Interview with Philip R. Lane, Member of the Executive Board of the ECB, conducted by Stephen Kinsella on 31 August 2023
5 September 2023
Let’s dive straight in and talk about the broad inflation outlook. What are we seeing?
We’re speaking today, 31 August, and this morning we had the flash data for August. Of course, in any given month, the change is relatively small, but what we did see is some easing in the elements of core inflation. So goods inflation and services inflation came down, which is very welcome. But, as I said, one month of data is only one piece of information; we need to see that continue. What I would say is: our calculations are basically that there were very strong price increases a year ago, which will fall out of the data this autumn. So we do expect to see this famous core inflation come down throughout the autumn. Let me now turn to the other big components of energy and food. Energy came in stronger. The first half of this year was dominated by this spectacular decline in the price of gas. Also, oil markets were contained. There has been some movement in recent weeks on those categories, which partly includes, in some countries, the lifting of subsidies, essentially. Now they are saying: at some point you have to remove the supports that were put in place last year. So we do expect, with energy and food, some bumpiness. But let me summarise that as: the latest data show that inflation is standing at 5.3 per cent overall, which remains high, but in terms of looking for signals of momentum and signals of directional change, I would underline the fact that there has been some easing in goods inflation and services inflation, which is a welcome development.
In terms of the key trends, are there any anomalies in this month’s reading?
Let me first of all emphasise that this is just the flash; the detailed inflation data don’t come out until the middle of the month. So we will have to see. But what is really important: some of the conjectures about the summer, whereby maybe we would continue to see quite strong services inflation, including coming out of the obviously still strong demand for tourism in many countries — services inflation remains significant, but the fact that there was some easing, I think, helps to limit that narrative. The most visible change over the summer was some turnaround in both the oil and gas markets. That remains a major source of uncertainty. As you know, in some of the gas markets, it has to do with some factors which may dissipate. But what I would say is that, after having seen a lot of welcome downward drift in energy inflation for most of the year, we will be keeping an eye on this situation.
Based on the flash estimate, give us a sense of short-term and long-term inflation forecasts and how those are aligned with your obvious mandate, the mandated inflation target.
Let me emphasise that the work to put together the September forecast is still ongoing. But if I go back to the June forecast, and indeed the basic narrative that has been there all year long, it is: remember where we were last autumn. Inflation was around 10 per cent last autumn. Essentially, what we had even at the end of last year was an assessment that a lot of this inflation would fall away in 2023, with further progress in 2024 and in 2025. And that basic shape of the inflation dynamic, where it’s a multi-year process, so 10 per cent inflation doesn’t collapse to 2 per cent very quickly, but the direction of a significant drop this year… It’s already halved from 10 to 5, with further progress expected this year. That reflects the fact that this 10 per cent inflation arose very quickly and had a clear connection to a number of supply factors. As those supply factors are reversed, you would expect to see some easing. But it’s also the case that it’s a multi-year process because there is a second round. And this year is really the year, we think, of peak second round. So this year, we are seeing significant wage increases. And again, the projections in June basically had an assessment that the overall wage adjustment would be multi-year: the biggest adjustment this year, where we’ve had average wages across the euro area growing north of 5 per cent; growing north of 4 per cent in 2024; and north of 3 per cent in 2025. So, well above historical trends but, crucially, on a declining path: five, four, three. In the calculations for September, we’ll have to see where we are on that. That inflation, again, does not collapse to our target rate quickly because that second round has to play out to some extent. The policy challenge is to contain it, to make sure that the second round is contained and does not become embedded. So the whole challenge here is: let’s get back to 2 per cent in a timely manner, to use our phrase. What is a timely manner? It’s sufficiently quickly that everyone understands that the current inflation episode is time-limited, it’s temporary, and therefore you should not change your longer-term behaviour embedding the idea that inflation would remain high. We want people to understand that this is a temporary inflation episode. It’s not going to disappear overnight, but at the same time it would be a mistake to extrapolate the high inflation we’ve seen into a longer-term projection.
