Navigating an uncertain outlook: the signals from the labour market − speech by Dave Ramsden

Speech

Thank you for inviting me to speak at the Barclays-CEPR Monetary Policy Forum, which gives me the opportunity to set out my latest thinking, in the light of last week’s decision by the Monetary Policy Committee (MPC). The MPC decided by a majority of 6-3 to leave Bank Rate unchanged at 4.25%, and I was one of the three members who preferred to cut Bank Rate by 25bps.

Over the last eight meetings, a year in MPC terms, we have gradually eased policy restrictiveness in the UK, cutting Bank Rate four times, reflecting the progress made in the disinflation process to bring inflation back towards the 2% target sustainably in the medium term.

Chart 1: Bank Rate has been cut four times reflecting progress on disinflation

Bank Rate, CPI and core inflation^(a)

Footnotes

• Sources: ONS and Bank calculations. (a) Core inflation is defined as CPI excluding food, beverages, tobacco and energy. CPI and core inflation data to May 2025, Bank rate to June 2025.

The persistence of domestic inflationary pressures has been our key focus and we have taken a gradual approach to reducing policy restraint to ensure persistent inflationary pressures in domestic wage and price-setting are squeezed out so that inflation returns sustainably to the 2% target. Since the start of this year we have stressed that we are taking a ‘gradual and careful’ approach in light of the increasingly uncertain global and domestic environment, which is leading to two-sided risks to the outlook for inflation.^{footnote [1]}

Today I want to set out how my thinking has evolved, building on a speech I gave in South Africa in February.^{footnote [2]} I’m going to focus on the labour market, since, notwithstanding all the global events, that is where I judge the most material developments have been in recent months.

Although the MPC’s focus on uncertainty has increased recently, uncertainty is not a recent phenomenon. Chart 2 shows that the Bank’s in-house uncertainty index, as measured by the nine indicators which make up the index, has yo-yoed over the last ten years in response to a series of shocks starting with Brexit. It has picked up sharply again in the last year, driven in the most recent episode by trade policy uncertainty.

Chart 2: The UK has been hit by multiple shocks in the last decade

Contributions to UK uncertainty index^(a)

Footnotes

• Sources: Baker et al (2016), Bloomberg Finance L.P., Caldara et al (2019), CBI, Consensus Economics, GfK, ICE Data Services, LSEG, S&P Global, and Bank calculations. (a) The UK uncertainty index is calculated as the first principal component of nine proxies of economic uncertainty, based on data starting in 1985. See Box A of the May 2025 MPR for further detail on the proxies. The final data points are to May 2025.

Uncertainty is typically thought of as widening the probability distribution of future events, making it harder to forecast what will happen next. More recently unpredictability has entered the MPC’s lexicon to refer to events that are unforecastable based on past experience. The MPC has employed the term with regard to the recent shocks to the outlook for the global trading system. There is lots of empirical evidence showing how challenging shocks and the resulting uncertainty (and unpredictability) are for businesses and households.

Increasing uncertainty and unpredictability is a key driver for the Bank for progressing its response to the Bernanke review; enhancing our forecasting infrastructure and toolkit and building out scenarios to illustrate the mechanisms at work and the risks around those is of even greater importance. In particular the extent to which demand or supply weakness is impacting the UK economy as these can have different consequences for the extent and nature of inflationary pressures. Scenarios help to test the robustness of policy decisions, helping us to arrive at policy which will be robust to a range of alternative plausible futures for the economy.

Developments in the UK economy …

Notwithstanding the prevailing uncertainty and the unpredictability of events, the main aggregates in the MPC’s baseline forecast for GDP, inflation, unemployment and spare capacity have not changed much so far this year, as shown in Table 1.

Table 1: Baseline forecast summary ^(a)

The main features of the outlook are common to both our February and May MPR baseline forecasts. Underlying growth in GDP is forecast to continue to be weak, in part because of restrictive monetary policy. GDP growth from 2025 Q2 onwards is projected to be much weaker than the headline numbers for Q1 where there is clear evidence of front-loading of activity in response to the threat of tariffs.

The MPC is forecasting a hump in inflation for the remainder of this year which is in part due to past energy price rises and a number of one-off increases in administered prices. Because of weak growth in demand – relative to estimated supply growth – a margin of spare capacity (excess supply or a negative output gap) opens up over the forecast supporting the disinflationary process. A gradual fall in headline inflation is projected through 2026 with inflation returning to the 2% target in early 2027. Unemployment rises to around 5% over the forecast period.

