Alan Taylor: we have flipped the switch from the ‘great moderation’

This is part of a series ‘Economists Exchange’, featuring conversations between top FT commentators and leading economists

Alan Taylor has an exceptional ability to bring contemporary lessons from rigorous study of economic history. For this reason, he is an ideal external member of the Bank of England’s Monetary Policy Committee, which is where he has been since September 2024.

Taylor is also professor of international and public affairs at Columbia University. He did his doctorate in economics at Harvard, where he studied under Jeffrey Williamson and former IMF chief economist Maurice Obstfeld, both celebrated economists. He has also held appointments at Northwestern University, the University of California, Davis, and the University of Virginia.

Taylor is a son of Yorkshire, born in Wakefield. He took his first degree at King’s College Cambridge. While there, he became a “wrangler” in mathematics, as John Maynard Keynes did before him, gaining first-class honours. He has held visiting positions at the London School of Economics and London Business School. He was a Houblon-Norman/George Fellow at the Bank of England in 2009-10.

Taylor has written or edited 10 books and more than 80 journal articles. With Obstfeld and Jay Shambaugh, he developed and tested the “trilemma” — the idea that one cannot have fixed exchange rates, free movement of capital and monetary policy autonomy at the same time. He has also co-authored work on credit cycles and contributed to contemporary ideas about “macroprudential” regulation of banks. He has also been an adviser to financial institutions, including Morgan Stanley and Pimco.

In all, he is exceptionally well equipped for his current role.

This conversation is in two parts. The first part, which focused on longer-term challenges, took place on 4 March 2025. The second, mostly focused on the impact of President Donald Trump’s trade war, took place on 23 May 2025.

Part I

Martin Wolf: How have you fitted in at the Bank of England?

Alan Taylor: I’ve been coming to the Bank of England on and off for the last 15 years and been looking increasingly at issues in macroeconomics, financial stability, exchange rates, interest rates, monetary policy and monetary transmission. So, it seems like a good fit. And I think I bring a new and different set of skills to the committee, including a historical perspective and also maybe more of an international perspective. I’m based in the US and I’ve done a lot of work over the years on the global economy. So I think these are useful additions.

MW: Have you found making monetary policy decisions, which are forward looking, a big shift from your focus on the past? Has it been a seamless shift?

AT: I think it’s been quite seamless. The famous saying is that it’s hard to make predictions, especially about the future. But it’s sometimes even harder to explain the past!

We have, thankfully, a great staff here to back us up and provide an incredible amount of insight and understanding of the data, the state of the economy and so forth. And so we’re well prepared to make each decision.

How is the 1970s different from today? The answer is the inflation targeting regime

MW: So let’s talk about monetary policy. You joined six months ago, well after the inflation shock. It was a fairly big surprise to most. It wasn’t so much of a surprise to me. But that’s because I’m a recidivist monetarist. Another surprise was that the “sacrifice ratio” — the cost of lowering inflation in terms of higher unemployment — then turned out to be relatively low.

AT: I’ll start with the operative word, which was “surprise”, also for most economists. Maybe there was more of a fiscal impulse in the US, and maybe more of an energy price shock in Europe. So a different mix of demand and supply characteristics in the shocks.

I think the world changed with the Russian invasion [of Ukraine].

An interesting exercise is to go back to the pre-invasion forecasts the MPC put out and plug in the actual price of energy. It turns out that when you put that into the models, the forecasts would have been much closer to the actual inflation path. And I think a similar exercise has been done at the European Central Bank.

So, I think that gives some reassurance that the framework for forecasting wasn’t wrong. And I think the Bernanke report made the same kind of point, by saying all of these central banks and probably many private sector forecasters made the same error for the same reasons.

The more interesting question is about the “sacrifice ratio” and how it played out after that shock. The way I think about that is that we’re not in the period in history when we tried to impose price controls or rationing, as we did in the 1970s. And so we let prices be set in markets.

That means that, when you have a shock like that, there’s going to be propagation of it. The first sectors to be hard hit are those that use energy heavily as an input and then this gets transmitted to other sectors, maybe food and so forth. And then it works its way to other goods and eventually to services.