I want to turn to the causes of inflation. So the concept of sellers’ inflation, which is very much associated with the economist Isabella Weber from UMass Amherst, and others of course – what’s your take on this? How do you see the potential for the kind of strategic price controls that she and others advocate on vital sectors as an emergency response to local shocks?
I think we need to break this down into some components. We absolutely do have to understand how firms make their decisions. Because, as you say, that is the proximate mechanism between whether prices rise quickly or stabilise: it is the decisions of firms, that is absolutely true. So let me then think about what the constraints facing firms are. One constraint facing firms is their cost base. And essentially, I think the scale of the increase in energy prices in Europe in 2022 just made it essential for many firms to raise prices. They would have gone bankrupt if they’d had to fully absorb the rise in energy costs. And that’s across the economy, many service industries – running a hotel is very energy-intensive, transportation, a restaurant… So there are many sectors which ended up being in our core measure, which are very energy-intensive. The cost element is there and you might say that’s not strategic, it’s an imperative. If a cost rises enough, you can’t absorb it and lower profits; you have to raise prices. Then there is a strategic element, which is: okay, I might have a cost increase but in the market conditions I face, can I as a firm also add on extra profits in that environment? Again, I think 2022 was a unique period in some sectors. In the energy sector, the way pricing is set is that it’s basically the marginal source of energy that determines the price. Essentially, there’s a windfall there for many utility firms, which might have lower-cost inputs, but they received that marginal price. So we saw significant profits in some types of utilities. In some manufacturing, and agriculture, there are a lot of global forces in pricing. In 2022 some of those global forces meant prices were quite strong and, again, essentially a windfall for the firms. That is reversing this year in many ways. So those firms now, if they want to be globally competitive, have to accept the different economic environment. Then we come to the domestic firms serving the national economy or the European economy. Here, again, 2022 was unique. If you go back to March 2022, you had this co-timing: the start of the war and the post-pandemic reopening of the European economy. So what we saw in spring 2022 and throughout the summer was essentially two forces. As I mentioned earlier, the increase in the cost of energy. But with the pandemic, there was a kind of reopening of many services, entertainment, hospitality, plus people had quite a lot of money in their bank accounts from not being able to spend during the pandemic. That is the perfect environment for strategic price increases. Going back to the question: in 2022, it’s clear that there was, I think, a significant contribution from strategic pricing.
This year? Our assessment for this year and the coming years is that the profit component is going to fall. That environment, which was so supportive of price increases, we don’t think it’s there and we think it’s going to get even more restrictive over time. And that’s where monetary policy is playing a role. Essentially, the interest rate policy we have is basically dampening demand. Firms know that, if they try to raise prices too much, they are going to lose too many customers. So there is a constraint on that pricing this year.
I think the history of price controls is that ultimately it’s not going to be the best way to fight inflation. They may contain inflation temporarily but we saw with the Nixon policies, for example, in the 1970s that essentially, once they are lifted, a lot of the suppressed pricing re-emerges. I think there’s a distinct debate – and let me emphasise that it’s distinct – which has played out in countries like France, for example, which is that it is not price controls, but essentially having various forms of fiscal intervention that limit the sharpness of the inflation increase. In turn, if you limit the intensity of inflation, then you may weaken second-round effects. There’s a lot of that still playing out and that remains an interesting issue. But price controls as such – I think I’d be in the sceptical camp that that’s the best way to deal with this. Again, I think many people recognise around the world, in many sectors, there is more market power than there used to be; there is more monopoly power. But that was true before the pandemic and is going to be true after the pandemic. So then the question is: how does market power interact with, essentially, the unusual environment? And I think that it may have played a role in 2022. But if it was just firms arbitrarily raising prices as they like, I don’t think we’d see the scale of disinflation that we are seeing already and that we expect to continue to.
I want to put to you a question that I’ve been asked a lot. It runs as follows: in the context of inflation driven by pandemics and wars, how effective do you believe interest rate hikes can be or would be? It’s often been put to me that this is not where you’re trying to dampen demand in one place. Perhaps it might be worth phrasing this in as simple a way as possible: why is it that, through no fault of their own, Irish homeowners who are paying a little bit more in interest rates, who have gone through the pandemic and the war – why, for example, are interest rate hikes dampening demand the solution?
