Diversion ahead − speech by Alan Taylor

Thank you for the introduction, and the kind invitation to come back and visit King’s. It’s an honour to speak here today, and I am looking forward to our fireside chat and taking your questions.

For me, King’s will always be a place of warm memories. It was on another October day, four decades ago, that I went up to Cambridge to read for the Mathematical Tripos. In 1984 I arrived at the gate from a small coal-mining village in Yorkshire with my happy but nervous parents in tow. I was the first in my family line to go to university, so it was a moment of great anticipation and anxiety, for them as much as for me, as I crossed that threshold. But we need not have worried—from day one, I could not have felt more at home in this college, nor more welcomed by students, fellows, and staff.

Arriving at King’s on that first day, and then as time passed, and even a little bit today, I always had a sense of wonder walking in through the Porters’ Lodge, thinking: I maybe do not know what comes next, what I will learn, or where I will go, but I sure am excited to find out. It’s a great institution that can make you feel that way, and I am very thankful.

Well, the journey, intellectual or otherwise, hopefully never ends – for me or for you – and I feel fortunate to be back here today following a career that has taken me around the world, studying and researching economics as an academic, and now, since last year, as a policymaker at the Bank of England on the Monetary Policy Committee.

The global economy at a turning point

Maybe drawing on recollections of forty years is a nostalgia trip, but when you get to my age you will also look back. And for me it also brings powerful reminders of the enormous changes we have seen in our lives since then, at home or across the world. That is true not least when it comes to the international economy, which has dramatically globalized in my lifetime. And that, as you might suspect, is the segue I need to introduce the topic of my remarks today.

I want to focus on one very important issue: the dramatic and disruptive developments we are now witnessing in international trade, and how they may impact the world and especially the UK economy in the months and years ahead, concluding with a discussion of the implications for monetary policy.

Specifically, I want to speak about trade diversion – what it is and why it matters.

First, an overview.

Trade diversion is a phenomenon where, due to changes in trade policy, i.e. tariffs, quotas, or other barriers, exporters are incentivised to re-direct goods to different markets when their usual ones become more expensive to access. The process may happen sequentially, if the set of countries imposing tariffs expands. It is therefore an almost certain outcome of the stream of trade policy actions we are now seeing unfold around the world, the only question being one of magnitude.

In total, however, when the dust settles, the world as a whole is almost surely worse off. Tariffs and other trade barriers divert activity away from the lowest-cost producer and towards less competitive suppliers. Although this is by design, in general, it does bring about an inefficient allocation of resources. In the near term, as we shall see, tariffs also tend to be inflationary for the country imposing them, and deflationary for partner countries, another potentially unwelcome set of shocks in the short run. And all this is before considering second-round effects from retaliation.

Recently, the US has imposed historically high and broad tariffs as shown in Chart 1, for example on steel, aluminium, autos and parts, but also blanket rates on all exports from particular countries. The current estimate, as of now, of the average effective US tariff on all imports is 18% (The Budget Lab at Yale, 2025), versus 2½% a year ago, an almost
10-fold increase. This includes baseline and sectoral tariffs, weighted by country and import volumes. For some countries, the current levels of US tariffs are much higher than this average. The estimated average effective tariff is 28% for China, where 30%-35% tariffs have been imposed on most goods, plus some sectoral tariffs, with higher levels postponed during negotiations. Other countries also face high tariffs, like India, where two rounds of 25% tariffs have been implemented. These tariffs make traditional export flows into the US more costly, pushing exporters, some of them very large in scale, to seek alternative markets.

Chart 1: US average effective tariff rates

Percent of value of total imports

Of course, even beyond the recent US tariffs, there is a broader trend of trade policy being driven more by concerns about national security and strategic autonomy. The word of the moment is geoeconomics (Mohr and Trebesch, 2025; Clayton et al., 2025). As part of this shock, we have seen various protectionist measures being applied but also export restrictions and recalibrations of supply chains, sometimes called “friend-shoring”. All these measures affect the balance of supply and demand across countries and so may amplify the forces of trade diversion. Trade is going through a `great rearrangement’ (White et al., 2025) and it remains to be seen how much of trade is at best diverted, at worst destroyed.

Some trade barriers might be avoided, leading to an increase in trans-shipment and other circumvention strategies on the side of exporters. When direct trade is blocked or penalised, exporters may use third countries to ship goods or re-declare their origin in order to evade tariffs or restrictions. But these are only partial ways to offset the disruption of trade barriers, and they are often imperfect and legally risky. Moreover, the development of circumvention strategies generates additional costs that will still tend to depress global welfare relative to an alternative free-trade ideal.

