Speech
I suspect it will be little news to the inhabitants of an island in the middle of the North Atlantic that we live in a world where big shocks can challenge the way we go about our lives and force us to adapt, build resilience and be ready to act. On some days, an icy wind blows from the North under a clear blue sky. On others, Autumn’s dismal rain moves in from the South under a sky packed tight with clouds. Occasionally, tectonic plates shift, and volcanos erupt. I am not going to belabour the metaphor. Suffice to say that we live in a world where big economic shocks can test the strength and resilience of the framework for monetary policy.
In the United Kingdom, where the weather can be to say the least variable – as here in Iceland – that monetary policy framework is inflation targeting. It is well-established and has proved its worth. The numerical inflation target and an independent Monetary Policy Committee with a diversity of views, expertise and experience, supported by expert analysis from staff, has been fundamental in anchoring inflation expectations and delivering price stability.
In the early days after inflation targeting was first established in the 1990s, however, the world was a different place. It was the ‘NICE’ (or ‘Non-Inflationary and Consistently Expansionary’) period, according to one of my predecessors – a time when, on the whole, fluctuations in economic activity were driven by small shocks to aggregate demand against a backdrop of a steadily expanding supply side.footnote [1] These were mild days of warm breezes and springlike rain.
In this world, monetary policy could no doubt be challenging, but the setting was reasonably predictable. Central to MPC deliberations at the time was to form a clear view of the outlook for aggregate demand in the 1-2 years ahead and set interest rates in a forward-looking manner such that it aligned with aggregate supply. In this way, inflationary pressures would be kept in check, and the economy could grow along a sustainable path with consumer prices rising at the target rate. This was a world that lent itself to one central projection as the basis for policy deliberations and to communicate the inflation and policy outlook to the outside word – with the central paths positioned within fan charts thereby rightly avoiding any suggestions of a spurious degree of precision.footnote [2]
The global financial crisis was the first major test of this approach. Demand was certainly affected in this episode, by uncertainty and the loss of wealth and income, but it was not the only part of the economy to suffer. The supply side of the economy was affected too, and sharp exchange rate moves affected imported inflation directly. Inflation targeting had to adapt to recognise and manage evident trade-offs between the speed with which inflation was brought back to target and the balance between aggregate demand and supply in the transition.
In the United Kingdom, this was expanded in the annual remit letter from the Government to the Bank of England in 2013. The remit recognises that there are circumstances in which returning inflation to target as quickly as the lags in the monetary policy transmission mechanism allow could cause undesirable volatility in economic activity and employment. Expanding this to large and persistent shocks was a sensible change, and ‘trade-off management’ was subsequently deployed in the response to the United Kingdom’s decision to leave the European Union.footnote [3] But it did not fundamentally change the way the MPC went about its business, with a central projection and a fan chart around it derived from past forecast errors at the centre of its policy deliberations and communication.
Over the last five years, the world has been very different, certainly compared to the tranquil years before the global financial crisis. As my former colleague Ben Broadbent put it, we went from NICE to NaSTY (‘Not-AS-Tranquil Years’).footnote [4] Whatever we call this period, a sequence of unprecedented global shocks has created a very challenging environment for monetary policy. The largest pandemic in a century, the largest war in Europe since 1945, and now a trade war between the world’s two largest economies – these are not small and simple disturbances to aggregate demand, and they come against a backdrop of low productivity growth and ageing populations. While it remains to be seen how recent changes to global trade policies will play out and what the effects on our economies will be, the effects of the pandemic and Russia’s brutal war on the Ukrainian people are fresh in our minds. Our economies have suffered, inflation has surged. These have been hard times for businesses and households, not least those on lower incomes.
As the economic effects of these shocks have faded and inflation has come down, we do also have a positive story to tell. The nominal anchor has remained intact. Inflation targeting – through forceful action to lean against second-round effects from the global shocks on domestic price and wage setting with a restrictive monetary policy stance – is working to return inflation sustainably to target. It is testament to the strength and resilience of this framework that we can say we are on course to put the inflation surge firmly behind us.
This is not to sound complacent. We must learn the lessons from the difficulties we have faced as policymakers and forecasters over this period. Our models, infrastructure and analytical frameworks were challenged by the sheer scale and unpredictability of the shocks that hit us. Underlying issues were revealed under the stress of these big unforeseeable events. Forecasting became much more difficult, irrespective of the specific models and approaches used. We need no reminder that the global economic environment is likely to continue to be challenging – and less predictable – than it was in the past. So we need to adapt and develop to ensure that our processes are nimble and robust, and that our monetary policy decisions are communicated effectively, while ensuring that we continue to act methodically in response to inflationary pressures.
That is why, in the Summer of 2023, we asked Ben Bernanke to lead an independent review into the Bank of England’s forecasting and related processes during times of significant uncertainty. We are very grateful to him for having taken on this work and for the dedication he put into it. As you would expect, the review Dr Bernanke published in April last year was a thorough and carefully conducted assessment of the relationship between our projections, monetary policy decisions, and their communication. It has been an excellent catalyst for a comprehensive programme of change.
There are many practical elements to this work. In the year since the report, we have made substantial investments to continue to develop key parts of the Bank’s model and data infrastructure. A significant programme is in train to deliver a state-of-the-art environment for working with data on the cloud, and to update our systems for accessing, analysing and visualising data accordingly. We have updated our core structural macro model to better capture the transmission of energy price shocks and to tackle the extreme data outturns of the pandemic period. And we are widening our suite of models including heterogenous agents, machine learning and threshold vector-autoregressive models as well as a new semi-structural model. Work is underway to bring this together in a modelling and data environment with enhanced capabilities for forecasting and policy analysis.