Let me again locate the role of monetary policy in all of this and let me come back to the basic pattern here: inflation rose to around 10 per cent at the end of last year and, as we talked about earlier on, we do think a lot of this will fade away just through the reversal of the supply shocks that had caused it. So then the question is: should monetary policy be kind of invariant? Just say that these supply shocks will fade away and therefore, monetary policy can just remain invariant. The reason why that’s not enough? Let me go into that. Number one: the most basic task is to avoid making the inflation problem worse. If you have a high inflation rate – even if you expect it to ease over time – and a very low interest rate, that creates a lot of pro-inflationary incentives. Basically you may say: “well, now is the time to borrow a lot of money because prices are going up. I should buy things now and then sell them later on.” That’s true whether it’s in the asset market or the goods market. So, to avoid pro-inflationary behaviour, it’s important that interest rates go up enough. That’s essentially the most basic reason why we normalised interest rates and have now moved into restrictive territory. We have published on our website our assessment that inflation would have been, I think, several percentage points higher than it is now if we had left interest rates super low, because that dynamic would have taken over. One counterfactual is that keeping rates super low, when inflation is going to be high on a multi-year basis, could have led to this kind of procyclicality, which would make life even worse.
Number two: I talked about a second round where firms raise prices because they have higher costs, and workers require wage increases because they’ve lost living standards. Then the question is: how do you make sure that process is contained? Because we know that if inflation gets embedded, it can be quite high for an arbitrarily long amount of time – this famous phrase of “tit-for-tat” inflation. The original shock is gone but, in any given quarter, firms are raising prices because they expect workers to be looking for higher wages, and workers are looking for higher wages because they expect firms to keep on raising prices. This is where dampening demand through higher interest rates means, as I said earlier, that firms know they have to limit their price increases, and workers can be confident that, because inflation is contained, their loss of living standards is not so much at risk. You mentioned that in the context of households and mortgages. That is definitely one channel but I think, in the euro area, the bigger channel is the behaviour of firms, especially in sectors which have a lot of debt: manufacturing, capital-intensive sectors, the tech sector. The tech sector is important here in Ireland. We know that for firms whose revenues are later but whose start-up activity is now, their ability to fund activity depends on interest rates. So the interest rate policy, we think, works through industry sectors and I wouldn’t overemphasise, in the European context, the mortgage channel. It is the most visible channel in some countries but I think that how firms respond to high interest rates is probably more dominant in the overall mechanism.
I want to talk a little bit about modelling. All models are limited. They’re obviously vague representations of reality. But could you elaborate on any limitations you perceive in the current ECB models for predicting inflation and give us a sense of how they are going to be addressed? In 2020 and 2021, the ECB was forecasting inflation around 2 per cent. It was already incorrect, even before the war. So what aspects of these models underestimated the eventual outturn of inflation? Was it the commodity forecast assumption? Was it the labour market? You mentioned wage growth already. What do we take from that going forward?
Let me try and describe a little bit about how the ECB forecast is assembled. Of course, we spend a lot of time on this, but let me at least give you a basic description. So we do have macro models, which indeed are calibrated for normal business cycles. Absolutely. But we also have two additional components. One are so-called satellite models. You might say, well, if we have a sense that the normal model might not be capturing everything, you can customise a satellite model and work through what might happen under different scenarios. Then the second element, which sort of connects that, is staff judgement. The forecast would only take a few seconds if we just pressed a button and ran the model. A lot of the effort in the forecast is trying to assess: in what way is the model wrong? And what kind of judgements should we include? Maybe I’ll come back to the various satellite models that you could run to help you form that judgement. They are informed judgements as opposed to just arbitrary judgements.
