Mixed signals, research findings, and policy judgements − speech by Catherine L. Mann

Introduction

As I think about my recent policy decisions, my assessment has evolved with incoming data and ongoing research. At the start of this year, moderation in price and wage growth, alongside slowing activity, had brought me closer to considering a reduction in Bank Rate. However, since the onset of the conflict in the Middle East, the reemergence of the trade‑off between inflation and activity has led me to place more weight on inflation persistence, shifting my view towards a longer hold, and potentially a need to lean against that risk.

This remains my view, as notwithstanding a sporadic reopening of the Strait of Hormuz, infrastructure rebuilding, future-proofing, and restocking supplies imply that higher energy prices will likely persist. Associated firm pricing reactions are also likely to keep inflation from moderating to the two percent target within the monetary policy horizon. That said, the discussion for today addresses the timing of a hike versus hold decision. The first topic is how an assessment of disaggregated data provides important nuances to the economic picture relative to the headline numbers. The second topic is how key research findings for both inflation persistence and monetary policy transmission inform my policy strategy.

The disaggregated assessment of the data shows mixed signals across key indicators. Realized inflation and expectations differ across households, firms, and financial markets, and have been influenced in the aggregate by fiscal measures. Wage developments differ across the public and private sectors, with further variation across industries. Public sector activity growth continues to outpace that of the market sector, alongside continued government budget overruns, including on account of interest costs. Volatility in both inflation and financial conditions has increased, affecting the degree of restrictiveness as well as the signal-to-noise assessment.

Interpreting mixed and noisy signals is nothing new for central bankers, but it is more challenging in the current environment where more frequent shocks are building on already persistent inflation with embedded second-round effects, amid higher volatility with elevated uncertainty (see Mann 2024b). In this speech, I will evaluate some of these mixed signals, explain how I interpret them and how – in light of research – they have shaped my recent decisions, especially the last one in June, and what they could mean for my monetary policy strategy going forward.

Uncertain underlying trend inflation

Prior to the conflict in the Middle East, inflation – as indicated in the February 2026 Monetary Policy Report – was expected to come down to our two percent target by the second half of 2026. However, that sanguine projection masked the underlying economic patterns, as increases in administered prices and national insurance contributions (NICs), which contributed 0.7 percentage points to the inflation hump in 2025, were assumed to dissipate in Q2 2026. The projected fall in inflation also reflected the energy bills package introduced in the Autumn Budget. The latter accounted for a reduction of nearly 0.25 percentage points to CPI inflation, as Chart 1 shows. Finally, slack in the economy was projected to weigh on inflation through the projected tightening of the fiscal position. The latter two factors underscore the importance of fiscal measures in getting us back to the inflation target.

Chart 1: Contribution of budget measures to deviation from inflation target

Percentage point deviation from the 2% inflation target

• Source: ONS and Bank calculations.
• Notes: ‘Administered prices, NICs, and electric vehicle (EV) mileage charge’ includes the estimated direct effects of changes to Vehicle Excise Duty, bus fares, VAT on school fees, and water charges. The NICs component represents Bank staff’s allowance for the effects of the recent increase in employers NICs on CPI inflation. The EV mileage charge component relates to the mileage-based charge on electric vehicles announced in Budget 2025.The ‘Energy budget measures’ bars represent the estimated direct effects of the energy bills package. The ‘Import prices’ bars includes an MPC judgement on the price of tradables (see May 2024 Monetary Policy Report) which has little effect on the profile beyond 2026. The ‘Other’ bars represent other deviations of inflation not attributable to the listed influences including the estimated effects of changes to tobacco and vaping products duties.

Of course, that projection was prior to the US-Israel-Iran war. The economy and our projections have evolved since then. For starters, the conflict resulted in a higher forecast for CPI inflation across all three scenarios presented in the April 2026 Monetary Policy Report. While outturns for price growth – so far lower than expected – continue to reflect lagged pass-through from pre-war moderation in commodity and domestic cost pressures, household inflation expectations have picked up materially since the conflict.

Because inflation shocks are amplified and attention to inflation rises when inflation is elevated – both with thresholds of around 3 to 3.5 percent CPI inflation – inflation expectations could pick up even more (Weber et al., 2025; Gaffney and Potjagailo, 2026).^{footnote [1]} Staff research also shows that the latest rise in households’ short-term inflation expectations move back into the non-linear region, even if not quite as far as in 2022 (Buckmann et al., 2025). While the most recent observation of the Citi/YouGov measure of household inflation expectations has decreased from its April peak, it remains within the region of these non-linear dynamics, and both short- and long-term expectations remain elevated relative to historical averages and their pre-Iran war levels.

