Reflecting on recent times – speech by Andrew Bailey

I want to use my time to look behind the latest state of monetary policy, at issues that have arisen from the experience of recent years, but with some reflections on where we are now.

The pandemic caused a sudden, coincident and precipitous fall in global demand and supply – one of these did not obviously lead the other. Moreover, in March 2020 we faced monetary policy and financial stability issues arising from the same source, namely the pandemic. Taken separately, the responses should be different, with a more exceptional temporary, targeted and typically maximum force intervention better suited to dealing with a financial stability problem. In contrast, a monetary policy response of the sort used in 2020 is typically undertaken over time. But when monetary policy and financial stability issues coincide, the judgement becomes more complicated. By engaging both of the core central bank objectives, the pandemic posed an unusual but not unprecedented challenge.

As economies started to adjust to the consequences of the pandemic there was a substantial increase in global demand for goods rather than services, at a time when the supply of goods remained disrupted and restricted. This was an asymmetric demand shock. Global goods prices rose as a result, akin to a cost-push shock for open economies like the UK. This was the context of the so-called transitory assessment of monetary policy, namely that such shocks should be short-lived in impact because supply chains should recover and inflation expectations should remain anchored in anticipation of that recovery. That’s the theory. The evidence suggests that, taken on its own, the global supply chain shock had run its course by the end of 2022. But a key question at that time was whether and to what extent there would also be catch-up effects in wages and services prices, and over what time period?

In the UK, the labour market did begin to tighten but it was hard to discern at the time by how much. In mid- to late-2021 the UK Government’s furlough scheme – a sensible policy in its own right – was creating uncertainty around the state of the labour market. We can now use tools such as the Bernanke-Blanchard approach to assess the timing and scale of this labour market effect.

Using this approach, and based on work done on the UK case by my colleague Jonathan Haskel, it became evident economy-wide in late 2021 and early 2022, putting further upward pressure on inflation. We started to tighten monetary policy from late 2021.^{footnote [1]}

Then, starting in February 2022, came the impact of the Ukraine War and Russia’s illegal actions. This created large supply shocks, in particular to energy and food prices. It substantially increased the risk of de-anchoring of inflation expectations and created the potential for larger second round effects in the form of more persistent higher inflation. As shock came upon shock, so to speak, the transitory judgement looked less appropriate.

Our response in terms of both the pace and scale of tightening was progressive but measured. Why? Four reasons stand out for me.

First, unlike in March 2020, in 2022 there was not a coincident financial stability shock. It could have happened, for instance in commodity finance markets, but in the event such a shock did not occur.

Second, the scale of the indirect or second round monetary policy effects was unclear at the time. There were several sources of such uncertainty. One was the transmission mechanism of monetary policy. Looking again at the UK case, in terms of the impact of a tightening phase on economic activity and inflation, the channels of transmission have changed over time. This no doubt reflects underlying changes in the economy since the previous tightening phase which was prior to the Global Financial Crisis, most obviously the shift from variable- to fixed-rate mortgages. But it most likely also reflects specific features of the latest tightening phase. One such feature is that it followed an extended period when interest rates were near the lower bound, along with the use of unconventional policy measures, both of which contributed to a relatively fast increase in longer-term market rates when the tightening came. A second feature of this tightening phase has been the size of the shocks we have witnessed and the potential non-linearities and asymmetries that have arisen in the monetary transmission mechanism. When we are dealing with relatively small shocks which stay within the locality around the inflation target where the linear approximations we normally make in our models hold, then these issues do not arise. That was the experience during much of the inflation targeting era.

But when there are big shocks that move outside this locality, such as those of recent years, then non-linearities and asymmetries can emerge. But, at the time, the impact of this was more uncertain.

The third reason for a measured response was that a very fast and large response of interest rates would have risked creating a severe economic shock of its own (a trade-off). I will come back to this point.

Fourth, while in Europe – and here I refer to Europe in a geographical sense - demand was recovering faster than supply, levels of activity were well below actual and potential levels pre-Covid, leaving substantial questions around how much slack there was.

To bring the story up to date. Headline inflation has since fallen sharply as energy and food price shocks in particular have fallen away. Global goods price inflation has also fallen back sharply, with supply chains restoring themselves and with signs of strong disinflation emerging in some key supplier countries most obviously China.