Then let’s think about what happened. I’d say part of it is they take some parts of the economy as exogenous, or by assumption. As part of this, it was assumed that the recovery in oil and gas prices compared to the worst of the pandemic would happen, but would be fairly measured. Honestly, even in late 2021 we saw a recovery in oil prices, but it was kind of orderly. What I would say is that, on the energy component, the main issue was the war. And I think it’s fair to say the war was an unexpected event. If you fed a model with a larger, more persistent energy shock, it would produce more inflation. So one issue is what you feed into the model. Maybe I’ll just pause there to say, again, if we return to a world where, essentially, energy shocks or other types of shocks become more normal, do we think of the pandemic as a once-in-40-year event? Do we think of the war as a once-in-40-year event? Where essentially you might say “we’re learning a lot about what happens in these rare events, but that doesn’t necessarily mean that how we think about normal times needs to be revised”. And this comes down to the second part of how the forecast is made: in real time, thinking about judgement. Because you can understand that we have, I think, very good modellers at the ECB. They’re well able to design a non-linear model, they’re well able to bring in the kind of strategic pricing we talked about – the understanding of what happens if you have a big enough shock that firms abandon their once-a-year revision of prices and they move to revising prices much more frequently; of what happens, similarly, if workers suffer a very large unexpected inflation shock, and therefore, for a given amount of unemployment or a given amount of profits, they need much bigger wage increases than normal. We can model that. But of course, while we now have ex post concrete evidence, that doesn’t particularly help you ex ante. What I would say more generally in terms of the modelling discussion is that there are many non-linear possibilities out there – so there is this setup where you have a basic model, which remains traditional, but you also have a range of alternative models in your toolbox which can handle various non-linear mechanisms. Then let me come back to the policymaking. Both in the pandemic and in relation to the war, we did publish some scenarios. In the end, those scenarios did say that you might get more inflation, but nowhere near the scale we’ve had. So it’s a fair comment to say that the scenarios we looked at were maybe too local. And there’s a fair question, which some of my colleagues have highlighted, where we need to widen the scope of what we think the distribution of risk is. But again, the big question is: is all of this to do with these very unusual circumstances? And how much will remain relevant on a day-to-day basis in the future? Let me make a link to the Jackson Hole speech by President Lagarde. What is very true is that we’ve had the pandemic, we’ve got the war, but we have a lot of other things going on. We have geopolitical fragmentation as a risk, we have the green transition. So let me highlight those two especially. We also have demographic transitions and so on. The 2020s is one of these big decades, I think. We do spend a lot of time thinking about how all of those factors may renew the focus on significant supply shocks. But again, once you have the shock captured or identified, then I think the macro models do a reasonable job. In other words, I think the big issue is not so much the models. It is the diagnosis of what’s going on. Once you diagnose what’s going on, the models can impose a useful discipline on thinking about how it’s going to play out. That’s really where the judgement has to come in. But again, let me emphasise for a central bank, we meet every six weeks. We meet eight times a year. It’s a continuous process. It’s not a case of making a one-time-only call. Over the last couple of years, clearly, we kind of changed our call. And I would say that over this year, the basic call has been that we will come down from high inflation, but it’s going to be a multi-year process. A number of months into that, a lot is lining up with that diagnosis, but a lot of uncertainty remains.
Could you just elaborate briefly, and I know it’s something we’ve spoken about before, about the ‘just transition’, particularly with climate change. What is the ECB’s understanding of climate change as a series of supply shocks, demand shocks? We talked about the widening distribution of risk. If we just confined climate change to damage to infrastructure caused by extreme weather – it’s an entirely reductive thing, but if we just confined it to that – if we have extreme wildfires across Europe in summer and we have flooding in winter, that alone damages enough of the productive capacity in the economy to require a monetary policy response. Would you agree with that assessment or is that something that needs further study? I know the ECB is working on this but it just strikes me, as you say, the just transition – you’re absolutely right, but what role does monetary policy have to play?
