The outlook for inflation and monetary policy - speech by Michael Saunders

Speech

In this speech, I want to discuss recent developments and prospects for the economy and monetary policy. I want to make five main points

• At the November monetary policy meeting, the MPC (myself included) judged that, provided the incoming data, particularly on the labour market, are broadly in line with the 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.
• Relative to that MPR forecast, my view is that (conditioned on the market rate path and assumptions on the pandemic and public health measures used for that forecast) risks are on the side of a more persistent period of excess demand and above-target inflation, reflecting greater domestic cost and capacity pressures.
• Given this, at the November MPC meeting I voted to tighten monetary policy by curtailing the asset purchase program and raising Bank Rate to 0.25%. If the economy develops as I expect, then some additional tightening, on top of such a move, probably will be needed fairly soon.
• But policy is not on auto pilot. The pace, and scale, of any monetary policy changes will depend on economic developments and the outlook. In particular, at the December meeting, a key consideration for me will be the possible economic effects of the new Omicron Covid variant, and the potential costs and benefits of waiting to see more data on this before – if necessary – adjusting policy.
• It is likely that any rise in Bank Rate will be limited given that the neutral level of interest rates remains low. Provided we do not delay too long, it should be a case of easing off the accelerator rather than applying the brakes.

The economy’s recent trends

Let’s start with the recent developments in the economy in terms of activity, the labour market, inflation and potential output.

At end-2019, before the pandemic, the economy was in reasonable balance, with inflation close to the 2% target, unemployment just below 4% and underlying pay growth at 3%-3½% YoY. Since then, the economy has been affected by two big developments. The first, of course, is the pandemic and the measures designed to tackle it. The other is Brexit (which occurred at end-January 2020) and the UK-EU Trade and Cooperation Agreement (TCA, which was agreed in late 2020 and provisionally applied from 1 January 2021), which have resulted in a marked reduction in the UK economy’s openness, notwithstanding trade agreements reached with non-EU countries.

The initial effects on the economy of the pandemic were far greater than those of Brexit. In Q1 this year, both GDP and total hours worked were 8-9% below their pre-pandemic levels. Moreover, the pandemic initially caused demand to fall more than supply. Hence, together with the demand-driven drop in global oil and energy prices (and temporary tax effects), CPI inflation fell well below the 2% target last year and early this year. With ample spare capacity, inflation well below target and tighter financial conditions (as a perceived rise in credit risk led to wider credit spreads) the MPC loosened monetary policy markedly during last year.^{footnote [1]}

As Covid-related restrictions eased, activity recovered substantially. The level of GDP in September was still slightly below the pre-pandemic level (ie Q4-19) although, allowing for the further growth in the economy since then, the gap now is probably smaller.^{footnote [2]}

Figure 1. UK – Percentage Changes in GDP, Employment and Hours Worked Since Q4-19

Footnotes

• Sources: ONS and BoE.

At the same time, even with the recent rise in gilt yields, financial conditions are considerably looser than late last year and early this year. In particular, bank lending spreads have fallen back to around pre-pandemic levels, reflecting the reduction in the perceived credit risk of borrowers, the decline in banks’ wholesale funding spreads, and surplus liquidity particularly in the ring-fenced banks.^{footnote [3]}

The recovery in activity continues to be uneven across several dimensions.^{footnote [4]}

Goods versus services. Among major advanced economies, aggregate spending on goods recovered much faster than spending on services late last year and early this year. More recently, there has been some rotation of spending back to services – which grew strongly in Q3 for the advanced economies as a whole – and spending on goods fell in the last quarter. Even so, aggregate spending on goods across advanced economies remains well above pre-pandemic levels, while spending on services is still below.

Figure 2. Advanced Economies – Level of spending on goods and services in pandemic and 2008/09 recession as per cent of pre-crisis level, solid line = pandemic, dotted line = 2008/09 recession

Footnotes

• Note: In the pandemic, data are shown relative to the average level in Q4-19. For the 2008/09 recession, they are shown relative to Q4-2007. AE average is weighed average of data for US, EU, Japan, UK and Canada. Sources: Eikon from Refinitiv and BoE.

Less global. The UK economy has become less globalised, with effects from the pandemic exacerbated by Brexit. For example, compared to the 2019 average, trade flows (imports plus exports as a share of GDP) in the UK have fallen by far more than in any other G7 country. Indeed in Q3 this year, the UK’s trade flows (as a share of GDP) were the lowest since 2009. By contrast, among the EU countries, trade as a share of GDP has largely recovered to pre-pandemic levels. Moreover, UK firms report greater frictions in the process of importing and exporting.^{footnote [5]} The UK labour market also has become less global: there has been a marked drop in the numbers of EU nationals working in the UK, and firms report greater difficulties in hiring EU staff to work in the UK.

