BoEMPR0003

Written evidence submitted by Dr. Swati Dhingra

Questionnaire on the reappointment of Dr. Swati Dhingra to the Monetary Policy Committee

Do you have any business or financial connections, or other commitments, that potentially give rise to a conflict or perceived conflict of interest in carrying out your continuing duties as a member of the Monetary Policy Committee?

No. As recorded in in the Bank’s published Register of Interests, I have a close family member working at Standard Chartered Indonesia and a close family member on the Labour Market Advisory Board. The Bank is satisfied that this does not constitute a conflict of interest. I attest that these records are fully up to date.

Do you intend to serve the full term for which you have been re-appointed?

Yes.

Do you have, or do you intend to take on, any other work commitments in addition to your continued membership of the MPC? How did you manage any other work commitments during your initial term, noting that you said at your appointment hearing that it may not be feasible to spend only three days a week on your MPC role?[1]

Yes. I continue to collaborate on research activities with the Centre for Economic Performance (CEP) at the London School of Economics (LSE). In my view, this is an integral part of my role at the Bank because it enables me to bring frontier evidence-based research to the deliberations of the MPC. I am a Trustee/Director of the Royal Economic Society (RES), the Royal Mint Museum (RMM), the Review of Economic Studies and the Institute for Fiscal Studies (IFS). These roles complement my research activities and contribute to the economics profession and public engagement more widely.

I had factored in spending more than the stated three days a week on the MPC role and expect that this will continue. I therefore remain on leave of public service from the Department of Economics, and I am grateful to the Directorate of the LSE for enabling me to prioritise my work on the MPC.

During my first term, I worked more than fulltime on the MPC role. I had been understaffed for extended periods of time which amplified the workload, and I have engaged with the Bank’s Monetary Policy Transformation (MPT) programme to examine and address the related issues. With an excellent economist and adviser in place and through changes in the deliberation process that are occurring through the MPT process, I am receiving greater support for the MPC work. This has allowed me to balance the work demands more appropriately and to make up for some of the lost opportunities that had occurred in engaging with the Agencies and media in the past. 

What is your assessment of the prospects for UK and global economic growth and inflation over the short and medium term, including specifically the impact of US tariffs? How do you see the balance of risks around loosening monetary policy prematurely as against those of remaining tight for too long?

Prospects

I joined the Monetary Policy Committee in August 2022, entering at a particularly challenging moment for monetary policymaking. Headline CPI inflation was accelerating rapidly, and would peak at around 11% by the end of the year driven by a succession of very large external cost shocks. A global pandemic and the policy response had already severely disrupted the UK and global economies, labour markets, and patterns in household saving and consumption. This was followed by the Russian war in Ukraine and the energy shock, which represented an unprecedented terms-of-trade shock for the UK economy.

At my appointment hearing, I suggested that examining how cost pressures would propagate the energy shock through firms’ supply chains would prove crucial to understanding the inflation process, and I explored these dynamics in my first two speeches as an MPC member[2]. Indirect exposure to energy as an intermediate input accounted in large part for the staggered pattern of sectoral inflation and subsequent disinflation. Energy and other imports accounted for 10.8 of the 16.8 percentage points of cumulative CPI inflation between 2019 and 2022, and 8.6 of 24.7 percentage points by 2024[3].

To put this in international context, some advanced economies, such as the United States and Canada, being net exporters, did not suffer a large negative terms of trade shock from the energy crisis. The work subsequently also found that while our European neighbours suffered in the way we did, they were somewhat less exposed and had a smaller shock. The share of energy in the consumption basket in the Euro area was not as large as the UK and, the energy price peak was much shorter and lower.[4]

There has been substantial progress on disinflation over the past two years, as the impact of global shocks has unwound, post-pandemic labour market churn has subsided, and the restrictive stance of monetary policy has dampened excess demand in the economy. That progress has allowed the MPC to withdraw gradually some degree of policy restriction, while maintaining Bank Rate in restrictive territory to weigh against any remaining inflation persistence.