I think there are a number of elements to it, but maybe the most important analytical issue is that all that you mentioned operates both on demand and supply. Sometimes you come up with calculations about the inflation impact that might be lower than you first thought. Essentially, this damage to supply has to be pro-inflationary. On its own terms, yes. But a firm that loses the value of its capital stock, a homeowner who was hoping to consume but now has to allocate her savings to retrofitting her home, means that the consumption demand also falls, together with the supply. So the net impact on inflation, and therefore the role of monetary policy, is maybe relatively contained. If you look at, for example, the IMF’s World Economic Outlook from last October, essentially, that was one of the big messages – the inflation effect is there, it’s not zero, but, basically, it is relatively limited. Let me come to two more significant scenarios. What I would emphasise is that, clearly, the transition has major implications for fiscal policy, has major implications across the economy. Everyone has to make significant adjustments. But I would not put the impact on inflation at the top of that list. One of the issues is that in the transition, there does need to be significant investment, both public and private. That can be supportive of economic activity, but then the funding of that investment is going to be very important, because of course, if essentially it requires rotation away from consumption and towards investment, that will lower consumption demand in the economy. If it calls for higher taxes, it will reduce disposable income. On the other hand, Europe has a significant current account surplus, so if the rise in investment is basically financed by running smaller current account surpluses, then the net stimulus to the economy will be bigger. So there are many conjectures. As you say, we’re working quite hard, but let me come to our absolute fundamental bottom line here: what Europe needs is significant, sustained action, because the worst-case scenario is not doing it and then, later on, when it becomes disastrous, having a massive recession because of the extremity of a high-stress situation. Then monetary policy would have a significant challenge, an unwelcome challenge if it’s coming from managing a disaster scenario. So pre-emptive, steady, significant investment to support the transition is absolutely what we want to see. As I mentioned to you, I don’t think the net effect on inflation is going to be zero, but because it operates both on demand and supply, it’s relatively contained.
I want to talk about labour market shortages. We currently have labour market shortages in many parts of the eurozone. And there’s a significant decline in working-age populations in places like Italy and elsewhere. How do you think about these changes in what you described (and I agree with this description) as a big decade? And how do these changes influence monetary policy decisions?
The demography element has multiple elements to it. It’s visible already. The demographic transition is quite predictable, and 20 years ago, 15 years ago, we could say “Oh, yeah, it’s going to kick in during the 2020s, we’re going to see it”. We actually are seeing it, less so here in Ireland because it’s still a relatively young population, but we know it’s coming. It’s visible in a number of countries. Then, how do we think about it? Yes, we do think that there’s a human capital dimension to the labour market issue, because you’re losing many people with decades of experience and skills. One of the questions is: do we have the education and training policies in place to make sure that the new entrants to the labour market can quickly work at that level? One of the big open questions is the margin of immigration. And we’ve actually seen a surprisingly strong return of immigration in the last year or two. One issue is improving participation rates. Here, I think working from home provides an important opportunity. We are seeing improved participation by older workers, by women as well. I think it’s reasonable to say that working from home helps that. All of those factors are relevant, but let me again come back to the supply and demand issue. Less supply of workers also means less demand, because as people go into retirement, they demand less, especially if you combine that with increases in longevity. An increase in longevity means, in basically every age group, people are revising upwards how long they have left to live. That is a very basic factor reducing consumption, because you have to save more to cover the fact that you happily expect to live longer. And all of that of course interacts with fiscal policy. My sense is that changing demography has huge implications for health systems, fiscal policy and the overall growth rate of the economy. But for monetary policy, it goes back to this basic issue: how much does it differentially affect demand versus supply? In the first order, it basically just acts in a similar way on both demand and supply.
Living in Ireland in the 2010 period, the ten-year bond spread became a thing that we all looked at. Now it’s not something we look at. It was very interesting that through the COVID period and beyond, peripheral spreads never widened appreciably. How has the ECB managed to avoid peripheral spreads widening? What implication does this have for future monetary policy?