Meanwhile, monetary policy has been leaning heavily against the indirect and second round inflation effects and the persistence thereof. The job of monetary policy – which it has been doing – is to squeeze the persistent element of inflation out of the system in a way that is consistent with returning inflation to its target on a timely and sustained basis.^{footnote [2]} When we think about inflation persistence, I would distinguish between extrinsic persistence (the duration of external shocks) and intrinsic persistence in the sense of capturing the impact of the echo effects of those external shocks owing to domestic responses (second round effects). Extrinsic inflation persistence is a setting where price and wage setting behaviour remains unchanged, but a succession of external inflationary shocks cause inflation to be above targe for a sustained period. Since the lags in the transmission of monetary policy are typically longer than the lags in the transmission of these shocks to inflation, some volatility of inflation is inevitable, and with the type of big shocks we saw, all going in the same direction, the impact will be large. Contrast this with intrinsic inflation persistence where price and wage setting behaviour does change, including in response to the unusual character of the external shocks. Here, the nature of the response of monetary policy needs to be different. The reality is that we experienced both types of persistence, though the scale of the intrinsic persistence has – almost of necessity – been harder to judge and therefore more subject to revision.

I think we are now seeing a revision down in our assessment of that intrinsic persistence, but this is not something we can take for granted. Important questions remain which in many ways set the framework through which we now view monetary policy.

On current evidence, and speaking of the UK, I would say that the persistent element is still with us but it is smaller in magnitude now than we expected a year ago, and considerably smaller than the type of persistence that was seen in the 1970s. However, that is not the end of the story. We still face the question of whether this persistent element is on course to decline to a level consistent with inflation being at target on a sustained basis and what it will take to make that happen. Is the decline of persistence now almost baked in as the shocks to headline inflation unwind, or will it also require a negative output gap to open up, or are we experiencing a more permanent change to price, wage and margin setting which would require monetary policy to remain tighter for longer? This framework is now prominent in our thinking on the MPC.

The first of these cases is the more benign – the persistence is essentially self-correcting with the degree of restriction we have in place today easing off over time. “Self-correcting” is the key phrase here: it depends on the credibility of monetary policy, which depends on the willingness of policymakers to act, and the best evidence to support this mechanism is well-anchored longer-term inflation expectations.

The second case is the intermediate one. Here we would need to maintain restriction for longer and thus open up more of an output gap.

The last case is least benign and would require more restrictive policy than the first two cases. It would suggest that there are structural changes in product and labour markets going on which are causing the supply side of the economy to change as a lasting legacy of the major shocks we have experienced. To be clear, as policymakers we can have all three of these cases in our expectations, with different weights attached.

Tentatively, it appears to me that the economic costs of bringing down persistent inflation – costs in terms of lower output and higher unemployment – could be less than in the past. This is consistent with a process of disinflation which is steady and more in keeping with a soft landing than a recession induced process. For the UK, this is consistent with how we have revised our outlook for growth, and the numbers themselves so far this year.

Inflation expectations appear to be better anchored, which I put down in good part to the presence of independent central banks with clear mandates and nominal anchors, usually in the form of inflation targets. But, crucially, policy does have to respond to ensure credibility is maintained. Moreover, while second round effects may be smaller that does not tell us how much smaller and less persistent they will eventually end up being than in the past. On this basis, at the moment I put more weight on the first case – self-correction - but some smaller weight on each of the other two. These weights can of course change over time.^{footnote [3]}

This is an important issue of the moment for monetary policy. It is directly relevant to the question of how challenging the last mile of returning to our inflation target level on a sustained basis actually will be. The well-known argument goes that an independent central bank operating a transparent low inflation target successfully over time will anchor inflation expectations more consistently. In this world, in the short-run monetary policy can respond to shocks with suitable flexibility, acting consistently but not always identically, to return inflation to target. In the UK case, the evidence suggests this may have worked insofar as we are seeing a lower level of inflation persistence than we expected a year ago. But, we need to be cautious because the job is not completed – we are not yet back to target on a sustained basis. Policy setting will need to remain restrictive for sufficiently long until the risks to inflation remaining sustainably around the 2% target in the medium term have dissipated further. The course will therefore be a steady one.