But more broadly, the challenge we face is to adapt our processes so that they assign more prominence to risks and welcome challenge to underlying assumptions, drawing on a wider range of analysis and exploring different economic shocks and mechanisms through which they affect the economy, while ensuring that it continues to serve to maintain the nominal anchor that is the Alpha and the Omega of inflation targeting.
The UK setup with an MPC with nine individually accountable members, each with an equal vote in monetary policy decisions, was never a good match for a single core model and a single central projection summarising one view of the outlook. Policy discussions on the MPC are open, frank and lively – as they should be. Expert views are exchanged, assumptions investigated, and questions posed. Three-way splits are not unheard of. I can honestly say that there is no groupthink on the MPC.
This is a system with great strengths. Diversity of expertise and experience, combined with expert knowledge, makes for better decisions given the complexity of monetary policy. Differences of views are inevitable consequences of the uncertainty we face. But equally, behind a split vote is often a high degree of communality on the qualitative factors shaping the outlook and the broad implications for the policy stance.
In agreeing on a central projection, the MPC has historically come together by forming what has long been described as its ‘best collective judgement’. This is a view of the outlook that all members can sign up to as reflecting the balance of views on the Committee. But exactly what the ‘best collective judgement’ is meant to represent has been left undefined and ambiguous, open for discussion and negotiation. Embracing this ambiguity has been the way the MPC has reconciled individual accountability with the approach of formulating monetary policy through a central projection for the economy and inflation.
Dr Bernanke challenged us to reconsider this approach, and instead of only deliberating our way towards a ‘best collective judgement’ of a central case to add alternative scenarios to our policy making process and communication. That, he said, would “help the public better understand the reasons for the policy choice, including risk management considerations”.
In response, as my colleague Clare Lombardelli set out in a speech last Autumn, we have been building scenarios into our processes, framed within a broader discussion of risks to the outlook.footnote [5] And we have started to see the benefits.
Scenarios have helped us not only to explore what would happen in case a particular shock, or constellation of shocks, should hit the economy, but also how any given set of shocks could affect the economy and inflation depending on the strengths of different economic mechanisms. And it has helped us consider how monetary policy should respond in different states of the world – or with a different balance of risks – as well as the implications of setting policy as if we are in one state of the world when in fact we are in another. These are considerations that enrich monetary policy deliberations.
In the Monetary Policy Report published yesterday, we presented two alternative scenarios along with a baseline projection. In the first scenario, global and domestic uncertainty could weigh on UK demand to a greater extent than in the baseline, in turn easing inflationary pressures. In the second scenario, recent energy price rises could lead to new second-round effects on domestic prices and supply could be more constrained, in turn increasing inflationary pressures.
These scenarios are meant to convey more than mere upside and downside risks to inflation. By setting out the mechanisms behind them, they explore why inflation may take a different path. And from a policy perspective, it matters whether inflation differs from the baseline because of demand or supply. Even if the difference in inflation is of a similar magnitude on the downside and on the upside, the size of the required monetary policy response might not be. A demand-driven downside scenario is likely to require a larger monetary policy response than a supply-driven upside scenario, simply because there is more of a trade-off to balance when inflation and activity move in different directions. These are nuances that an articulation of the mechanisms behind the scenarios can help us bring out and clarify in our communication.
But the choice of these two scenarios – proposed by Bank staff – should not be taken to mean that MPC members, individually or collectively, put a larger weight on a downside risk to inflation from demand and an upside risk in inflation from supply than the opposite constellation with an upside risk from demand and a downside risk from supply. Nor should it be taken to mean that inflation risk is skewed in one direction or the other, or that we see the risks to the path for Bank Rate to be skewed. The scenarios are only two examples from many possible paths the economy could take. That is important to emphasise.
What the scenarios also do, by exploring important judgements underlying the projection, is to serve as articulations of elements of the outlook that individual MPC members can use to position themselves within our material. This is an additional benefit in the UK context where MPC members are individually accountable for their votes and are expected to explain their positions to the wider public. Scenarios, combined with a broader set of analysis, can help support the explanation of alternative views without the need to sign up to a ‘best collective judgement’.
This is the direction of travel for our monetary policy communication. We are moving away from one central projection reflecting the ‘best collective judgement’ of the MPC, set within fan charts to illustrate risks around it. This was a good approach in a world where fluctuations in activity and inflation were largely driven by relatively small disturbances to demand.
But it does not work as well in the world we now live in where we are exposed to big shocks to supply as well as demand – and in particular in a context where individual members of a policy committee are individually accountable for their decisions. So instead, we are putting greater weight on the key judgements behind our view of the outlook, emphasising the underlying economics as much as precise numbers, and framing our discussions within a broader discussion of risks and drawing on a wider range of analysis.
We will maintain a baseline projection, based on a staff proposal, one that a majority of the MPC agrees is a reasonable baseline, rather than one that meets an elusive notion of the MPC’s ‘best collective judgement’. And we will use scenarios as vehicles for exploring risks around the baseline and accommodating differences of views on the Committee.
Over time the scenarios we look at will evolve as our capabilities advance. And as we build out the scenarios, we will be able to develop the explanation about how we have factored them into monetary policy decisions. So our public communications will evolve along with this process.
These changes that will help us build resilience into our inflation targeting framework and secure the nominal anchor for the future, whatever it may bring, come wind or rain. Our commitment to the 2% inflation target is unwavering.
I would like to thank Richard Harrison, Robert Hills, Karen Jude, Martin Seneca and Matthew Waldron for their assistance in preparing these remarks, and Fabrizio Cadamagnani, Alan Castle, Swati Dhingra, Clare Lombardelli, Catherine L. Mann, Huw Pill, James Proudman, Fergal Shortall, James Talbot, and Iain de Weymarn for valuable comments.