Navigating the economy through the Covid crisis

## Navigating the economy through the Covid crisis

Speech given by

Dave Ramsden, Deputy Governor for Markets & Banking

The Strand Group, King's Business School 14 July 2021

With thanks to Tom Smith and Michael Yoganayagam for their assistance in preparing these remarks, and to numerous colleagues, including Saba Alam, Andrew Bailey, Nick Bate, Max English, Greg Kidd, Andrew Hauser, Maggie Illingworth, Sarah Illingworth, Sean Maloney, Alex Rattan, Andrea Rosen, Michael Saunders, Fergal Shortall and Silvana Tenreyro, for their many helpful contributions.

## Introduction

It's very good to be back speaking at this 51 st Strand Group event and in my role as a Visiting Professor at King's. This is the third Strand Group event I've spoken at but my first time speaking to you virtually.

I've been involved in forecasting throughout my career, both in my current role as a Monetary Policy Committee (MPC) member and in my previous civil service career at HM Treasury. 1 Outside of work I'm also a keen mountain walker - most recently being lucky enough to spend some time on the Isle of Skye - and it's hard not to see parallels between the two activities. Forecasts act as a map for policymakers of where the economy is heading, showing what adjustments to the current path are needed to avoid impending hazards and ensure they safely reach their policy objectives. 2

I should be clear at the outset, to borrow a well-worn quotation, that 'the map is not the territory'. 3 The forecast summarises what is likely to happen on the journey, but it is the economy itself that we as monetary policy makers must navigate, using the forecast alongside many other inputs, including a wide range of data and intelligence from the Bank's network of agents. And as can sometimes happen when walking in the mountains, different MPC members can interpret the territory in different ways, supplementing the use of a map with the equivalent of a compass and GPS as well as their own judgement. 4

The job of economic forecasting has become increasingly tricky over recent years, as the economic territory underfoot has become less and less familiar. The 2007-09 financial crisis was characterised by very large impacts on the supply side of the economy, as well as complex financial sector interactions and new forms of monetary stimulus in response. That was followed the decade after by Brexit, which combined acute political and economic uncertainty with a multidimensional and shifting set of actual and anticipated impacts on demand, supply and the exchange rate. And most recently we have had the Covid crisis, a one-in-ahundred-year health and economic crisis, which has left monetary policy makers firmly in unknown territory.

Of course the MPC have not been the only ones to find ourselves in this situation. Other policy makers have faced similar challenges, and have found highly innovative and effective ways to overcome them, such as the Treasury's furlough scheme. And of course the intellectual challenges faced by economic policy makers have been tiny compared with the very substantial challenges faced by the healthcare and other key workers who have had to tackle the Covid pandemic itself, as well as those who have suffered from the disease.

1 My previous Strand Group lecture, 'The First 50 years of the Government Economic Service', touched on some of that earlier experience.

2 My MPC colleague Silvana Tenreyro made a similar observation in her 2018 speech on models in macroeconomics.

3 Or to give the quote from Korzybski (1933) more fully: 'A map is not the territory it represents, but, if correct, it has a similar structure to the territory, which accounts for its usefulness'.

4 It's not for nothing that the Bank of England's central forecasting model is known as COMPASS.

Figure 1: The MPC's approach to its forecasts has evolved during the pandemic

## An evolving approach to the outlook

The MPC, supported by the efforts of Bank of England staff, has continued to make use of forecasts throughout the Covid period. But our approach to those forecasts has evolved as we have gradually oriented ourselves in the new economic territory we have found ourselves in.

In our first Monetary Policy Report (MPR) following the outbreak of the pandemic, published in May 2020 (shown in orange), the impact of Covid was still so uncertain that we felt unable to publish a single central forecast. Instead what we published was a 'plausible illustrative economic scenario', together with estimates of the sensitivity of the economy to some key variables. The scenario and accompanying estimates were nothing like our usual detailed map of the outlook. But they served as an initial outline of what a future map might look like.