What about prices and inflation? For third countries, trade diversion in reaction to an increase in trade barriers is most likely to have disinflationary effects because, from their perspective, incoming goods supply increases in the near term. This may not be entirely welcome. For example, concerns about trade diversion from the US to European markets have led the EU to recently create an import surveillance task force which is monitoring trade flows to identify “dumping”. Some EU measures are in place and more may follow, as we shall see. Logically, that could divert trade even more into fourth countries, like the UK.

In short, if trade diversion happens on a material scale in the UK, or in our closest major trading partner, the EU, then it is likely to have substantial bearing on prices faced by firms and households in the UK economy, for both final and intermediate goods. Those developments in turn would have implications for future economic prospects, and most importantly for the future path of inflation. And that is what matters for the MPC and its key mandate for low and stable inflation at a 2% target.

With that summary overview of trade diversion, let me try to look at the question in a bit more detail. And as I often do, as a way of organizing my thoughts, I will discuss trade diversion through three main lenses: economic theory, historical evidence, and the evidence from today in real time, such as it is at this point. Let me take each in turn.

Trade diversion in theory

In terms of theory, I want to walk through the basic idea behind trade diversion, which dates back 75 years to pioneering work by the economist Jacob Viner (Viner, 1950). He was living in another era, but it was also a time when large-scale trade disruption was afoot. The world was emerging from a protectionist phase in the interwar period, and the postwar multilateral trading order began to be built around GATT (later the WTO), bound together by the glue of non-discriminatory most-favoured-nation (MFN) rules. But Viner had glimpsed something new, and he had identified a weak link – that, as an exception, blocs of countries might form customs unions or free trade agreements, as a group of six European countries would in 1957, marking the origin of today’s EU.

Viner derived the implications, new ideas that could not be seen in a classic two-country trade model. For his logic, you needed at least three countries. And there then followed a set of classic economic lessons epitomising the Theory of the Second Best: trade might be created within the bloc as internal barriers fell, and that creation effect should be welfare improving; but trade would also be diverted, as trading partners outside the bloc might see their trading frictions increase, and that diversion effect would be welfare reducing. These countervailing forces had to be weighed up.

The proof, formally, takes a lot of algebra to derive the implications for quantities, prices and welfare outcomes in a general equilibrium setting. But this isn’t a PhD course in international trade. So, informally, let me derive a sketch of a result in a partial equilibrium setting, using a simple supply and demand diagram, and two important simplifying assumptions which we can think about later. We will have three countries and assume a global numeraire in the background and balanced trade. Suppose we are the home country 1, and we import from country 3, as does country 2. Initially there is free trade with no frictions.

Chart 2 shows the initial world equilibrium with no tariffs for this market at point A, where S3 (export supply from 3) meets D1+D2 (import demand from 1 and 2). Absent tariffs, the world price is P and is the same in all countries, and country 3 exports X3. Focusing on the home country, at point B the imports of country 1 are M1, and that leaves the remainder (the distance between A and B) to be absorbed as imports by country 2.

Now suppose country 2 imposes a positive import tariff T. The price P will now denote the price in the free-trade countries 1 and 3, but the post-tariff price in country 2 is higher at P+T. Thus, as a function of P the actual demand for imports by country 2 shifts in to D2’(P) which is equal to D2(P+T). The new world equilibrium is at point A’ and the demand drop induces a lower price P’, seen in countries 1 and 3. The post-tariff price in country 2 meanwhile rises to P’+T. Total exports from 3 decrease to X3’ but total imports in country 1 increase to M1’. Globally, trade is falling (trade destruction), but locally for the bystander country, imports are actually rising (trade diversion). Home import quantities rise and import prices fall.

I think that, despite its simplicity, this is a helpful device to think about the first-order impact of asymmetric tariffs on trade flows and prices in a multi-country setting. And I will argue further that for the bigger and more important policy question – could these effects be potentially large? The answer is yes, and I will try to make that point two ways.

But I said I would get back to our assumptions, so let me detour to spend a moment on two important caveats about the theoretical impacts of tariffs. This is a newly invigorated field of research in international economics, after a period of relative quiet. Against a broader Whiggish undercurrent, a sense once prevailed that conditions of relatively free, or freer, trade would be the norm, so research moved on. But history is not linear, and new work has started to engage with the changing conditions we now face, building on older classic works in the field.