In the MPC’s baseline forecast, CPI inflation is projected to rise to a peak of 3.5% on average in 2025 Q3, before gradually falling back to target in the medium term. Annual GDP growth is projected to remain close to its current level, of just over 1%, before picking up towards the end of the forecast period. With potential productivity assumed to recover some of its recent weakness, a margin of slack is expected to continue to open up over the coming years, sufficient to remove any remaining excess inflation persistence. Conditional on the market-implied path for Bank Rate, monetary policy was expected to weigh on aggregate demand, with fiscal consolidation to play a greater role in the latter part of the forecast.[5]

On balance, the risks to inflation and growth appear to me to be tilted to the downside. The most significant contributions to the near-term pickup in headline inflation reflect developments in household energy bills and past energy shocks, and to a lesser degree, regulated price increases, rather than an imbalance in underlying supply and demand pressures. Unfavourable demand conditions, a consequence of subdued consumer spending and weak investment intentions, and a labour market in balance reduce the likelihood of renewed second-round effects. Along with domestic demand shifts and emerging slack, recent global developments in energy and trade policy point to potential downward risks to global growth and world export prices. I have therefore voted to reduce Bank Rate at consecutive meetings this year.

US Tariffs

I set out my early view of the consequences of US tariffs in February of this year.[6] On the overall impact on inflation in the UK, the direct effect of US import costs and potential dollar strengthening are likely to be offset by reduced global price pressures. Our imports of goods from the US, that could possibly see costs rise in the US and then passed through to UK consumers and businesses, are a much smaller share of the consumption basket than the overall influence of imports from the rest of the world.  Among the goods we buy from the US, energy is a key import and unlikely to see price-increasing effects from tariff impositions. Reduced import demand from the US is more likely to put downward pressure on global export prices.

Expectations of increased trade barriers and uncertainty over them are likely to dampen global economic activity, and to subdue global price growth. In the May MPR, conditional on developments in trade announced so far and an appreciation in the sterling effective exchange rate, the MPC’s central projection suggests a modest drag on UK growth and inflation over the forecast period. While some direct sectoral vulnerabilities have been addressed in the new trade deal with the US, sectors that have rigid supply chains that are highly integrated with the US and its trading partners, such as pharmaceuticals, could slow activity in high value-added export sectors and dampen UK growth. The scope and duration of the impacts on economic activity and inflation will be contingent on the timing, magnitude and nature of trade policies globally, and without clarity on the policies, it remains too early to conclude over what period and to what degree different economic effects could materialise.

If tariff impositions and retaliatory actions proceed in an orderly way, countries will be able to undertake a deliberate and systematic reconfiguration of trade relationships. While relative prices and economic activity would need to re-adjust to new trading realities, inflationary pressures would remain relatively contained through targeted tariffs and countermeasures. Increased trade barriers would dampen income growth and put offsetting downward pressure on prices from increased trade costs.

If heightened tariffs precipitate stricter barriers on the foreign content of imported goods, global supply chains could fracture along geopolitical lines, with far-reaching consequences for economic growth. A disorderly fragmentation, where in an extreme scenario, several large economies decide to impose trade barriers similar to those proposed by the US, would put severe strain on a few sources of supply. Price spikes and volatility would arise as economies lose diversified supply relationships, particularly in critical inputs. These could amplify into broader inflation through the increased complexity and interdependence of supply chains. While I remain cognisant of these extreme geoeconomic scenarios, current reports suggest that large economies, such as the European Union and trading partners like ourselves, are looking to negotiate with the US administration and hence the world economy seems to be moving closer to an orderly fragmentation, possibly of a milder form than one with very punitive countermeasures.

Beyond tariffs, the prospects for inflation and growth will need to account for exchange rate dynamics and financial conditions that will be affected by spillovers from the US. Based on the movements seen on days with tariff announcements in the run-up to the impositions, the dollar would have been expected to appreciate in the short term. This is a key price-increasing channel for studies that have modelled the impact of US tariffs on the UK economy. But in my view, the longer-run attractiveness of the dollar has recently been working against this channel and I will continue to monitor how this develops over time and the net impacts of world prices, exchange rate movements and financial conditions for the inflation outlook. The Financial Policy Committee is also keeping the MPC informed on potential financial spillovers from the world economy.