On the firm side, research surveys reveal that the recent inflationary episode spurred an increase in the share of firms setting prices in response to changes in economic conditions – so-called state-dependent pricing – instead of adjusting their prices at fixed intervals (Bunn et al., 2026a). State-dependent firms increased their prices more quickly when inflation rose but also raised their prices by less relative to time-dependent firms as inflation decreased (Chart 2) as the time-dependent firms’ adjustment caught-up. A higher frequency of price adjustments yields an upside inflation bias as well as more volatile inflation, as prices respond more quickly, more often, and more strongly to positive inflation shocks.

Chart 2: Firm output price growth by price-setting behavior

Own-price growth and expected year-ahead own-price growth, percent

• Source: Bunn et al. (2026a).
• Notes: Chart depicts trends in own-price growth (solid lines) and expected year-ahead own-price growth (diamonds), split by firms’ price-setting behavior. Firms are split into state-dependent and time-dependent pricing based on each firm’s most recent response to this question (as of April 2026). The series are three-month moving averages, with the latest data up to May 2026. Percentages in brackets denote share of firms using state- and time-dependent pricing.

To get a better understanding about the current conjuncture and prospects for nominal conditions, I look at firms’ responses to the Decision Maker Panel (DMP) survey, particularly the distributions of responses to questions about expected CPI inflation and expected one-year-ahead price- and wage-setting. Following the outbreak of the Iran war, the survey data show that firms’ CPI inflation expectations quickly responded, (left panel in Chart 3). Although the distribution has shifted to the right in the latest quarter (purple) relative to the three months preceding the Iran war (orange), it remains below 2022 (aqua) and is also much less dispersed. This implies that the pattern of firms’ responses does not yet exhibit the same degree of uncertainty about future inflation as it did during the
Russia-initiated inflation surge.

The distribution of firms’ expectations for their own-price growth in the year ahead has shifted modestly to the right, too. This is most apparent for consumer-facing firms and goods producers, while services firms and business-facing companies have adjusted their expectations by less. These changes in firms’ expectations and price-setting behavior are a key ingredient for the evolution of CPI inflation itself.

The DMP survey data also show that some firms have modestly altered their realized prices in recent months, with the distribution for goods-producing firms exhibiting a bit more mass in the right-hand side of the distribution, suggesting that there could be more inflation in the pipeline. Input and output PMIs have also been elevated.

Chart 3: Distributions of firms’ inflation and wage growth expectations

Expected one-year-ahead CPI inflation (left panel) and expected year-ahead wage growth (right panel)

• Source: Decision Maker Panel.
• Notes: Charts show kernel density plots for expected one-year-ahead CPI inflation (left panel) and year-ahead wage growth expectations for 2022, December 2025 to February 2026 (three months prior to the Iran war), and March 2026 to May 2026. Distributions are weighted by industry and employment shares.

Widening the aperture on potential wage dynamics from the firm-level survey data, current official wage statistics show mixed signals. Despite a continued easing in private sector regular pay growth (left panel in Chart 4), which is now within the Bank staff’s estimates of target-consistent wage growth, whole-economy total pay has not decelerated to the same degree. We therefore continue to see a wedge between the two pay measures, as illustrated by the right panel in Chart 4.

Over the past year, that wedge has largely been driven by stronger public sector wage growth and the impact of bonuses. In recent months, the contribution from the public sector has eased somewhat. Drawing on the evidence from February 2026 Monetary Policy Report on heterogeneity in wage setting, this pattern is consistent with stronger pay growth in sectors characterized by more structured or collective bargaining where inflation and inflation expectations play key roles. This includes the public sector and similar “union-type” segments within the private sector, as well as cost-minimizing firms, which together account for around 80% of private-sector workforce.

Pay settlements have largely been finalized for the year, and there is limited evidence (as shown in Chart 3 above) from firms in the DMP of a pick‑up in what they expect to offer their employees next year. Looking ahead, the risk is that wage pressures are deferred rather than dissipated, particularly if future pay settlements – including in these more collectively bargained segments – are set against the backdrop of higher realized inflation and elevated expectations in the second half of this year.

Furthermore, even if energy prices moderate, the temptation will be to maintain the own- price and wage strategies to partially rebuild margins, which, according to the survey data, have been absorbing the current shock (Bunn et al, 2026b). All told, distributions are important indicators to evaluate whether inflationary pressures, the wage setting process, and resulting price setting feed into overall inflation.

Chart 4: Regular pay growth and contributions to the wedge between whole-economy and private sector pay growthThree-months-on-year growth (left panel) ^(a) and percentage point contributions to the difference (right panel)^(b)

• Source: ONS and Bank calculations.
• Notes: Left panel (a) depicts three-months-on-year public and private sector regular pay growth in percent and the range of target-consistent estimates for regular pay growth. Right panel (b) depicts percentage point contributions by bonuses and public sector regular pay growth to the wedge between three-months-on-year whole economy total and private sector wage growth. Latest data: April 2026.