As time passed, and as Figure 1 illustrates, we began to fill in the picture. By August 2020 (shown in grey) we felt able to publish a forecast once more; but it was a forecast that came with a higher-than-usual degree of uncertainty and a larger-than-usual downside skew, as captured in our published fan charts. By November (yellow) some of those downside risks had crystallised, in the form of a new rise in Covid cases and an associated increase in restrictions. And while by February 2021 (dark blue) the vaccination programme had begun, it was too early to be certain about how successful it would be, and the risks remained to the downside. The map had been gradually filled in, revealing a more complex and bumpy economic landscape than had been depicted in our initial outlines, perhaps most apparent in the evolution of our inflation forecast.

The stance of monetary policy during the Covid period has been guided by that evolving economic outlook. In the initial phase of the crisis the outlook was very uncertain, but the direction for policy was clear: the priority was to support economic activity and employment, both with the goal of returning inflation sustainably to the MPC's 2% target, and to provide a bridge through the Covid crisis and minimise the potential for longer-term damage or 'scarring' to the economy's potential. To that end the MPC deployed its full range of policy tools in March 2020, cutting Bank Rate to 10 basis points, launching a new Term Funding Scheme with additional incentives for SME lending, and announcing additional very substantial programmes of asset purchases, first in March, and again in June and November 2020.

As the economic outlook became somewhat clearer, the focus of monetary policy also evolved. While our ultimate goal remained to return inflation to target by bridging through the Covid crisis, our priority shifted to ensuring that the economy remained on track and that we were not blown off course as further risks crystallised. We made this explicit in November 2020 when announcing our most recent £150bn programme of QE purchases, which we consciously decided to continue for an unusually long period through to 'around the end of 2021': we said in our Minutes that 'risk management considerations implied that policy should lean strongly against downside risks to the outlook.' That purchase programme is still under way, with over £90bn of purchases completed as of today.

For the same reason we gave forward guidance in November that we 'did not intend to tighten monetary policy at least until there was clear evidence that significant progress was being made in eliminating spare capacity and achieving the 2% inflation target sustainably.' And we highlighted the asymmetry in our assessment of the risks to activity and inflation by noting that 'if the outlook for inflation weakened, the Committee stood ready to take whatever additional action was necessary to achieve its remit'.

Figure 2: The MPC's May MPR forecasts

Our latest forecast, published in our May 2021 MPR and shown in Figure 2 , represented a further evolution of our view of the outlook. The economic news since our February 2021 Report had been substantially positive. Activity had been stronger than expected and Covid restrictions had been eased slightly earlier than expected, partly due to the success of the vaccination programme in the UK. Reflecting that news, as well as the extension of the furlough scheme, GDP was projected to be higher, and unemployment lower, over the entire forecast period. Inflation was also expected to rise a little higher than in February, peaking at around 2½% in 2021Q4 before falling back towards the target.

The MPC also revised its view of the risks to its forecast. We reduced the uncertainty around our GDP projection, retaining a downside skew to GDP only in the first year of the forecast period. And, importantly, we updated our guidance to make it more symmetric: rather than saying we would take additional actions if the economy weakened, the MPC said we would 'take whatever action was necessary to achieve its remit'. That improving but still risky outlook for the economy, mapped out by our forecast, is what underpinned the MPC's decision to leave Bank Rate and the target stock of QE unchanged in May.

What I want to do in the rest of my remarks today is set out what I have learned since we produced that May forecast and draw out what I think that means for our map of the economy. In particular I want to explain why in our June Minutes the MPC said we now see 'two-sided risks' around the central path for the economy, and set out how I think those risks might evolve, and how the contours of our map might change, in the coming months.

## Developments since May

The MPC's next Monetary Policy Report is not published for another three weeks, and we have not yet begun to discuss what our updated August forecast might look like. What I am going to discuss today will be the MPC's initial assessment, as set out in the minutes of our June meeting, of what we had learned about the economic outlook since our May MPR, as well as my own interpretation of the very latest data which have come out over the last three weeks since the June MPC meeting.

Global growth appears to have been somewhat stronger than we expected in May, largely due to upside news in the euro area. Activity in the US so far looks to have been broadly in line with expectations; but CPI inflation in the US has now reached 5.4% - its highest level since 2008, and 0.5pp above market expectations at the time of release.

