UK Monetary Policy – ‘Crossing the river by feeling the stones’ - speech by Huw Pill

Given at the Confederation of British Industry (CBI)
Published on 26 November 2021

In his first speech as our Chief Economist, Huw Pill explains how he views the impact Covid has had on the economy. He says that, provided the jobs market continues to be strong, he thinks interest rates will need to gradually increase in the coming months, to make sure inflation comes back down from current high levels. 

However he points out that the economic picture is still uncertain, so the Bank of England can’t give precise guarantees on what will happen to interest rates, especially further into the future than the coming months. That will depend on how the economy performs. And with high uncertainty, the Bank of England should take a cautious approach to policy, assessing each decision on a step-by-step basis.


Having only joined the Bank in September, this is my first public speech as a member of the Bank’s Monetary Policy Committee (MPC). With inflation having risen to well above its 2% target and forecast to approach 5% next spring, it is a challenging time for monetary policy. Joining the Bank at this moment has proved something of a baptism of fire.

To describe how I think we at the Bank should meet these challenges, I thought I would take advantage of my debut speech to review the MPC’s most recent policy decisions and explain how I see their rationale.

The MPC’s November decision

At its November meeting, the MPC decided to maintain Bank Rate at 0.1%. It also decided to continue with the existing asset purchase programme, although this will anyway come to an end next month.

At the same time (and somewhat unusually), the Committee gave a steer about the outlook for coming MPC decisions, announcing that: ‘provided the incoming data, particularly on the labour market, are broadly in line with [our] central projections in the November Monetary Policy Report, it will be necessary over coming months to increase Bank Rate in order to return CPI inflation sustainably to the 2% target.’

What I thought would be useful today was to explain why I have supported these decisions. This entails placing those decisions in the wider context of how the UK economy and the MPC’s monetary policy have evolved over the past few years.

Setting the scene

Let’s start by winding back the clock back to the last few months of 2019. An MPC member speaking here in Newcastle two years ago would have been expected to focus on the UK’s withdrawal from the EU and its implications for the economy and inflation prospects

It is unlikely that an MPC member would have focused solely on this concern, to the exclusion of all others. But it is equally unlikely that their speech would have mentioned at all the possibility that a global pandemic could shut down large swathes of the UK economy. That was a topic for questionable disaster movies, not monetary policy discourse.

And yet for the bulk of the past eighteen months, it has been the impact of Covid-19 that has dominated the discussion and decisions of macro policy makers, both in the MPC and elsewhere. This would not – could not – have been anticipated two years ago.

There is an important lesson here. Who knows what the future will bring? Or, as an economist might say: it is difficult to make forecasts – especially about the future. Monetary policy naturally had to change course once the impact of the pandemic became apparent. But this was not foreseeable. Well-intentioned plans laid out ahead of the pandemic needed to be redrawn.

As this example demonstrates, monetary policy needs to retain the discretion to address unforeseen events, even if that discretion itself needs to be constrained by the overriding imperative of achieving the inflation target. The resulting policy flexibility is necessary to cope, in a pragmatic way, with unexpected economic disruption and disturbances.

That has important implications for central bank communication. Any guidance offered by monetary policy makers about the policy outlook always needs to be conditional on how circumstances evolve. And, given the often substantial uncertainty surrounding what the future holds, in general such guidance cannot amount to much. With this in mind, observers need to guard against interpreting any conditional signals offered by central banks about the policy outlook as hard pre-commitments to a pre-announced course.

These are themes to which I will return when I conclude my remarks.

The impact of the pandemic

But first let me recall how our world has been turned upside down over the past couple of years.

Tragically, the human cost of the pandemic has proved to be substantial.

The economic impact has also been significant. Economic activity in the UK dropped by roughly a fifth in the spring of last year – and 2020 saw the largest annual fall for almost a century.