Monetary Policy

The MPC voted to reduce Bank Rate by 25bps in May, but I voted in favour of a larger reduction, so where I come down on the balance of risks around the pace of withdrawing restriction is clear enough. Judging the pace at which we renormalise the policy stance involves consideration of the risks of proceeding too quickly or too slowly. While the risks of moderating the policy stance too quickly are well-rehearsed and remain at the forefront of MPC deliberations, undue delay entails its own risks in achieving our inflation target sustainably, and comes at a cost of weighing on living standards more than is necessary to safeguard against risks to price stability.

It is very difficult to identify in real time to what extent restrictive policy is suppressing demand and to what extent supply. Ultimately though, a protracted period of contractionary policy may have deleterious effects on supply capacity and productivity. This can occur as a result of reduced investment, through financial conditions being restrictive or demand conditions (consumption weakness, uncertainty) remaining unfavourable for too long. In the Bank’s standard forecast treatment, business investment is one channel through which supply responds to monetary policy, and Staff work suggests it has contributed to weaker headline productivity over the recent tightening cycle, in line with survey responses from the Decision Makers’ Panel.

A continued monetary policy stance that was too restrictive risked inflation deviating from the 2% target on a sustained basis, by suppressing demand and disincentivising investments in supply capacity. To avoid this, I have cautioned against excessive reliance on late-cycle indicators around turning points, as this could lead policymakers to become too backward-looking.[7] For instance, producer price inflation of items in the consumption basket has helped to anticipate movements in headline CPI, whereas AWE wage growth has lagged inflation and labour market pressures. Despite conflicting reads from AWE on future wage-push pressures, incoming wage settlements have so far been close to the Agents’ annual pay survey figure of 3.7 percent for the end of 2025, and are starting to approach rates that are consistent with stable long-term pricing.

In addition to these factors, continued weakness in consumer spending and firm investment suggests that there is sufficient room to alleviate monetary restriction without risking second-round effects. Even if neutral rates have risen for hard-to-evidence reasons and even if the recent memory of high inflation makes it more natural for households to be sensitive to price increases, the context of a weaker UK and global outlook and a smaller near-term uptick in inflation makes this a very different environment from the recent high inflation episode. Yet we have kept monetary policy very restrictive for a long time and have only just begun to ease some of the restriction.

At the actual and market implied path for Bank Rate, we would average 430bps across meetings during the upcoming period where inflation is expected to average 3 percent (spanning 1.5 years from 2024 Q4 to 2026 Q1). In my view, this is a disproportionate response in light of the dynamics seen during the recent high inflation episode when inflation returned to target from a very high average of 6.9 percent under an average Bank Rate of 450 bps (for a period of 1.75 years from September 2022 – my first vote – to May 2024 when inflation returned to target briefly). I have therefore voted to reduce Bank Rate by larger decrements, because gradual Bank Rate reductions would still keep us in restrictive territory all year.

You have been outvoted on the appropriate level of Bank Rate at 16 of the 21 MPC meetings you have attended, the highest proportion of any current or previous MPC member. What are your reflections on whether monetary policy has been set rightly over your tenure to date? How might inflation and growth have developed had Bank Rate been set in line with your votes?

I would have preferred Bank Rate to have followed a different path over this tightening cycle. In my view, we could have contained inflationary pressures with a lower peak in rates, and avoided the strain we have put on living standards and business investments, without risking the disinflation process.

I have held a minority view on inflation dynamics observed following the energy cost shock. The general consensus has been that strained capacity and relatively strong demand fuelled inflation. I agreed with this to a degree, as there was an overhang of forced savings and support from the pandemic that contributed to some demand pressures, along with the fact that Bank Rate started from a historically low level. My main view though was that the source of inflation persistence, including in services, was price transmission from supply chains and the labour market churn that had occurred over the pandemic period. Judging from how disinflation has unfolded, the assessment appears to have been borne out.

