I wanted to start with a general question to all of you. Obviously, there are four of you. Committee members will try to direct questions to individuals, but of course, if there is something you particularly want to add to an answer, please feel free to do so, but please equally do not feel that you have to answer every question, otherwise we could be here for a very long session. I know that people in the Bank are watching this session, but they may not want to be watching for the next six hours.
I wanted to start with the Governor’s appearance before this Committee last month, when he told us that inflation is “more likely than not” to peak in October or November. Do you agree with him and, if not, when do you think that it will peak? Sir Jon, perhaps we will start with you.
Sir Jon Cunliffe: I am still pretty much happy with that forecast. It has been a bit lower than we had forecast over the autumn, but past experience on the pass‑through of inflation, first into import prices and then that pass‑through into domestic prices, suggests that the peak is in the last quarter of this year, so I still think that is what will happen. Where, exactly the peak is—0.1, 0.2, one way or the other—is quite difficult to forecast.
Q2 Chair: Does anyone have a contrary view? No?
Dr Gertjan Vlieghe: It is important to add that we are likely to see the peak of inflation around now if the exchange rate stays around the current level and if oil prices stay around the current levels. It is very much conditional on that.
Q3 Chair: If there were to be changes in either of those things, that could affect the rate of inflation, or the peak.
Dr Gertjan Vlieghe: Yes, absolutely.
Q4 Chair: I also want to ask about the remit that the MPC has from the Chancellor, which requires the committee to trade off the inflation target against output and employment in exceptional circumstances such as those prevailing now. However, it does not specify how to weight that trade‑off, so that appears to come down to members’ personal choices. Please explain your personal preference for trading off inflation and unemployment. I might have set you an essay question there that could take quite a long time to answer, so if you could answer as briefly as possible, I would be very grateful. Dr Vlieghe, perhaps we could start with you.
Dr Gertjan Vlieghe: An important point to make off the bat is that even though in exceptional circumstances we have some leeway to vary the timing over which we return inflation to target, there is no permanent trade‑off. You cannot say that you would like a little lower unemployment for some permanently higher inflation.
In the end, the inflation target always prevails, and so the discussion is really about whether you try to bring inflation back to target over a period of 18 months to two years, or whether you are in exceptional circumstances such that you can take slightly longer, which so far, for us, has meant up to three years. We have not really gone very far beyond that. In the end, it is really a matter of timing, as opposed to how much weight you put on employment versus how much weight you put on inflation. Inflation always ends up being the dominant factor in the long run.
Q5 Chair: Mr Saunders?
Michael Saunders: I would agree with that. I am not sure there is that much difference between us over the trade‑off. Over the course of this year, inflation has risen well above target. It is likely to remain so for some time, even if we are close to the peak. At the same time, the amount of spare capacity in the economy has fallen faster than expected. That trade‑off has shifted and, for me, that was the reason why my vote on interest rates, in turn, moved. The economy developed in a way where there was less trade‑off than there had been previously, with the prospect that spare capacity will continue to fall. I felt that it was right that we move to a less stimulative stance.
Sir Jon Cunliffe: I have the same general framework for assessing the trade‑off, and the Governor made a long speech on this called “Lambda”, after the Greek letter in the equations. For me, the issue is that clearly, spare capacity in the labour market—slack—has gone down because unemployment has been going down, and the economy probably has to run at a slower rate than it used to, because its potential is lower, without creating inflationary pressure.
The key for me is that the inflation target applies at all times. That means that even after this burst of inflationary pressure from the sterling depreciation has passed through the system, we still have to hit the 2% target. In looking at our forecasts and what we do on lambda, my concern is that there will not be enough domestically generated inflationary pressure in the second half of the forecast to mean that, when the externally generated pressure passes, we are pushed back up to the 2% target. However, the overall framework—the trade‑off—I think I share with the rest of the committee.
Ian McCafferty: Of course, it is the mandate itself that says that inflation is our primary target, and the concern that we have for supporting the Government’s economic policies via activity and employment is subject to hitting that target. The only thing I would add would be to say, in addition to considering spare capacity, we also need to consider that, if we are to exercise a certain degree of tolerance about above target inflation of the sort that Jan described, in terms of returning inflation to target over a rather longer period than normal, we have to be very mindful about inflation expectations and whether that degree of tolerance therefore changes the inflation outlook itself.
Q6 Chair: Thank you. I wanted to turn to the most recent committee, and obviously the increase in the interest rate and the votes that were cast on that. Sir Jon, we received your homework slightly later than the others.
Sir Jon Cunliffe: My apologies.
Q7 Chair: However, thank you for your report. Obviously, you were one of the two members of the MPC—and obviously, we saw Sir Dave recently for his appointment hearing—who did not vote to raise the bank rate at the last meeting on 1 November. Please explain the reason—you have already begun to touch on it—for your dissent.
Sir Jon Cunliffe: My view at the last meeting was essentially around this question of domestically generated inflationary pressure, and within that, essentially, pay pressure. We have seen a tightening in the labour market. We have seen, at the beginning of my time on the committee, the fastest drop in unemployment for 40 years. We have now seen unemployment go down to 4.3%. For a while, we thought that drop in unemployment was not generating the pay pressure we thought it might, because there was just more slack in the labour market to do with under‑employment—people who the statistics were not picking up—and that was certainly true in the first couple of years that I was on the committee. In 2015, we had compositional effects. There was a range of explanations.
However, the concern I have—and it is a concern, rather than a proven fact—is that pay growth has just become less sensitive to the rate of unemployment in the economy and, as a result, I do not have the confidence that maybe I would have had two or three years ago that at these low levels of unemployment, over the forecast period, pay growth is going to come up to the 3%. The 3% figure that we have is not a huge figure. Now, some of the low pay growth has been to do with productivity, but this productivity, if you like, nets out. If workers get higher pay because of productivity, the economy can produce more, so inflationary pressure is not the result.
However, it is this pressure between the supply of labour, the demand for labour and what that does for the price of labour—wages—that I just have more doubts about. There is a big academic discussion about whether something fundamental has changed in the world. I am not saying that it has, but I am saying that in a world where some of those fundamental relationships have not operated for a while—it could just be that it is going to happen and it has not happened yet but it will do before my term ends—it could also be that something more has changed. I just want to see more evidence before I make the decision, and that is why I voted to hold rates, and that has been part of my thinking on the committee consistently for the last two or so years.
Q8 Chair: That is very interesting. Perhaps I could push you a little further. You talked about whether this is perhaps a new normal, and somewhere in the evidence that we looked at, one of the suggestions was that people are not moving jobs, which is something that drives up pay growth, in sufficient numbers, although perhaps that move in the employment market is beginning to come back. Whilst you said you wanted more evidence, do you think that we have reached some sort of new normal—that things have changed so dramatically that we are not going to go back to seeing the pay growth that we have seen in the past?
Sir Jon Cunliffe: Unemployed vacancy rates are back to where they were in 2015 and almost back to pre-crisis averages—a little below. Quit rates, people leaving jobs, went down after 2015 and have now come back up again. Separation rates, which is mainly people leaving jobs for involuntary reasons or “reason not given”, have remained fairly flat. If you look at the surveys, they are showing recruitment difficulties. The agents are picking them up.
I could have said all those things in 2015; in fact, we did say all those things in 2015. After 2015, these indicators went down again, and now they are coming back up again. Yes, I certainly take some information and some signal from them. It is a question of how much weight I wanted to put on that signal.
Q9 Chair: As a final question from me, perhaps I could ask the other three members—because you did all vote for an increase in the rate—where you are on this issue about pay growth. If some members want to see more evidence, I suppose the question would be have rates been raised pre‑emptively?
Dr Gertjan Vlieghe: It is a very difficult judgment, and I agree with all of the concerns that Jon has. In the end, the bit that has united the committee for some time is that, since the beginning of the year, we have been saying that over the forecast period rates were likely to have to rise somewhat, and really the difference is about exactly when you have accumulated enough evidence to start that process. That is a matter of pure judgment. There is no science to that.
What has persuaded me is that, first, compared with 2015, the unemployment rate is now another big step lower. We are down at 4.3%. I put a lot of weight on these measures of under‑employment that Jon talked about. This is people who have part‑time jobs but who would like a full‑time job, or people who are not working as many hours as they would like. However, that has fallen too, and as a matter of fact it has fallen a little faster than the unemployment rate, so that gap is also closing.
Then, we are hearing a lot, both in surveys and from our contact directly from firms, that they are now finding it more difficult to recruit, not just in isolated sectors but in terms of more broad‑based pressure, and that they are starting to respond to that by paying a little more. I am also hearing from a lot of people that their employees seem more confident to be willing to move jobs for higher pay, whereas previously they might have just stuck with what they had.
It is early days, but I do think there is some evidence that this process of labour market tightness leading to upwards wage pressure has now begun. It is not across the board. Like Jon said, we have been in a position a few times when we thought we saw early signs and then they faded again. If you wait until all the signs are lined up to support the decision, you will almost always be too late, so there is some judgment as to, “Do I wait slightly longer, or is now the right time?” For me, enough evidence has accumulated over the course of the year.
Q10 Chair: Mr Saunders, do you have anything to add to that?
Michael Saunders: I agree with all of that. To be sure, over the last few years pay growth has repeatedly under‑shot the Bank of England’s forecasts and external forecasts and, reflecting that, we cut our estimate of the equilibrium unemployment rate at the start of this year from 5% previously to 4.5%. It is possible that it may even be slightly lower than that, but that slack has been used up, with the jobless rate now falling to 4.3%—the lowest for 42 years.
Now, to be sure, pay growth this year has remained subdued, but there are lags between tightening in the labour market and pay. It usually takes several quarters, perhaps a year, for pay to respond fully to the recent tightening in the labour market. There are also significant composition effects that are bearing down on the published figures for pay growth, and those effects in the past have been quite temporary.
There have also been quite large rises in non‑wage costs over the last couple of years, and the firms, to an extent, may be economising on pay to compensate for those. However, of course, in terms of overall inflation, those non‑wage costs matter as well. I would also highlight the slow‑down in inflows of foreign workers into the UK, which may make pay growth here more sensitive to the tightening in domestic conditions than it has been previously.
Q11 Chair: Thank you. Mr McCafferty, do you have anything to add?
Ian McCafferty: There are two things to add. Most of the parts of the data that influenced my decision have been touched on by both Jan and Michael. The two other points to make are, one, that we are already seeing some pick‑up in particularly private sector wages. Clearly, the public sector is still affected by the public sector wage freeze, or the 1% per annum cap, but certainly private sector pay went through a very weak‑ish period over the course of last winter, as did growth in private sector employment.
Now, one can attribute that—and certainly the information we have from the agents, as well as our own conversations with businesses, does so—to the surprise, if you like, of the referendum decision, which led to what I have described in the past for a similar period, a few weeks or a few months after such a surprise decision back in 2001, with 9/11, as a paralysis of decision‑making on the part of all economic agents. That then, as I say, led to a weakness in hiring between the third and fourth quarters of 2016, and a weakness in pay growth between Q4 and Q1 of that period.
That is already starting to disappear as people have become more used to the decision. Some of that surprise has now worn off, and if you look at the quarter‑on‑quarter annualised data on private sector pay, that is already running at well over 3.5%. In terms of the annual calculation, as that weak period last winter falls out of the equation, we will see a natural acceleration in the rate of annual pay growth, even if nothing more happens to the shorter term pressures from now.
Q12 Kit Malthouse: I just wanted to ask a question about the effect on pay of auto‑enrolment, and whether the movements in pay, or the slow growth rates in pay, might have been affected. We are now into our third year of auto‑enrolment being phased in, which obviously adds a significant amount to payroll costs, albeit not to net pay. Are the figures inclusive or exclusive of auto‑enrolment?
Ian McCafferty: AWE is exclusive of auto‑enrolment specifically. That comes under non‑wage costs. The only evidence we have relates to 2016, when there was a quite sizeable increase in the numbers affected by auto‑enrolment. That had, at least from the agents’ discussions—it is very difficult to see in the data—a small impact in terms of some firms saying they were offering less in terms of pay rises as a result of having to incur the costs of auto‑enrolment.
Now, of course, the impact of auto‑enrolment varies from year to year. My understanding is that the next sizable increase is 2018 rather than this year. We do not know yet, and the agents have very little information as to whether there will be an impact on pay, or whether the fact that unemployment is now so much lower, and therefore labour market tightness is so much higher than it was back in 2016, means that firms will have to both offer pay and bear the costs of auto‑enrolment. We will not know for some time.
Sir Jon Cunliffe: You can see that unit labour costs have gone up. Higher pension costs have been part of that—the ONS revision—and as higher pension costs come in again in the future and they have to be paid for, that will also show in unit labour costs. The question is precisely whether employers are simply saying, “Well, lower pay increase this year because you are getting something on your pension,” or whether the pressure that you get from the labour market at these levels is just less than it used to be.
The only other point I would make is that I do not think there is an issue about whether we are using up capacity in the labour market. We clearly are, because recorded unemployment is going down, and measures of under‑employment, as I said, have gone down quite fast. There is maybe a bit left there, and the natural rate may be lower than the 4.5% that we have estimated—it probably is.