Figure 3. UK and Main EU Countries – Trade Flows (Imports Plus Exports) as per cent of GDP

Footnotes

• Note: Main EU Countries are Germany, France, Italy and Spain. Data measured in real terms. Figures for 2021 are the average for the year to date. Sources: ONS, Eikon from Refinitiv and BoE.

Public versus private. Within the UK, there has been a marked rise in public sector employment, reflecting in part the expansion of pandemic-related health spending.^{footnote [6]} As a result, total employment has recovered significantly more than private sector employment.

Across the economy as a whole, it appears that demand has recovered more than supply, such that – even though GDP remains below its pre-pandemic level – the expansion has been limited by widespread capacity constraints. For example, CBI surveys suggest that the share of firms reporting output is constrained by shortages of skilled or professional labour is relatively high in both manufacturing and services. Likewise, high shares of manufacturers also report shortages of plant capacity and materials. Conversely, the share of firms in services and manufacturing reporting activity is constrained by a lack of demand or orders is the lowest for decades.

Consistent with this, the labour market has tightened further and indeed now appears stretched by historic norms. Unemployment and under-employment have fallen back close to pre-pandemic lows, and short-term unemployment is at a record low.^{footnote [7]} The ratio of vacancies to unemployment, and the BoE Agents’ survey measure of firms’ recruitment difficulties, are both at a record high. There are signs that the tightening in the labour market is being reflected in higher pay, especially in new hires (see for example, the REC/KPMG survey). Underlying AWE growth (adjusted for composition effects and flows on/off furlough) across the economy has risen to around 4½% YoY, clearly above the pre-pandemic pace. The recent rate of underlying pay growth is estimated to have been above the MPC’s expectations and higher than implied by models of labour market fundamentals. It also is probably a little above the pace consistent with the inflation target, assuming hourly productivity growth remains close to the pre-pandemic trend of around 1% YoY.

Figure 4. UK – Measures of Labour Market Tightness

Footnotes

• Note: The blue line is the average of various surveys of labour market tightness, derived from business surveys by the CBI, BCC, REC, and BoE Agents. Each survey is measured as standard deviations from its 2000-21 average. Sources: CBI, BCC, REC, ONS and BoE.

Figure 5. UK – REC Survey of Pay Growth and YoY Change in Average Weekly Earnings

Footnotes

• Note: REC series have been mean-variance adjusted to AWE private sector regular pay over 2000-2019. Underlying AWE private sector regular pay excludes furlough and compositional effects. Sources: REC/KPMG Report on Jobs, ONS and BoE.

Against this backdrop, inflation data have remained strong in recent months. The QoQ annualised pace of core CPI inflation has picked up from below 2% in Q1 this year and just below 3½% in Q2 to around 4½% in the latest data, and is around the highest pace of the last 20 years. The QoQ annualised rate of headline inflation is slightly higher, around 5%.

As was the case a few months ago, the pickup in headline CPI inflation partly reflects higher energy prices, especially household gas prices. The rise in wholesale gas prices over recent months will, unless it quickly reverses, feed through to another sizeable rise in the Ofgem price cap in April, lifting YoY CPI inflation further. The rise in gas prices has little effect, of course, on core CPI inflation.

Inflation (both headline and core) also has been lifted by global capacity pressures in manufactured goods, caused by buoyant global spending on goods (reflecting the rotation of demand towards goods during last year and early this year, discussed earlier) and some disruptions to global supply chains. These effects have continued to feed through to the UK through strong gains in prices of non-energy consumer goods (especially consumer durables) in the CPI.

Figure 6. UK – QoQ Annualised Rates of CPI Inflation

Footnotes

• Note: Data are seasonally adjusted. Latest data are for the three months ended October. Sources: ONS and BoE.

But the rise in inflation has broadened in recent months. As the labour market has tightened, there has been a further rise in domestic cost and price pressures. For example, a range of measures suggest that core CPI services inflation – which is less affected than goods prices by global cost trends – has risen above its 2018-19 average. The same applies to service sector producer price inflation and the BoE Agents’ scores of service sector inflation. At the same time, guides to longer term inflation expectations (especially those for households and financial markets) have drifted up.