Let me divide that into several phases. Clearly, Europe suffered this sovereign debt crisis, both vis-à-vis larger countries and also smaller European countries, and we noticed it very much here in this country. If you go back to 2012, especially, there was a lot of pessimism about the future of Europe. Number one, we did have the actual recovery from that crisis. That crisis led to many reforms, including the Single Supervisory Mechanism, much higher bank capital to the extent they are a lock on future banking crises, macroprudential rules, including mortgage rules here in Ireland, and a rebuilding of fiscal capacity through much lower deficits. Also in the private sector, much more risk aversion. Households are deleveraging quite a bit. So we went into the pandemic in better shape. The pandemic itself, initially in March 2020, led to a lot of pressure, but of course, there were two intertwined responses. One is we launched the pandemic emergency purchase programme (PEPP), which was important not only in terms of an overall stabilisation bulwark, but also its flexibility was important to contain the flight to safety dynamic. It’s important to underline that, in a common currency area, the risk factor at a national level is higher because people can run without taking currency risk. You can run out of any individual country without taking currency risk. The PEPP was, I think, well designed to contain that. What was also needed – and it remains very important now – is that it went hand in hand with Next Generation EU, which basically said: we’re not going to ask each country to handle the pandemic on its own. We’re going to have this layer of pooled fiscal resourcing, which will be directed to those countries that are most exposed. This played a very important role and it continues to do so, because that money is now being spent in countries like Spain, Italy, Greece, and so on. That’s a very important type of stability. So I would say that period was very important in the policy response. Then if I turn to right now, let me emphasise: we have a cyclical inflation problem, not a long-term inflation problem. Short-term rates have gone up, people expect us to keep rates high for quite a while, but not forever. Going back to ten-year rates, people expect rates to come down later in the decade – but not to pre-pandemic levels. I think the surveys show the market thinks that our policy rate will come down to about 2 per cent. So it’s not going back to zero or negative, but it’s a lot lower than it is now. So, if you price a ten-year bond, it’s several years of high rates and then, further out in the decade, it’s something around 2 per cent. If you have that cyclical view of the problem, this is why I think those spreads are contained. Let me emphasise: this is why it’s so important that we do keep these short-term rates high for sufficiently long. Because if we did not do that, then people, I think, would build their expectations that inflation would basically become de-anchored, and then those ten-year rates would go up. Then those countries that are most vulnerable, I think, would disproportionately suffer. We often say it, and I think it’s fundamentally correct, that medium-term price stability, getting back to 2 per cent, is the cornerstone of a more stable financial system, including stable sovereign debt markets.
Final question. As the ECB chief economist, you’ve lived through interesting times. What have you learned? What would you have done differently?
We’re learning all the time. Absolutely, a central bank is very much learning all the time. We have a lot of analysis going on. It’s a very rich environment for an economist to work in. What I would say, however, is that we’re midway through this inflation adjustment. What I outlined here earlier on is this vision, this narrative, is a multi-year perspective but inflation will come back to our 2 per cent target. There’s a risk around that and if you think about the 1970s, having launched so much economic research, so many retrospective assessments, what we know now about the 1970s is very different from what you might have written in 1975. I think a lot of the ultimate assessment will have to wait a couple of years to know how this turns out. So I’m not going to give you a major conclusion now. Much will depend on our basic assessment – that the conditions are there, both through the monetary policy but also through the reversal of supply shocks, for inflation to come back to our target in a timely manner in a couple of years. At that point, we can, I think, be better timed to do a full retrospective. More narrowly, we have been focusing our retrospective on the forecast issues. It goes back to what I said earlier: a lot of it was basically that the energy shock was not predicted. It’s a shock that is not intrinsic to the model. And then this basic idea that, essentially, we incorporated some deviations from the normal model, but not enough. But again, whether this was so obvious ex ante, I’m not sure.
We need to write more papers. Spoken like a true academic!
For sure. Maybe it’s a good way to finish on this. One of the main messages from President Lagarde’s Jackson Hole speech was humility. In line with that, we very much have to emphasise the uncertainty. We do have lots of good analysis, but in the end, the analysis depends on the diagnosis. And the diagnosis is really where the judgement calls come in. It’s important for all of us to think back to that period, late 2021, early 2022: given and rerunning those analyses, what could have been plausible alternative decisions?