The question for me is not so much about that. It is not about where you hit the bottom, as it were. It is more about whether the sensitivity of pay to inflation has changed—whether that relationship has changed. Even for a tight labour market, workers just feel less confident in their pricing power. There is some evidence that may have happened in advanced economies, but that is the point, rather than a point about labour supply—for me, anyway.
Q13 Wes Streeting: Good morning. I have a series of questions about how the public make sense of decision‑making within the MPC, and in particular how they plan for the future. To begin with, if a consumer today were making a decision about whether to take out a variable versus a fixed‑rate mortgage, what duty do you think the MPC has to provide that consumer with information, to help them understand the potential future path of interest rates?
Sir Jon Cunliffe: Maybe I can start from the other end. The FCA, and to a certain extent the FPC, has a duty to make sure that the consumer realises that, if they take out a variable rate mortgage, interest rates can go up and to ensure that the mortgage can be afforded even if interest rates go up. That is why the FPC asks for new mortgage applicants to be tested, if you like, at a stress rate of 300 basis points or a 3% increase.
For the MPC, our job is bringing inflation to target and keeping it there. We forecast as best we can, but forecasting is as much an art as a science. We do not know what the balance is going to be in the economy, so we would not really be in a position to tell a consumer, “You are better off taking a variable rather than a fixed,” because although we read the economy and forecast where it is now, and we publish our forecasts for inflation against the market path of interest rates, things change, and we cannot give a guarantee as to where interest rates are going to be in six or nine months. The average fixed is two years, so somebody will be looking over that period, and I do not think we could give that sort of assurance.
In a way, it is not really our function. Our function is to be able to move interest rates—to bring inflation to target and to hold it there. Borrowers need to make sure that they can deal with a range of possibilities around interest rates, which is what the stress tests and affordability tests are supposed to do. However, the MPC cannot give guidance as to where variable rates are going to be over the period.
Q14 Wes Streeting: Is there any divergence from that view or anything other members of the panel want to add?
Michael Saunders: No, I agree with that. In a way, it is tempting to think that the central bank could promise where interest rates are going to be in the future and to think that might reduce uncertainty. Actually, it would make uncertainty in the economy greater, not less. Monetary policy, in the end, responds to what the economy does. It is not possible anywhere to forecast the economy with perfect certainty. As a result of that, you cannot forecast exactly what interest rates are going to do. If you were to set a path for interest rates and stick to it regardless of what the economy does, the result would be greater economic instability—and, I suspect, greater uncertainty—among borrowers and companies.
Different central banks give different forms of guidance about interest rate prospects. The Fed do their dot plots. Some central banks publish their forecasts at that point of where their policy might go, but what you see is that, in practice, the actual path of interest rates—the path that evolves in the future—is often very different from what they had said previously, because in the end policy responds to what the economy does. The same would apply to us. If we were to seek to provide certainty, I suspect that, in practice, policy might well differ from what we had said previously—for good reason.
Q15 Wes Streeting: I take your point, not least because I lost count of the number of times in advance of your meeting setting the rate that even the most expert analysts and commentators predicted a rate rise, and we eventually got there. You led me on to my next question, which was specifically about the Fed’s dot plots. I think, Mr Saunders, you have given your view that that would be an undesirable development; correct me if I am wrong. It seems to me that you think that it would be an undesirable development for the MPC to go down that path, but does any of the rest of the panel—or Mr Saunders, if you have anything to add—think that taking the dot-plot approach would be a desirable way forward for the MPC?
Ian McCafferty: I would support Michael. It would be an undesirable way for the MPC to go forward. It provides a certain degree of simplicity. It is a single point on a graph for an individual member of the FOMC, but at the same time it lacks any explanation as to what individual members of the FOMC think about the economy, or why they are forecasting that particular point on that particular graph at that particular time. It gives you no indication of the reaction function of the individual member of the FOMC.
The way we try to provide some form of forward guidance, if you like, is by discussing our individual reaction functions: what does the data need to do for us to react? How do we see the data going forward, and therefore how do we believe we would react as the economy evolves? As Michael says, policy will follow the way the economy evolves, so if we can explain where the economy is likely to go in our central forecast, what the likely risks around that are and how we are likely to respond, that gives more information to the public and the market than simply a point on a graph that tells you what the individual member of the FOMC might think at any point but not why.
Q16 Wes Streeting: Are there any other views to add?
Sir Jon Cunliffe: The Swedes publish a forecast interest rate path. What is important is that you have a way of communicating to the markets and the public how you see the economy evolving. We do that using the market path and show where we expect growth and inflation to be on the market path. There are different ways. It is not really about how you communicate it. The market may well take a different view, and in the US, it does. The committee has published its dot plot, and the market has got a much lower path. Just because you publish your view does not mean a) that you are right—it is your best forecast and b) that people outside will agree with you.
The other thing that we do, which in the US and elsewhere is less clear, is that we have a committee structure where we publish the votes and explain our votes. In that way, you can see the balance of arguments and what goes underneath that forecast. I do not necessarily think there is a right or wrong way. Personally, I prefer the way we do it to the US method, but that just may be a parochial preference.
Q17 Wes Streeting: We have the advantage of receiving Bank of England bulletins, so any time someone significant within the Bank makes a speech, we have the advantage of being immediately notified and have the opportunity to read the speech. We obviously read your minutes, and they form part of our judgments as a Committee. As members of the Treasury Committee, or market participants making judgments, we are a very different set of people with different considerations from members of the public. Do you think it matters that market participants will find it easier to comprehend your decisions and to interpret your speeches than one of my constituents thinking about whether to take out a fixed or variable rate mortgage?
Sir Jon Cunliffe: It matters a huge amount that your constituents—households and firms, as well as market participants—understand what we think and why we have done what we have done. We are putting a lot of effort into overhauling our communication: to try to use social media more, and to try to start with what we call the layer 1 message—the message that gets across using icons what we have done and the reason for it. You can see that on our website for the November decision. There is then a layer underneath that that explains it in more detail, and then—for those who really want to read the Inflation Report and drill down—a third layer is there.
We are really reversing the presumption that you start with the technical detail and then simplify it. We start with the essential message we want to get across, and then put the detail there for people who want that. That is enormously important. We need the markets to understand, as well, because in the end it is the market that determines the interest rates that firms and households face, but it is hugely important that we get that message across, and that we make speeches—as far as we are able—in relatively simple language that people can understand.
Q18 Wes Streeting: Sometimes, even members of the Treasury Committee need help with interpretation. One of the recurring phrases in MPC minutes is as follows: “Monetary policy cannot prevent either the necessary real adjustment as the United Kingdom moves towards its new international trading arrangements, or the weaker real income growth that is likely to accompany that adjustment over the next few years.” Why has the MPC felt it important to repeat this exact form of words, and what do you think is gained by each subsequent repetition?
Dr Gertjan Vlieghe: We want to make the distinction between the things that are within the power of monetary policy and the things that are not. If you have demand that is temporarily elevated or temporarily suppressed for whatever reason, monetary policy can lean against that to make sure that is a short‑lived effect and that the inflation effect that would otherwise come from it is also short‑lived. That is roughly what we can do.
If you have a permanent structural change in the economy, which can happen for many different reasons—the changes related to Brexit is just one example—there is nothing that monetary policy can do about that. At best, we can ease the path to that new state, but we cannot prevent the change, and that is the point that we keep making. If there are things happening in the economy that will lead, for example, to a permanently lower level of income, there is nothing monetary policy can, or should want to, do about that, because if we try to, we just end up generating inflation.
Q19 Wes Streeting: Finally, just for clarity, in the latest set of minutes, alongside that now‑familiar phrase that I repeated to you, the MPC has added, “It can, however, support the economy during the adjustment process.” In terms of understanding that nuance, what was the intention of the MPC when you inserted that phrase? Was it to reinforce the point that you have just made—to set out the limitations of what the MPC and monetary policy are able to do?
Dr Gertjan Vlieghe: There are two points. One is to make the point about the limitations on what monetary policy can, and cannot, achieve. The other point is that with the bit that we have some control over, which is the transition to that, we have a little bit of influence over whether an adjustment to income happens, for example, all through wages or partly through employment as well. In this case, we wanted to support employment, so that less of the hit would come via employment. That is, indeed, going back to our decision last August, in which we explained very specifically that in this transition we want to support the economy, so that we can achieve this change with a higher level of employment and jobs than would otherwise have been the case. However, ultimately, a real-income adjustment is a real-income adjustment.
Sir Jon Cunliffe: I agree entirely with that, and I am old enough to remember a period when monetary policy tried to change long‑term growth permanently, and it ended very badly in terms of inflation. If there is a permanent structural change, we cannot offset that.
Our remit is to bring inflation back to target. We can trade off, as we discussed before, between inflation and unemployment to a point, but whether we will be able to support the economy as Brexit materialises will depend upon this other phrase, which is also, I admit, a bit convoluted: it depends on how the supply and demand sides of the economy adjust to whatever Brexit and transition emerge, and how the exchange rate adjusts. If demand suddenly falls and the supply side does not, those are the classic circumstances in which the central bank supports demand, because we are opening up an output gap, we are opening up unemployment, and we want to support the economy that way.
If, however, the demand side of the economy just keeps on demanding, because people do not think it is a bad thing, but the supply side of the economy contracts, then we have inflationary pressure and we have to act in a different way. If the exchange rate goes up, that will, all things equal, reduce inflationary pressure with a higher exchange rate, but if that is the result of a very good outcome—markets are surprised and sterling bounces up—it may be that consumers and households also bounce up. We will have to measure those things, so it is about how demand adjusts, how supply adjusts and how the exchange rate adjusts that tell us whether the Brexit that emerges and the path that emerges will be disinflationary or inflationary. We will have to adjust to that. Supporting the economy through the transition is dependent on those things as well.
Michael Saunders: The remit gives us exceptional circumstances. In the exceptional circumstances since the Brexit vote, we are seeking to achieve a reasonable trade‑off between above‑target inflation, which is triggered by the drop in the pound, in turn triggered by the Brexit vote, and the amount of spare capacity in the economy.
The amount of spare capacity in the economy is small and shrinking. Against that background, we have been providing considerable stimulus. The rise in rates that we have done means that policy is less loose, but it is still fairly loose. We are not setting policy in a way that seeks to push unemployment higher, or to create rising spare capacity. We have no need to do that: the domestic economy has not yet overheated. We have set policy in a way that we think will probably still cause spare capacity in the economy to shrink, with the jobless rate stable or perhaps even slightly lower over the next couple of quarters. In that sense, policy is continuing to provide support for the economy. Policy is still looser than neutral and has no need to be tighter than neutral, because the domestic economy has not overheated.
Chair: That is a good point to move, I think, from Sir Jon’s comments on to QE.
Q20 Alison McGovern: Could I ask one follow‑up question very briefly on that, Chair, to test your patience, especially to Dr Vlieghe? You have collectively given an excellent description of the role of monetary policy. Do you think that the Bank’s actions post the Brexit vote itself were a good example of when monetary policy can have a smoothing effect?
Dr Gertjan Vlieghe: Yes, absolutely. The circumstances that we were faced with around the time of the vote were that we saw a very sharp deterioration in confidence. We saw a deterioration in a number of indicators that usually relate fairly well to aggregate activity, and we wanted to support activity, because that was a very sudden adjustment. If we had not provided support, we think that would have led to a rise in the unemployment rate, which at that time—given, also, what the inflation outlook was and given our remit and the trade‑off—we did not think was desirable. That was a good example of how to smooth that adjustment.
Q21 Alison McGovern: On QE, could I start by asking how often the MPC discusses QE?
Ian McCafferty: In terms of our vote, we vote every month on whether we will change the level of assets that we have purchased. It is one part of the voting structure that we do, and you will see that in the minutes: in recent times, we have voted to maintain unchanged the level of purchased assets at £435 billion. In one sense, we discuss it to that level every month. More broadly, we discuss it as and when either we feel it necessary or when there is something to be said. It is not necessarily pre‑ordained.
Q22 Alison McGovern: Is there often something to be said?
Ian McCafferty: To my recollection, the last time we discussed it within the committee in any detail was when I raised an issue that suggested that some of the potential impacts of unwinding QE, whenever it happens, and it is unlikely to happen in anything like the near future, may well be different from those that prevailed on the way in, as it were—that the multipliers, the transmission mechanisms, may well be somewhat different. Of course, at that time we had dysfunctional markets and a good deal of disruption in terms of business and consumer confidence that we do not face, and as a result, the multipliers—the transmission mechanism—may well be different.
I made the suggestion in the committee that the staff be requested to start looking at that issue and undertaking some analysis, such that we would have a better understanding than we do currently about what might happen were we ever to decide to unwind. When I say “ever”, it is simply that we have not made any decisions within the MPC as to the timing, other than to say it is unlikely to be in the near future. I mentioned that point—and I am quite pleased to be able to put the record straight again—in an interview with the Times, and that was misinterpreted as saying that I am calling for an immediate withdrawal of QE.
That is incorrect, on the basis that I am a strong believer—as, I believe, are other members of the committee—in what we call the marginal instrument argument, which is that bank rate is our most important instrument. We need to get bank rate to a position in which it could be moved materially in either direction to make sure that it is our most important instrument in the event of a need to cut, and there is no need to do anything to QE until that point has been reached.