The first important caveat concerns balanced versus imbalanced trade. Classical trade theory usually starts from an assumption of balanced trade, that is, countries do not run persistent trade deficits or surpluses, whereas international macroeconomics now treats intertemporal trade and imbalances as the central case. A classic paper by Obstfeld and Rogoff (2000) was a rare attempt to draw the link and explain how lowering frictions in trade would tend to go hand in hand with wider global imbalances. Higher frictions go the opposite way – and, trivially, tariffs so high as to be prohibitive would mean autarky: and then, not just trade balances, but trade itself would be forced to zero. By implication, local changes in tariffs near zero might have second-order effects on imbalances, but the response has to be nonlinear as we go further. New work by Costinot and Werning (2025) digs deeper into this problem and finds that large tariffs can have wealth effects which impact trade imbalances, and that responses depend on the sensitivity of consumption patterns to wealth changes.

The second caveat concerns how the results may be undercut by changes in exchange rates. In the simple analysis above, the exchange rate is ignored. But the classic exchange rate argument has always been that a fixed savings-investment imbalance on the macro side would assert itself: If we want to consume more than our income, a tariff would be offset fully by exchange rate appreciation, restoring the quantity imbalance. But in 2025 the forecast of dollar strength based on this thinking went awry. Now, ceteris may not be paribus, but the classic argument may just not be right after all. Recent research has made the point that the exchange rate need not fully offset tariffs (Jeanne and Son, 2024). And, as above, a prohibitive tariff clearly would override this logic, too. But we also now understand better the connection between wealth effects to exchange rates, given net and gross international asset position (Gourinchas and Rey, 2005). Recent work by Itskhoki and Mukhin (2025) highlights this mechanism, noting the large US dollar liabilities abroad, and incorporating other drivers of exchange rates and risk premia. Despite the old conventional wisdom, in these more complex settings, the exchange rate offset may not operate, which matches what we have been observing over the course of this year.

Diversion then: In search of lessons from history

Thus, with these caveats in mind, the basic economic analysis can be a guide. And in that setup big tariff shocks will make a big impact through trade diversion. But how big is big?

To make a forecast of trade diversion and its impacts, economists need to calibrate a model. That involves two problems: First, it means getting the right model, from a wide menu. And second, it means getting the right calibration to suit the world you wish to simulate. The Bank of England, like other central banks and researchers, is using a wide range of models and calibrations to assess the likely impact of tariffs and conduct sensitivity analysis. You can see some of the results of this work in our recent Monetary Policy Reports and there will be more to come.

As we weigh up that range of evidence, I think history can also be used as a guide. And the obvious place to look is the experience of the 1930s, the last time that tariffs were raised around the world to the kinds of levels we are now seeing again today. And it was also a time when the mechanics of trade diversion went to work in a big way.

That is not to say that trade policy was the only economic problem in the 1930s. The interwar economic disaster was over-determined. The constraints of the gold standard imposed deflationary pressure in the 1920s and inhibited policy action after 1929; more deflation came from overshooting commodity supply in many markets; lasting currency misalignments after First World War caused imbalances and painful adjustments; in many countries a credit boom turned into a financial crash after 1929, with painful real effects from bank failures around the world (Kindleberger, 1973; Bernanke, 1983; Eichengreen, 1996).

Real economic activity collapsed in the early 1930s, and trade volumes thus fell in part as a result of the contraction. Some observers have therefore seen trade policy as a smaller part of the overall narrative of the Great Depression, but most analysis was aggregative and missed some important details on the composition of trade relevant for our question.

But recent research (de Bromhead et al., 2019) has studied UK trade flows in the interwar period at a more granular level to provide sharper evidence on the role of tariff policies and trade diversion. In the 1930s, the world trading system was rapidly becoming insular through multiple and often retaliatory beggar‑thy‑neighbour policies. The US imposed the Smoot-Hawley tariffs, at levels similar to today, but other trading partners were also putting up barriers. At a 1932 conference in Ottawa, the UK brought in Imperial Preferences, lowering tariffs within the Empire, seeking to boost intra-Empire trade and offset a slump, while raising tariffs on foreign imports from outside the Empire.