Macro policy makers sought to support the UK economy in the face of this sharp and unexpected downturn. As part of this, and in pursuit of its statutory objectives, the MPC eased monetary policy significantly. In March 2020, Bank Rate was lowered from 0.75% to 0.1% (then considered its effective lower bound) and large-scale asset purchases were re-started. These measures sought to support market functioning, ease monetary and financial conditions, and support the economy. Subsequently, asset purchases were further increased, as the MPC sought to buy some insurance against the threat that the UK could be caught in a self-sustaining trap of low growth, weak demand and sub-target inflation. That, at least, is my interpretation of UK monetary policy over the eighteen months prior to my appointment at the Bank, when I – alongside you – was observing from outside.

In the face of the pandemic-induced downturn, these initiatives sought to build a bridge across the valley in economic activity to a post-COVID world. By sustaining productive relationships in established supply chains and labour and credit market matches, permanent scarring of employment and activity following from the sharp downturn could be minimised.

I recognise that we are still to re-attain pre-pandemic levels of activity and employment in the UK. And risks to public health remain.

Nonetheless, the bridge provided by supportive macro policies to the post-pandemic world now has secure foundations. We have made substantial progress in traversing it. With the progressive re-opening of the economy, momentum in private demand continues to build. Savings accumulated during lockdown are available to be spent. As uncertainty recedes and demand recovers, investment intentions are strengthening.

In short, the post-pandemic recovery is now well-established (and even starting to mature).

These are successes. But they beg the question of whether the measures taken by the MPC at the peak of the COVID crisis in mid-2020 are still necessary. Or, to express it another way, whether the strongly supportive setting of monetary policy established in the face of the pandemic remains appropriate, now that the recovery in economic activity has achieved some momentum.

Supply / demand imbalances during the recovery

Before I address that central policy question, I want to describe how the economic recovery – although of course very welcome in itself – has brought with it a new set of challenges for monetary policy.

One element of the MPC’s assessment of UK price developments seeks to identify imbalances between supply and demand that create strains on resources and lead to a build-up of inflationary pressure.

Usually the supply side of that equation is quite predictable. Supply potential is pinned down by the size of the labour force, the available machinery and equipment, and the efficiency of business processes – all things that are likely to evolve reasonably slowly. Our analysis can then focus on demand, taking supply as largely given, at least over the horizons relevant for monetary policy.

But this has changed over the past twenty months. Pandemic lockdowns, furlough schemes and fiscal measures have created large and uncoordinated shifts in both aggregate demand and aggregate supply.

The gap between demand and supply that is relevant for inflation prospects – and thus monetary policy – has therefore proved volatile and unpredictable. The difference between two large and volatile quantities is inevitably fickle, and therefore difficult to measure or forecast: assessing imbalances between supply and demand has got much trickier.

What’s more, beyond the volatility of the aggregates, the composition of both demand and supply has also varied substantially.

On the demand side, consumption patterns initially shifted from consumer-facing services to durable household goods. If lockdown meant that cinema visits were no longer feasible, families might watch films at home on a new large screen TV.

Now that the lockdown restrictions have been receding, there is tentative evidence that demand patterns are starting to revert. But whether we ever return to the ‘old-normal’ in terms of consumer spending patterns remains an open question.

On the supply side, global supply chains have proved vulnerable to factory shutdowns and transportation dislocations caused by pandemic restrictions. Shortages of workers with the relevant skills and difficulties in sourcing crucial materials and intermediate inputs weigh on output, leading to longer delivery times. The seminal example provided by the global lack of HGV drivers is instructive here. And the complexity of the supply network implies that shortages can ripple through the economy in unanticipated ways. After all, one firm’s longer delivery time is another firm’s sourcing problem.

In sum, the evolution of both aggregate demand and supply, as well as changes in the composition of each, have created unexpectedly strong inflationary pressures in recent months. Both globally and here in the UK, supply bottlenecks and resource pressures have emerged.

From the MPC’s perspective, the key question is whether these inflationary pressures will dissipate of their own accord, or whether they will trigger wage, cost and price dynamics that would prove more persistent unless monetary policy actively contains them. In other words – and to use the jargon that has become fashionable among central bankers – whether current inflationary pressures prove ‘transitory’. The answer to this question is central to whether Bank Rate needs to rise.

Transitory or persistent? – Four considerations

Perhaps understandably, the word ‘transitory’ has become somewhat notorious of late, as central banks’ characterisation of recent inflation developments has been challenged by repeated overshoots of their inflation projections.