Along the way, wage contracts are being renegotiated. So we take that as given. And the question is: what can you do in response, to minimise how large the transmission is from the shock and also ensure that it doesn’t get embedded into expectations?

So I think that brings us back to how the 1970s is different from today. And the answer is the inflation targeting regime.

Has it faced a big test along the lines of the 1970s? Maybe not, until now. But it’s now happened. As you say, probably the surprise to many people was the low sacrifice ratio, so far. It was more of a “soft landing” than most people had expected.

I put that down to expectations being more firmly anchored. This has not been a replay of the 1970s, when the anchor was dislodged and then it took the better part of a decade or two for it to be fixed again. So, hopefully, future historians will admit that it wasn’t perfect — It had to deal with the realities of shock propagation in a market economy — but it’s played out well enough, so far.

MW: Surely, a part of the problem is that we didn’t stress that one of the implications of inflation targeting is that bygones are bygones. And therefore, one might find that while one thinks that on average inflation should be 2 per cent a year, actually the price level might be up some 20 per cent over three years and that this will never be reversed.

Yet if a shock like that were to happen again, the confidence we have in the the anchoring of the inflation target might collapse, because people will then say we can live with one such surprise, but not another one.

AT: I completely agree. One of the interesting side effects of the 1970s was that economists and political scientists began running election predictions models, to see if economic factors could predict votes, and for the first time, they had a large enough sample and found that people really disliked inflation and that tended to mean incumbent parties lost.

MW: It’s not really surprising, is it?

AT: Not at all. And I think one way to read this is that this is the reason we’ve ended up with inflation targeting. The public wants low and stable inflation. And therefore governments tell the central banks to devise a way of delivering that.

We know enough about economic history to show us that we could have had price-level targeting, instead, or we could have had nominal GDP targeting, or we could have had the gold standard.

It’s not for the Bank of England to decide what’s best. That’s a political decision, with costs and benefits. So, with targeting of the price level, you could think of the deflation in the UK of the 1920s and the “debt deflation” of the 1930s in the US as reflecting that sort of approach. So would we want to not let bygones be bygones and correct them and go back to the price level target? That’s been debated, but it’s not where we are now.

There could be a future debate over inflation-targeting regimes and monetary policy. I’m sure there will be. And reviews are constantly happening. So, it’s definitely the case that the public wants inflation to be controlled and we have to try and deliver it. Finding the regime that does this best is, I think, one of the main goals of macroeconomics.

MW: I suppose the big point is that, if you look back over an extended period, economies get hit by big shocks, some of them monetary and financial and some of them real. So we do know that this regime will be tested from time to time. The evidence we have is that this regime minimises the unhappiness associated with very high unemployment or very high inflations. I think that’s still the case.

The global financial crisis was the start of a sequence of unfortunate events

AT: Unemployment was the other factor that economists and political scientists have found to be a reliable predictor of unhappiness, along with inflation

MW: So probably we want to minimise the variance of both inflation and unemployment — and that’s something inflation targeting does relatively well.

So, how do you see the current environment?

AT: Extremely uncertain — and that’s been building up over my months in the job. Risks are very two-sided and the standard deviation of possibilities is expanding. So I think that makes it a big challenge.

MW: So let’s talk about “the age of uncertainty “— the period from, say, 2007, with the global financial crisis.

AT: The US stands out as having a higher growth rate. But the US has a tech sector unlike anywhere else. Its demographics are more favourable and its capital stock is following the latter at a more rapid pace, as well. But there’s also been disappointing growth in total factor productivity relative to the pre-2008 trend.

So what could we put that down to? Some people might say that we’re running out of ideas — the nature of technological change has shifted in some way. But I think we’ve entered a period of uncertainty, which I think has been amplified by multiple shocks.

Periods of uncertainty and shocks like this one are pretty unusual. So, the global financial crisis was the start of a sequence of unfortunate events — the Eurozone crisis, Brexit, Covid and the Russian invasion of Ukraine. So that’s five shocks. And I think that has an effect in terms of greater precaution.