Despite that, global yield curves, led by US Treasuries, have first risen and then fallen back in recent weeks; the ten year gilt yield is now 10bps lower than at the time of our May forecast, while ten year US Treasury yields are 20bps lower. Bank staff models suggest US factors have been the dominant driver of the movements in UK yields since May. Equities however have remained robust with some indices reaching

record highs last week, a trend which the Bank's Financial Policy Committee drew attention to in its July Financial Stability Report.

In the UK, the GDP releases for March and April revealed notably stronger growth than we had expected in May, with monthly growth rates of 2.1% and 2.3%. At the time of our June meeting Bank staff had revised up their expectations for 2021 Q2 GDP growth to 5½% on the previous quarter, from 4¼% at the time of the May Report. That would be consistent with output in Q2 being less than 4% below its 2019 Q4 level. The monthly figure of 0.8% growth for May, released last Friday, contained some offsetting downside news. Within that however consumer-facing services continued to grow, with the drag coming from sectors like cars and construction where demand was constrained by supply shortages - a theme I will come back to.

The UK labour market has also outperformed our May expectations, with unemployment falling to 4.7% and employment growing 0.3% in the three months to April. That continues a trend of downside surprises to our unemployment forecasts over the past year, even after allowing for the extensions to the furlough schemes. The falls in unemployment have however been accompanied by higher inactivity. Other labour market indicators have also been strong, with a 200,000 increase in HMRC payroll numbers in May and an increase in job vacancies to above pre-pandemic levels.

Most strikingly, private sector regular pay rose by 5.6% in the April data - though it is important to note that this is boosted by compositional effects, and by a base effect from spring last year. Aiming off for these transitory factors 'underlying' pay growth appears close to pre-pandemic levels. As Andrew Bailey set out in his recent speech, the compositional and base effect in the data meant that even if the level of average earnings was flat for the rest of the year, the growth rate would peak at nearly 8%. Putting all that together I expect to see another high reading in the May earnings numbers when they are published tomorrow. The latest REC survey for June reported labour demand growth at record highs, staff availability declining at a record rate, and pay growth rising fast. Against that backdrop it is not surprising that the labour market is a recurring theme of the intelligence on economic conditions that we receive from our agents and from our (virtual) regional visits.

Finally there has also been significant news in inflation. UK CPI inflation rose to 2.1% in May, and then to 2.5% in this morning's June data release. That represented material upside news to our May short-term projections, and as a result also represents a faster move to above-target inflation than we had been expecting. Other costs have also continued to rise, reflecting an underlying theme of supply constraints and bottlenecks: the oil price has now reached $76 per barrel, non-oil commodity prices have continued to rise, and the sustained rise in shipping costs that we have seen since 2020 has also continued. And those cost pressures appear to be passing through to manufactured goods. That said, some of the most extreme price pressures, for instance in the lumber futures market, do now appear to be easing.

Table 1: Estimates of Q2 economic performance

|------------------------------------|----------------|----------------|-----------------------|

| GDP growth (% YoY) | 20¼ | 21½ | 22¾ |

| CPI inflation (% YoY) | 1.7 | 1.7 | 2.1 |

| Private sector regular AWE (% YoY) | 4¾ | 7¼ | 8 |

When thinking about how these latest developments might affect the economic outlook, it's important to keep things in perspective. The projected GDP growth rates shown in Table 1 compared with a year ago show just how unusual this period is economically, with growth of 22¾% expected in 2021Q2 relative to the trough in activity a year ago . And relative to that, the forecast revisions between the November 2020 MPR and the staff's latest 'nowcast' for GDP growth to 2021Q2, while noteworthy, are relatively small.

Nevertheless in the near term the signal is clearly to the upside. It is now conceivable that the pre-Covid peak in output in 2019Q4 will be restored in the course of the current quarter, and quite possible that unemployment could peak lower than our May MPR forecast. And inflation now looks likely to peak well above 3% and maybe nearer to 4% - the biggest news to our earlier inflation projections is yet to come. Reflecting all this I am somewhat less confident than I was that there remains a margin of spare capacity in the economy.