Rather than define ‘transitory’ here, I will outline four key questions that frame how I have made – and will make – the crucial assessment of whether the current high level of inflation will prove transitory or persistent: (1) Can monetary policy offset the immediate impact of shocks to inflation? (2) To the extent that is possible, should monetary policy do so or simply wait for the impact of those shocks to dissipate of their own accord? (3) Will inflation prove persistent even as the direct impact of such shocks recedes? And (4) does the stance of monetary policy itself contribute to the build-up of inflationary pressure, independent of the other shocks influencing price developments?

First, the ‘can?’ question. Monetary policy – the tool available to the MPC to meet its inflation target – famously influence inflation only with ‘long and variable lags’. To the extent that the unpredictable supply / demand imbalances I have described affect price developments more quickly than a monetary policy response to those imbalances can do so, it is simply impossible for monetary policy to offset their impact on inflation. By the time our tools gain traction, the inflationary impact of such shocks might have been and gone.

As a result, there is always going to be some unavoidable short-term volatility in inflation around the MPC’s 2% target. And if those unanticipated supply / demand imbalances are big (or there are other large jumps in inflation, such as those stemming from international energy price shocks), the scale of this volatility can be large. Monetary policy always needs to adopt a medium-term orientation, focusing on horizons at which it can exert an influence on price developments and thereby offset the impact of underlying disturbances on inflation.

In this respect, we in the MPC simply have to live with some short-term volatility of inflation – unfortunately, there isn’t really much we can do about it. What we can and will deliver is inflation anchored at its 2% target over the medium term – as the UK experience within an inflation targeting regime since 1997 demonstrates.

Second, let me address the ‘should?’ question. The emergence of a temporary bottleneck in the UK economy – such as a shortage of the HGV drivers needed to deliver in-demand household goods – is also likely to build domestic inflationary pressure. Were monetary policy to tighten to contain that pressure, demand throughout the economy would be squeezed, introducing volatility into aggregate economic activity. Such output volatility is itself likely to be costly. A trade-off emerges between, on the one hand, reducing the volatility of inflation around its target and, on the other hand, reducing the volatility of economic activity.

This trade-off is foreseen in the MPC’s remit, which provides scope for inflation to be returned to its target gradually, so as to avoid costly volatility in output. Indeed, managing this trade-off in the face of temporary supply disruptions is the essence of the flexible inflation target strategy pursued by the MPC. Moving back to target more slowly than might be possible can be justified if it avoids other forms of economic disruption – and does not interfere with the lasting achievement of the inflation target over the medium term.

Third, even if the initial trigger of inflation is temporary, above-target inflation itself might prove persistent if rising prices and costs feed off one another in a self-sustaining cycle. These are the pernicious ‘second round effects’ that have long worried central bankers.

After all, one firm’s price is another firm’s cost. Provided that firms enjoy some independent pricing power, feedback from one’s cost to another’s price and back again can sustain inflation, even after the initial shock recedes. Analysis of such price / cost interactions naturally extends to wages and the labour market, which has long been the object of such discussion in the UK given its inflationary history in the 1970s. This all remains in the MPC’s focus.

However, the MPC’s recently published forecast foresees relatively little persistence in either the stronger-than-expected wage growth we have seen of late, or the high CPI inflation we are expecting over the next few months. This view relies on our assessment of developments over the past twenty-five years, which exhibit little evidence of second round effects in wage and price setting or self-sustaining ‘cost/price spirals’ that could lead to more persistent above-target inflation, even as demand and supply conditions normalise.

Of course, this view is surrounded by risks. Might the coming years differ from the previous twenty-five? You don’t have to embrace an alarmist view that 1970s style stagflation is around the corner to imagine that wage, cost and price developments could evolve in a less benign manner. Identifying changes in firm’s pricing power, in the evolution of costs and, in particular, in the behaviour of labour markets after the end of the furlough scheme remains at the centre of the Bank’s inflation analysis.