Alison McGovern: Thank you for laying that out there. You will have a lot of sympathy from this Committee for being misunderstood in your role.
Chair: It is an occupational hazard.
Q23 Alison McGovern: Indeed. Could I just go to the substance, Mr McCafferty, of the point that you were making? Essentially, you were saying it would be helpful if the Bank started the work, because more knowledge is better than less. Do you think that they have done that work, and do you think there is more to be done? I am not asking for sharp‑edged criticism of the staff, but where do you think we are in terms of that?
Ian McCafferty: The Bank is starting to undertake some of that research. Clearly, it is going to take some time in order to try to do that. We do not have a very long period on which to look at data, and therefore there is a good deal of work that has to be done in preparation for coming up with any answers on that.
As I say, my instinct would be that the multipliers will be somewhat lower, and that is based partly on the fact that, looking at what has happened in the United States, the reaction in markets to the announcement that the Fed was withdrawing some of its QE stimulus was far less than had been feared in advance. The movement in the treasury curve has been something like 10 basis points, whereas on the way in similar announcements were having moves of about 60 or 70 basis points on the treasury curve. It may well be that the multipliers, the transmission mechanism, will be lower on the way out, but there is quite a lot of work that needs to be done. My understanding is that it is under way, but we have yet to discuss it at the MPC.
Sir Jon Cunliffe: Could I just add two points? First, I emphasise that we are seeing an experiment play out in the US now, which is a bigger market—a bigger QE—where the withdrawal, which Janet Yellen has said will be like watching paint dry because it will happen over an extended period, has not led to the moves in markets that one would have expected. There is a lot of interest, among both market participants and central banks, as to why that has not happened.
The other point is that it is not quite right to say that we did QE and just forgot about it. The committee has looked, on a number of occasions, at whether the impact of maintaining that stock of assets is waning. There is an argument that the impact of holding those assets erodes over time, and so this kind of stimulatory response that you get from the stock of QE is less. As a result, on the exit, you will see less pressure.
When we have tried to look at where interest rates are relative to some concept of the natural rate, we have also tried to look at the impact of that stock of QE that was held constant until August 2016, and then where we are now after that increase—where that has gone. It will need a lot of work and, as I say, the last point at which we said we would look at it would be 2%. We are way off that, I think.
Q24 Alison McGovern: Is there anything else that anyone wants to mention on the American comparison? If not, I just have one final question on QE, which is about an interesting comment from former Treasury Permanent Secretary Nick MacPherson. He described QE as “like heroin”, meaning that you started off with a small bit, and ended up needing ever bigger hits that had ever bigger negative side effects. Do you have any response to Nick’s comments?
Sir Jon Cunliffe: I do not want to go into lurid analogies.
Chair: Please be as lurid as you like.
Sir Jon Cunliffe: You have to wait until you have left until you can do that, I think.
Chair: Or until you join Twitter.
Sir Jon Cunliffe: Nick was never that lurid when I knew him in the Treasury.
The Bank did QE during the crisis. We then stayed at £375 billion of gilts for quite a long period. I would argue the referendum, as we talked about, was a pretty exceptional event, and then we had to come in and support the economy again.
It is not an addiction. In Nick’s analogy, you get more and more hooked and you need more of it. I have just explained that the impact of it—the stock of it—may well be waning, so I do not think we are addicted to it. We will see in the US, but, when the time comes, this may be something that you just unwind very slowly, because you do not want to cause disruption, and the marginal instrument you want to use is interest rates.
Michael Saunders: I think he is wrong. The QE provided useful stimulus at a time when the economy needed it. The alternative would have been a weaker economy via unemployment and, for us, the prospect of an inflation under‑shoot. That would have been an undesirable outcome.
Now, if there are adverse side effects of QE, they come chiefly from the very low level of bond yields. You can see the effects of that on pension deficits, for example. However, consider the world in which QE had not been done. The economy would be weaker. The chances are that bond yields would have been as low anyway. We would not have solved the problem of pension deficits, and you would have had a weaker economy as well. That would have been an undesirable outcome.
Ian McCafferty: There is a lot of mythology around QE. QE is a way of lowering the effective interest rate once the normal interest rate instrument has got close to its lower bound. The fact that we have had to maintain the level of QE after that stimulus of 2009, and then add a little at the time of the Brexit referendum, is a reflection of the path of the real economy over the course of the last 10 years. The real economy has remained relatively weak, which is a reflection of the need for low effective real interest rates, and that is why QE has been maintained over that period.
Q25 Mr Jack: Sir Jon, you are clearly an independent thinker—either that or you did not get the “wear a white shirt” memo, which I also resist whenever I can. Before the MPC’s decision to raise interest rates, did the Bank analyse how far a rate hike might be used by banks and building societies to increase their net interest margin, and did they analyse how far that might be passed on to savers?
Sir Jon Cunliffe: We do our calculations on the basis that the rate is passed through to savers and to borrowers. We look at the way in which borrowers respond to a higher interest rate, and savers likewise. Of course, the pass‑through does not happen instantaneously. I think 60% of mortgages are fixed‑term; only about 20% are on the variable rate, so the pass‑through in the stock of mortgages, you would assume, would take some time, and you would expect the flow of new mortgages to adjust. With deposit rates, as well, there are different products out there, so we do not assume an instantaneous pass‑through. We do assume, over time, the pass‑through is there.
This question of banks’ net interest margins and profitability was an issue when we cut rates in August 2016—not because we have a particular brief to maintain bank profitability. We do not, but we wanted to make sure that the interest rate stimulus was passed through, and if banks were in a position where they could not cut the deposit rates, they might not cut. Experience suggested that they might not cut the borrowing rates as well.
Our evidence is that the TFS that we put in place there has not really changed banks’ NIMs at all. It has done what we hoped it would do: it allowed the rate increase[1] to be passed through without boosting bank profitability or decreasing it, and that was really its purpose. When we are talking about an interest rate increase, you do not need to do something. You are not dealing with that restriction in the transmission mechanism of monetary policy in the same way.
Q26 Mr Jack: Given your role at the Bank for financial stability, do you consider the effect of a rate rise or fall on bank profits when you are taking MPC decisions? You have just answered the second question regarding bank profitability.
Sir Jon Cunliffe: We look quite a lot at bank profitability, and the FPC publishes quite a lot on it. There, there is an interest, again, because we want banks to be able to build up capital. Clearly, if they are not making profits—if their business models are damaged and they cannot accrue capital—they are less resilient, so we look at bank profitability, and we looked a lot in the FPC at this question of banks’ net interest rate margins.
We discovered that, unlike what you see on the continent, where they have become quite compressed, over quite a long‑run period—more than 10 years—although it would be different for individual banks, banks’ net interest margins have been around two and a quarter. They have fluctuated a bit, so we did not think we had a huge problem with banks having much lower interest rate margins, which, as I say, is a worry in other parts of Europe. However, for the specific August rate cut, because you were getting down to effective zero, we thought there could be a problem there, and the evidence so far seems to be that the TFS broadly did what it was supposed to do.
Ian McCafferty: As far as the MPC is concerned, the key consideration is, of course, the extent to which the transmission mechanism can be expected to work with any change in bank rate. That is why, as Jon has very well explained, we took the specific action last August, while bank rate was as close as it was to the zero lower bound. Other than that, we have no brief on the MPC per se to consider bank profitability. That is left to the FPC, and should be left to the competitive market.
Q27 Mr Jack: Can I move on, then, from the impact on people’s debt to the impact on people’s savings? The Prime Minister’s spokesman was quoted as saying, “Following the rate rise, we would expect to see higher interest rates … passed on to savers,” and also that “some banks have already said they will increase rates on their savings products. We would expect others to follow suit.” Do you agree with that statement?
Ian McCafferty: Yes. I have been quoted as saying, as Jon has pointed out, that these things do not happen overnight. The mechanism of the competitive market within the banking sector takes a bit of time to work, in terms of savers reacting to changes in banks, so we would expect to see that competitive process force changes on all banks in the fullness of time, but it does take a little time for any change in bank rate to be fully reflected, both in terms of savings and in terms of borrowing rates.
Q28 Mr Jack: As you know, Royal Bank of Scotland, NatWest and Lloyds have all struggled to push that margin up, so to what extent do you think the role of the MPC is to push for banks to raise rates, rather than to analyse when that might happen?
Ian McCafferty: It is the role of the competitive market to determine what individual banks charge, and you have quoted a number of banks that have not yet fully reacted. There are a number of banks, usually bracketed in the challenger bank category, where we are seeing a bit more reaction, and I hope and fully expect that over time the competitive nature of the market—as far as savers are concerned—would lead banks that may be a little slower to react to have to react to that competitive pressure.
Q29 Mr Jack: Does anyone want to add to that?
Sir Jon Cunliffe: I would agree with that. The word “expect” has two different meanings. I think over time it will be passed through. There is more to come and to work through, and to some extent banks have not been able to charge the higher interest rate on their lending yet because of roll‑off and the like. It probably will get passed through.
The question is, then, what should happen, which is the other use of “expect”. We want a competitive banking system where that does happen and, if savers feel it has not happened, they go on the best buy tables. It has now been made much easier to move your account; it is not the Bank of England’s job to do that, but the Government have put a lot of effort into making it easier for bank accounts to move. People look at the best buy tables and say, “Well, if I can get a better rate elsewhere, I should do that.” That is what I would expect, in that sense of the word.
Q30 Mr Jack: Sir Jon, as you know, the prudential and conduct regulation of the banks has been delegated, so is it helpful to the integrity of these arrangements that No. 10 might seek to micro‑manage the retail banking sector?
Sir Jon Cunliffe: I did not think it was a micro‑management of the banking sector. I thought it was maybe just an articulation of what most people would think is reasonable.
Chair: Your previous career is coming out here, Sir Jon.
Sir Jon Cunliffe: On the political side, people want to state what the expectation is, in the sense of what should happen, and it is perfectly possible to say that from No. 10 without—as a member of the PRA board, now the PRC—my feeling that my independence was compromised.
Q31 Chair: Just to follow up from Wes Streeting’s comments about the way members of the public understand and know about your very important work in terms of supporting the economy that every household and individual relies on in this country, you have talked about issues around employment and wages, which are critical, and we are conducting an inquiry on household finances and what has happened to the household balance sheet, rather than the nation’s balance sheet.
However, the point that I think Alister is seeking to bring out is that many savers are older people. They rely on their interest rates on their savings. They have had almost no interest for a long time now. Is it not reasonable that, when the base rate does change, banks should react to that and treat their savers to that increase? I appreciate the point about a competitive banking market, but if that is not working, would you not like to make it clearer—as people who are interested in the economy and how individuals are reliant on that—that you think that banks should pass those interest rate rises on?
Mr Jack: That is absolutely right. Think of it like your golf handicap. It goes down faster than it goes up.
Sir Jon Cunliffe: I would say that a) we are not at the end of this story yet, and there may well be reasons why. There have been lags in the past; these things have taken time to go through. As a member of the MPC, I care about understanding the transmission mechanism and making sure it works. As a citizen, I might well agree that that is what I would expect to happen, but, from a MPC point of view, we have enough responsibility already without, if you like, pushing into those areas.
Q32 Chair: I just want to follow up on the communications point, because it is really important. One of the questions you were asked to address in your annual reports—which you all did, so thank you very much—was the level of communication that you had engaged with. One of you—it might have been you, Dr Vlieghe—had done 22 visits to universities and schools. I really applaud that; that is very important.
However, your evidence in answer to Wes Streeting’s comments earlier on has so far been described on Twitter this morning as “complacent”, in terms of the way in which the MPC’s decisions are communicated to members of the public. It is going to have to go further than social media and infographics, and I would have thought that you could be making it clear perhaps in the ways that you are able to—whether it is through a raised eyebrow of the Governor, or something else—that these interest rate increases should be passed on to savers, who have borne the brunt of decisions. Mr Saunders, what do you think about that?
Michael Saunders: The Bank of England has enormous powers over different areas: monetary policy, banking regulation, and so forth. I am not seeking to expand those powers. At the moment, we have no tool with which we could force banks to change their interest rates, and I am not asking for any such tools. Our remit is very clear, and so I stick to that. If you were to ask me with a non‑MPC hat on, I might agree with you that I would expect it would get passed on over time through competitive pressures, but I am not seeking the tools with which to force that.
Q33 Chair: Do we need to beat the banks up?
Ian McCafferty: If there is evidence that the competitive market is not functioning properly, such that these changes in bank rate do not get passed through in both directions over the course of time, it is a recourse to the competition authorities.
Sir Jon Cunliffe: I should say that this is our first experiment in trying to push this out on social media and the like.
Chair: No, I welcome it. I think it is a great idea.
Sir Jon Cunliffe: I would be delighted if it were a huge success first time, but it is something that we are going to develop and carry on doing.
Chair: We will all be retweeting, I can assure you.
Michael Saunders: If I could come back to this question of guidance over interest rates, I was arguing against the false precision of seeking to give you a path of interest rates that you know for certainty will be followed. I do not think that is possible. I do think it is useful for the Bank to communicate widely to households and businesses a general sense that interest rates are likely to go up over the next two or three years if the economy evolves as we expect. We cannot give you a precise path, but that general guidance is very useful for them.