Here the evidence for UK trade diversion is compelling, based on disaggregated UK trade data for 258 goods and 42 trading partners over the period from 1924 to 1938. The authors use state-of-the-art empirical techniques for calibration and they find a large impact of UK trade policies during this shift towards protectionism. The changes in tariffs can account for about one quarter of the decline in UK imports between 1929 and 1933, in line with results also found for the US. But much more importantly for us, the discriminatory tariffs and quotas imposed by the UK shifted the origins of imports towards the British Empire. The central estimate suggests that the protectionist shift in UK trade policy can explain over 70% of the increase in the Empire’s share of UK imports between 1930 and 1933, and well over half of the shift toward Empire by 1938.^{footnote [1]}

And that shift in trade patters was profound, for the UK as well as for other trading blocs as shown in Table 1, reproduced from the paper. The share of UK imports from formal and informal Empire trading partners rose from 30% in 1929 to 42% in 1938, a significant change in trade flows. Similar shifts were happening across the globe, of course. In France, for example, the import share from formal and informal Empire trading partners more than doubled from 12% to 26%. As each bloc turned inward, the effects of trade diversion on themselves and on others got further amplified.

Table 1: Share of formal and informal Empire in trade, 1929-1938

Percent

• Source: de Bromhead et al. (2019).

Trade diversion showed itself on a large scale in the 1930s – just not immediately, but slowly and inexorably as exporters and importers adjusted over a period of years. The first historical lesson here is that the effects of trade diversion can be large when tariffs of this kind of magnitude are imposed. The second historical lesson is that these effects can take time to gradually materialize, as the rearrangement of trade patterns on this scale happens gradually. Prices and quantities will grind slowly to the new equilibrium.

History does not repeat and the 1930s are not a simple predictor of what will happen today, as so much else is different. Rather, this focused study warns us that today’s tariff levels, which are of a not dissimilar magnitude, could also cause a gradual but substantial rearrangement of today’s world trade flows. The impacts would be felt on quantities and prices and hence would have implications for monetary policy. So let me now turn from past to present and ask what, if anything, we see of trade diversion in the real time data coming in so far this year.

Diversion now: Tracking trade shifts today in real time

I now want to draw from a current, imperfect, and constantly expanding assemblage of real-time information and analysis that can give us the best sense of the extent to which we are now seeing the forces of trade diversion at work.

My main assumptions here are that the process will be slow moving, adjustment will go on for the next year or two, at least, and we will need time to see the effects play out. The effects will be seen first in the tariffing economies, and later in the tariffed economies. Under the presumption that the UK will not erect major new trade barriers we will be essentially a third country, an innocent bystander in this maelstrom of trade disruption.

How will we be affected? If some past exports from non-US locations may no longer get into the US given new barriers, where will they go? It is not just about our trade with the US. Some of our export goods will be affected, given less demand from the US, though the US share of our exports is narrow. That means lower prices for our exports diverted away from the US, all else equal. But it is also about our trade with third countries. Here a much wider range of our import goods will be affected too, where less demand from the US means more supply left over for us to absorb. That means lower prices for our imports, all else equal, as these other countries’ exports are diverted away from the US. The major countries or regions that matter here are the large exporters to the US that have been subject to even higher tariffs than the UK, with the major examples by trade weight being the EU bloc and China.

And digging deeper into policy actions and responses, there are a panoply of potential second round effects to think about. For example, the EU and China have taken measures in response and may do more. China has imposed large retaliatory tariffs on the US, even as the EU has mostly held back. Those measures will divert US trade into other locations, including the UK, adding further downward pressure on UK prices. But the EU has also set up monitoring of trade diverted into the EU, and has taken some measures to limit what they see as a potential dumping threat, especially from highly tariffed exports out of East Asia that can no longer make it into the US. Those EU anti-dumping measures will then entail even further diversion of goods into the UK, lowering our prices further still. The knock-on effects can stretch out as we wait for the new global equilibrium to emerge.

So what evidence do we see at this early stage of trade diversion at work? Let me point to a few pieces of incoming data.

First, let us look at the origin of these new tariffs. In Chart 3, I show the value of US imports of Chinese goods as reported by the US Census Bureau since the year 2000. We can clearly see the growing importance of trade between China and the US after the Chinese accession to the WTO in 2001, only briefly interrupted by the Global Financial Crisis. That trend then flattened off, Covid happened, but still, tens of billions of dollars’ worth of goods were flowing from China into the US every month.

But we also see right at the very end the profound impact that the newly imposed tariffs between China and the US have had on bilateral trade flows. In the span of a few months, imports from China just about halved and now sit at pre-GFC levels in seasonally adjusted terms.