Behind these concerns lies the threat of a de-anchoring of inflation expectations from the MPC’s 2% target. But, unlike in the 1970s, we now have a strong institutional framework that protects against such self-sustaining cost/price dynamics. The Bank and MPC have been given independent powers to pursue that target, and, through parliament, are accountable to the public for our performance in this regard.

Drawing on evidence from a variety of survey and market measures, we continue to believe that medium to longer-term inflation expectations are anchored at levels consistent with our target. But we cannot be complacent in this respect. Were such de-anchoring to threaten, a strong and prompt monetary policy response would be required.

Fourth, it is possible that inflation persistence may result from an inaccurate calibration of the monetary policy stance. However competent and well-intentioned, policy makers are neither infallible nor omniscient. Were the strongly supportive setting of monetary policy established in the face of the pandemic to be maintained for too long, as policy makers sought further reassurance that the recovery in economic activity is securely established, then eventually above-target inflation would emerge over the medium term.

In this case, the fault lies with the stance of monetary policy itself. The appropriate policy response is to re-calibrate the policy stance correctly. Monetary accommodation would need to be withdrawn to reflect the changed circumstances.

Assessment as I joined the Bank in September

As I joined the MPC almost three months ago, I weighed up the prospects for UK inflation and monetary policy against the background of these four questions. No single consideration offered a satisfactory diagnosis of the outlook. All four played some role.

In large part, I agreed with the Bank’s published analysis that argued that the bulk of the above-target inflation anticipated from September would prove transitory. I believed that the uncomfortably high inflation rates we will face in the coming months mainly reflected temporary factors, such as the pandemic-driven supply bottlenecks I discussed earlier, which are likely to ease as the health situation improves.

My starting point was that monetary policy should largely ‘look through’ these effects.

Lags in policy transmission – which are typically estimated to range from 12 to 24 months – imply that policy tightening now would prove largely futile in containing the prospective inflation spike to 5% foreseen for next spring. And the supply-side character of many of the disruptions currently driving higher inflation, if temporary, motivate a somewhat more gradual approach to bringing inflation back to target.

Yet, at the same time, I questioned whether the stance of monetary policy established at the onset of the pandemic remained appropriate, now that the recovery in economic activity was increasingly well-established and inflation was set to rise well above target. Even though I believed longer-term inflation expectations remained anchored at levels consistent with the MPC’s 2% target, neither the historically low level of Bank Rate nor the historically high size of the Bank of England’s balance sheet seemed sustainable in the face of emerging two-sided risks to the economic outlook.

On that basis, as I joined the Bank in early September, I saw a low hurdle to withdrawing some monetary policy accommodation during the following months.

Given this assessment, it is natural to ask: if that was your assessment in September, why have you not been voting to raise rates over the past few months?

My response has three elements.

First, to emphasise, I do not see an immediate threat of UK inflation de-anchoring from its 2% target at the policy-relevant medium-term horizon. Given the still incomplete and potentially fragile recovery from the pandemic, the monetary policy adjustment required at this stage involves taking the foot of the accelerator rather than slamming hard on the brake.

Having being forced by ‘emergency’ circumstances (arising from the global financial crisis and Covid pandemic) to implement aggressive easing measures on a number of occasions in recent years, we need to be wary of acting equally aggressively in the opposite direction when circumstances do not justify.

Such ‘over-steering’ with monetary policy runs the risk of adding to economic volatility rather than containing it. In my view, what is required is a well-prepared, purposeful and measured shift in the stance of policy, reflecting the changing circumstances we face.

Second (and building on my previous remark), shifting to a stance more reflective of the emerging two-sided risks to the outlook required careful preparation after several years of aggressive policy accommodation.

On joining the Bank, I inherited a form of forward guidance on interest rates that ruled out hikes until a number of conditions had been met. Assessing these conditions required the exercise of judgement. The Committee was split on whether these economic thresholds had been reached, and thus on whether circumstances permitted a change in policy stance consistent with the standing guidance.

I admit that I am not an enthusiast for such guidance at the best of times: on my reading of experience across a variety of central banks, even if it starts well, communication of this form tends to end in confusion. But whatever its merits in general, by the time I had joined the Bank in September the MPC’s specific formulation was starting to look ill-suited to a situation where two-sided risks to the economic outlook were becoming more evident. I welcomed the explicit retirement of that guidance at the MPC’s September meeting.