Mr Jack: The last question was on the sensitivity of interest rates to households and firms, but that is picking up on it, and Sir Jon answered it in an earlier answer to Wes, so thank you very much.
Q34 Stewart Hosie: Mr Saunders, you are very clear in what you are saying about not being prescriptive about where interest rates will go, but that rather cuts across the Governor’s forward guidance policy, which pretty much seeks to tell the public, businesses and banks where interest rates will be for quite some time in the future, does it not?
Michael Saunders: I do not think it does. The forward guidance was launched in mid‑2013, and then the aim was really to say that interest rates would not be going up for a while, because there was plenty of slack. Obviously, as time has moved on, the guidance that we are giving is a general sense that, if the economy evolves as we expect, the market path of interest rates has a bit further tightening over the next two or three years and, while not seeking to endorse that exactly, that is not unreasonable.
Q35 Stewart Hosie: You would be unhappy, then, if the Governor came in front of us and gave his forward guidance that was a little bit firmer than the market‑driven approach that you are suggesting might be the case.
Michael Saunders: I am not quite sure as to what guidance it is that he would give in that case.
Q36 Stewart Hosie: I am just curious to see if there is a debate about forward guidance. You seem absolutely determined to have the market decide, full stop, but we have had—certainly over the difficult period—a slightly firmer view.
Michael Saunders: Actually, no. If you go through this year, the MPC said from early this year that we thought, if the economy evolved as we expected, interest rates over time might well need to rise faster, to a greater extent, than markets at that stage priced in. I forget exactly when we started saying that; I think it was around May.
Dr Gertjan Vlieghe: It was May.
Michael Saunders: We said that steadily, and then in September we intensified it in a way that signalled the possibility that the rates might rise rather earlier. That guidance was, I think, very useful. That does not mean that we always have to be doing it. What we have said now is that we expect that interest rates, if the economy evolves as we expect, might well rise further over time.
Stewart Hosie: Indeed, and that signalling was very clear. I am just trying to find out where the debate is.
Sir Jon Cunliffe: Forward guidance gets misunderstood a bit. The committee talks and thinks a lot about its communication, so that statement in May and August that the market might be under‑pricing interest rates, or would be if the economy evolved as we expected, was something that the whole committee signed up to. The statement in September, where I was not part of that majority, was a very clear signal of where the majority was—a stronger signal, because it was necessary to prepare people for the possibility of an increase.
If you go back to the 2013 forward guidance that you mentioned, unemployment was 7.8% or 7.9% and falling quickly, but 7.8% is pretty high. Ex‑members of the MPC and commentators in the markets were saying, “Interest rates need to be tightened; the Bank of England is well behind the curve and letting inflation get out of control.” Of course, we just control the overnight interest rate, as it were. The market sets the rate that businesses and households face, and there was a real danger that the market would see that drop in unemployment, apply a pre‑crisis psychology to it, and push up market interest rates at a time when we had a lot of spare capacity in the economy.
That so‑called “state contingent guidance”—we will not consider raising rates until unemployment has come down to a certain level—was designed to avoid the market raising rates at a time when the economy was weak. Those were specific circumstances. Would you use that again in similar circumstances? My own view is that unemployment dropped much faster than we thought, but the guidance was successful in ensuring that we did not see a rise in market interest rates at that time, and that sort of communication was part of the flexible armoury of a central banker, but you do not use the same thing in every situation.
Q37 Stewart Hosie: Indeed, and the advantage of that policy is that the interest rate was not automatically changed on an unemployment trigger. It simply triggered a reconsideration, which allowed it to be changed or not, and the rest is history.
Sir Jon Cunliffe: Yes.
Q38 Stewart Hosie: Mr McCafferty—I am progressing badly here—your annual report in November describes the questions that you focus on in determining your interest rate view. In particular, you ask, “What is the balance of inflationary pressure between exchange rate and domestically driven factors, and hence the profile for inflation over the policy horizon?” Can you briefly tell the Committee how you answered that question most recently?
Ian McCafferty: Most recently, the balance of that question has shifted quite significantly, relative to where I would have made my judgment earlier in the year. That explains why, over the course of the year, my voting pattern changed, such that I have been voting for a rate rise since June 2017.
The balance has changed in two respects. One is that I think the degree of spare capacity, one part of our trade‑off, has been eroded even slightly faster than I had expected, and I also have concerns that, if anything, the degree of effective spare capacity may be even slightly smaller than we have positioned in our central forecasts, such that over the course of the forecast the output gap will disappear maybe a little bit earlier than we have in our current forecasts. That would then mean that, in terms of the way in which we consider this trade‑off, the trade‑off itself disappears, because if there is no spare capacity, you have no trade‑off to make and the tolerance of above‑target inflation disappears.
My second element was the degree of risk around the inflation forecast, where in the early part of last year, given the surprise of the Brexit vote and the uncertainty around quite how consumers, as well as businesses, would react, there was a certain downside risk to the economy. Although that risk was manifest in the surveys that we saw initially after the referendum result, it tended to improve somewhat as we got through the winter. It is clear that the economy has held up a little bit better over the course of 2017 than we would have expected immediately after the vote.
That leads me to a belief that there may well be a bit more upside risk to inflation than I thought was possible at the beginning of the year, not least because, although I am very happy with the way in which we estimate the degree of inflationary pressure that comes from changes in import prices—that is, the degree to which we judge pass‑through—there are a number of possible channels that are far more difficult to measure, each of which might add a little to inflation pressure.
There is one I have talked about in the past that you might call the Microsoft or Marmite effect, whereby products that are made in the UK using UK ingredients still see some inflationary pressure as a result of a change in the exchange rate. This is simply because the company that makes them has to report its profits in a foreign currency to a foreign parent, and as a result it comes under pressure to put up the prices as a result of the change. That is very difficult to measure, in terms of building into a pass‑through model, but it does suggest that there may be some upside pressure. Overall, the balance of that trade‑off shifted from one in which the risks were balanced to one in which the output gap was closing rather faster; there were some upside risks to inflation, and that led to my change in my vote.
Q39 Stewart Hosie: These issues—spare capacity, the Brexit risks, the Marmite effect and so on—are very practical. One of the other questions that you posed to yourself was the effective degree of stimulus from the bank rate, given the likely cyclical change in the equilibrium rate of interest against which the level of stimulus is judged. This is a slightly more philosophical question—again, please be brief, because I am certainly conscious of my time here—but how did you answer that question in the most recent round?
Ian McCafferty: The first thing to say, without going into huge amounts of detail—as I am conscious of the time too—is that, of course, it is impossible to observe independently what the natural rate of interest is at any one time. However, it is a fair judgment that relative to the very depressed levels that we saw a few years ago immediately after the crisis and during the recession, it has probably risen, even though it is still probably significantly lower than it was prior to the crisis. That then means that, were one to keep nominal bank rate at any particular position relative to a gradually rising r‑star, you would therefore be adding to stimulus, not necessarily intentionally. That is another factor—although it is very difficult to be precise in arithmetic terms—in terms of my decision‑making over the last year.
Q40 Stewart Hosie: When you talk about adding to stimulus, are you effectively talking about an increase in the money supply, or the continuation of a relatively low cost of money?
Ian McCafferty: I am talking about the cost of money in terms of the interest rate, and if one does not change the nominal rate in line with any understanding of a change in the underlying r‑star, you will be changing the effective level of stimulus, not necessarily intentionally.
Q41 Stewart Hosie: Let me move on to the subtext of that question. We are talking about the cost of money, and that is perfectly sensible in terms of the interest rate, but what about the impact of money supply itself? What consideration, if any, do you give to that, in terms of the future path of inflation?
Ian McCafferty: We watch the money supply data closely, because ultimately, of course, inflation is, and can be, a monetary phenomenon. Over the course of the last year or so, most of the measures of money supply, both broad and narrower measures, have been growing roughly in line with nominal GDP, so from that point of view I do not think they are providing any undue stimulus or undue drag on the economy. However, it is something we watch as part of our monitoring of the economy.
Q42 Stewart Hosie: Let me ask others, starting with Sir Jon.
Sir Jon Cunliffe: I agree with that. I would just say a bit from a historical perspective. A Canadian central bank governor said in the early 1990s or late 1980s, “We did not abandon monetarism; it abandoned us.” I remember all the attempts, very much going back to what Jan said at the beginning. You want monetary policy to deal with nominal things—the rate of inflation, money supply or, in some countries, the exchange rate—rather than real things. However, which nominal things you choose depends on your being able to establish some reliable relationships, and right throughout the 1980s and 1990s or whatever, and now, the relationship between the money supply and inflation and growth in the short run does not give you a very good read. That is why the Treasury switched to inflation targeting in 1992. It is also targeting a nominal thing, but we had more success with being able to, if you like, adjust policy to control that thing, whereas using the price of money, the interest rate, rather than the supply of money was trying to control the monetary aggregates—sterling M3 and M4 plus now—just did not give us a reliable relationship for a number of reasons, and that is still pretty much the case, I think.
Q43 Stewart Hosie: Dr Vlieghe and Mr Sanders, do you share those views?
Dr Gertjan Vlieghe: Yes. The point I would make about money supply is that it is useful, but it is not useful in the simple, short‑term inflation dynamics: “Do interest rates need to go up a bit or down a bit?” The link between the money supply and inflation is far too loose for that.
Over very long periods, money supply will be most useful in telling you that monetary policy may be making a mistake. If, for many years, money supply is growing faster than you think would be consistent with your inflation target, it may be telling you something about either incipient inflationary pressure or some excess building up somewhere in the financial system that you should be looking at and focusing on. However, that is a very different point from saying, “The money supply over the last quarter accelerated by 0.5%. Maybe I need to worry a little more about inflation,” because at that level it does not work at all.
Chair: Kit, did you have anything you wanted to add on this section?
Q44 Kit Malthouse: I did. I am interested in that, because earlier in your evidence, I think, Sir Jon, you said that QE was effectively designed to lower the cost of money: to flood more liquidity and supply into the market, to lower the real rate of interest that was being charged. However, that must, therefore, have had an inflationary effect. If not, it has had a mis‑allocation effect.
Sir Jon Cunliffe: As I say, there is a lot of work on QE, and I think academics will be studying the QE of the last 10 years for a long time, trying to work out more precisely how it works, but it works through a number of channels. The channel that was most powerful in the UK was not an increase in the money supply. My recollection is that the money supply went down over the period that we were doing QE—the broad money.
The most powerful channel was the so‑called portfolio effect. The Bank bought the safe assets that insurance companies and pension funds held. They then had to invest. They invested in corporate assets, and that portfolio effect brought the price of lending for companies down. That is one of the reasons why I said earlier on that this effect might be eroding. If you buy a large proportion of the safe assets—the gilts—then pension funds and insurance companies have to go into corporate bonds. That has an impact.
Since then, of course, the Government have issued a lot of gilts, so the leverage that you get from that stock of gilts changes over time, but there are a number of channels. Also, if we look at the first instance of QE, it was more powerful than the second QE. Just as a signalling effect, when the Bank did that it signalled to people that the Bank would keep interest rates low for a long time, and that is why the first QE announcement seems to have had a much bigger impact on market risk than the second. I do not think it is primarily a money supply thing.
Q45 Kit Malthouse: However, it has had a similar effect, in that it has driven down the cost of money.
Sir Jon Cunliffe: But it did not have the effect by the link between money and inflation activity that the policies targeting money supply did.
Q46 Kit Malthouse: Forgive me if I am simplifying this too much, but the fundamental theory was that if you have too much money, the value of it drops, which means that inflation rises, so you need more money to buy the same thing, and that effectively QE has had the same effect. We have certainly seen that in asset prices. You need more money to buy the same house than you did 10 years ago.
Sir Jon Cunliffe: Yes, although house prices have followed a logic of their own.
Kit Malthouse: Yes, but I am talking about asset inflation across the board.
Sir Jon Cunliffe: Certainly, you put the price of safe assets, gilts, up. However, to me, that is a rather different thing from trying to steer inflation by the money supply.
Q47 Kit Malthouse: You do not think there is a connection, therefore, between that decrease in the cost of money and the inflation in asset prices feeding through to general inflation.
Sir Jon Cunliffe: I think there was a link with lowering the cost of market interest rates, which is what QE did. As I say, some of it was not just the actual purchases; it was the signalling effect that reinforced the Bank’s commitment on interest rates.
Q48 Kit Malthouse: That might have been the case in the early days, but there is quite a lot of evidence now that it is pushing real interest rates in the market very low.
Sir Jon Cunliffe: There is a discussion about why long‑term real interest rates are very low, and the primary drivers of long‑term interest rates are not QE that is happening in the UK and elsewhere. There are other things that are driving long‑term interest rates lower.
Michael Saunders: I think QE lowered market interest rates as a whole, but the route through which that leads to higher inflation in the end—or less of an inflation under‑shoot—is that the drop in market interest rates boosts growth, investment, hiring and consumer spending, and helps use up the economy’s spare capacity. If that had not been done, the chances are that inflation would have been under‑shooting. Any inflation over‑shoot risk probably only comes about once the economy has recovered significantly, and of course we have the ability to reverse policy to prevent an inflation over‑shoot happening on a sustained basis.
Kit Malthouse: I can see that.