We seem to have flipped the switch from the pre-financial crisis “great moderation”. Now we think that every three years we’re going to get a new shock. And I think that’s been very wearing for the consumer and business.

In behavioural terms, you’d understand this as “recency bias”. And I think that probably has a bearing on willingness to take risks and so investment. It’s probably depressed demand in a lot of places. So we’ve gone from not having much resilience to having an abundance of precaution, lots of resilience, but less of the risk-taking essential for growth.

Fundamentally, I’m a technological optimist and an optimist about human capital. There could be a lot of potential for growth to resume. But at the moment it feels like there is a sense of “let’s wait and see”. I think the lesson of economic history is that the effects of these major crises last much longer than a normal business cycle. It’s not quite the same thing if shocks dislodge people’s beliefs and expectations in a fundamental way. Then it can take quite some time to repair.

MW: An argument I made between, well, 2008 to the middle of the last decade was one I took from John Maynard Keynes, which is that if you have a huge financial shock, which leads to higher precautionary saving in the private sector and a reduction in the propensity to invest, that is when governments have to be relaxed about running deficits. And if they pursue austerity policies instead, which we did and, of course, the Eurozone did, too, one ends up with years of weak demand. So, we made a big policy mistake.

AT: Keynes said the time for austerity is the boom, not the slump. I wrote a paper with that as the title. So, I can’t disavow my research. I think that’s another example of the value of studying economic history. The only experimental laboratory we have is history. So, I think there’s a tremendous amount of insight there.

MW: Quantitative easing was very controversial. Some people believed it was the ruination of nations, because it distorted a fundamental price — the interest rate — and terrible things happened. I tend to the view that it was far better than the alternatives. Now, it’s more than 15 years since we started it. What’s your view on what it did and didn’t do?

AT: I’d separate this into two pieces. First, should the QE tool exist — that is, is it potentially useful? And when, how much and where should we think about using it? And what does the evidence show on this?

So I think if we get into periods of extreme stress and financial market malfunction, such as at the height of the global financial crisis, or the initial moments of the Covid pandemic, QE is the big bazooka. It’s an instrument you don’t want to say we will never use in such moments. And I think the research has shown that it did have an impact there. And I think future economic historians will agree with that. But it was used many more times.

MW: And it was used over very long periods.

AT: And I think there’ll be an evaluation of that too. The evidence as I read it is the effect of QE in calmer times, more normal times, is less dramatic. It may have temporarily shifted yield curves, but not that much or not in a very persistent way. So I think there are probably better ways to supply liquidity outside of those extreme events.

I have technological optimism, but also human capital optimism

MW: The other policy instrument that was used in those times, albeit with very varying degrees of enthusiasm by different central banks, is negative interest rates. Do you have any sense now of what we should learn from experiences with negative rates?

AT: There’s been research with mixed conclusions about whether that led to adverse effects on the financial system. It’s a relatively brief episode. I’m not sure that we’ll get conclusive evidence.

MW: I would like to shift to prospects — the tailwinds and headwinds of today. Do you have strong views on the neutral interest rate? Do we know the neutral rate? Does it matter?

AT: I think it’s endogenous. I think one of the reasons we’re in a low neutral-rate environment is because we have abundant savings. Some of that is because of slow moving forces, like demography. So, life is getting longer. People understand the need to provide for that. Societies are becoming wealthier around the world. So more and more people are thinking beyond the idea of a working life followed by a very short retirement. I think that’s a big difference versus 150 or 100 years ago. So that’s a stream of savings. And maybe the initial reaction is that this will give us an enormous pool of savings to invest.

So, I think my interpretation of where we’ve gone in the last 10 or 15 years is that “R-star” [the “neutral” rate of interest] is unusually low, for reasons having to do with that flood of savings. And that’ll be with us for a long time. But it seems that a lot of wealth is sitting on the sidelines right now. So I see that as an upside. Right now, my estimate of R-star is pretty low, not as low as during the Covid years, but similar to 2018 or 2019.