Figure 7. UK – Measures of YoY Services Inflation

Footnotes

• Sources: ONS and BoE.

At first glance, it may seem surprising that capacity pressures in the UK have worsened to this extent when, as noted, GDP is probably not above its pre-pandemic level.

It is of course possible that recent GDP data will be revised up, as often happens.

But the broader point is that since Q4-19, potential output has been significantly reduced by the twin impacts of Brexit and the pandemic (including the measures designed to tackle the pandemic).

Some of this has come through lower productivity. The MPC has judged that Brexit is likely to reduce the level of potential GDP by about 3¼% over time, largely through adverse effects on productivity from reduced investment and trade openness.^{footnote [8]} Much of this loss is likely to be concentrated in the early period after Brexit (ie since January 2020), as firms adjust to the new trading relations. The pandemic has probably also caused some loss of productivity through lower business investment and erosion of workplace skills, perhaps offset partly by gains from more widespread working from home.^{footnote [9]} In all, GDP per hour worked in Q3 this year was just 0.1% up from Q4-19, undershooting even the modest trend in the preceding years.^{footnote [10]}

In addition, the workforce has been reduced by an outflow of EU nationals and lower workforce participation.^{footnote [11]} The participation rate among the 16-64 age group has fallen from 79.5% in Q4-19 to 78.8% in Q3 this year, with notable declines among people aged 18-24 years and the over 50s.^{footnote [12]} This decline in participation has cut 270k (0.8%) off the workforce over that period, and the shortfall is even greater compared to the rising pre-pandemic trend in participation. At the same time, LFS data suggest the number of EU nationals in the workforce has fallen by 8% (200k) since Q4-19, cutting 0.6% off the UK workforce.^{footnote [13]} In all, the workforce in Q3 this year was 250,000 (0.7%) below the Q4-19 level, and 750,000 (2.2%) below an extrapolation from Q4-19 of the previous rising trend.^{footnote [14]}

Figure 8. UK – Workforce (Millions of People)

Footnotes

• Sources: ONS and BoE.

Moreover, mismatch may have lifted the medium-term equilibrium unemployment rate (ie the NAIRU). As noted, the composition of demand has changed during the pandemic, in terms of occupation, skills and geography. The expansion of public sector employment has added to competition for workers for other sectors. The changes in labour supply discussed above also have been uneven.^{footnote [15]} As a result, there may be greater mismatch between supply and demand for labour, including shortages of critical skills. With convex Phillips curves, the scale of inflation pressures in sectors with excess demand are likely to exceed the scale of disinflationary pressures in sectors where there is still spare capacity.^{footnote [16]} The November MPR was consistent with a small reduction in potential output from these effects in recent quarters (through a higher NAIRU).

On top of this, just over 1 million jobs were furloughed on average in Q3. Allowing for sectoral differences in average hours worked and productivity levels, the effects of furlough probably reduced potential GDP in Q3 by about 1%: this effect unwound in Q4 with the end of furlough.^{footnote [17]}

In the November MPR, the MPC judged that potential output in Q3 this year was slightly more than 2% below the Q4-19 level, and about 5% below a simple extrapolation that starts with the 2019Q4 level and assumes a 1.5% annual growth trend (the average annual growth in potential GDP in the three years pre-Covid). It is not really possible to definitively separate some of the effects of Covid and Brexit on potential output, and some of the shortfall (eg lower population growth due to outflow of EU workers, weakness in business investment) could in part reflect both factors. For Q4-21, with potential GDP lifted by the end of furlough, the MPR estimate was that potential GDP was about 1¾% below the Q4-19 level (and about 4½% below the trend described above). In the MPR forecast, most of the shortfall of potential output versus its previous trend was expected to persist. The November MPR forecast implied that potential GDP at the end of the three-year forecast period would be about 3½% below the 1.5% annual growth trend described above, and about 2% below the level implied by the MPC’s pre-pandemic projections.

Even with uncertainties around these potential output estimates, two key points follow. First, given the sizeable persistent loss of potential GDP caused by Covid and Brexit, it probably is not feasible to expect the economy to fully recover to the pre-Covid trend. The shortfall of GDP from an extrapolation of the pre-Covid trend does not all represent spare capacity: much of that shortfall probably reflects a persistent loss in potential output. Second, given that the UK economy has also been significantly affected by Brexit, the loss of potential output since 2019 in the UK may well have been relatively large compared to other major economies.