Michael Saunders: I do not think there is a link between money supply and inflation that does not go through growth and the erosion of spare capacity.
Q49 Kit Malthouse: I understand, but there is a basic cost equation: if you have more of something, it costs more to exchange it for something else. This was the basic theory back in the 1980s, or whenever it came out. I mean, this is what is behind the rampant inflation in places like Venezuela: pumping money in to try to solve the problem, and in the end you just create massive inflation.
Dr Gertjan Vlieghe: It is correct to say that QE had an inflationary impulse. It is important to remember that it was an inflationary impulse that was there to counter a severely deflationary shock to the global economy. Not everyone did QE in the same size at the same time with the same level of aggression, and there is very clear evidence now, if you look across countries that were hit with roughly similar shocks, that those who went in earliest with the most aggressive QE and were determined to keep it there until inflation stabilised had the best results of anchoring long‑term inflation at target. The others suffered a drift of inflation expectations lower, and therefore bond yields followed.
Q50 Kit Malthouse: But it was an inflationary impulse; I think that was the phrase you used.
Dr Gertjan Vlieghe: Yes, it is an inflationary impulse to counter a deflationary shock. That is the point.
Q51 Kit Malthouse: Okay, so you are effectively saying that QE has a role in inflation.
Dr Gertjan Vlieghe: Absolutely, but when people say, “Where is the inflation?” the inflation is the deflation that you did not see.
Q52 Kit Malthouse: When you consider your decisions about rate rises, how concerned are you that, by having ultra-‑low rates for a prolonged period, you have moved us into a trap because there has been a mis‑allocation of rates? Sir Jon, you talked about bank margins being broadly the same, but if banks are going to grow profits they have to increase volume on the same margin on a lower amount loaned, and the only way you can increase volume in banking is, effectively, by impairing quality if you want to do it over time and in a sustained way. How cognisant are you that you might have created a trap now in that there has been a huge mis‑allocation of capital, effectively, in banking—what, in any normal era, would be un‑commercial loans—and therefore movements in rates start to make that house of cards look deeply unstable?
Sir Jon Cunliffe: I would say a number of things. One, you can increase profits in banking by reducing costs, and net return on assets depends primarily on the costs you have; two, you can increase profits in banking by avoiding misconduct charges; and three, you can increase profits in banking by sorting out investment banking models that do not work.
When I look at the price-to-book ratio of British banks compared with some of their competitors in the States, I would say that their lower profits for the period have been much more to do with misconduct and sorting out investment banking models than to do with return on assets. However, those that have a low cost-to-income ratio are also the ones that are generating the most profits. I am not sure that I would take the argument that the only way to increase bank profits is to increase, if you like, the amount of lending they do.
Q53 Kit Malthouse: But that is the whole point of QE. You wanted them to increase lending.
Sir Jon Cunliffe: Sorry, I thought the question was whether we had increased banks’ profits and forced them into doing that.
On this question of whether there is a very large stock out there of poor‑quality loans, either to corporates or to households—and I say this from a FPC point of view and a PRC point of view—looking at our stress tests and the rest, that is not what they show.
Q54 Kit Malthouse: Well, the last stress test required quite significant uplifts in capital.
Sir Jon Cunliffe: The stress test subjects the banks to a severe but plausible stress. The current stress tests—and the results will be coming out at the end of the month—have a 35% drop in house prices, which we have not seen in the UK before; a 40% drop in commercial real estate; a 4.5% drop in GDP; and China going into a recession. These are very severe stresses, and we are saying to banks, “We want you to have enough capital to come through a stress event that is more stressful than the last crisis, and still have capital to lend to the banking sector.”
I certainly would not want to take the stress test results as saying that banks are sitting on a lot of difficult loans. We stress them to really very large falls in those asset prices, to make sure they have the capital. I am not sure I would take the argument that QE has led to a lot of lending that is sitting there and is vulnerable to a change in conditions.
Michael Saunders: In broad terms, household and corporate balance sheets are in much better shape now than they were 10 years ago, and banks’ balance sheets as well. Late in the last decade, the UK had a very high ratio of private sector debt to GDP; it had risen very sharply over the extended period since the early 1990s. That was a sign of fragility. Since then, the private sector has deleveraged substantially. The ratio of private debt to GDP is down from over 190% to just over 160% of GDP. It is now similar to the advanced economy average. I do not buy this idea that the economy is a house of cards that is waiting to crumble.
Sir Jon Cunliffe: On credit, before the crisis, credit in the economy was growing at over twice the rate of GDP for a number of years. It is now growing broadly in line with nominal GDP.
Q55 John Mann: Just on that, Mr Saunders, I presume that you go with the Governor, who told us at our last meeting with him that the personal debt burden was overstated, as opposed to Mr Brazier, who says that household debt is dangerous to borrowers, lenders and everyone else in the economy.
Michael Saunders: I have a bit of concern over the level of household debt. I talked just now about the overall level of private sector debt, combining households and companies. The UK’s ratio of household debt to income is lower than it was but still relatively high compared with other countries. Debt to income is 134%; I think it peaked at 147%. Consumer credit, unsecured lending, has grown at about 10% year to year. I have a little bit of concern over that. The number of personal insolvencies is up 25% over the last couple of years.
I do not think that we are yet at a point where the household sector is fragile like it was in 2006-2007, but if you like this is a movie that we have seen before in the UK. We know how a long period of rapid debt growth ends, and it is not good. We are not at that point now, but it is important—and this is Jon with his FPC hat on—that they act, as they are, to ensure that we do not end up in those conditions.
Q56 John Mann: Dr Vlieghe, you have done eight regional visits, according to your report. How many businesses, on average, did you meet on those visits, approximately?
Dr Gertjan Vlieghe: My visits are usually about a day and a half or two days long, and so I do about three or four events that have maybe 10 to 15 people each, so I see roughly 40 or 50 people on each visit.
Q57 John Mann: How many of those are decision makers when it comes to investment or pay increases?
Dr Gertjan Vlieghe: All of them. They are usually CFOs or CEOs.
Q58 John Mann: How many are from the public sector?
Dr Gertjan Vlieghe: A minority. Usually, they are private sector or what we call third sector businesses.
Q59 John Mann: How many are self-employed?
Dr Gertjan Vlieghe: We see some very small businesses. We see SMEs, including some people who run a company with fewer than 10 people, but a single, self-employed person with no employees? No, I cannot say that I have met those on my regional visits.
Q60 John Mann: Mr McCafferty, in terms of the growth in jobs, a dramatically increasing proportion over the last five years have been self-employed, and if you are self-employed, one could not pay National Insurance; one could, temporarily at least, have savings by not renting premises but working from home. How well is the growth of self-employment built into the labour market models that you are using?
Ian McCafferty: All those points that you have made, Mr Mann, are to do with government policy, not to do with Bank of England policy. The tax system as affects the self-employed, as opposed to any other sector of the economy, is nothing to do with us. We build changes in the economy into our models. Clearly, we have to have data, and therefore we have to see these observations. We cannot necessarily react immediately to any changes in the structure of the labour market; we have to have some proof that these things are under way, but we do build changes to self-employment as best we can into our models.
Q61 John Mann: You comment a lot on productivity, so it absolutely is to do with you, because they are potentially timelines in terms of productivity and it makes productivity harder to analyse, this so-called puzzle.
Ian McCafferty: We can analyse the puzzle, but we do not have the policy tools to make significant changes.
Q62 John Mann: Absolutely not, but I am questioning the analysis and how that leads to your decision-making.
Ian McCafferty: Sorry, I am not sure what your question is, therefore.
John Mann: It is about the labour market, and whether the tools for understanding the labour market are there, or whether, in fact, they are rather old tools that are being used, in terms of assessing how the labour market operates.
Dr Gertjan Vlieghe: We have a lot of analysis, and we have all given a number of speeches highlighting the fact that, if you want to assess labour market developments, it is not enough to just look at employment growth and the level of unemployment. We have talked at length about the composition as between self-employment, part-time employment, full-time employment, the gig economy, and the idea that people who might superficially—in some ONS statistics—look like they are fully employed are not fully employed, so this is something that we have all been analysing and commenting on for a number of years. I am not sure where you are seeing the great lack between what is happening in the economy and our understanding of it.
Michael Saunders: The Bank does try to reach out to all parts of the economy. We all do regional visits, and there is probably variation in which types of company we see. I was in the North East a few weeks ago, and did a session at a centre for IT startups, one-man businesses; I saw a whole string of those. I saw another group, which is basically a local circle of self-employed. They get together once a month; I went along to talk to them as to how things are going. The Bank does try to reach small firms and self-employed as well as large firms to hear what is going on.
Regarding your point about changes in the labour market, I absolutely agree with that, and the rise in self‑employment; you also have the gig economy and various changes in the tax and benefit system. All of this has helped to make the labour market significantly more flexible than it used to be, and that has been part of the story of subdued pay growth, even while the jobless rate has fallen significantly over the last few years.
I have to say that what I am hearing now is that firms are finding it much harder to get staff. Recruitment difficulties are rising. You can see that in a range of surveys, and even though the labour market is more flexible than it was, there are still limits as to the amount of labour supply. We may be reaching those limits now.
Sir Jon Cunliffe: On this modelling point and old tools, or whether the tools are up to date, we do a supply stock‑take, as we call it, every February. We look at the supply side of the economy; we look precisely at some of these issues, and the example alluded to before—that we dropped where we thought the natural rate of unemployment was—was actually not to do with those things but a lot to do with educational attainment, which meant that people spent less time in unemployment. If you went into unemployment, you found a job more quickly, so frictional unemployment was reduced. That is one example of trying to keep up with the way that the labour force is changing.
The self‑employment rise started in the 1980s; it has pretty much doubled between then and now, but other things, such as zero‑hours contracts, have moved more quickly in the last years. It is not that we have not observed the changes; it is just that it is impossible to know how that sort of labour market, with those changes, will function with a very low level of unemployment, because we have not seen unemployment at this level since the 1970s.
Yes, the changes are there; yes, we try to capture them. It tends to be a bit backward‑looking, because if the models are not capturing what is happening, you try to find ways to adjust them. We try to look for those changes, but we are now seeing very low unemployment, and seeing how the labour force reacts to that in terms of pay and other things. It is almost certainly true that we will learn something from going through that process.
Q63 John Mann: However, understanding that, the concept of overtime has virtually disappeared. People are simply required to work more hours, whereas it is not that long ago that, in engineering, overtime decisions required third‑party validation if the workforce was going to agree them, on every single occasion. That is a huge shift, culturally.
Now, I am not interested in your views about whether that is a good thing or a bad thing, but in terms of how you assess how the labour market works and therefore the responsiveness of employers, it is, it seems to me, rather critical. Therefore, which employers you are talking to and how you are getting that data in, including at the micro level, and hence this shift to self‑employment, also seems to be rather critical. Otherwise, this puzzle is going to remain a huge puzzle. Who should be providing additional research?
Sir Jon Cunliffe: At the micro level, the agents go and ask these questions, and the agents feed in to every monetary policy round. We have special surveys. Certainly, when I go around the country and I see self‑employed and others, the question I always ask is, “Can you make a pay increase stick? Can you make a price increase stick? What has happened to pricing power? If you are self‑employed, could you put your prices up?” Those are questions we ask.
Even if we all go around the country, the sample is pretty small. Using the agents, you try to broaden that out, and then you take that and try to look and see what the top‑down data is telling you, and try to bring those two things together. That is what we try to do once a year when we go through the supply side of the economy, but the economy is a huge and complex thing made up by transactions of millions of people. Part of this is backward‑looking: you see what has happened, and then you try to see how you can explain it and how you can get a better understanding of it.
Q64 John Mann: However, one of the consequences of extreme flexibility in the labour market is that it has an impact on investment decisions. Do you think the relationship, Mr Saunders, between investment and low interest rates has vanished as an economic concept?
Michael Saunders: Vanished? No. In the immediate aftermath of the crisis, you saw a shift in the labour/capital mix. It was relatively cheap to hire people, because unemployment was high; there was ample availability of foreign workers, and at that point credit availability was tough, and so borrowing to invest was difficult. Essentially, firms chose to expand by hiring, rather than investing. That has been there through the last few years. Now, I am not sure that is continuing so much. I am hearing more talk from firms that, as it is getting harder to get new staff, they are looking to invest more to increase efficiency—labour‑saving technology—in a way that they have not had to do in the last few years, because the availability of labour was there, and is less so now.
Ian McCafferty: I would add to that that the cost of capital is only one factor in any decision by a CFO or a CEO to make an investment, and probably, in most cases, is not the most important factor. It is an important factor, but the most important factor, in my experience, is the degree of certainty about the economy going forward, and the level of demand expected for the product for which that investment is destined. From that point of view, we can change the cost of capital; we can help encourage investment but, rather like a piece of string, you cannot push it; you can only pull it. To the extent that we have had high levels of uncertainty prevailing in the economy for some years now as a result of the financial crisis and conditions thereafter, that will have had an effect on decision‑making as well as the cost of capital.
Q65 John Mann: Would your conclusion to that, Mr McCafferty, be that the very weak response of industry in investing as a consequence of sterling depreciation, which you would not normally, theoretically, expect to happen, is a delayed response, waiting for certainty in the economy, should the level of sterling remain pretty much as it is now?