There had been concerns that the disinflation process would occur asymmetrically – inflation changes would be slower on the way down compared to the way up. There were also concerns that real wage growth would spark consumer spending and create inflationary impulses, or that businesses would not pass through disinflation from easing external cost shocks to consumer price inflation, due to margin rebuilding or additional labour cost pressures. While these were perfectly plausible reasons to worry about the inflation outlook, the evidence to support these reasons was also consistent with my alternative interpretation, as was a set of other pieces of granular data that I placed a greater weight on than the rest of the MPC.

I had been monitoring price transmission through the supply chain of the consumption basket, which suggested that cost push inflation was easing rapidly and being passed on across stages of the supply chain. I had also been examining granular measures of wage inflation, and they clearly revealed that the numbers were much more noisy than usual, causing implausibly large differences between wage growth observed from the much larger administrative payrolls data and the AWE monthly wages and salaries surveys. For these reasons, I did not vote with the majority to raise Bank Rate by 50bps when wage inflation peaked. I explained these reasons in my speeches in June 2023 and later in February 2024 when I first voted to reduce Bank Rate.[8]

Since that time, many of the assessments are supported by a cumulation of evidence – disinflation in headline CPI was strikingly symmetric on the way down, consumption spending remained subdued even against a backdrop of real wage growth, and disinflation along the supply chain continued in services – slowing from a peak of 6.5% in Q2 2023 to 2.9% in Q4 2024 (the last quarter for which services producer prices are available). That being said, nominal wage growth has been stronger than I would have anticipated, even taking the lower RTI wage growth numbers instead of higher AWE growth.

My general view, in the summer of 2023 when AWE inflation peaked, was that after a period of very high inflation, second round effects would naturally follow. Not seeing nominal wage inflation after such a large rise in consumer prices was not tenable. And, because wages tend to be slow-moving, I expected the momentum to stay for a couple of years, during which goods price disinflation would partly offset wage-push inflation. I attributed much of the wage dynamics to real wage resistance and to a lesser degree to the broad-based churn from the pandemic and recovery period. While this reasoning was consistent with the role of inflation expectations in wage decompositions and observations of cost-of-living payments by firms, subsequent co-movement of wage deceleration and normalisation in vacancy postings suggests that the context of the post-pandemic churn might have been a more important factor than was evident early on.

Against the backdrop of a looser labour market now, the fact that wage growth has come in around the range suggested by the Decision Makers’ Panel and the Agents’ Pay Survey suggests that further wage-push pressures do not pose an upside risk to the inflation outlook. Staff work shows that vacancies have fallen back to around trend levels. Initial fears about labour supply constraints, such as from inactivity and net migration, have not materialised as intensely, with firms reporting an easing in recruitment difficulties for some time along with more muted wage growth in skills pockets that previously showed greater strength. Nonetheless, there remains a clear risk that the new near-term reacceleration in headline inflation could give impetus to second-round effects once again. The ability of inflation expectations to sustain second-round effects depends crucially on the wider economic context, that is much weaker now in terms of labour demand and has not shown the expected pick-up in consumption for a considerable period of time.

Given this assessment and the channels through which monetary policy might be expected to operate, I would not have raised Bank Rate as far as 5.25%. I would have preferred to hold Bank Rate at a lower level for as long as was required to return inflation to target sustainably. The monetary transmission channels are quite concentrated on certain sub-populations and take time to build. For example, about 40 percent of outright homeowners are not being subjected to the mortgage cash flow channel and fixed rate mortgages are making the transmission process of tightening much longer. I would therefore have traded off sharper intensity for a more modest peak for longer to attain greater coverage and build-up over time for monetary tightening to propagate. I would have also preferred a longer hold at a more modest peak because it gives time to learn about the economy to calibrate policy more precisely. Depending on the state of the economy, I would have preferred to see Bank Rate settle at a higher nominal level after the tightening cycle, even if forces that may have raised the neutral rate were absent, to avoid the lack of policy room for rate reductions and reliance on unconventional monetary policy tools that are less well-understood. I have needed to adjust to the higher peak by having to move to changes of larger magnitudes to make up for the sharper peak in tightening.