The November MPR Outlook

In the November MPR, the MPC judged that there was a small margin of excess demand in Q4, of roughly ¼% of potential GDP. In that forecast, demand will be underpinned by a decline in household and corporate saving rates and, in the first year or two of the forecast, by support from fiscal and monetary policies.^{footnote [18]} But the economy will also face significant headwinds. The MPR projects that, in the near term, activity will continue to be restrained by supply bottlenecks. Moreover, household disposable incomes will be squeezed by higher inflation in the near term, and a marked slowdown in AWE growth (from 3½% YoY in 2021Q4 to 1¼% in 2022Q4). This expected slowdown in AWE growth partly reflects the unwinding of composition effects (including those from furlough), but also reflects lower underlying pay growth. Later in the 3-year forecast period, the rising market path for rates and tighter fiscal policy restrain demand further.

The MPR projects that potential output will be lifted by the end of the furlough scheme in Q4, and also by reduced labour market mismatch in coming quarters. Participation regains most of the lost ground in the next couple of quarters.^{footnote [19]} As a result, in the MPR forecast, supply and demand will be roughly in balance two years ahead and there is a modest margin of excess supply three years ahead. In the MPR forecast, the jobless rate rises marginally in Q4 this year, by 0.1pp, as furlough ends, falls to around 4% during next year, but then drifts up close to 4½% three years ahead as demand growth cools later in the forecast.

The MPR forecast is that CPI inflation will rise to about 5% in Q2 next year as household energy prices increase further. Nevertheless, in the MPR forecast, global goods price inflation slows markedly from this quarter onwards, as supply bottlenecks ease and demand rotates further away from goods and back to services. Later in the forecast, rising unemployment keeps pay growth subdued (with AWE growth of 2¼% two years ahead, 2¾% three years ahead). Hence, the MPR projection is that, as effects from higher energy prices and gains in global costs fade, CPI inflation will fall slightly below the 2% target three years ahead. In contrast, the alternative MPR forecast with Bank Rate unchanged at 0.1% shows CPI inflation staying above the 2% target two and three years ahead.

That forecast assumed that there will be no further new Covid-related severe restrictions or lockdowns in the UK, although the pandemic is assumed to continue to weigh more substantially on economic activity in some other countries with lower vaccination rates. I viewed those as reasonable assumptions at the time, and still do at the moment. But inevitably they are uncertain, especially in light of the new Omicron Covid variant.

Risks around the outlook for growth

There are two-sided risks to the MPR forecast for GDP growth from developments in private sector balance sheets, but (conditioned on the assumptions above) I think the balance of these risks is more to the upside than downside.

A major upside risk is from the possibility of a bigger rundown of the excess savings accumulated by households during the pandemic, which has now reached £250bn (17% of annual consumer spending) and has continued to increase in recent quarters. The November MPR assumes that 10% of these savings will be run down gradually over the next three years, supporting demand. But there is considerable uncertainty around this. I think it is likely that over the three-year forecast period, risks are probably skewed to the upside (ie bigger rundown of savings and stronger consumer demand), especially if tail risks relating to renewed lockdowns and/or a significant rise in unemployment recede, reducing reasons for households to hold a high level of precautionary savings.^{footnote [20]}

Figure 9. UK – Change in SME Bank Deposits and Debts since End-2019 (£bn)

Footnotes

• Note: The distinction between ‘smaller businesses’ and ‘medium-sized businesses’ is as defined in the UK Finance data. The distinction is not exact and is generally based on business turnover (with smaller-sized SMEs being up to £1m/£2m and medium-sized being up to £25m). Sources: UK Finance and BoE.

On the other side, there is a possible downside risk from the increase in corporate debts, especially among SMEs, during the pandemic. Aggregate bank debt among UK SMEs has risen by 20% since end-2019, with increases of more than 40% for hotels and restaurants, as well as recreational and personal services. However, while some individual companies may have become more fragile, in aggregate UK corporate balance sheets have strengthened during the pandemic, because the corporate sector has run a large financial surplus and accumulated liquidity. For example, in aggregate, UK companies’ bank deposits have risen by £145bn (34%) since end-2019, well above the rise in their sterling bank debt (£34bn) and total net finance raised by companies (£77bn). This point – deposits have risen by more than debts – also holds in aggregate for SMEs (bank debts have risen by £36bn since end-2019, bank deposits by £70bn), as well as for most industry sectors (including among SMEs).^{footnote [21]} Moreover, much of the rise in SME debt reflects the BBL scheme, for which the repayment terms are relatively generous, with low and fixed interest rates (and scope to defer payments if needed).^{footnote [22]} Hence, while the risk of heightened corporate vulnerabilities needs watching, currently I would not rate it as highly as the upside risks from accumulated household savings. Indeed, at present, firms’ intentions for hiring and investment remain extremely strong.