Ian McCafferty: I am not sure that you can say that the investment performance of the last year or so is specifically related to short‑term movements in sterling. There are considerable lags between moves in sterling and the adjustment, both to the trade account and, as a result, to investment patterns. I would say that the relative weakness in investment that we see at the moment—and investment is growing, but it is growing rather less than I would expect, given the strength of the global economy and the low cost of capital—is a reaction to the longer term high levels of uncertainty about the economy that we have seen over the course of the last 10 years.
Sir Jon Cunliffe: Just to add one point, the Bank did some research last year. We surveyed 5,000 companies, trying to get a good match for regions and sectors in the economy—apart from the innovative tech sector, which is more difficult to get to.
The survey showed that the hurdle rates—the rates of return on investment that a company was requiring, or the rules of thumb that small companies were using to decide whether to invest—had not really changed in 15 years. Interest rates were very low, and therefore you might think, “The interest rate is 0.5%; I will look at an investment that gives me 7% or 8%, rather than still wanting a 12% return, as I did before,” but that did not seem to have happened. We asked quite a complicated question: “Why have your hurdle rates not dropped?” One of the answers that came back was “uncertainty”: “I am more uncertain now that there will be the future demand there, and that is holding me back.” There was also a lot of inertia. It is quite an interesting study. It is trying to get underneath some of these questions, but this uncertainty came back as one of the big responses.
Michael Saunders: Many of the background conditions for investments at the moment are tremendously positive. The rate of return on capital in the UK is high; the cost of capital is relatively low; capacity use is fairly high.
One thing that I hear from businesses, and you see this in surveys—for example, by Deloitte and the CBI—is that uncertainty over Brexit is probably holding back investment from responding fully to those conditions. We will see how this evolves over the coming year. At some point, we will have greater clarity on end‑state and any transition, and that might affect business confidence either way. It might also have implications for household confidence and the exchange rate. We will have to wait and see how that goes.
Chair: One hopes so, but we are going to save Brexit until the end of this session.
Q66 Catherine McKinnell: Actually, I was going to mention Brexit. I am quite interested in what you are saying about investment, and I did want to explore that a bit further. Mr McCafferty, you mentioned the long‑term uncertainty in the economy and the impact that has had on investment decisions. It was not me who raised Brexit; it was one of you. I think it might have been you, Mr Saunders. To what extent do you think the current uncertainty surrounding Brexit is having an impact on investment decision‑making in particular?
Ian McCafferty: All of the survey evidence that exists—both surveys that we have run via the agents and those coming from outside, and Michael mentioned Deloitte and a number of others—does suggest that there is an impact. Now, survey evidence tends to be qualitative rather than quantitative. It is in balances of responses rather than any fixed number, so it is impossible to say exactly by how far.
However, just to pick up on Michael’s point about underlying conditions and our position in the cycle, the economy is running pretty close to full employment, the output gap has closed, global conditions have improved markedly over the course of the last year, the cost of capital is still relatively low and the outlook for demand is probably improving. At this point in the cycle, you would normally expect relatively fast cyclical rates of growth for business investment. In previous cycles, we have seen investment growing by double digits, year on year, at this time in the cycle.
At the moment, we are seeing investment grow—business investment, at least—by 4% or 5% per annum, at least in terms of the recent past and in terms of our expectations in the forecast, so there is an impact. Now, you can argue whether that forecast is precise. The links between investment intentions in the business surveys and the numbers that you get in the ONS investment survey are not absolutely precise, but there is clearly an impact from that elevated level of uncertainty.
Q67 Catherine McKinnell: Do you forecast investment will continue to rise post Brexit, or is there any ability to forecast that in any way?
Ian McCafferty: The most recent Inflation Report forecast suggests that business investment will continue to grow over the course of the next couple of years—the forecast horizon that we put in the IR—by a roughly similar pace to that we have seen over the recent past, so, again, about 4% to 5% per year. That suggests that there will continue to be some uncertainty, but it is still growing.
Sir Jon Cunliffe: The Governor made this point when he was here. Our forecast assumes some average end state for Brexit that is some average between WTO, the single market and a trade deal, and it also assumes a smooth transition. The forecast is under those conditions.
Q68 Catherine McKinnell: You mentioned that you have travelled and met regionally. Michael, you mentioned that you had been in the North East recently. Do you take into consideration the potential differences in terms of the regional impact of uncertainty? Take the North East as an example; that is my region. There is a big anticipation of EU investment funding that is currently on its way through, and as that goes presumably it has an impact on investment decisions in different ways in different regions. Are you able to factor that in to your forecasting?
Ian McCafferty: The regional visits, and our understanding of different conditions in different parts of the country, provide depth and flavour to our understanding, and as such, therefore, are important in feeding through both to our forecasts and to our policy judgments. However, I would have to say that, of course, monetary policy is a single instrument, and applies at a national level to the macro economy. Therefore, while we will try to understand as best we can sectoral and regional implications, we can only ultimately affect the macro economy through that single instrument.
Q69 Catherine McKinnell: You have also mentioned the drop in sterling and the modest correction that we have seen to the UK’s trade deficit. One of your key judgments in the last forecast is that net trade will support growth as a result of devaluation, but we saw back in 2008-09 that that did not necessarily occur. What is going to be different this time?
Sir Jon Cunliffe: The other point that is different this time is that the world economy has, for the first time this year, exceeded our and the IMF’s forecasts. It is growing more strongly. World trade is up. The euro-area economy seems to be recovering; the US is pushing ahead, so with that stronger world picture, you get an impact from a lower price: that is a depreciation. You also get an impact from what has happened to demand, and in that 2008-09 period, world demand was pretty low. Now, both things are moving in the same direction.
Michael Saunders: As this year has progressed, we are seeing that reflected in the strength in surveys of export orders, and in the ONS data for export volumes. Manufacturing employment grew substantially in the first half of this year. You can see that boost coming through.
Dr Gertjan Vlieghe: The bit that has not happened, just simply because it takes time, is that initially when sterling drops, the things you import simply become more expensive. After a while, you start trying to look to increase production of those things at home, which boosts economic activity, but that reorientation towards finding domestically produced substitutes is a much slower process, and you would not expect to see that so quickly. So far, we are not off track in terms of the reaction of the economy to the exchange rate that we would expect, but it just takes a long time.
Q70 Catherine McKinnell: There are some anxieties that, with Brexit uncertainty holding investment back, there is a risk that we will see the downsides, in terms of higher consumer prices from the fall in sterling, but without that increase in exports. However, you do not seem to share that concern.
Dr Gertjan Vlieghe: The data so far has exports growing at about 5%, and that is a reasonable acceleration over the past year. It is a combination of weaker sterling and, as Jon said, a significant pick‑up in global demand over that period, so that part is intact. The bit that has prevented, if you will, the trade balance from adjusting is the fact that we are still importing a lot, because the things we are buying are more expensive, and we have not yet begun that process of looking for domestic alternatives.
Sir Jon Cunliffe: I do not want to stray too far into the world of Brexit possibilities and scenarios, because there is a very wide range, but to the extent that supply chain relationships will no longer work or be profitable under whatever assumption businesses are making—businesses on both sides of the Channel and the Irish Sea—many of these contracts are multi‑year. If a five‑year supply chain contract is coming up, the question is whether, if there is Brexit uncertainty, either side will be happy with that contract. As Brexit emerges, we will get a clearer idea of that, but if it looks as if there will be some frictions in the trade, that will lead to changes.
It takes time to build new supply chains. I have met domestic companies here that have said that other UK companies are looking to make their supply chain more UK‑based, and it has been a contract for them. I have heard about the opposite as well, but that effect is just now starting as people are renegotiating multi-year contracts, and that will affect trade. However, what we do not know—because it is really difficult to know—is how big that impact is going to be, and of course it will depend on people’s perception of the end state and the path to that end state.
Michael Saunders: It is worth noting, also, that this import substitution and growth of supply chains in the UK is really about the intermediate goods sector. In the last CBI survey, the share of firms in intermediate manufacturing that say they have skilled labour shortages was the highest for more than 30 years. That may be slowing, or inhibiting, this process of import substitution.
Q71 Stephen Hammond: Gentlemen, good morning. You will remember, obviously, that in your February 2017 Inflation Report you adjusted the equilibrium rate of unemployment, and, Mr McCafferty, you will remember that we had some interesting discussions on that point back in February. Given where the rate of unemployment is now, what is your view on what the rate should be, and do you think that your 4.75% is too high? What is your current view?
Ian McCafferty: Just to clarify, of course, the committee decided that the rate that we would plug into the model was 4.5%.
Stephen Hammond: Yes, but you argued strongly for 4.75%.
Ian McCafferty: I felt a slightly smaller adjustment back then, at that supply stock‑take, was possibly a better decision. The reason for that was two‑fold. One was that there are a number of different ways in which one can approach the measurement of the equilibrium rate of unemployment, and different models give you different numbers. Those that were more related to wage performance suggested that the equilibrium rate was lower, whereas those that were looking at physical heads and other physical measures within the labour market were suggesting slightly higher than that 4.5%.
I believe that there are some other reasons why wage growth had been weak, other than simply that the equilibrium level of unemployment was lower than we had expected, which would suggest that maybe a smaller adjustment was right. I also believe that in terms of forecasting technicalities, with these fundamental parameters within a model that are directly unobservable, it is good practice to change them only in smaller increments than larger increments, because otherwise you get a problem of continuity and communication, in terms of explaining why your forecast is changing.
However, since then, it is clear that unemployment has continued to fall relatively rapidly. It is now even below our 4.5%, and I look forward to the supply stock‑take that we do in February. We will probably come up with some similar findings at that stage—that those measures of equilibrium unemployment that relate directly to wages will suggest that we should be pushing it down a little further. Maybe those that look at skill shortages, difficulties in recruitment, the number of vacancies and so on, all of which remain at elevated levels, would suggest that we have already reached equilibrium unemployment, so there is going to be a balance of judgment as to whether it is lower than 4.5% or not. I would want to look at the evidence in February before making any further decisions.
Q72 Stephen Hammond: Given your earlier answer to Mr Hosie about the tightening of slack, presumably you are open‑minded about it needing to be lower than it is.
Ian McCafferty: I am open‑minded but, as I say, there will be a number of different ways in which we can approach this, which may not give us a single answer. There will be judgments still required.
Q73 Stephen Hammond: Dr Vlieghe, you were obviously supportive of a move to 4.5%. Looking at the forecasting horizon as it is now, what do you feel about the equilibrium rate of unemployment?
Dr Gertjan Vlieghe: I do not want to pre‑judge all the analysis that we are going to do in February, which is the analysis to take into account precisely the sort of changes that Mr Mann is rightly pointing out. The importance of it is that we are open to changing these things periodically in response to new evidence, and that is absolutely the right thing to do.
The thing that strikes me is that, for a number of years, as Michael has also pointed out, we were under‑predicting the rate of wage growth. Part of that was because we were too optimistic on productivity. Part of that was also because we thought, given the low rate of unemployment that we had already seen, there would be more wage pressure, and there was not. Since we made the last change down to 4.5%, we have not been particularly surprised any more in either direction on wage growth. Wage growth has roughly done what we thought would happen, so the wage growth is not going to give you a huge steer in either direction.
As Ian said, we want to look at a number of other measures about skill shortages and about allocations in different sectors, and what firms are seeing for difficulty in hiring. At this point, I do not want to judge whether that means up a bit or down a bit, or the same. That is what we will come to a collective view on in February.
Q74 Stephen Hammond: Mr Saunders, Dr Vlieghe has rightly pointed out that you have said that wage cost growth in 2017 is unlikely to pose much upside risk to inflation. Given you made a speech earlier in the year saying that you thought that unemployment would need to fall to 4% before we saw any real impact, what changed your mind to vote for a rate rise?
Michael Saunders: Let me correct that: I did not say that unemployment would have to fall to 4% before we saw an impact. I thought that the equilibrium jobless rate was lower than 5%, but I was quite careful not to pin it down to a single number.
Q75 Stephen Hammond: I thought the speech said that the jobless rate would need to fall to 4% or lower to lift pay growth to 4%.
Michael Saunders: Yes, but if you take the sentence before that, it says, “If you were to make some assumptions,” which I then argue are probably not the full story, “then it is possible”. As my colleagues have said, we look at this through a range of approaches. The more extreme one, on one side, has said that perhaps equilibrium could be as low as around 4%, but there are other methods that pointed to a slightly higher figure. It is conceivable that the equilibrium jobless rate is slightly below 4.5%. I doubt if it is a long way below, and in any case, with the jobless rate currently at 4.3% and falling, the implication would be that slack is small.
What changed my vote since the start of the year was, as I was saying earlier, inflation well above target and slack falling faster than expected. The terms of that trade‑off between above‑target inflation and the amount of spare capacity in the economy have shifted significantly. Policy responded.
Q76 Stephen Hammond: The equilibrium rate of unemployment is still around 4.5%, in your judgment.
Michael Saunders: To me, it could perhaps be slightly lower. We are going to do more on this in the next few months, but let us say, just for the sake of argument, that it was 4.25%. With a jobless rate of 4.3%, we are there.