Chart 3: US imports of Chinese goods

Seasonally adjusted level

• Source: LSEG Datastream and Bank calculations. Latest observation: August 2025.

So, at a first glance, a key part of our theory appears to hold: that of a large buyer removing themselves from the market – causing the inward shift of the demand curve (for Chinese goods in this case).

The next question now must be, where are all those goods going, if not to the US? Contrary to what one might expect, Chinese exports to the world are up about 4% in the year to August despite the drop in US demand. In the next chart I am decomposing growth in Chinese exports into different regions. in Chart 4, exports to the US are in orange and clearly contribute negatively since April of this year. We can also see, however, that they have been more than offset by increasing exports to the rest of the world, mainly other Asian economies but also Europe.

Going even more granular in the next panel, Chart 5 shows the contribution of some selected countries and regions to Chinese export growth in February to August 2025, that is, the months after the introduction of tariffs, over the same months in 2024.

Again, a second part of our theory stays intact: Instead of restricting production, Chinese exporters are finding alternative buyers to sell to. The EU plays a nontrivial role here, as does the UK once we adjust for the scale of the receiving region.

Chart 5: Contributions to Chinese export growth by trading partner

Percentage point contributions to growth Feb-Aug 2025 over Feb-Aug 2024

• Source: Bloomberg Economics.

But what about prices? There are good reasons for why we should see trade diversion show up first in quantities and in prices only with a lag. After all, prices are sticky and determined by often only infrequently renewed contracts. In the short run, importers and exporters, and wholesalers and retailers, might all wish to smooth prices, buffering out of profits, until only later passing them through to their counterparties – a phenomenon we also see often in the case of domestic tax policy changes.

All that is in contrast to the equilibrium theory where everything adjusts at the same time. Still, there is some evidence already that prices are adjusting to the new reality of world trade.

In the next panel, Chart 6, we see prices for Chinese trade goods both from the exporter’s (LHS) and importer’s (RHS) perspective. For the exporter, prices are little changed since the start of the year, and, since the renminbi has been flat vis-à-vis the dollar, so are Chinese export prices in dollar terms. However, the renminbi has depreciated significantly versus European currencies. So, for a UK or euro area buyer, Chinese goods have become relatively cheaper.

Chart 6: International export and import prices indices

Index, 2023 = 100

• Source: Office for National Statistics, Eurostat, LSEG Datastream, and Bank calculations. Latest observation: August 2025.

We can begin to see some of that effect show up in domestic import price indices (RHS). There now is a large gap between price developments for all US import goods (solid aqua) and those imports coming from China (dotted). It is plausible that tariffs have led to a change in composition so not all of the gap might be due to within-product price changes but to a shift towards relatively cheaper goods. Notably, however, going the other way, European imports from China have also already become relatively cheaper. So it seems like there are tentative signs that the final theoretical prediction, too, may be starting to come through: that tariffs between some countries spill over into price drops in third countries due to trade diversion.

And there is narrative evidence to confirm that the redirection of trade flows to the EU, and the lowering of the prices of these goods, has not gone unnoticed by the EU authorities at a time when a new import surveillance task force is monitoring trade diversion. Their latest heatmap shown in Chart 7 from September (European Commission, 2025) speaks to growing concerns about import surges, with trade flows from China drawing the most attention.

The rise in EU trade policy actions is also notable. In the news section of the website the Commission is listed as taking six actions in the entire calendar year 2024. It took less than two months to take six actions in 2025. And over just the first nine months of 2025, the total number of actions so far is 23, with China frequently mentioned.

It is worth noting here that these actions raise the prospect of a double diversion phenomenon, or a cascade of diversion. First, the US raises barriers on imports from
low-cost producers, who then redirect their goods to third countries, like the EU, who in turn respond with further barriers to those low-cost producers, who then move on again to direct their large flows of exports to an ever-smaller target group of open export markets. Naturally, the UK comes to mind as one of those potential targets.

So finally, let’s turn to the UK. Now, in the grand scheme of things, the UK may seem to play only a minor role—you may have spotted it in the decomposition earlier, but it is hardly visible. Next to the three large trade blocs, China, US, EU, we are a tiny country after all. But that does not mean that what happens elsewhere does not matter for UK prospects, quite the opposite. Most of our imports actually come from Europe such that price pressures there will also mean price pressures here. UK goods import prices have been trending down for a few years now, weighing on aggregate inflation (Chart 8). They are of course much higher than before Covid, as are world export prices. But I would expect this overall trend to continue and even to quicken as trade diversion picks up around the world, at least so long as the UK remains relatively open and takes few protectionist steps of its own.