Risks around the outlook for potential output

In terms of potential output, I suspect that risks are to the downside of the MPR forecast, chiefly because I doubt that workforce participation (and hence labour supply) will recover as much as the MPR projects. In other words, I think it is likely that much of the drop in participation during the pandemic will persist even as the labour market tightens.

As noted, the drop in participation has been most pronounced among people aged 18-24 years and the over 50s. The LFS data suggest that the main reasons for the drop in workforce participation (among people aged 16-64 years) are long-term sickness, early retirement, and because more people (especially those aged 18-24 years) are in full-time education. These effects have outweighed the drop in the numbers who are outside the workforce because of caring responsibilities (possibly reflecting upward effects on participation from more widespread working from home).

There is little sign that lower participation reflects a lack of job opportunities or a “discouraged worker” effect. For example, the number of job vacancies is at a record high, and the LFS data show a marked drop – to a record low – in the numbers of people (aged 16-64) who are outside the workforce and say they want a job.^{footnote [23]} Conversely, the number of people (aged 16-64 years) that are outside the workforce and do not want a job has risen by 410,000 (1.0% of the 16-64 population) since Q4-19.

Among those who are aged 16-64, outside the workforce and do not want a job, there has been a 70k rise since 2019 Q4 in the numbers who are retired. But the main factor is a sharp rise in the numbers who report they are outside the workforce and do not want a job because of long-term sickness: this category has risen by 250k (from 1.52 million people in 2019Q4 to 1.77 million people in Q3 this year), and in absolute terms is the highest since data began in 1993. As a share of the 16-64 population, the numbers of people who are outside the workforce due to long-term sickness and do not want a job is the highest since 2005 (having fallen from 2005-14 and risen slightly during 2017-19).

Figure 10. UK – Change since Q4-19 in Number of People (Aged 16-64 Years) That Are Not in the Workforce, Split by Reason and Whether They Would Like a Job

Footnotes

• Sources: ONS and BoE.

Figure 11. UK – Per cent of Adult Population (Aged 16-64) Who Are Outside the Workforce Because of Long-term Sickness and Do Not Want to Work

Sources: ONS and BoE.

The downward effects on participation from long-term sickness and retirement are most pronounced among people aged 55-69 years, although there have also been effects among younger age groups.^{footnote [24]}

I suspect that, especially among the over 50s, the recent jump in the numbers reporting they are outside the workforce because of long-term sickness reflects health risks around Covid – either worries about the possible effects of catching Covid or people who are suffering effects from Long Covid. ONS data suggest that more than 1% of the population report their activity is limited by Long Covid, with higher figures (above 1½% of the population) among age groups 35-49 and 50-69 years. Knowledge of Long Covid is inevitably incomplete at this stage. The ONS survey suggests that some people recover over time, but the outflow rate from Long Covid appears to have been relatively low.^{footnote [25]} Hence, it is possible that many of those currently reporting adverse effects from Long Covid will still be affected two or three years from now (and the total with Long Covid will probably be topped up by new inflows of people).

Figure 12. UK – Per cent of People Reporting Their Activity is Limited By Long Covid

Footnotes

• Note: The chart shows the per cent of each age group that report their activity is limited “a little” or “a lot” by effects from Long Covid. Source: ONS.

It also is possible that some people in the 50-64 age group have chosen to retire because increases in asset prices over the last decade make retirement more affordable and the pandemic has led them to re-evaluate their work-leisure preferences.^{footnote [26]}

Among the 18-24 age group, the drop in participation reflects the jump in the numbers in full-time education (FTE) – a group that typically have relatively low workforce participation because they are studying full time.^{footnote [27]} The rise in FTE numbers may reflect the lack of job opportunities during the lockdowns, and a sense that the pandemic has highlighted the greater long-term rewards (in terms of job opportunities) from higher education.^{footnote [28]} With higher education (ie university level or equivalent) typically taking at least a year and more usually 2-3 years, participation among the 18-24 age group is likely to remain below pre-Covid levels for some time.

Overall, therefore, I expect that the pandemic will have a bigger and more persistent adverse effect on workforce participation than in the November MPR forecast, hence implying a lower path for potential output.