Q77 Stephen Hammond: Sir Jon, I noted your speech last week on the Phillips curve. Can I ask you three very quick questions? One, is it your view that the relationship has completely broken down or has moved to a different relationship—perhaps a flatter relationship rather than a curve? Or can I take it, given your answers to Mr Mann, that you think it has broken down, and that the reason why it has broken down is that the labour market has changed so substantially that it is almost impossible to draw any sensible observations from that from a micro to a macro level?
Sir Jon Cunliffe: First of all, I should just apologise for croaking at the Committee, and coughing. I have lost my voice a little, but I am just rediscovering it.
Chair: It is that time of year. At least you are not giving a Budget tomorrow.
Sir Jon Cunliffe: I know. No, I did not mean to say that I thought it had broken down. It probably is flatter. There have been periods in the past when this curve is not a fixed thing; it is quite difficult to estimate. What I tried to say is that the questions over it have been increasing. There have been three periods when this relationship between unemployment and pay have been called into question. One was the Great Depression, when unemployment went up and the labour market should have cleared, and did not. Keynes tried to explain that but, clearly, supply and demand did not work, because there was a huge amount of supply, but changing the price did not work. That is where the Phillips curve tried to explain why that had happened.
Then we saw it in the 1970s, when unemployment went up and pay went up, if I can put it that way. Pay proved to be quite insensitive to unemployment, and people explained that through bringing in inflation expectations, which have become an important part of the story. We are now going through a period where the relationship does not seem to have worked particularly well, and the question is, “Can we explain that?” You can add things to the Phillips curve framework that explain why it is a bit flatter, or maybe it has dropped—because u-star is lower or fluctuating.
Q78 Stephen Hammond: If you look at the data blocks between 2011 and 2015, it is almost inverse.
Sir Jon Cunliffe: Well, indeed. You saw what we published in the Inflation Report, but ask my colleague Ben Broadbent, who published a little Phillips curve chart in a speech he made last week. If he puts inflation expectations back in and looks at it in a certain way, he can re‑establish the curve.
I do not have the evidence to say that it has broken down. I do not think it has operated as we had expected it to, which is part of the reasons why our equations have not worked. We can get to that by lowering it; we can get to that by tilting it. My own view is that it probably has flattened. What I was trying to say, though, is with that uncertainty around this fairly fundamental part of our thinking, I am prepared to look for more evidence. As Jan has said, that makes forward‑looking policy more difficult, because if I wait until I have everything, it is almost certainly too late, but that has just changed the trade‑off.
Q79 Stephen Hammond: You have slightly anticipated my final question to you, which is: given the nature of the changes and the uncertainty about some of the changes and lack of explanation, how does that make it a useful tool for policy deliberation?
Sir Jon Cunliffe: I am not ready to throw out the idea that supply and demand do not work in the labour market[2]. There are some academics who say that the supply/demand relationship does not work there as a basic relationship, but I think that it means it is a tool that you use with more caution. The reason we have nine members on the committee is so that we can all take different views, but for me it is a tool you use with more caution, because it depends on unobservable things and estimations of unobservable things. At a time when you think relationships may have changed, you might just want a bit more evidence to back that up.
Q80 Alison McGovern: Specifically on your answers to Catherine, the Governor has previously spoken at length about regional imbalances and described them as a great, problematic imbalance in our economy. I accept what Mr McCafferty said earlier about having one tool, which is national, but what is problematic about that is that there is a wide variation in terms of productivity, capacity and existing patterns of growth in the different regions.
Can I just invite you to comment a little more about how the committee gets an understanding of that and considers that? Under the shadow of Brexit, the regional imbalances in our country are laid absolutely starkly bare now, so whilst you are a national body, how do you take into account the disaggregated impacts of your aggregate policy?
Dr Gertjan Vlieghe: I am going to make a quick point, and Jon, no doubt, is going to give you some detail. We do pay a lot of attention to various distributional aspects of the economy: whether everything is close to the middle of the distribution, or whether some regions, groups of firms or groups of people are particularly badly off compared with other people. The reason we pay attention to that is because sometimes that has consequences for how the economy behaves, and so we pay attention to it even though we do not think we can particularly fix it. It is not in our mandate to fix it.
In the context of, for example, Brexit, it is a very different thing if you think that everybody is going to be subject to a given change, versus whether you think that some companies are going to be subject to a very large change and other companies to a very small change, because that large change may then have onward consequences that make the whole behave differently. We do pay a lot of attention to these things. We have all mentioned distributional consequences several times in communications, but it is not because we think we can address them. It is because we want to understand whether they are affecting the way the economy as a whole behaves.
Q81 Alison McGovern: Just to be clear, you feel you have absolutely no remit to account for the specific impact of a particular area of our country if the aggregate decision has to be made—
Dr Gertjan Vlieghe: Yes, unless it has financial stability consequences, in which case it would be outside the MPC.
Sir Jon Cunliffe: We do look at the composition of the economy, where changes are happening, because that gives us, as Jan said, a reading on the economy as a whole. However, it is asking an awful lot of unelected technocrats to ask us to take a decision on whether it is one part of the country rather than another. To be quite honest, if I had to make choices about which part of the country required stimulus or which part required cooling‑off, I would be ill‑equipped.
I can make monetary policy or technical choices based on the aggregates, but how I should balance one set of interests against another would just be very difficult for me. Obviously, you could give us a remit that said, “Favour this part and not that part,” but, for us, monetary policy is already quite a lot of power in the hands of unelected people. Parliament in the UK, quite rightly, sets the inflation target, so it could change that, but saying, “You have to make those judgments,” is much more difficult for us than saying, “This is the target to hit,” on a national measure. We try to understand the economy on a regional, sectoral basis, of course, but you deal with that. Then if we need to offset what you have done because it has had consequences geographically or distributionally, we will use other tools to do that.
Q82 Alison McGovern: Can I just come back to understanding? The productivity problem that we have in the disaggregate is quite different for different sectors of our economy. How far are you able to see into areas of the economy on a sector basis that have seen lower productivity growth and understand what that amounts to—what the problem is?
Sir Jon Cunliffe: We get regional breakdowns; the regional data is from the ONS, but you can see where productivity growth appears to be lower. We can understand why that has happened and then—going back to Jan’s question—use an interest rate tool to try to get to that, because those are the only tools we have: interest rate and, as Mr Malthouse said, QE. Using those tools to get to productivity is applying the wrong tool to the problem. Productivity is a much longer run, structural thing, and that is for government policy.
Ian McCafferty: There is some evidence to suggest that the differences in productivity performance within sectors, between the best and the worst‑performing in any particular sector, are equally as great as those between sectors. I am not even sure that it is necessarily the case that you can look at it purely on a sectoral basis. You almost have to look at it on a firm-by-firm basis in order to make a full diagnosis, at which point you have to then conclude that monetary policy is not the most effective tool at dealing with that sort of thing. There are plenty of other policies—industrial policy, fiscal policy, all sorts of things—none of which are in the hands of we unelected technocrats, which would be much better addressed for these differences in productivity and the problem as a whole.
Chair: Point taken. We are going to move on now and resume looking at Brexit.
Q83 Rushanara Ali: Good afternoon, everyone. I am not going to apologise for focusing on Brexit. Just to kick off, do you agree that at the end of the Article 50 process it is overwhelmingly likely that the UK will either be in a status quo transition with the EU or in a no-deal scenario, trading on WTO rules?
Sir Jon Cunliffe: The Government’s policy is, I think, that we will leave the European Union at the end of the Article 50 process, although there has been some debate in Parliament about that and that is really for Parliament, not the Bank of England, to decide, I am pleased to say. I observe the negotiation in the same way that members of the Committee and the public do. Every negotiation is full of noise, lightning and thunder and the like. I do not have a better view and I do not think the Bank of England has a better view of the likelihood of one outcome rather than another. This seems to be a process that is not really for economists. We do not bring any special expertise to bear in analysing where this negotiation will come out.
The Government have said they want an implementation period. They have talked about two years for that period and I assume they have talked about starting that on 29 March 2019, so that is the working assumption that I would put into any thinking about the economy. However, it is not within our gift to determine that.
Ian McCafferty: From the point of view of monetary policy, the important thing is not what we think might happen but what the main agents in the economy—consumers, businessmen, investors in the foreign exchange market—think will happen. As a result, the way in which we are conducting monetary policy is to observe closely those behavioural decisions and any behavioural changes that come as a result of the flow of news on the negotiations. We are watching consumer confidence, business confidence, investment intentions—all of these factors—and the impact of the exchange rate. It is those indicators that are more critical for monetary policy than any view I have, as an individual, or that the Committee has on the likely outcome of the negotiations and how the economy will perform.
Q84 Rushanara Ali: What would be an ideal likely outcome, in your view?
Ian McCafferty: I am not sure I am qualified enough to talk about that in any detail.
Q85 Rushanara Ali: Does anyone else want to add to those points?
Dr Gertjan Vlieghe: It is not for us to say what the best outcome is. It is for us to say that, whatever the outcome, we will support the economy to meet its inflation target. That is all we can do and that is all our remit is.
Sir Jon Cunliffe: Just to add to what Ian said, it is looking at what is in the economy now, because there clearly is something in the economy and something in the exchange rate now. It is also being clear among ourselves, to use an overused word, what framework we will use for thinking about how to deal with more clarity when it emerges: what supply will do, what demand will do, what the exchange rate will do, and how we put them together. That is why, for us, developing that framework, which we did after and before the referendum, and sharing it is really important.
Q86 Rushanara Ali: I know that you do not want to get into the business of predictions and so on, but in the inflation report the MPC says “projections assume that, in the interim, households and companies base their decisions on the expectation of a smooth adjustment to the new trading relationship”. You may not want to address this question, but what are your expectations and what are the contingency plans in the event of there not being a smooth transition or a smooth adjustment?
Sir Jon Cunliffe: From the MPC’s point of view, the point we are making is that households and firms seem to be expecting that, certainly households: if you look at consumer confidence, it is down, but you do not get the sense that households are expecting a particularly bumpy transition to a particularly bad endpoint. Foreign exchange markets have a more pessimistic view; firms are probably somewhere in between, and that is what we see. As news from the negotiation emerges about how we get there and where we are going, if I can put it that way, then we will see how those things change. We need to stay nimble with monetary policy, because it is the fastest acting macroeconomic instrument, and be looking, as Ian says, at those things that tell us about expectations and then being ready to act.
If I could add one point to this, which is just worth getting on the record, there is quite a debate in academic literature around policy reversals. If a central bank raises rates and then brings them down again, does it cost credibility or do people say, “No, no, the world keeps changing and what you did yesterday was sensible then and what you do today is sensible now”? Certainly speaking for myself, when we are dealing with Brexit, because it is an unusual thing to have a change that you cannot predict in your trading relationship with a major partner, I do not particularly worry about policy reversals or the like. We will have to respond quickly, as we did on what we saw in August 2016, when all the indicators were confidence was going down. Forward‑looking indicators suggested quite a drop. Nimbleness, or agility, is key to me.
Michael Saunders: It is important to appreciate that the implications from Brexit for monetary policy could go either way. Let us say for the sake of argument that Brexit turns out better than households, businesses and markets are currently expecting. You might expect that consumer and business confidence would rise, spending, investment and hiring would increase, but probably also the pound would appreciate. The net effect of that on monetary policy, which is where this framework of supply, demand and the exchange rate comes in, could go either way. The same applies on the other side.
Q87 Rushanara Ali: The Governor has said that the MPC will only reassess the approach to forecasting Brexit when there is a material change to circumstance. Can you give an example of what would constitute a material change in circumstance? Who would like to start?
Dr Gertjan Vlieghe: I can try. By way of example, there is a Reuters survey out that asks investors what they think is the likelihood of a disorderly Brexit, and they put that at 30%. There are also betting odds on not having a comprehensive trade deal by March 2019, which is not quite the same question, and that is roughly 50:50. As we said, the financial markets right now seem to take a fairly pessimistic view of things, meaning that they put a lot of weight on a bad outcome and then put some weight on a good outcome.
If we were to see, because of political developments or the way the negotiations go, that lots of people now think it is really likely that there is going to be a very disorderly outcome, what you would expect to see is probably another big move in financial markets and quite possibly a big move in business confidence surveys or household surveys, or both. That might be the kind of material change where we say our assumption that everybody out there is working on the assumption of a smooth transition is clearly not tenable anymore. At that point, we might consider it, but that is just one example; there are others.
Q88 Rushanara Ali: Would you say that a failure to make progress in the Article 50 talks by December or March at the European Council would fit into that sort of—
Dr Gertjan Vlieghe: No, not necessarily at all. This goes back to if people see that failure and then say, “Well, that is just how negotiations go and it is going to take a few more months”—
Q89 Rushanara Ali: It depends on how they respond to it.
Dr Gertjan Vlieghe: It depends on how everyone reacts to this, exactly, and we are not going to take a view about: “What is the likelihood now? What is the likelihood now?” We are just saying, “Are these circumstances changing in a way that is having a material impact on the economy already, or not?” So far, that has not been the case.
Michael Saunders: We are certainly not going to give a running commentary on the ups and downs from week to week.
Q90 Rushanara Ali: At what point and in what particular circumstances does no deal become the central forecast?