We will publish updated baseline forecasts in August, which will also have to take account of the rapidly moving geopolitical developments in the Middle East.

…with a focus on the labour market

Given the relative stability in the baseline forecast what is it about the outlook that has influenced my thinking? The short answer is the cumulative evidence of a continued material loosening in the labour market. As a result while I still think the risks to inflation continue to be two-sided I’m now attaching more weight to the downside risks in the medium term.

Assessing the state of the labour market has been challenging in recent years. This is in part due to issues with the official Labour Force Survey (LFS) series but also because some of the apparent trends, for example in the decline in participation and, more recently, the strength in the official earnings series have been very difficult to explain based on past experience. As a consequence, we have increasingly examined the trends in a wide range of labour market indicators. And I want to use the next section of my remarks to set out my latest assessment in three parts – first what we are seeing in terms of labour market quantities, second what we can glean from flows data, and third what is happening on the nominal side of the labour market.

The vacancies to unemployment (V/U) ratio has become our primary indicator of conditions in the labour market. As shown in Chart 3, the V/U ratio has come down significantly from its peak in mid-2022, with the majority of the move coming from falls in vacancies. We tend to use 2019 as a comparison point for the state of the labour market, given domestic demand and supply were seen to be broadly in balance, and relative to that pre-Covid reference point the V/U ratio is now materially lower.

Chart 3: The V/U ratio continues to decline

Vacancies to unemployment ratio and estimated equilibrium values^(a)

Footnotes

• Sources: AA/WARC Expenditure Report, ONS and Bank calculations.(a) The equilibrium V/U ratio is estimated using an error-correction model over the period 1982–2024. The real cost of vacancy posting and hourly labour productivity are included as long-run determinants for the level of vacancies. The model also includes controls for short-term movements in these variables (Stelmach et al, 2025). The final data points for both series in the chart are 2025 Q1, while diamonds represent projections for 2025 Q2.

As vacancies have exhibited an upward trend over time Bank staff have estimated an equilibrium V/U ratio, also shown on Chart 3. The difference between the observed and equilibrium V/U ratio – the V/U gap – gives a measure of the degree of slack in the labour market. In the last couple of years the V/U gap has shown a reasonable correlation with Bank staff’s estimate of the output gap, as shown in Chart 4. On that basis, the V/U gap suggests that a degree of economy-wide slack has now opened up.

Chart 4: The V/U gap has been positively correlated with the output gap

The gap between the actual V/U ratio and its equilibrium level and the MPC’s estimate of the output gap^(a)

Footnotes

• Sources: ONS and Bank calculations.(a) The final data points for both series in the chart are 2025 Q1, while diamonds represent projections for 2025 Q2.

This loosening is also increasingly apparent from the employment numbers. The latest PAYE data for payrolled employees, notwithstanding the potential for revisions, indicate that private sector employee numbers are now clearly in contractionary territory, as shown in Chart 5. Growth rates in the public sector are also falling back. Retail and hospitality, the sectors particularly impacted by the National Living Wage (NLW) and National Insurance Contributions (NICs) increases in April 2025, have fallen particularly sharply. Notwithstanding these early indications it will be some time before we can be more certain about how NLW and NICs increases will feed through into firms’ hiring decisions, versus being passed on to consumers in price increases, or reduced wage growth or simply absorbed into firms’ profit margins.

Chart 5: Private sector employee numbers are in contractionary territory

Growth in the number of payrolled employees

Footnotes

• Sources: Pay As You Earn Real Time Information from HM Revenue and Customs and Bank calculations. Latest data to May 2025.

The Bank Agents’ measure of recruitment difficulties and the REC survey, which has been a pretty reliable leading indicator of the V/U ratio, points to some further loosening in the pipeline, as shown in Chart 6.

Chart 6: Survey indicators point to further loosening in the pipeline

Agents’ reported recruitment difficulties^(a) and REC survey’s proxy for tightness

Footnotes

• Sources: Bank of England Agents; KPMG/REC UK Report on Jobs, ONS and Bank calculations(a) The Agents’ scores are judgement-based assessments of economic conditions, based on the Agents’ conversations with businesses. The recruitment difficulties score measures the scale of general recruitment difficulties across the economy. Latest data to May 2025.

It is important to stress that what we are seeing in terms of the outlook is a sustained cumulative loosening rather than a more abrupt weakening. The redundancy rate and notifications have picked up only very gradually and remain at a very low rate, though the notifications series shown in the left hand panel of Chart 7 does not cover the smallest businesses. A potential leading indicator of a more material increase ahead is the rise in Google searches for redundancy, shown in the right-hand panel of Chart 7.