Risks around the outlook for inflation

Turning to the inflation outlook, one key uncertainty concerns energy prices. In line with the MPC’s recent practice, the central forecast in the November MPR assumed that global energy prices follow the path implied by futures markets (averaged over the 15 days prior to the forecast cutoff date) for the next six months and are stable thereafter. However, futures markets imply gas prices will fall further beyond six months ahead. An alternative scenario, which assumes that global energy prices follow the path implied by futures markets for the full 3-year forecast period, would imply a significantly lower inflation path two and three years ahead (with higher real GDP growth).^{footnote [29]} My view is that either assumption would be reasonable, and so (in line with the MPR) this implies a downside risk to the inflation forecast two and three years ahead (with upside risks for real wages and real GDP growth).

Other than that, my view is that (conditioned on the pandemic assumptions described earlier) there is a series of risks that point to more persistent cost and capacity pressures and hence a greater and more persistent overshoot of the 2% inflation target.

First, there may be greater near term pass through from recent strength in non-energy costs. The MPR projects that the QoQ annualised pace of core CPI inflation will moderate from 4½% in Q3 to 3¾% in Q4 and below 3% on average in Q1 and Q2 next year. However, most indicators for near term inflation remain strong. Figure 13 shows a pyramid of price and cost indicators, ranked in terms of their correlation with near term swings of a measure of core CPI inflation. While the inflation data can be volatile from month to month, there is little sign that the near term pace of price increases is set to slow in my view.

Figure 13. UK – Pyramid of Price and Cost Indicators

Footnotes

• Note: The series are detailed in Saunders (2021b).

Second, world trade prices may again overshoot the MPR forecast, further lifting cost pressures through the prices of tradable goods. The MPR assumes that, after strong recent gains, world export prices will rise only slightly in Q4 and then fall slightly over the next three years as demand rotates further away from goods and supply bottlenecks ease. It may be reasonable to expect these supply/demand imbalances to ease eventually. However, I see little sign that they will ease as soon as the MPR implies. So far, spending on goods remains well above pre-Covid levels, manufacturers’ delivery times continue to rise rapidly, and manufacturers’ price expectations are very strong both in the UK and elsewhere.^{footnote [30]} More generally, stated government policy measures may support investment spending on goods (eg investment to limit climate change, UK Corporation Tax super-deduction, US infrastructure investment plans). Indeed, to the extent that people and firms recently have been unable to buy consumer and investment goods because of supply bottlenecks, there may be pent-up demand that will underpin spending on goods in the period ahead. Given lags, continued gains in global export prices would add to UK inflation over the next year or two.

Third, in my view there are upside risks to the MPR forecast for pay growth throughout the forecast period.

As noted, the MPR projects that underlying pay growth (adjusted for composition effects and flows on/off furlough) will slow in 2022, despite falling unemployment. This reflects the view that underlying pay growth has recently overshot the path driven by labour market fundamentals, and hence will now undershoot that path even with falling unemployment. I am sceptical of this argument. In my view, a neutral assumption would be to lock in the overshoot in the level of pay that has occurred, and project that pay growth will match the path implied by labour market fundamentals going forward. To be sure, the labour share in GDP has risen slightly compared to the relatively low levels in the years just before the pandemic. But it is now similar to the average over 2000-19 and not unusually high. Moreover, the current policy stance is highly accommodative, such that cost increases are being passed on to selling prices rather than resulting in a margin squeeze that might prompt firms to cut back. I doubt that that underlying pay growth will slow in the coming year unless labour market slack rises significantly. That currently seems unlikely, given strength in vacancies and surveys of firms’ hiring intentions, unless there is some other significant downside shock to demand.

Figure 14. UK – Labour Share in GDP

Footnotes

• Sources: ONS and BoE.

Indeed, rather than a slowdown in underlying average earnings, it seems more likely to me that pay deals will pick up in the coming year, because the labour market is tight and seems likely to be tighter than the MPR forecast – with a lower path for unemployment over the forecast period.