Chart 8: Price index of UK imports

Index, 2023 = 100

• Source: Office for National Statistics, CPB World Trade Monitor. Latest observation: July 2025.

Let me end with just an anecdotal example of this trend, and that is the phenomenon of imported vehicles, and specifically battery-electric vehicles (BEV). As the BBC recently reported, in the latest vehicle registration data for September, roughly 300,000 cars were registered in the UK; of these, more than half were electric, and about one quarter were BEV only (BBC, 2025). This is the first year when hybrids and BEVs have reached these milestones, and the growth in the hybrid and BEV segments is particularly strong, over 30%. And, significantly, among the ten best-selling cars in the UK were two Chinese models. With the US not welcoming such cars, and the EU applying protective measures, especially against the lowest-cost Chinese EVs, these trends may well intensify.

This anecdotal example only refers to automobiles, but I believe the same price pressures will unfold for many manufactured intermediate and final products. These kinds of developments are of a piece with the trade diversion story I am trying to tell. And I believe there is more diversion ahead.

Policy implications

Let me now turn to the monetary policy implications of trade diversion, and how I see them in the context of my view of the current macroeconomic outlook.

My current view is one of a preponderance of downside risks, which I see as having increased steadily over the last 12 months. The UK economy has slowed, and output has fallen below potential; slack has opened up in the labour market; confidence indicators of businesses and households have drifted to very low levels; and the trend of underlying disinflation has continued. Notably, disinflation has been led by the strong reductions in wage settlements, which look to end this year in the mid-3% range, and are likely to be closer to if not below 3% into the next year. As I see it, in an economy with rising unemployment and weak demand, wage settlements will be pushed down, and wage-led domestic inflation will not re-kindle an upward spiral.

Let me briefly show this chart which was also shown by my MPC colleague Catherine Mann last week (Mann, 2025). Every line is a probability density for the GDP nowcast in a particular quarter. Whereas, for the previous 3 quarters, probability mass was concentrated in positive territory – indicating an expectation of expansion – in the latest quarter, the aqua line, more mass has shifted left. From the perspective of this model, the likelihood of contraction in the third quarter of 2025 is now 27%, the highest this year.

Chart 9: Distributions of GDP nowcasts from a quantile MIDAS model

Density over quarterly GDP growth

• Source: Mantoan and Verlander (2025). Notes: The chart shows fitted non-parametric probability distributions for quarterly GDP growth from the quantile-MIDAS model of Mantoan and Verlander (2025). For the mapping between quantile regression and density forecasts see Mitchell et al. (2024). Monthly labels refer to the nowcast of the respective quarter of the year that the month falls in. For example,
• December 2024 represents quarter-on-quarter GDP growth for Q4 2024 based on data until
• December 2024.

Set against that, in terms of upside risk, most inflation expectation measures remain moderate. The inflation hump in 2025, while yet to dissipate, should do so in early 2026 when the taxes and administered price hikes from earlier this year roll off. Still, I see the upside risks to inflation as low compared to the downward trajectory in output and inflation fundamentals, and this view has led me to dissent in five of the last seven votes on the MPC in favour of a lower path of Bank Rate.

I now see three plausible scenarios in 2026.

The first scenario is the soft landing. I saw this as the most likely outcome in January as described in my first speech in Leeds. But the developments I just described mean that this outcome is receding in terms of probability. By maintaining what I think is a too restrictive path of interest rates, we may have braked too hard, such that inflation cannot smoothly return to target with the economy close to potential, as my votes have indicated.

The second scenario might be called the bumpy landing, one which I think is increasingly likely, a downside scenario, where inflation undershoots, and goes below target in late 2026, and the economy moves into a weakened state for a sustained period, with output and employment below potential, leading to undue damage to economic activity. Part of this scenario, in my mind, could end up resulting from some of the trade diversion pressures that I have described today: if we underestimate the forces of trade diversion washing up on our shores in the next year or two, our inflation forecast will miss the mark.