So I think some of the models are running neutral rates in Europe, real neutral rates, at around zero and maybe close to 1 per cent, in the US and UK. But as I said, they’re endogenous. They could be rising if those opportunities for growth materialise.

MW: Let’s talk about what might make growth go well. Technology?

AT: And I also think the upside to human capital. I think if we ask whether we have reached our full potential, in terms of human capital, I don’t think we’re even close. So I have technological optimism, but also human capital optimism.

MW: Is there not the argument against that? Compared to the opportunity we had half a century ago or so when many children left school at 14 or 15, and their education was pretty basic, and 5 per cent of each generation went to university in this country. Relative to the opportunity we had then, it’s much smaller today.

AT: It’s definitely smaller. And we have those previous investments to thank for the progress we have made. But I would think the journey is by no means complete either in terms of quantity, but especially in terms of quality. So, I think that’s going to be the next challenge.

MW: One not insignificant issue is that the technology we’re all talking about at the moment is AI and there’s a big debate whether it’s predominantly complementary with human capital or a substitute for it. Do you have strong views on that?

AT: I don’t have strong views. But there have been many technological developments of the last couple of hundred years where people said: “Oh, this is going to displace labour and this will have potential downsides.” But we’ve always found with the passage of time that new employment opportunities open up.

MW: Over the last wave of technological progress, the evidence seems to be that it’s been predominantly skills-biased. So it’s been better for people with quite sophisticated skills than for people who aren’t highly educated, the sort of people who suffer with deindustrialisation. Even if you’re right overall, we do have to worry about its social impact

AT: I think we saw similar issues in the 19th century with the arrival of the machines that initially replaced skilled workers. But I think people looking at this think it will have differential impacts throughout the skill distribution: it may not be as simple as skilled versus less skilled.

Part II

Martin Wolf: What is your perspective, in terms of lessons from the past, on what’s going on with US trade policy today?

Alan Taylor: When economic historians look at the past and ask when was there a shock to trade policy, a small number of major events come to mind. You could look at successive rounds of trade negotiations under the Gatt or WTO, which have been gradual changes. Recently, people have been interested in whether these were pro-growth. What did they do for income distribution, or other outcomes.

In terms of going in the direction of higher protection against imports, there have been even fewer and bigger events. The biggest experiment was the 1920s and 1930s. And so, that’s a case where protectionism increased very rapidly, and then it took decades after the second world war for this to reverse in gradual steps.

So, I think that’s one thing we can look back to. But it’s unfortunate that it’s a sample of one, and it’s contaminated by lots of other things going wrong at the time.

I think the other thing that we can do is learn about trade diversion. Many WTO or Gatt rounds involved simultaneous lowering of tariffs by lots of countries, so that’s a general global reduction of barriers for many bilateral pairs. The US experiment is unusual. It’s one country deciding to raise tariffs, some other countries retaliating bilaterally and many other countries not wishing to put barriers up among themselves.

A trade war is going to be negative for growth

MW: I wrote quite a bit about this issue of diversion back in November 2024. So, let’s assume that the UK ends up with this 10 per cent general tariff against it by the US and let’s suppose that the EU and other countries end up with 30 per cent. Well, that looks as though it’s going to create trade diversion in our favour relative to other competitors.

So that’s a good thing for the UK, three cheers for the US. But, of course, there’s this protection by the US. So, while we might be favoured against EU competitors, we’re disfavoured vis-à-vis US competitors.

And that’s complicated enough. But then what might happen, given that we have a free trade agreement with the EU, is that stuff will come into the UK and be changed a little bit, or maybe not be changed at all and then be exported from the UK as British stuff.

And so, the Americans are going to have to create rules of origin governing trade with us, as if we were in a free trade agreement. And one can multiply this across the world. This is going to be the most complex imaginable set of rules and regulations governing trade among a couple of hundred countries. It sounds horrendous, doesn’t it?

AT: In terms of trying to model or think about where we land, if you’re thinking a year or two out, where will we be? So, in the case of the UK, we have a five-page document which can’t get into all those details right now. And there’s an intention to make a deal, but we’re not close. There are things that are still ongoing in negotiation. Will other sectors be added? What will the tariff rates be on them?