Chart 7: Redundancies have seen little movement though early signals of a rise

Measures of redundancies^(a)(b)

Footnotes

• Sources: Insolvency Service Redundancy Payment Services Database, Google Trends, ONS and Bank calculations.(a) Employers are required to submit an HR1 form at least 30 days before the first dismissal where 20 to 99 redundancies are proposed, and at least 45 days before the first dismissal where 100 or more redundancies are proposed.(b) Google searches show the seasonally adjusted average of trends for the terms: ‘redundancies’, ‘redundancy calculator’, and ‘redundancy insurance’.

Turning to the LFS based flows data I’d like to start by recognising the ONS’ efforts in
re-introducing these data series, which help inform understanding of labour market dynamics. Although it is important to acknowledge that the relatively low sample sizes, while rising, mean we shouldn’t over-interpret what we take from these data.

With that said, I’d like to draw out what signal we can take from the inflows to unemployment, the ‘separation rate’ and the outflows from unemployment, the ‘job-finding’ rate. Typically, at the start of a more pronounced downturn there would be an uptick in the separation rate, as workers become unemployed. In parallel outflows from unemployment typically tick down slightly from the start of a downturn. Chart 8, with the job-finding rate inverted, suggests initial signs of that process re-emerging over the last year.

Chart 8: LFS based flows data suggest more loosening to come?

Cumulative changes in the flows-consistent 16-64 unemployment rate

Footnotes

• Sources: ONS Labour Force Survey and Bank calculations.

If the recent trends in flows persist then we may see signs of a more substantial weakening in the labour market emerging, which would risk unemployment rising more over the next year than in the MPC’s baseline forecast.

What are the implications for the nominal side? The private sector regular Annual Weekly Earnings (AWE) data does continue to be volatile, but the April reading of 5.1% was below our recent forecasts, as shown in Chart 9.

Chart 9: Developments in wages are on track

Short-term projections of annual private sector wage growth^(a)(b)

Footnotes

• Sources: Bank of England Agents, DMP Survey, HMRC, KPMG/REC UK Report on Jobs, Lloyds Business Barometer, ONS and Bank calculations.(a) Private sector regular pay growth in the aqua line shows the ONS measure of private sector regular average weekly earnings growth (three-month average on same three-month average a year ago). Bank staff’s indicator-based model of near-term private sector regular pay growth is quarterly and uses mixed-data sampling (or MIDAS) techniques. Latest data points are for the three months to April 2025 for private sector regular pay and 2025 Q2 for the indicator-based model estimates.(b) Diamonds indicate projections or expectations for pay growth. Definitions of wage growth vary between each of the measures. The Agents’ pay survey diamond shows respondents’ expected average pay settlements in 2025, weighted by employment and sector. The DMP diamond shows average expected pay growth one year ahead for respondents to the May 2025 DMP Survey. Feb MPR pay growth projections are for 2024 Q4 to 2025 Q4, May MPR pay growth projections are for 2025 Q1 to 2026 Q1.

This follows the concerning pick-up in AWE growth rates in the second half of last year. Our indicator-based model estimates didn’t take too much of a signal from the pick-up but the previous downward trend in wages did level off. Given these uncertainties the MPC anchored its expectations for earnings to the Bank’s agents pay survey, which has had a good recent track record of predicting actual pay settlements. The survey was consistent with settlements falling from 5.4% in 2024 to 3.7% in 2025, which would be approaching rates consistent with the MPC’s inflation target.^{footnote [3]} The latest intelligence from the Agents had continued to suggest average pay settlements for 2025 of 3.5% to 4%. Further encouragement comes from the Bank’s and Brightmine’s settlements databases, which point to actual settlements since the start of the year in a range of 3% to 4%, although the sample is still partial at this stage.

Developments in inflation

Headline CPI increased to 3.4% in May from 2.6% in March, so we are now seeing the hump in inflation that we had been expecting. The headline CPI number was in line with our short-term forecast for 3.4% in Q2. This shows inflation peaking at 3.7% in September before easing back to around its current level by the end of the year. The main components of inflation are on different paths, as set out in Chart 10.

I have previously drawn attention to the extent to which we can draw conclusions from these components,^{footnote [4]} noting that the degree of symmetry in the respective rise and fall of each wave gives a gauge on the extent of inflation persistence. The key services ex. rents disinflation process is now clearly slower than its earlier inflationary rise and at 4.4% is uncomfortably high. It will stay elevated in the short term due in large part to the effects of various administered and one-off price increases. But given the high labour intensity of services the increasing evidence of a loosening in the labour market should continue to provide a disinflationary impulse.