In the near term, the jobless rate in Q4 may well undershoot the MPR forecast, with a smoother exit from the furlough scheme than expected. At the time of the MPR, there were some early indications (eg ONS BICS redundancy expectations, HR1 redundancy notifications, BoE Agents’ survey) that most furloughed workers had returned to work at their existing firm. There was also some evidence (eg rising total for online vacancies, surveys of recruitment difficulties) that the labour market tightened further in October, after the end of furlough. The central forecast in the MPR is for a modest rise in unemployment in Q4 (from 4.4% in Q3 to 4.5% in Q4), but with upside risks to the unemployment projection in the first year and a half of the forecast. Since the MPR was published, ONS data show that unemployment in Q3 fell to 4.3%, undershooting the MPR forecast by 0.1pp. Moreover, from that lower starting point, various indicators suggest that any boost to the jobless total from the end of furlough was probably less than expected in the MPR forecast.^{footnote [31]} Further ahead, the downside risks to participation discussed earlier, and upside risks to demand, would make it less likely that labour supply will rise enough to create rising unemployment over the forecast period, and more likely the economy will remain in excess demand.^{footnote [32]}

A higher path for nominal pay growth would probably tend to be reflected in some mix of higher real wage growth and higher inflation than the MPR forecast.

A fourth upside inflation risk is that Brexit, by reducing the economy’s openness, might steepen the UK wage and price Phillips curves, such that pay growth and inflation react more strongly to the current period of excess demand than previously. This possibility is still uncertain at this stage. But, it is likely that the increased openness of the UK and other economies flattened Phillips curves over recent decades.^{footnote [33]} Now, with less scope for imports of goods or workers to ease specific capacity pressures, it seems more likely that the Phillips curve will steepen than flatten further. A steeper Phillips curve might, of course, imply downside risks to inflation at the end of the forecast period if the MPR forecast that spare capacity emerges at that horizon three years ahead is correct. But if, as I expect is more likely to be the case, excess demand persists, then a steeper Phillips curve would increase inflation throughout the forecast period.

A fifth upside risk for inflation is that if these various external and domestic cost pressures coincide, then inflation expectations could drift up further, potentially further reinforcing domestic cost and price pressures.

To sum up this section, my own views are that, conditioned on the yield curve used for the November MPR and the MPR assumptions on the pandemic, risks are on the side of a persistent period of excess demand and above-target inflation two and three years ahead. This reflects a mix of a lower recovery in potential output, resilience in demand, higher pay growth and a continued effect from global costs.

At the same time, it is within the range of possibilities that the new Omicron Covid variant will significantly affect the economic outlook. Apart from the direct impact of any additional public health measures on economic activity, aggregate demand and supply could also both be affected by increased precautionary behaviour as well as effects on supply chains and the composition of spending.

Implications for monetary policy

In setting monetary policy, our focus is usually mainly on the inflation outlook roughly two years ahead, because it typically takes more than a year for changes in monetary policy to have their main impact upon economic activity and thus on prices. There is little monetary policy can do to address factors that have only transitory effects on inflation – affecting the near term outlook, but not the outlook for inflation at the policy-relevant horizon.

Some of the current and likely near term inflation overshoot falls squarely in this transitory category, especially effects from energy prices. Inflation pressures from non-energy global costs carry more significance, because – given the lags with which these pass into the CPI – they can affect UK inflation at the policy-relevant horizon. But, I put most weight on trends in domestic cost and capacity pressures, especially in the labour market and services inflation, because these tend to have more persistent effects on inflation.

As discussed, my view at recent meetings has been that, with the current policy stance, risks were on the side of a persistent inflation overshoot, including through the tight labour market and higher services inflation.

Last year, with ample slack, tighter financial conditions and inflation well below our 2% target, the MPC loosened monetary policy aggressively, in order to support activity and reduce risks of persistent below-target inflation. I fully supported those actions. Indeed, my view was that, with the economic risks we faced then, it was better to err on the side of providing too much stimulus rather than too little. In my view, that previous willingness to loosen aggressively when needed has always been mirrored by a willingness to tighten again if (as now) there are widespread capacity pressures and inflation is well above target.

Against this backdrop, at recent MPC meetings I have voted for tighter monetary policy in order to return inflation to target on a sustained basis. At the August and September meetings, I voted to curtail the current asset purchase program. At the November meeting, my view was that the need for some withdrawal of stimulus had strengthened, given the more widespread evidence of domestic capacity and price pressures, especially in the labour market. At that meeting, I again voted to curtail the asset purchase program and also voted for a hike in Bank Rate to 0.25%.

The current asset purchase program will be completed by the next MPC meeting in December. At that stage, consistent with MPC guidance, initial options for tightening monetary policy will focus on a higher Bank Rate rather than a reduction in the stock of purchased assets.^{footnote [34]}

I do not think it is sensible to use code words to signal my likely policy vote at any particular meeting. Nor am I a fan of giving precise forecasts for the future path of Bank Rate several quarters ahead. I expect that such forecasts would give the undesirable impression that policy is on autopilot, whereas in practice – as you would expect – the path of policy will depend on the economy and the outlook and hence is inevitably uncertain.