And in terms of all the other ancillary non-tariff barriers, rules of origin, etc, all of that is going to be a much longer document. Typically, free trade deals are not one page or five pages. They’re thousands of pages. And it will take weeks, if not months, to get to the final clarity on what just the single bilateral US-UK deal will look like, and similar amounts of time for any other deals, an EU deal, a China deal.

And many countries, maybe there’ll just be some unilateral tariff announced. That may be the direction, given there’s such a pressure of time.

So, all of that is going through your head week by week. As you’re hearing new announcements, you’re trying to think, where are we likely to land? And that, as you said, means many different sources of pressure on a third-party economy like the UK. Where will it stand in relation to its exports to the US? Where will it stand in terms of trade diversion into the UK from other countries?

And I’m thinking a lot about how we model that. Everybody’s thinking about that in policy institutions around the world. But, at the moment, it’s like a “quantum state”.

MW: But could we summarise? We don’t know where it’s going to go but it would be reasonable to make two suppositions. That trade will globally be less free than it was before. And linked to that, a lot of that will be because we have moved, as you said, into a quantum state.

AT: I think the increase in frictions and the increase in uncertainty, those are both likely to be contractionary, at least in the short run. And depending on how long it takes for agreement on the new trading rules to be reached, there could be uncertainty for quite some time. And I would agree that’s going to be a drag.

MW: I was looking recently at what Consensus Forecasts have been showing just for the first months of this year. For the US, they were showing the forecast for GDP growth this year down from 2.3 per cent in January to 1.1 per cent in April.

AT: Yes, that’s something the UK government has drawn attention to, that a trade war is going to be negative for growth.

MW: When people look back on the trade war and its consequences, is this going to be small compared to other big shocks?

AT: Of course, this comes with the caveat of what we said in the first five minutes, which is there’s such a range of uncertainty. So, I think your baseline was 10 per cent for the UK and maybe 30 per cent for others. Some people say, well, the pause will just be forever. So a baseline of 10 per cent, with some exceptions for some countries.

Some people worry that we’ll get to the end of the pause and very big numbers will go back on.

I think, again, there’s a range of possibilities, depending on where the needle goes on those tariff settings. And as part of this, there’s going to be discussion on other things, like rules of origin or non-tariff barriers and so forth. Maybe some of those will go in a more liberalising direction, depending on the countries involved. I don’t know.

The other, maybe more optimistic, thing to say is that when we’ve seen trade policy experiments in the past, even some quite big ones, the effects are not usually immediate. So, in the case of liberalisation by a developing country, you’ll generally see the pay-offs building up over a decade or two, through some improvement in growth performance relative to whatever the control group is. So, it’s not like we generally see a sudden step change, as a result of changes in trade policy. It’s not like what happens if there’s a financial crisis or a pandemic. So, in that sense, one wouldn’t be expecting to see something dramatically bad in a very short space of time.

MW: My reaction to that is, yes, sort of what I thought, that we’ve got through all these shocks. But what’s sad about this one is it’s completely unnecessary. It’s completely the decision of one person.

AT: As a policymaker, you have to sort of take the world as it and say, OK, what can we do? What are the best responses?

MW: Well, I’m still an editorialist, of course. So, speaking as God’s editorialist, I regard this as a bit of nonsense.

I have one final question that might be too embarrassing for you to answer, but do you think the central bank and treasuries in general should worry about the net worth of central banks? Does it affect its ability to conduct monetary policy in any way?

AT: You’re right, it’s too embarrassing to answer.

MW: I thought it might be. You see where I’m going?

AT: I can see right through you, Martin.

MW: I thought I could get that past you. Never mind.

AT: I didn’t put my foot in it? Good.

MW: You’ve become more and more like a proper central banker. And having dealt with them now for 35 years, I have found they’re almost always impeccable in not answering the questions in the way you want them to. But that’s all right. It shows you’re becoming a true professional. Treat it as a compliment.

AT: Exactly. Thank you.

The above transcript has been edited for brevity and clarity