Looking further ahead there are three key considerations which are shaping my assessment of the outlook and the risks around it:

1. Are developments temporary or more persistent?

2. Are they economy-wide or sectoral?

3. What they mean for the balance of supply and demand in the economy?

As set out in the MPC's June minutes, the Committee's expectation was that 'the direct impact of commodity prices on CPI inflation would be transitory'. Their central expectation was for 'a temporary period of strong GDP growth and above-target CPI inflation, after which growth and inflation would fall back'. I agree with that initial assessment - though it's important to stress here that 'transitory' does not mean 'here today, gone tomorrow'. I learnt recently for instance that some businesses are bringing forwards their shipping arrangements for Christmas stock. That will put further pressure on global capacity, extending an increase in shipping costs that started in late 2020 and had already exceeded my expectations, meaning that the current increase in shipping costs could have an impact on consumer prices at least until the end of the year.

I also agree with the Committee's assessment in June that there are now 'two-sided risks' around that central path, for both GDP and inflation. Indeed for me that updated assessment is one of the most significant implications of developments since May, when the MPC still judged the risks to GDP to be skewed

to the downside in the near term. For that reason I want to spend some time setting out what I see as the key risks and how they might evolve - a survey, if you like, of what new features future editions of our map of the economy might need to incorporate.

## Two sided risks

I should say right away that the downside risks to GDP and inflation from Covid remain clear and present. Although the UK's vaccination programme has proved successful, the pandemic in the UK and the wider world is not yet over. It remains to be seen what the full impact of the new Delta variant will be and how households and businesses respond to the combination of the current rise in infection rates and the impending lifting of remaining restrictions.

Starting with my first consideration, whether recent strength proves persistent or temporary. Even absent a further impact from Covid, it is unlikely that the current pace of GDP growth can be sustained for much longer. Indeed if current strength represents a bringing-forward of the recovery that we had expected later in the year, then subsequent growth might if anything be expected to be weaker rather than stronger. The very latest indicators suggest the pace of underlying demand growth may indeed be slightly moderating, though as I mentioned previously the read from the latest GDP data itself is clouded by supply constraints.

But there are certainly upside risks to household spending, if consumer confidence continues to rise or if households run down their accumulated aggregate savings faster than we have assumed. The current strength in the housing market is suggestive in this regard, both as a potential indicator of household confidence and due to the consumption of things like durable goods that goes alongside housing transactions. One of the early legacies of the pandemic appears to be an increase in the number of people taking advantage of the ability to work more remotely and moving house for lifestyle reasons, a trend which I think could be sustained for some time; like many of the long-term effects of Covid, that is likely to have differential impacts within regions and in different parts of the country.

I gave a speech last November setting out by views on the longer-term effects of Covid more generally, on what we buy, what we make and how we work. I don't have time today to repeat everything I said then; and indeed with only eight months having elapsed, there is a limit to how much more we have learned on the subject. That part of the map remains sketchy at best. But I will mention a few points that are particularly relevant to my subject for today and to the three considerations that I outlined earlier.

The first point is about the sectoral impact of Covid. One of the most striking economic features of the pandemic has been the differential impact between and within sectors. At the peak of the crisis the pandemic induced a material change in consumption patterns. Much of that sectoral shift has unwound as the economy has recovered; but some of it has persisted, and some further shifts have occurred - it is that sectoral imbalance between demand and supply that accounts for many of the bottlenecks and cost increases that I

mentioned earlier, and that explains why some firms are continuing to hold workers on furlough even as others are posting vacancies. How rapidly the balance of sectoral demand and supply adjusts will be one factor that determines how quickly inflation returns to target - my second consideration.