In the light of the experience of the past three years and also the Bernanke Review, what is your assessment of the Bank of England’s forecasting capability, and how forecast outputs are presented and communicated?

All central banks saw forecast performance deteriorate over the inflation period. Despite employing different suites of forecasting tools/models and facing varying degrees of the energy shock, forecasting performance, as measured by the one-year-ahead forecast error, relative to scale of energy shock was fairly similar.[9] The Bank of England’s performance did not stand out in this respect.

It’s not the size of the misses in absolute terms (as these will be larger when absolute changes in inflation are larger) or performance relative to our counterparts in other countries that is of greatest concern to me; it is that they consistently came to the downside over the period in which inflation was accelerating. The persistence of forecast misses in a single direction and not being able to detect turning points in an unprecedented period are my main concerns. There is some inevitability to this – while forecasts are most useful during such periods, their performance is also most likely to suffer. Judgements are valuable during periods in which there is not historical precedent or other extraordinary circumstances (about which past experience and forecast errors are uninformative) but these must be disciplined and re-evaluated critically.

I welcome the ambitious programme of work already underway, which promises to overhaul the Bank’s deliberations, modelling and data environment with enhanced opportunities for debate and greater capabilities for forecasting and policy analysis. On the first of these – deliberations of the MPC, Deputy Governor Clare Lombardelli and Staff have already put in place experimental alternative formats for meetings and we are seeing the benefits of a higher quality of exchange of ideas and assessments in the MPC meetings. This is a very positive development and will enable meaningful MPC discussions in the future.

On forecasting and policy analysis, I remain of the opinion that I expressed in a speech last year, that investment in data will have the most benefit. Investments in measurement and related analytic tools, rather than tweaks to forecasting models or communications, will advance our understanding of the economy and ultimately lead to better decision-making. Better data and research design can reduce forecast uncertainty by providing more accurate forward-looking measures and by enabling tests of different mechanisms driving inflation. For example, modelling of the monetary transmission mechanism is enriched by the Bank’s access to administrative mortgage contracts data, that enables precise calibration of reductions in the distribution of disposable incomes of households. Other examples of forward-looking measures and granular analysis include company dissolution registries and local labour market analysis, that I highlighted in my speech because they provided more accurate information on future outcomes and enabled deeper insights into the extent to which demand and supply forces were driving inflationary dynamics.

The imprecision of forecasting medium-term outcomes, particularly during large shocks, means investing in the data and analytical infrastructure would offer much better preparation for future uncertainties – whatever the state and configuration of our modelling. I do not disagree that models can be improved, but the greatest gains to forecast performance will come from improving the quality of the data on which our models rely.

As far as the presentation and communication of forecast outputs are concerned, I support the shift away from the MPC’s ‘best collective judgement’ and towards scenarios rooted in economic mechanisms. I would appreciate a greater role for expert Staff views to inform these scenarios and to enable individually accountable members to articulate their alternative views on how the economy may evolve and the shocks to consider. This process would enable members to quantify their alternative views and to have them cross-checked by expert Staff input on their economic significance. I understand that this would not be possible for every member in every decision meeting, but it would address key points of difference when uncertainty clouds assessments of the economy and hence allow for more robust decision-making and greater accountability of individual members’ views.

In your appointment questionnaire in 2022, you were concerned that the MPC providing a view on the likely future path of interest rates could “lead to a costly erosion of credibility,” and that if individual members give views it could reduce “clarity”. What is your view now, in the light of your three years on the MPC and the Bernanke Review?

The MPC has used forward guidance in the past. Any future rate path is contingent on future developments; if the economy deviates from its forecast, it follows that an alternative policy response may be appropriate. The risk is that this conditional statement would be interpreted inevitably as an unconditional commitment. It was this – providing such a view lends itself to misinterpretation – that I suggested might lead to a “costly erosion of credibility”. This still holds true.