Before considering challenges going forward, I want to pull out one further point from recent experience. In an effective monetary policy regime we should have flexibility in the short term on how quickly we return inflation to target. Looking back over recent years, I think it is common ground that the Covid and Ukraine shocks could not have been anticipated, and certainly not in sufficient time for any consequent monetary policy actions to have had an effect on monetary conditions. But there are two follow-up questions that I get asked. One starts with the words “with the benefit of hindsight” what would you have done. The second, asks about actual policy choices once the shocks took effect.

This is where so-called trade-offs come into the picture. The language of the Bank of England remit states that “the inflation target holds at all times”, which is essential. But it also states that “the actual inflation rate will on occasion depart from its target as a result of shocks and disturbances”.

It further adds that “attempts to keep inflation at the inflation target in these circumstances may cause undesirable volatility in output”. In 2013 the remit was amended by adding that where shocks are particularly large or the effects of shocks may persist over an extended period (or both) then“the Committee is likely to be faced with more significant trade-offs between the speed with which it aims to bring inflation back to target and the consideration that should be placed on the variability of output”.

This is what we call the trade-off language. We are not alone in having this type of language in our remit. The language applies when the economy is hit by temporary cost or supply shocks. In these circumstances we have to judge the appropriate balance of inflation and output volatility, when judging how quickly to bring inflation back to target. Moreover, such judgement must ensure that, again using the language of our remit, “inflation expectations are firmly anchored in the medium term”.

Applying this framework over the Covid and Ukraine period has not been easy. As policymakers dealing with the here and now, balancing the objectives of applying the inflation target at all times and dealing with trade-off situations requires us to decide whether to look through a transitory shock or respond because it could have
quasi-permanent features. In doing so of course we must distinguish between these two shocks in real time. Good luck with that as they say. It has required judgement on whether the shocks were temporary or not, and whether they were expected to be so or not by participants in the economy. The further complication is that the configuration of shocks – with no gaps between them - effectively meant that the judgements had to be reached on the shocks as a collective more than individually.

The UK was hit by a severe terms of trade shock, raising the rate of inflation in imported goods and services relative to domestic prices, leading to a sharp fall in household real incomes, something that could not be offset by monetary policy. But let us say we had seen all of this coming in sufficient time to act given the lags in the transmission of monetary policy. In that case, the remit trade-off language would have to be taken into consideration. A negative terms of trade shock involved a sharp fall in real incomes and wages.

If monetary policy had acted to offset this shock, it would have required a much larger fall in domestic prices and wages in order to counteract inflation in import prices. It would have required domestic prices and wages to fall in nominal terms. Achieving this effect would have called for a substantial rise in interest rates creating a strong probability of a deep recession and a steep rise in unemployment. Moreover, it would not have led to inflation being at target on a sustained basis – which is our objective. Rather, as energy and food prices fell back, inflation would fall below target.

Put simply, responding in this way to short-lived shocks is not consistent with our remit nor is it the way to ensure the policy framework remains robust. That said, the trade-off element of our remit leaves us with the judgemental challenge of how to assess and respond to the impact of supply shocks on inflation, particularly when they are accompanied by demand pressure.^{footnote [4]} What is the best way to make sure inflation expectations are anchored, and how do we judge in real time whether they are anchored, accepting that short-run expectations will move with headline inflation measures? The judgement therefore remains one of to what extent central banks can look through such shocks, or not, all of which will be state contingent.

So, let me draw out what I see as key challenges for central banks arising from these experiences. Modern monetary policy with its emphasis on independent responsibilities, clear policy objectives and nominal anchors typically in the form of an inflation target is the product of tackling the great inflation of the 1970s. More recent experience defines at least three challenges to that orthodoxy.

First, how to operate monetary policy near to the lower bound of interest rates when the challenge is to raise not lower inflation. Moreover, if the response has to go beyond moving the official short-term interest rate, the policies adopted to counter disinflation will take longer to wind down and have implications that are more spread out and open to contestation.

Second, in a more uncertain and volatile world, how to set policy in an environment where shocks may or may not be short-lived – either individually or collectively - and where they may create conditions where judging trade-offs between inflation and activity play a large part.