The second point is about potential scarring effects. The economy has proved remarkably resilient to the effects of Covid, and the impact on the supply side has so far been smaller than we might have feared at the aggregate level. That partly reflects, however, the fact that most of the government support schemes for businesses, in particular the furlough scheme, remain in place. As those schemes come to an end it is likely that at least some businesses turn out no longer to be viable, and there is a risk of more substantial impacts. If the effect is unbalanced across sectors then there may also be more substantial labour market hysteresis than assumed in our central May forecast. In addition to any Covid effect, the long run effects of Brexit on the supply side of the economy, in particular on productivity, are largely yet to come through;

One additional supply uncertainty that has emerged has been the question of whether temporarily inactive workers will resume their search for jobs - if inactivity remains at a persistently higher level that would reduce the supply of labour and could contribute to wage pressures. Again it is likely that we will not know the answer to this for some time. But it is the balance of demand and supply that will ultimately determine the medium term inflation impact - my third consideration.

## Inflation expectations

One final risk that I will be monitoring closely is the risk that the expected transitory increase in inflation and wages generates expectations of a more sustained rise in inflation. As of now the near-term strength in UK inflation is expected to be transitory, and surveys of households, businesses and professional forecasters are consistent with UK inflation remaining well anchored.

One measure of inflation expectations that has picked up since the start of 2021 is the five year UK inflation swap rate, five years forward: a key indicator of longer-term financial market inflation expectations and one to which the MPC naturally pays close attention. As Figure 3 shows, this measure is historically relatively stable for the UK. In fact some of the larger historical moves have reflected uncertainty about possible changes to how the underlying RPI index is calculated, rather than fundamental shifts in inflation expectations. And as Figure 4 shows, it fell by much less, and rebounded more quickly, than equivalent measures in other economies.

That history of stability means that while the recent pick-up in the indicator is not especially large in a historical or international context, it is still worth noting. Unlike some of the other risk indicators that I have been discussing, it has also decreased rather than increased since May, in line with the broader falls in nominal yields that I flagged earlier. And while shorter-horizon market inflation expectations did increase in

response to this morning's CPI release, the reaction in the five year five year rate was more muted. Nevertheless it remains about 15 basis points higher than at the start of the year.

Taken at face value, the increase since the start of the year might suggest some pass-through from nearterm to longer-term UK inflation expectations. Indeed interpreted literally it would imply markets expected inflation to be a full 15 basis points higher on average over the five-to-ten year horizon relative to the start of the year. And stepping back from the month-on-month variation, it is noteworthy that the UK measure, unlike those for the US and euro area, stands above rather than below its historic average (shown by the dashed lines).

This indicator is, however, a difficult one to read. UK inflation markets are sensitive in general to non-fundamental factors - although it is less obvious that these have played a role this time. 5 Bank staff also estimate that the increase this year reflects an increase in inflation risk premia, which could in turn reflect a reduction in the risk of disinflation; that would be a more benign development than the alternative of an increase in underlying expectations. And the recent fall-back in rates makes me somewhat less concerned than I might have been before. Nonetheless, the increase in this measure since the start of the year does suggest all else equal there may be some risk of more persistent inflation than the MPC had factored into its central expectation.

Figure 3: UK 5 year five year inflation swap rates have historically been stable…

Figure 4: … including during the Covid pandemic

## Looking ahead

As I've already stressed, the MPC's November forward guidance said that we did not intend to tighten monetary policy at least until there was clear evidence that significant progress was being made in eliminating spare capacity and achieving the 2% inflation target sustainably. That guidance still applies today: we have repeated it in the Minutes of each subsequent meeting, including our most recent one in June.

I have thought since we first introduced it in November that our guidance would become more salient as the economy recovered, and that is being borne out by current developments. Inflation is already overshooting the target, and it is less clear, at least to me, that there remains a degree of spare capacity. But as the MPC's guidance makes clear what matters is whether we expect inflation to return to target sustainably. At the same time, when we launched our ongoing £150bn QE programme in November it was explicitly for risk management reasons. Since then the degree of uncertainty, as measured by the width of our fan charts, has decreased significantly, while the balance of risks has shifted as I have set out to being two-sided.

While the range of potential outcomes is, as I have described, a complex and multidimensional one, and is still beset with considerable uncertainty, I find it helpful to map out three broad scenarios which for me summarise how the risk factors may play out, and map across to what course the economy might take in future.