On the merits of individual MPC members making public their preferred rate path, my view has evolved, and I’ve become more sympathetic to the idea in principle. Such a device would facilitate a better internal discussion – permitting Committee members to discuss strategy in a more straightforward way – and provide greater clarity to an individual member’s view and vote in the context of how they expect the economy to evolve. It would bring out differences in members’ rate strategy, which can be reduced oftentimes to their immediate vote at the next meeting, which obscures differences in assessments of the economy. Moreover, individual rate paths would provide greater accountability for a Committee that is forward-looking and focused on the medium-term.

In my appointment questionnaire, I identified the possible confusion between individual and collective future rate paths as a challenge for communication that might reduce clarity. This still represents a formidable challenge. Furthermore, such an effort would draw comparisons with the US Federal Reserve’s dot plot. Individual members of the Fed have entire regional departments to support them, we have two members of Staff to support each external member.  So ultimately it’s not at all practicable for individual members, especially external members, to each produce their own forecast. As a result, any proposed rate path of an individual member would need to be understood as being qualitative and not a precise quantitative commitment.

At your appointment hearing, on the magnitude of the impact of the UK’s departure from the EU you said that “we are going to have to wait a little bit longer to tease apart how much.”[10] What is your assessment three years on?

For the first time in half a century Britain must develop a trade strategy. As highlighted in the Economy 2030 Inquiry, our trade strategy will influence what families and firms buy from abroad and what gets produced domestically.[11] Our jobs, productivity levels and, ultimately, living standards will depend on its success. The stakes are high, especially a time of heightened geopolitical tensions.

To design and develop a successful trade strategy, we need to understand how global trade developments, and Brexit in particular, have been shaping economic outcomes in the UK. As I explained at my appointment hearing, the initial impact of the introduction of the TCA was obscured to a large extent by the pandemic and related labour market and supply chain disruption, as well as the Russian war in Ukraine and the surge in the price of natural gas. Disentangling the effects of concurrent shocks and establishing a convincing counterfactual for Brexit requires granular data in addition to aggregate statistics. Unfortunately, data of sufficient granularity becomes available only with a lengthy delay.[12] Even where detailed firm-level statistics have been released, research has been hindered by access constraints.[13]

And this matters a lot. Three years on, with pandemic impacts and inflationary pressures receding, we’ve been able to scrutinise the data to discern what we can of the longer-term impacts from Brexit. My main assessment is that leaving the EU has not strengthened our comparative advantage in the world economy and the sum total of what has occurred since the referendum has left us economically weaker in several respects. However, we still must proceed cautiously on specific performance outcomes and remain tentative when relying on only a few years of aggregate data to gauge their evolution. UK exports provide an instructive example: trade patterns shifted dramatically after 2022, with a reversal from rising goods exports during the prolonged recovery from the financial crisis to declining goods exports after the TCA, that have fallen further after a recovery in 2022, along with average services export performance with some weaknesses becoming apparent despite the worldwide services boom. This serves to underscore the inadequacy of aggregate data in this period. Conclusions based on data up to 2022 – the last year for which many granular series are available – gives at best, an incomplete account and at worst a misleading picture of the UK’s performance in the world economy.

Nonetheless, I will survey what we have come to understand from detailed analysis, where possible, and aggregate statistics more recently, and clarify some of the issues that have surfaced in public discourse. I have discussed how some of these issues relate to wider fragmentation in global trade and consequences for monetary policy in a recent speech.[14]

Brexit had made itself felt immediately in financial markets. Following the announcement of the referendum result on membership of the European Union, sterling depreciated sharply overnight and remained 10 percent below its pre-referendum value over subsequent years.[15] To this day it has not fully recovered. A weaker exchange rate led to a rise in prices for food and imported consumer goods in the UK that directly reduced purchasing power of households by an estimated £870 per household per year on average.[16] Depreciation was a headwind for businesses too. Imported inputs that UK firms rely on became more costly without offsetting relief from increased export revenues for importers. This slowed nominal wage growth, particularly in services sectors, that were more reliant on dollar imports. In work at the Centre for Economic Performance (CEP), we estimate that this resulted in an aggregate fall in real earnings of 3 to 3.6% cumulatively over the three years after the referendum – or about £770 to £850 per year (in June 2016 values), relative to the absence of such shocks.[17]