That said, if the economy develops in line with the MPR central forecast or my expectations, the direction of travel for Bank Rate during the next few quarters is clearly likely to be upwards. I would characterise such an interest rate outlook as a withdrawal of monetary policy stimulus in the context of a low neutral rate rather than a move to a tight policy stance.^{footnote [35]} Of course, if easing is required, the MPC has options to support the economy if needed – but this is not my central expectation.

In considering if and when to adjust rates, there is always a case to wait and see more data. At present, given the new Omicron Covid variant has only been detected quite recently, there could be particular advantages in waiting to see more evidence on its possible effects on public health outcomes and hence on the economy. But continued delay also could be costly. If the economy continues along its recent path, then maintaining the current highly accommodative policy stance would probably allow the labour market to tighten further and, with inflation well above target, reinforce risks of a further rise in long-term inflation expectations. This could require a more abrupt and painful policy tightening later. For me, the balance between these considerations is likely to be a key factor at the December meeting.

Monetary policy can do little either way to resolve the supply side issues facing the UK. I am not convinced by the view that it would be possible to lift labour supply (or significantly lift overall potential output) by aiming to “run the economy hot”. It seems unlikely that the drop in workforce participation reflects a lack of job opportunities, given that job vacancies are at a record high, and the number of people who are outside the workforce and would like a job is at a record low. A policy stance that generates excess demand would, in my view, be likely to result in a persistent inflation overshoot and rising inflation expectations rather than cause a matching rise in potential output.

Whichever way the economy develops, the MPC will, as always, remain focussed on ensuring that inflation returns sustainably to the 2% target, in a way that supports output and jobs. The UK’s framework of an independent central bank with a clear inflation target and remit, and effective monetary policy tools will ensure that the UK does not face a persistent inflation problem.

The views expressed here are not necessarily those of the Bank of England or the Monetary Policy Committee. I would particularly like to thank Matt Swannell and Katie Taylor for their help in preparing this speech. I have received helpful comments from Andrew Bailey, Harvey Daniell, Michael Goldby, Jonathan Haskel, Enrico Longoni, Nick Mclaren, Ryan Murphy, Huw Pill, Fergal Shortall, Silvana Tenreyro, Rupert de Vincent-Humphreys, Carleton Webb, and Chris Yeates, for which I am most grateful.

References

Bean, C (2006), 'Globalisation and Inflation', speech given at LSE Economics Society, London School of Economics, London.

Brewer, M and McCurdy, C (2021), 'Post Furlough Blues - What happened to furloughed workers after the end of the Job Retention Scheme?', Resolution Foundation report, Resolution Foundation.

Brewer, M, McCurdy, C and Slaughter, H (2021), 'Begin again? Assessing the permanent implications of Covid-19 for the UK’s labour market', Resolution Foundation report, Resolution Foundation.

Bunn, P, Chadha, J, Lazarowicz, T, Millard, S, and Rockall, E (2021), 'Household debt and labour supply', Bank of England Staff Working Paper, Bank of England, No. 941.

Carney, M (2017), '[De]Globalisation and Inflation', speech given at the 2017 IMF Michel Camdessus Central Banking Lecture, Washington.

Coile, C, and Levine, P (2010), 'Reconsidering Retirement: How Losses and Layoffs Affect Older Workers’, NBER.

Coile, C and Levine, P (2011), ‘Market Crash and Mass Layoffs: How the Current Economic Crisis May Affect Retirement’, The B.E. Journal of Economic Analysis & Policy, vol 11(1), pages 1-42.

Crossley, T, F, Fisher, P, Levell, P, and Low, H (2021), ‘MPCs in an Economic Crisis: Spending, Saving and Private Transfers’, IFS working paper, Institute for Fiscal Studies, No W21/26.

Saunders, M (2021a), 'Supply and Demand During and After the Pandemic', speech given at online webinar.

Saunders, M (2021b), 'The Inflation Outlook', speech given at online webinar.

UNCTAD (2021), 'Review of Maritime Transport 2021', United Nations Conference on Trade and Development.

Understanding Society (2020-21), ‘Covid-19 study – Waves 1-8’, The UK Household Longitudinal Study, Understanding Society.