Another inheritance was the ongoing quantitative easing (QE) programme, expanded in November 2020, and scheduled to end in December this year. At its August meeting – and thus before I became a member of the committee – the MPC made a significant investment in setting out a plan for the withdrawal of monetary policy accommodation, outlining how ending asset purchases, changing Bank Rate and starting to shrink the Bank’s balance sheet would be sequenced and managed.

Policy discussion in the preceding months had entertained the possibility that QE be halted early as a signal of vigilance to rising inflationary risk. By September – and thus in the specific circumstances created by the MPC’s August announcements – I felt that an early cessation of asset purchases would be more likely to confuse rather than clarify our stance. I favoured waiting for the QE programme to reach its natural end in December, as a result of which Bank Rate would be clearly established as the active instrument of monetary policy going forward.

With that in mind, I welcome the prospect of QE coming to an end next month. It has served its purpose in supporting the economy during the Covid crisis. It remains part of our policy toolkit standing ready to be re-activated, should conditions change. But the articulation of our stance and rationale for our policy decisions will be simplified and clarified by returning to a focus on Bank Rate.

Third, my assessment of the inflation outlook rests on a cumulative assessment of the incoming data, seeking to separate the policy-relevant signal embodied in lasting, underlying trends apparent across a wide set of indicators from the noise evident in the month-to-month gyrations of any single indicator. In the absence of a dramatic shock (such as financial crisis), building that cumulative assessment takes time. But I felt – and continue to believe – that this is an investment well made.

In particular, in line with the MPC’s collective communication in September, I recognised that uncertainty surrounded how the end of the furlough scheme would affect labour market developments. A number of puzzles were evident: for example, how could the labour market tightness evident in high levels of vacancies be reconciled with large numbers of jobs remaining on furlough just ahead of the scheme’s end? There was value in waiting for this uncertainty to be resolved – or at least reduced – before deciding whether to take the first steps in withdrawing monetary policy accommodation.

Recent news and its interpretation – The current conjuncture

But if preparing the ground was necessary and further evidence of sustained recovery were needed in September, where do we stand now, a few months on?

The available indicators, surveys and Agency intelligence give us a strong steer that the end of the furlough scheme has not led (and will not lead) to a substantial spike in unemployment. On the contrary, most recent indicators point to a further tightening of the labour market. On this basis, the end of furlough is, of itself, unlikely to dent the high level of vacancies, ease bottlenecks or weigh on the elevated level of underlying wage growth that characterise the labour market at present.

At the same time, in the MPC’s latest projections, inflation is set to approach 5%pa next spring. That is substantially higher than foreseen earlier this year. The bulk of the upside surprise in CPI inflation stems from the unexpected rise of energy prices in recent months, in particular wholesale gas prices. Given the idiosyncracies of the UK gas market, both the amplitude and the persistence of the resulting hump in CPI inflation have increased as a result.

Yet, even if uncertainty around their future evolution is substantial, I don’t see reasons to expect continued rises in energy prices, at least not at anything near their recent pace. As a result – and as long as any second-round effects are contained by the credible anchoring of medium-term inflation expectations at target – the impact of energy prices on inflation will prove temporary.

Nonetheless, at 4.2%pa in October, UK CPI inflation is well above its 2% target. This is clearly uncomfortable for a monetary policy maker with the objective of price stability.

To sum up, in my view, the ground has now been prepared for policy action. The QE programme will reach its natural end next month, clearing the way to a simpler presentation of policy decisions. In my view, incoming data support the conclusion that the recovery is continuing. Although supply disruptions weigh on activity, they also create inflationary pressures. The labour market is tight.

Taken together, these developments were sufficient for me to support the MPC’s November steer that, should the incoming data continue to be consistent with the projections published in the committee’s latest Monetary Policy Report, it will be necessary over coming months to increase Bank Rate for the inflation target is to be achieved in a sustainable manner.

In other words, given where we stand in terms of data and analysis, I view the likely direction of travel for monetary policy from here as pretty clear.