Economic uncertainty rose sharply after the Salzburg summit in September 2018 when the EU did not accept the UK’s Brexit proposal, raising the likelihood of a no-deal Brexit. Elevated uncertainty around the post-Brexit arrangements had a wait-and-watch effect on businesses. Firms in the Bank’s Decision Maker’s Panel reduced their investments by about 11% over the three years following the vote, and aggregate productivity in the UK fell as businesses reorientated to determine impacts of a post-Brexit landscape.[18] After the referendum, UK performance on a number of metrics suffered compared to its peers – GDP, productivity, trade and purchasing power.[19] With the agreement of the Trade and Cooperation Agreement (TCA) December 2020, this especially uncertain period was brought to a close.

The February 2023 Monetary Policy Report contains an early stocktake conducted by Bank staff to assess the impact of the introduction of the TCA on UK-EU trade.[20] The general consensus had been that increased trade barriers brought about by EU withdrawal would reduce bilateral trade and weigh on UK productivity over the course of the following decade. However, Bank analysis of trade in goods data revealed unanticipated softness in the data, suggesting that the impact on trade had occurred somewhat more quickly than previously estimated. Taking a steer from the data at the time, the MPC’s previous forecast for the effect of EU withdrawal on productivity was revised down from about -1.7% to -3.25% by bringing forward the 15-year long-term impact to occur within the forecast horizon instead. This was corroborated by industry views from the Decision Makers’ Panel that suggests larger productivity losses of about -3.5% by 2022, possibly because more EU exposure channels, such as migration and investments are included in these surveys. While the salience of Brexit can also be leading to larger effects in the panel, research studies find business investment stalled after the referendum in manufacturing firms by an estimated £15bn over the five-year period from 2016 to 2021, though possibly linked to the sterling depreciation for importers.[21] Perennial data issues notwithstanding, the MPC continues to monitor evidence on the overall size of the impact.

Goods trade statistics led to different conclusions for the role of Brexit and the TCA in the decline in trade seen in the UK. Early studies found almost no effect from Brexit – bilateral trade between the UK and the EU was not appreciably lower after Brexit than other bilateral trades between countries.[22] However, UK’s openness – measured typically by trade as a share of GDP – had fallen compared to peer countries.[23] More recent updates of these studies find that UK exports to the EU have fallen, but by less than half of what was expected (-13% compared to an expected -22% to -38% reduction in UK-EU trade over ten years after Brexit),[24] though when the UK’s performance is set against that of peer countries, the estimate for reduced trade is already in the range of the expected long-term reduction by 2023 (-27%).[25]

Naturally, these are substantive differences and understanding them is important for determining the extent to which Brexit has affected the UK economy through one of the most direct channels of international trade in goods. One easy point of reconciliation is that 2022 showed some recovery to reduce the divergence with peer countries,[26] but new aggregate trade data for 2023 and 2024 has shown greater weakening. The main takeaway point is that it is hard to attribute the entire slowdown in UK exports to the pandemic, and Brexit is a plausible explanation for why the post-financial crisis recovery in exports of the UK stalled in the post-TCA period. This comes out strongly in industry views and in industry-specific studies that are the focus of recent research.[27]