The third scenario is the hard landing, a deeper worry. This was a remote and low probability event a year ago, but the risk is rising. In this scenario, weak demand at home can lead to a more forceful downturn, where recession dynamics start to kick in that can be very difficult to contain or even reverse. The economy has been flirting with zero growth, and the realisation of negative readings could easily change the future path for the worse. The probability of this outcome is now not trivial. This would be the ‘downside to the downside’ scenario and it would lead to an even more dramatic inflation undershoot than the second scenario. To end up here would be a mistake.

I will be looking carefully at the domestic dynamics, especially in the labour market and wages, as well as output and business sentiment indicators to see where we are on these different possible trajectories. But I also remain alert to the risk that, as in our long history as an open trading nation, we may also be buffeted by significant shocks from the rest of the world.

Thank you.

The views expressed in this speech are not necessarily those of the Bank of England or the Monetary Policy Committee.

Acknowledgements

Thanks to Lennart Brandt and Vitor Dotta for help preparing this speech, and to Craig Botham, Sarah Breeden, Ambrogio Cesa-Bianchi, Nicole Gorton-Caratelli, Emil Iordanov,Simon Lloyd, Catherine L. Mann, Giulia Mantoan, Kirti Pandey, Daniel Dorey Rodriguez, Rana Sajedi, and Jessica Verlander for their comments and help with data.

References

Bernanke, B. (1983). ‘Nonmonetary Effects of the Financial Crisis in the Propagation of the Great Depression’, American Economic Review, 73(3), pp. 257-276.

British Broadcasting Corporation (2025). ‘Sales of electric cars hit record high in September’, accessed on 6 October 2025.

Clayton, C., M. Maggiori, and J. Schreger (2025). ‘Putting Economics Back Into Geoeconomics’, NBER Working Paper, No. 33681.

Costinot, A. and I. Werning (2025). ‘How Tariffs Affect Trade Deficits’, NBER Working Paper, No. 33709.

De Bromhead, A., A. Fernihough, M. Lampe, and K. Hjortshøj O’Rourke (2019). ‘When Britain Turned Inward: The Impact of Interwar British Protection’, American Economic Review, 109(2), pp. 325-352.

Eichengreen, B. (1996). ‘Golden Fetters: The Gold Standard and the Great Depression, 1919-1939’, Oxford University Press.

European Commission (2025). ‘Monitoring Trade Diversion’, accessed on 6 October 2025.

Gourinchas, P.-O. and H. Rey (2007). ‘From World Banker to World Venture Capitalist. U.S. External Adjustment and the Exorbitant Privilege’, in R. H. Clarida (ed.), `G7 Current Account Imbalances. Sustainability and Adjustment`, University of Chicago Press.

Itskhoki, O. and D. Mukhin (2025). ‘The Optimal Macro Tariff’, NBER Working Paper, No. 33839.

Jeanne, O. and J. J. Son (2024). ‘To what extent are tariffs offset by exchange rates?’, Journal of International Money and Finance, 142, 103015.

Kindleberger, C. P. (1973). ‘The World in Depression, 1929-1939’, University of California Press. Reprint: 2013.

Lloyd, S. P. and S. Solomou (2020). ‘The impact of the 1932 General Tariff: a difference-in-difference approach’, Cliometrica, 14, pp. 41-60.

Mann, C. L. (2025). ‘Explaining the consumption gap’, speech given at the Resolution Foundation.

Mantoan, G. and J. Verlander (2025). ‘Beyond point estimates: quantifying risk around the near-term UK GDP forecast using a new quantile-MIDAS model’, Bank Overground.

Mitchell, J., A. Poon, and D. Zhu (2024). ‘Constructing density forecasts from quantile regressions: Multimodality in macrofinancial dynamics’, Journal of Applied Econometrics, 39(5), pp. 790-812.

Mohr, C. and C. Trebesch (2025). ‘Geoeconomics’, Annual Review of Economics, 17,
pp. 563-587.

Obstfeld, M. and K. Rogoff (2000). ‘The Six Major Puzzles in International Macroeconomics: Is There a Common Cause?’, NBER Macroeconomics Annual, 15, pp. 339-390.

The Budget Lab at Yale (2025). ‘State of U.S. Tariffs: September 26, 2025’, accessed on 6 October 2025.

Viner, J. (1950). ‘The Customs Union Issue’, Carnegie Endowment for International Peace. Reprint: Oxford University Press, 2014.

White, O., J. Condon, C. Bradley, M. Birshan, S. Smit, and C. Lamanna (2025). ‘The great trade rearrangement’, McKinsey Global Institute.