The first potential scenario I foresee is broadly in line with the MPC's 'central expectation' set out in the June minutes: that the economy will experience a temporary period of strong GDP growth and above-target CPI inflation, after which growth eases and inflation fall back towards the 2% target. Of course as I stressed earlier, 'temporary' doesn't necessarily mean 'here today, gone tomorrow', particularly for inflation. The supply side pressures and their interaction with demand might take some time to disperse. In the mean time we could see very high inflation in certain sectors and products, such that at a whole economy level I wouldn't be surprised to see the whole economy CPI inflation rate potentially rising as high as 4% for a period later this year, before falling back towards the 2% target. That would not be an unprecedented event: we saw similarly high outturns in 2008 and 2011. But the shock this time is very different and perhaps less predictable.

The second potential scenario I foresee is a more persistent inflationary one, in which supply pressures do not ease and demand continues to pick up. The further easing of restrictions could, for instance, act as a further boost to household and business confidence, even as bottlenecks on the supply side persist. That could lead to a more sustained and widespread increase in inflationary pressure, particularly if the temporary inflation overshoot led households and businesses to expect high inflation to persist for longer. A variant on this scenario could see demand weaken in part in response to sustained supply pressures.

My third potential scenario is one in which demand growth does not just moderate but instead weakens more substantially, alongside easing supply pressures. That could happen for instance if household confidence falters in response to rising Covid cases or if further new variants emerge. A combination of weaker demand and easing supply pressures could see inflation falling back faster in the medium term, not just towards but potentially back below our 2% target.

For me although my starting point for navigating the economy is the May MPR forecast overlaid by what has happened since, there is still real uncertainty as to what path the economy will take. That was one reason why I thought the conditions to consider tightening set out in the MPC forward guidance were not yet met at our June meeting. But based on the rapid pace of developments since we published our May forecasts and the shift in the balance of risks I can envisage those conditions for considering tightening being met somewhat sooner than I had previously expected.

That reflects my current assessment that on balance I put more weight on my inflationary than my disinflationary scenario. But there should no presumption that the recent strength in demand and inflation will be sustained, and while I do currently put less weight on my disinflationary scenario, it is too early to rule it out. Along with the rest of the MPC I will continue to watch the data and newsflow closely.

As and when conditions for considering tightening are met there is also the question of how we might tighten policy. The monetary policy toolbox has come a long way since the days when it contained a single tool, the

Bank Rate. 6 As the terrain has got more complicated to navigate, the tools have become broader in range and more complex, including Bank Rate, forward guidance and asset purchases. Indeed as things stand we have an asset purchase programme in progress which is not due to complete until the end of the year. Back in February this year the MPC asked the Bank to take steps, and the PRA to engage with PRA-regulated firms, to ensure negative rates could be implemented in practice if needed; that work is due to complete in August. And the MPC have begun work to reconsider our previous guidance on the appropriate strategy for tightening monetary policy, as we committed to in our February minutes.

Of course I am just one of nine members of the MPC navigating a path for the economy on this journey different MPC members will have their own views about where the economy is heading and what the appropriate path should be. The personnel on the expedition are also changing; at the August MPC meeting we will temporarily be down to eight members and there will be two new members joining later in the year.

Back in March last year the outlook for the economy was in pretty uncharted territory. Over time prospects have become clearer and we have become more able to rely on the map provided by our forecasts. Now that the recovery is becoming more established, and the risks are once again two sided, we are heading back to more familiar, but still challenging, ground. The map is becoming more detailed and the contours for the immediate path ahead are filling in, but at the same time choosing the precise direction of travel is becoming less straightforward.

As always the great strength of the MPC is that while we bring different perspectives on the best way to get there, we share a common goal, consistent with our remit, of returning inflation sustainably to the 2% target. And as always I am looking forward to our debates and discussions as we prepare to publish our latest map of where we are heading in the form of our August forecast.

6 I covered this in detail in my speech last October, 'The Monetary Policy Toolbox in the UK'.