In terms of services, aggregate trade with the EU was affected by weakness in the tourism industry following the pandemic. By now, much of services trade has recovered from the pandemic shifts. Worldwide services export values have boomed, rising by 69 percent over the ten-year period from 2015 to 2024.[28] Recent trade in services have reduced the extent of the early divergence observed in trade falling much more with the EU. While the UK remains a `services superpower’, services growth has been normal compared to the OECD but much below that of the European Union or the Euro area in 2024.[29] Our new deep agreement with the EU contains cross-border restrictions on trade and investment that have been applied to many services. In work at the CEP, we find that these restrictions have reduced the likelihood of UK firms exporting to the EU by 6 percent, relative to exporting the same service to non-EU countries.[30] This occurred at the same time as when intra-EU trade has increased sharply in services.[31] And our analysis of OECD services trade data suggests that by 2023, UK export values to the EU were 8.6 percent lower than those of other OECD countries in services that received trade restrictions under the TCA. However, it is hard to put a more precise number to the loss in overall services exports because granular ONS trade values are not publicly released for disclosure reasons and 2021 is the last year for which services trade and foreign investment data are available in the secure data service.

Overall, we do not observe changes in the composition of trade and investments from the EU to non-EU after Brexit on a scale comparable to those we have seen in international migration to the UK. Even so, there has been a discernible weakening in export performance, accompanied by a slowdown and a slight pivot away from the EU in goods and services trade. At one extreme, all of this slowdown after 2022 can be attributed to Brexit because the recovery from the pandemic seemed almost complete on imports by 2022. And in the absence of granular data to tease this apart more precisely, a more definitive conclusion is that Brexit has not given the UK any productivity advantage from being able to set its own trade policy and becoming Global Britain.

In a recent speech, you said that “in a world where external supply shocks become more prevalent […] monetary policy action alone […] is not well-suited to address systemic price shocks in key sectors such as energy and food [and] may even be counterproductive.” Is it your view that monetary policy should not respond to external supply shocks in most circumstances?

It’s not possible to be so categorical when discussing the policy response in such general terms. Assessing the need for monetary policy action would require careful evaluation of the nature of the shock, its sectoral impacts and the time horizon over which it would be expected to unwind. This, in conjunction with an assessment of prevailing economic conditions, will dictate the extent to which monetary policy should respond.

In a narrow sense, it is the case that a contractionary response to a negative external supply shock would be counterproductive. External supply shocks confront policymakers with an inflation-output trade-off that does not arise with demand shocks, against which monetary policy is more readily and effectively deployed. In a broader sense however, a monetary policy response may well be required to manage the subsequent adjustment to external supply shocks. Given the backdrop to the recent inflation episode (low-rate environment, supressed demand and forced savings from pandemic-era restrictions), I voted to raise rates to lean against second-round effects from the global shocks on domestic price and wage setting with a restrictive monetary policy stance. Demand management was necessary to reduce any excess demand on strained capacity. My point in the cited passage was that the response needs to be calibrated carefully, lest we curtail investment that would enhance supply resilience and, in doing so, exacerbate future vulnerabilities.

Our inflation targeting framework has served us well over the last few years. The massive terms-of-trade shock we faced took time to propagate through supply chains and for relative prices to adjust. Had we tried to prevent all of the first-round (and some seemingly second-round) inflationary pressures with monetary policy tools, it would have been a massive blow to the economy with possibly deep scarring effects that would have resulted in a permanent dent to our supply capacity. The flexibility of our framework allows us to avoid such extreme deployment of our tools and to focus on dampening second-round effects that can threaten price stability while allowing relative prices to adjust to provide the signals needed for balancing demand and supply in a market economy.

Where excessive adjustment costs arise from monetary policy, a coordinated policy response in the face of an external supply shock becomes essential. It is not the place of an MPC member to opine on matters of fiscal policy, but it suffices to observe that, in a turbulent world, our macroeconomic framework will best support output growth and price stability if monetary policy works in concert with fiscal policy in certain respects. In the recent episode, the energy price cap is an instance of this; similar measures have been successful elsewhere (and Switzerland that widely deployed price controls during the inflation episode has been commended for its performance in containing second-round effects). Fiscal policy has greater traction on supply side responses, with a greater range of more targeted policy levers, whereas MPC tools are more effective in demand side management. In my view, systemic price shocks require relative price shifts, that are best managed in their short- and long-term adjustments, by a combination of monetary and fiscal policy instruments.

16 May 2025