Turning to the other components of the CPI, rent inflation, having stayed stubbornly high, has finally started to fall relatively steeply.

Chart 10: Components of inflation on different paths

Inflation rates of the components of CPI inflation

Footnotes

• Sources: ONS and Bank calculations. Data to May 2025.

At the same time what is clearly evident from Chart 10 is the pick-up in food price inflation from the middle of last year. At 4.4% in May it was above our forecast and the level of food prices is now 36% above its level four years ago in spring 2021. This is significant because there is some evidence that food prices are particularly salient for households’ perceptions of, and expectations for, inflation.^{footnote [5]}

Core goods price inflation has remained subdued, tracking slightly below our short-term forecasts. To complete the picture, even before the latest geopolitical developments, energy price inflation was close to turning positive again in May after the best part of two years in negative territory.

Uncertainty in the outlook – weak demand or weak supply?

When weighing up the risks to the inflation outlook as they have evolved in 2025, I’ve focussed on two qualitative scenarios for weak demand and weak supply. To more systematically address the mechanisms underpinning the main risks, Bank staff developed two quantitative scenarios which we included in the May MPR. In the staff’s weaker demand scenario uncertainty played a bigger role than in our baseline forecast, weighing particularly on investment but also on household saving and leading to domestic inflationary pressures fading more quickly. In the weaker supply scenario, the current hump in headline inflation led to additional second round effects and more inflation persistence, underpinned by less loosening in the labour market and amplified by weak potential productivity growth.

More detail on the scenarios can be found in the May MPR^{footnote [6]} and in speeches from other MPC members. For the purposes of my remarks today the main evolution in my thinking is to put more weight on the mechanisms at play in the weaker demand scenario, which will be reinforced by the risks of a more significant loosening in the labour market.

I remain alive to the risk that we will not see the recovery in productivity envisaged in the MPC’s baseline forecast. And I am not ruling out that the wage setting process has become less flexible, for example through wage setting having become less contestable. Nor am I ruling out that mismatch may have become more pronounced in certain sectors. We will need to stay vigilant about the changing composition of inflation and the risk that rises in the most salient parts of the consumer basket, food and energy, increase the chances of second round effects from the current hump in headline inflation. Similarly, the MPC needs to pay close attention to any evidence that the later stages of the disinflation process, whereby lower wages growth feeds through into lower services prices growth is stalling.

These are all plausible considerations. But in my assessment of the risks to meeting our inflation target in the medium term they have been outweighed by the cumulative evidence across a wide range of indicators of ongoing labour market loosening.

Implications for monetary policy

In terms of the MPC’s June decision I assessed that the signals from the labour market provided a sufficiently strong case for a reduction in Bank Rate to 4%, to guard against the increasing risk that, in the medium term, inflation would settle at below our 2% target. Given the myriad uncertainties relating to global developments but also the UK outlook any policy approach has to be robust to alternative states of the world.

Before finalising my vote I challenged myself on whether sequential cuts in Bank Rate in May and June would be appropriate if it subsequently became clear that risks to inflation were actually weighted to the upside. It was a finely balanced judgement for me but I concluded that my approach was robust for two main reasons. First because even at 4% I assess that monetary policy would remain clearly in restrictive territory. So if evidence emerged that pointed to higher inflation in the medium term then Bank Rate could be held higher for longer than would otherwise be the case. Second given the environment we are operating in I think it is important that monetary policy is outlook dependent and should respond and be seen to respond when the evidence on the outlook requires it. I continue to take a watchful and responsive approach to setting policy. I think this is in line with the emphasis the MPC puts in its collective communications on policy not being on a pre-set path and that we will consider the degree of restraint which is required from meeting to meeting.

I see no inconsistency with my latest vote and the MPC’s gradual and careful approach to the withdrawal of restrictiveness. This has served the MPC well over the last year as we have cut Bank Rate four times by a total of 100bps in total. And I believe it will continue to serve us well in these times of increased uncertainty and unpredictability, as long as the disinflation process is judged to be continuing, in order to return inflation back to the 2% target sustainably in the medium term.

With thanks to Ed Kent and Michal Stelmach for their assistance in preparing these remarks, and to Andrew Bailey, Josh Lillis, Andrea Rosen, Vicky Saporta, Martin Seneca, Fergal Shortall, Carleton Webb and Jan Zacek for their helpful comments and contributions.