Sir Jon Cunliffe: I suppose no deal becomes the reality when you get to the end of the Article 50 period and there is no deal. There is a possibility of a deal all the way to the end. The nature of the deal will change. Clearly, if you agree a transition period at one minute to midnight, you will get expectations moving around before that much more than if you agree it in advance. It comes back to this point: even if we knew now what the outcome was, we do not know how supply, demand, the exchange rate will adjust to it, because there are different expectations out there and they cannot all be right. Somebody is going to be disappointed; somebody is going to be proved right. The person who is disappointed may well cut back on economic activity; the person who is proved right may boost it. Even if somebody said to us now, “This is the Brexit; this is how consumers will react; this is how the exchange rate will react,” they would have to tell us the whole picture. Therefore, it is not so much it is the safest thing to do; really the only thing to do is to see how economic agents are filtering that news and then try to react to that.
Q91 Rushanara Ali: In relation to the transition deal and reports that 60% of businesses would trigger their contingency plans if there is no clear plan for a transition deal by the end of this year or early next year—anything ranging from December to March—how well prepared do you think they are in the event that there is no transition deal and that it is not an orderly transition or Brexit negotiation process and deal, and what will be the effects on the economy?
Sir Jon Cunliffe: Maybe I will talk about that from an FPC perspective, because we have looked at it in relation to the financial sector, where we have a specific responsibility. On our understanding of other sectors, I have heard everything on regional visits from, “We are taking a lease on a factory in Ireland and we are moving production there, because we cannot take the risk,” to “I do not need to do anything and I am not doing anything.” I have heard the full range.
Q92 Rushanara Ali: “Anything” up to when?
Sir Jon Cunliffe: At all. As far as the financial sector is concerned, there are a couple of points worth making. One, “triggering” makes it sound like it is a gun that goes off and that is it. In fact, for the financial sector, this is phased implementation. They have already started the trigger. A number of the financial firms in the UK have located where they would want to be in the event of there being no deal. They are thinking about what permissions they need to operate in those countries. They have started to look at premises; some have taken premises; one or two have tweeted. That sort of thing has happened. Over the next year or so, they will increasingly implement those plans, so there is not one trigger, but as you get closer there are some decisions that, for the financial sector, are pretty binary.
Q93 Rushanara Ali: Will it cost a lot of jobs?
Sir Jon Cunliffe: If you move families, you have to have school places. You are tied to the school year, which is an important thing, so there are points when those jobs and other things will move, but it is not: “If I do not get something by a certain date, we go from one state to another.” This is a progressive series of implementation. I would imagine the further we get in, the more you will start to see people having to cross points of no return.
Q94 Charlie Elphicke: Just turning to migration, particularly in the context of Brexit, my first question is whether you agree that, with unemployment now at 4.3%, the path for labour supply is a crucial determinant of the amount of slack in the economy and hence the monetary policy decisions that are made?
Sir Jon Cunliffe: On migration and employment, I would just take you to a more general level. Because migrants provide demand and they are in the economy and they provide supply and demand, what matters for inflation is if there is any difference between, if you like, the labour supply they provide and the demand they supply to the economy. Generally speaking, a 10% increase in the labour force leads to quite a small increase in inflation, because you are moving both supply and demand at the same time.
What is important is, as with all these things, whether you get abrupt changes. If an industry that is dependent on a workforce from abroad suddenly does not have access to that labour—it cannot find alternatives quickly and it cannot quickly invest to find machinery to make up—then you could get some impact. How quickly inward migration adjusts, particularly in the areas where industries are dependent on it, could well have an impact on the balance of supply and demand in the labour market and, therefore, what happens to inflationary pressure. However, at the overall level, you get an effect both on more labour supply and on more demand, so they tend to balance out.
Ian McCafferty: This is a very good example of where there may well be significant differences between individual sectors. From that point of view, we can do our best to understand those differences, and the extent to which those differences then affect the whole, were they to affect food prices in particular, for example, they would have an impact on inflation. At the same time, though, while we can understand those and see their impact on the macroeconomy, there is very little we can do using interest rates to make those changes.
Q95 Charlie Elphicke: Looking at migration, my constituency in Dover are quite passionate on the subject of migration and controlling our borders. That was a key reason that so many voted to leave the European Union, and, indeed, a key reason why almost all the North East voted for Brexit as well. They believe that if there is less migration, there is a better chance of a pay rise. Mr Saunders, are they right?
Michael Saunders: The effects on the whole economy of migration are pretty complex. All of the growth in the workforce in the UK in the last five years is from people born outside the UK. It is from inward migration. That has been the source of labour supply. As that slows, workforce growth, labour supply growth, is now slowing quite sharply, but those foreign workers in the UK are consumers. They add to demand and, seeing that demand, firms invest. They are workers who work in our export industries, so they boost growth in the economy as well as boosting demand for labour.
Now, as to the end state from lower inward migration and whether that means that pay growth is higher or lower, frankly I am unsure; it could go either way. You certainly should not just say the reduced supply of labour means everybody gets a pay rise, because you probably also see slower growth in the economy.
Dr Gertjan Vlieghe: There are two important views here. One is there is a lot of evidence—and really a lot—across many different countries, many different periods and many different types of migration that constantly refutes the idea that when migrants come they take the jobs of the locals and reduce the wages of the locals. It just does not happen on any meaningful scale; the effects are tiny. The easiest way to understand that, which is also a big part of the academic literature, is to make the distinction between substitutes and complements. If a plumber comes to the UK, then he may well have a small downward effect on the wages of plumbers, but he will have an upward effect on the wages of people who work in the plumbing supply store, and the net of those two things is approximately zero.
Q96 Charlie Elphicke: Looking at where we are in terms of wider industry, let us say you have a fall in net migration and there is a constant labour shortage. Would that incentivise firms to make business investment in kit and therefore drive productivity, and could that have a potential longer term positive effect on employment and unemployment?
Ian McCafferty: I have heard stories going in both directions in terms of that argument. Some firms have said, “Yes, we will need to look at labour‑saving investment in capital.” Others have simply said, “No, we will simply slow down our production.” I was in Lincolnshire not that long ago and many of the firms, particularly the agricultural firms I was speaking to, were saying, “No, it is not cost effective to do that. We would simply let the crops rot in the ground.” It is very difficult to say, net‑net, in advance of these changes exactly how it will affect the labour market.
Michael Saunders: The thing that you have to take account of here is that foreign workers coming into the UK, especially those from other EU countries, on average are better educated than the national population. A higher share of them are graduates, and part of the attraction for high‑skilled, high‑paying firms to locate in the UK has been the availability of high‑skilled labour, not just from the UK but from a range of other countries. In that sense, the inflow of migrants has helped to expand economic opportunities and the prosperity of the UK and, for some firms, the loss of that ability to hire people easily might well be a reason to invest less.
Sir Jon Cunliffe: You also may see a productivity change, because one of the ways you get diffusion of skills and certain techniques and the like when you do not have them is you bring in people who do have those skills and they transmit them to the workforce over time. Lower migration in those areas may have a longer run effect, but that is quite a slow‑acting thing.
Q97 Charlie Elphicke: Let us say that we have a sudden flowering of investment in automation, autonomous vehicles—the technology that was expected in the next five or 10 years suddenly arrives next year. What would the effect be on inflation? Productivity would spike, I would imagine. The economy would, presumably, expand at a faster trend rate. What would be the inflation and wider effects of that kind of revolution?
Ian McCafferty: What we do know about those sorts of revolutions is that they take considerable time. Electricity was first adopted in the 1870s. It did not really affect factory processes until the 1920s, so this notion that somehow we will find a sudden change that will be adopted across the piece over the course of the next couple of years or so is slightly implausible, if I may say so. To some extent, it is going to depend on, therefore, the pace of adoption; it is then going to depend on where the benefits from that investment lie and whether shareholders are benefiting or whether workers are benefiting. There is going to be a whole series of things, but to imagine that somehow, deus ex machina, some technology is going to come along over the next couple of years and significantly change the performance of the economy is perhaps hope over experience.
Sir Jon Cunliffe: As well, it depends how flexible the economy is and how well it can redeploy resources, people and capital, to different things. If you had a technological change that suddenly threw a lot of people out of work, the impact on demand would probably come much faster than any improvement in supply, and you would see, if you like, demand efficiency in the economy, more unemployment and less inflation as a result. What really matters is how quickly, as an economy, you can move resources, retrain people and reallocate capital.
Michael Saunders: A common pattern that you see in countries where you have had a gradual increase in productivity is that, perhaps if wages do not respond to it straight away, unit labour costs weaken and company profitability rises. Employment strengthens; firms hire more people, and they invest more. Wages pick up and people, feeling more confident about their future income prospects, borrow in anticipation, so you see stronger consumption and investment. In other words, the improvement in productivity growth feeds through to real incomes and stronger economic growth. The outlook for the economy under that case would be better. The balance between supply and demand, which is what we can affect, frankly is ambiguous; it could go either way.
Q98 Charlie Elphicke: Turning to the Brexit monetary response, obviously there has been quite a lot of discussion about that already. Let us say a deal is agreed. The Government say, “We will pay 40 billion; we will agree a free trade agreement that allows limited or no divergence.” In that scenario, would you think it would be, broadly, monetary business as usual?
Sir Jon Cunliffe: It goes back to the point we have been making. In that scenario, I imagine much of the pessimism that is in the exchange rate would reverse, so sterling would go up. That would be bad for exporters, who are depending on that exchange rate, so demand would go down, but inflationary pressure would also go down, because we would not be paying the current prices for our imports. That is one effect.
What happens to supply? Supply probably has not adjusted, though it is starting to adjust now and, generally, it tends to adjust in a more slow‑acting way. I guess where people were not investing or being pessimistic about supply contracts, there would be a boost there. However, there might also be a boost in demand, and consumers and firms who had been holding back would really get much more optimistic than they are, relative to the foreign exchange markets.
When you put those things together, it may be that demand increases faster than supply and the exchange rate is not big enough to offset that inflationary pressure, in which case monetary policy would respond by tightening to bring supply and demand back into kilter. It could well be that households do not respond, because that is what they are assuming anyway, and the foreign exchange markets do respond, because they have been pessimistic, and as a result it is disinflationary, in which case you are going to have to respond in the same way. I am not giving you an answer, because I do not have one.
Chair: The answer is that no one knows.
Sir Jon Cunliffe: All I have is a way of thinking about this and things that we are very carefully watching and being ready to act on.
Q99 Charlie Elphicke: Let me give you a different case, because you obviously have a foot in both camps, having not so long ago been UKRep as well as wearing your current hat. Let us say we go for divergence and maybe there is a limited form of deal. What is your personal sense of short‑term effects and longer term effects of such a strategy for this country?
Sir Jon Cunliffe: I will not answer as UKRep. It has been four years since I did that and the world has changed.
Chair: Very sensible.
Sir Jon Cunliffe: Markedly. If there is a deal that means that we have a lesser trading relationship with the EU, what will the impact of that be? It clearly depends partly on if we can offset with larger trade elsewhere or more activity elsewhere. My short‑term view goes back to the point I was making earlier. When you move resources around the economy, it does not mean that the end result is not better. It may well be, but that act of moving people around the economy, moving capital, normally costs you and it costs you in productivity particularly, because people have developed skills and some of our capacity will be facing the wrong way. It will be facing into European markets, where the trade is no longer as easy. We are pretty flexible in the UK—one of the most flexible of the European economies—but reorienting that capacity does not happen overnight and you normally take a hit in economic activity and productivity.
Q100 Rushanara Ali: Sir Jon, I am not going to make you present the alternative doomsday scenario to Charlie’s question, but given what he said, and of course we want to see that sort of world—a positive outcome in the light of the referendum result—does it follow that we do need a very good deal in order to get to that scenario that you have painted as opposed to crashing out and ending up on WTO rules?
Sir Jon Cunliffe: Part of this reorienting of resources around the economy depends on lots of other things: how much support is there in the world economy in general; how quickly can you do trade deals with other countries. I am not a trade negotiator; I do not have a view on that.
Q101 Rushanara Ali: But your points were put in response to Charlie’s question predicated on us having a deal, as he set out. In an ideal world, we have a deal that allows us to trade without restrictions and so on.
Sir Jon Cunliffe: Sorry, I misunderstood. It is the same answer. I thought the question was we get a deal but it is not the relationship we have for the moment.
Rushanara Ali: I am sorry—I mean before that.
Sir Jon Cunliffe: A fortiori, if there is a deal that has a lot more change to the current relationship, then there is just more readjustment to do.
Q102 Rushanara Ali: The issue currently, as you well know, is there is a great deal of concern about not having a deal and crashing out and the consequences of that, as you say, in terms of readjustment and so on. I just wanted to get a clarification on that point.
Ian McCafferty: The point to make is that, irrespective of the nature of the deal, sudden change is difficult for business to deal with. Therefore, the better they can understand both the endpoint and the transition to that endpoint, the better.
Sir Jon Cunliffe: We have said at the FPC that we need a transition.
Rushanara Ali: No, I meant beyond that.
Chair: We have asked about that.
Sir Jon Cunliffe: We are on the record.
Chair: Thank you very much indeed, gentlemen. We very much appreciate your evidence this morning and your being here. We will see you in a couple of months’ time. For now, thank you very much.