After all, the constant Bank Rate scenario presented in the November Monetary Policy Report – in other words, the MPC’s best collective judgment of where we would end up if there were to be no change in Bank Rate – leaves inflation above target throughout the forecast window. And, crucially, forecast inflation remains appreciably above target at the policy-relevant horizon, when lags in monetary policy transmission unwind.

If we ‘do nothing’ with Bank Rate, inflation will end up ‘too high’ relative to our target. That is why the MPC collectively felt able on this occasion to offer a steer about policy prospects in the coming months, something it has generally refrained from doing.

Nonetheless, reflecting the need to retain flexibility that I mentioned earlier, the MPC’s steer remains a conditional statement. We will see what the future brings. And anyway, I cannot make promises on what the MPC will decide: I speak only for myself. The final decision on policy rates will be taken by a committee of nine people, each bringing their own independent views.

That said, for me at least, the burden of proof has now clearly shifted. While in September I was still seeking data to confirm my assessment of the strength of the post-pandemic recovery and accumulation of inflationary pressures, now I scan incoming information for challenges to that view.

Beyond ‘the coming months’

Finally, let me turn to the likelihood, pace and magnitude of policy action after any initial withdrawal of accommodation. In other words, the outlook for monetary policy beyond the coming months.

My main message here is that we should be cautious in offering guidance on the path of Bank Rate at these longer horizons. Better to focus our analysis and communication on an assessment of the economic situation and its implications for inflation as it unfolds, as reflected in the MPC’s forecasts published each quarter.

However much we flag that any guidance on Bank Rate is conditional on how circumstances evolve, the danger exists that guidance will be interpreted as a commitment, and the necessary policy flexibility will be compromised.

If you allow me a short digression into economic jargon, we need to be wary of embracing concepts such as the ‘neutral real interest rate’ or R-star in our communications. These are labels for the level of Bank Rate at which monetary policy neither adds to nor subtracts from inflationary pressure and the momentum of economic activity.

In principle, such measures are appealing. They claim to offer an anchor and signpost for guidance about interest rates down the road. But, in practice, R-star is conceptually and empirically elusive. Any specific point estimate is too uncertain, variable and fragile to offer much of a framework to steer policy design and communication.

Beyond the coming months, the MPC will need to assess the impact of each policy decision it takes on a step-by-step basis, and then evaluate if and when further action is needed. If data evolve unfavourably and inflation is forecast to fall below target at the policy-relevant horizon, we can remain on hold (or reverse course). If data strengthen and inflation is forecast to remain persistently above target, we can continue to raise rates.

Rather than signalling where we expect Bank Rate to be at longer horizons, this framework focuses attention on our analysis of economic conditions and the inflation outlook; in particular, on the MPC’s assessment of how the balance of two-sided risks around our forecast evolves over time. In doing so, it emphasises the data-dependency inherent in each ‘finely balanced’ decision the committee takes.

Deng Xiaoping famously characterised China’s economic liberalisation as an attempt to ‘cross the river by feeling the stones’. That metaphor feels appropriate to describe the coming challenges for UK monetary policy.

We know where we are headed: towards the 2% inflation target.

We know that the passage there be challenging, with strong currents and hidden undertows all threatening our progress.

But rather than identifying a point mid-river that we forge towards with no regard for the fast-moving eddies encountered along the way, it is better to adopt a step-by-step approach, where we ensure our feet are securely anchored on the slippery river bed, before inching further forward.

Reacting to the ebbs and flows of the current. Showing caution rather than bravado. Learning by doing. These are the secrets to safely reaching the other side.

That is the task that lies ahead for UK monetary policy.

I owe many thanks to Jack Meaning, Rich Harrison, Martin Seneca, Olga Maizels and Ed Millar for helpful discussions in the preparation of this speech. I would also like to thank Andrew Bailey, Jamie Bell, Ben Broadbent, Alan Castle, Jonathan Haskel, Andrew Hauser, Catherine L Mann, Nick McLaren, Dave Ramsden, Michael Saunders, Fergal Shortall and Silvana Tenreyro and for their comments. Opinions (and all remaining errors and omissions) are my own.