Third, how to ensure that we can at all times meet both our monetary policy and financial stability objectives, including when they may be pointing in opposite directions, and in particular that central banks have the necessary operational tools and can use them at all times if required. Recent experiences indicate the larger role of the financial stability objective in modern central banking in contrast to the orthodoxy of the post-1970s era when monetary policy alone came to the fore. This rebalancing in some ways takes us back to the earlier classical gold standard era of central banking. But in modern central banking, the separation principle is much more important than it was in the classical era. By this I mean distinguishing between, and communicating the distinction between, monetary policy and financial stability actions.

Looking at recent experience in the UK, the so-called LDI event was clearly a financial stability not monetary policy one, and the challenge was to structure the intervention and communicate it in a way that reinforced this point. The March 2020 “dash for cash” was more complicated because it arose in the context of both financial stability and monetary policy events (the latter in the sense of the severe and unexpected recession). How to structure and communicate the policy response by central banks in this – hopefully highly unusual - situation deserves more attention. That said, in the normal state of affairs we are moving to a world where our standard operations meet the system’s demand for liquidity in normal times, and not more than that. Against this backdrop, monetary policy can most effectively set the official interest rate. In normal conditions, meeting the demand for liquidity – most obviously through the level of reserves banks choose to hold – should also support financial stability. Beyond that, we must then have a set of effective and usable exceptional tools for financial stability stresses including when they arise in the non-bank part of the financial sector.

The classical gold standard was an era of prolonged monetary stability punctuated by the need to respond rapidly and forcefully to financial stability shocks, and in doing so preserve monetary stability. Walter Bagehot’s critique of the Bank of England in the classical gold standard era was for not responding in a timely and forceful way to financial stability shocks. In the Covid-Ukraine era central banks did respond rapidly and forcefully. Some would still say we did too much. I disagree with that assessment. But I do think that going forwards – and as we experienced positively with the LDI crisis – rediscovering the distinctive roles of monetary and financial stability actions will help us. Financial stability issues themselves create macroeconomic trade-offs which can affect monetary policy.

But these will naturally fit within the framework of monetary policy. This is what we did in March 2020. That said, distinguishing the handling of primary monetary policy and financial stability shocks, and having more targeted tools in our boxes, will I am sure assist with the inevitably hard job of policy making.

To conclude, the challenge for monetary policy of the Covid-Ukraine era could not have been to prevent the inflation happening. To attempt to do that would have been to ignore the trade-off element of our remit. Put more directly, as Ben Bernanke did in his report, it would have led to a depression and thus a repeat of the errors of the 1920s.^{footnote [5]} Rather, the task for central banks was to produce an orderly return of inflation to target, with expectations anchored.

Recent experience leads me to be cautiously optimistic that inflation expectations are better anchored as a result of the regimes we have in place. The second round inflation effects appear to be smaller than we expected. But it is too early to declare victory. Policy does have to react – the regime works because we use it.

Economies - and the UK is a good example here - are in a stronger and more resilient position as a result of the actions taken over recent years. A key point here is not to do with the immediate situation and prospects. It is that the underlying resilience of economies and financial systems should be sufficient to enable sizeable market movements and risk asset price corrections to occur without threatening stability. This is the lesson from Bagehot in the nineteenth century. His diagnosis was that flaws in the response to financial stability events undermined monetary stability, at that time in the form of the classical Gold Standard. Similar diagnoses have been made around the Great Depression and the Global Financial Crisis. This explains why we have done so much to build stronger financial stability buffers over the last fifteen years since the GFC. This is the best way to support resilient and competitive financial systems and strong economies. Market events like those of two weeks ago or so will happen; the test is not whether they happen but whether they trigger wider instability. As central banks we operate within systems that are framed by law and institutions to create and preserve stability and prosperity in the public interest.

All that said, communicating policy is harder in uncertain conditions with large shocks. Communication is complicated where we have to operate to achieve both our monetary policy and financial stability objectives, with the latter typically being a matter of operating in an emergency in such conditions. Likewise, communicating when we decide to accommodate short-run shocks and/or there is a trade-off between inflation and activity involved is essential but difficult. These are all areas that will benefit from further assessment.

Thank you.

I am grateful to Nat Benjamin, Fabrizio Cadamagnani, Jonathan Haskel, Karen Jude, Clare Lombardelli, Katie Martin, Rhys Phillips, Huw Pill, Vicky Saporta, Martin Seneca, Fergal Shortall and Paolo Valenziano for their assistance in preparing this speech.