The trade-off re-emerges

Alongside the higher projected paths for inflation in the April Monetary Policy Report is a slower projected pace of GDP growth across all scenarios relative to February (Chart 5). The trade-off between inflation and activity has re-emerged, as I have noted in previous speeches (see Mann 2025a, 2025b).

Chart 5: Real GDP in the February and April 2026 Monetary Policy Reports

Index (2025 = 100)

While headline GDP growth remains subdued, looking beneath the aggregate number reveals a divergence between the role of public and private sector activity. While GDP growth has recently been primarily driven by market sector output (MSO, right panel in Chart 6), public sector output has grown by around 4.5 percent since Q4 2023, outpacing the cumulative growth of market sector output by around one percentage point (left panel in Chart 6).

Chart 6: Growth in market sector and public sector GDP and their contributions to GDP growth

Cumulative change since 2023 Q4 (left panel) and contributions to change since 2023 Q4 (right panel)

• Source: ONS and Bank calculations.
• Notes: Notes: The government sector comprises public administration, education, and health, while the market sector comprises all other industries, including agriculture, wholesale and retail trade, accommodation and food services, and other market-based activities. Latest data: April 2026.

Why do I highlight the different growth rates of government and market sector GDP? The MPC’s forecast is conditioned on announced fiscal policy, as reflected in the Office for Budget Responsibility’s (OBR’s) projections. In simple terms, fiscal policy is funded by a mix of public sector borrowing and taxes. As noted in the Bank’s recent Forecast Evaluation Report, “looser-than-expected fiscal policy has […] contributed to there being less spare capacity” (Bank of England, 2026b). As shown in Chart 7, this reflects higher outturns for public sector borrowing than embodied in the OBR’s forecasts, consistent with the OBR’s own forecast evaluation (Office for Budget Responsibility, 2026). Using a structural VAR model, Brignone and Piffer (2026) demonstrate that upward revisions in inflation forecasts following the post-pandemic inflation surge can be attributed to a combination of expansionary shocks on the demand side, revisions to past shocks, and contractionary supply-side shocks. Going forward, the question is whether there will be upside surprises in fiscal policy,^{footnote [2]} which could provide additional support to public sector activity and can have broader implications for productivity, aggregate demand, inflation, and financial conditions.

Chart 7: Successive forecasts and outturns of public sector net borrowing

Public sector net borrowing as a percent of GDP

• Source: Office for Budget Responsibility.
• Notes: Data shows successive forecasts (grey) and outturns (orange) of public sector net borrowing as a percent of GDP. Latest forecast: March 2026.

Labor market data also show a mixed picture, with implications for the strength of demand. Beneath the rise in headline unemployment in recent years and its projected increase, the dispersion in unemployment rates across industries has also increased. Chart 8 illustrates how unemployment rates across different industries have changed since 2022. The shaded area (aqua) shows that the dispersion of unemployment rates across different sectors has increased since mid-2025, suggesting that unemployment is more unevenly distributed across industries. For instance, the unemployment rate remains higher in accommodation and food services and has risen faster in that industry compared to overall unemployment since mid-2025. By contrast, the professional and technical sectors are closer to the lower end of the distribution of unemployment rates across sectors.

Some of these differences may reflect varying exposure to recent changes in fiscal policy, such as changes to national insurance contributions and the national living wage, as well as sector-specific demand conditions, including exposure to the energy shock. The result is a labor market that is far from uniform, indicating that some industries are more resilient than others, and that headline unemployment may overstate weakness in the labor market. Therefore, demand may be more resilient than projected by the aggregate numbers.

Chart 8: Dispersion in unemployment rates across industries

Unemployment rates by previous industry (percent)

• Source: April 2026 Monetary Policy Report.
• Notes: Data show ONS estimates of unemployment by previous industry. This measure will not fully capture new entrants to the labor market who do not find a job as they have no previous industry, nor will it reflect the capacity for workers to transition across industries. The shaded area shows the range of industry-level unemployment rates by previous industry over time. Given data limitations and relatively small sample sizes, it can be difficult to draw firm conclusions from changes in these data. These data are not seasonally adjusted. The latest data shown are for the three months to April 2026.

Financial conditions are volatile

I also considered financial conditions in forming my June decision, as has always been the case. Global spillovers, as discussed in one of my speeches last year, have importantly contributed to the volatility of financial market variables, particularly in a small open economy like the UK. That remains the case today (Chart 9).

Looking at the time-varying volatility of asset price factors extracted from a daily structural VAR based on Brandt et al. (2021), we see that the Iran conflict has been associated with a material increase in the volatility of international monetary policy factors but also in the volatility in the UK-specific risk factor. In fact, the UK-specific risk factor is now furthest away from its historical normal. By contrast, volatility in the model's macro factors and global risk (which is a flight-to-safety factor) is little changed. This divergence suggests that markets are interpreting the conflict less as a shock to the growth outlook and more as a source of uncertainty about inflation and the associated monetary policy response, as well as perhaps the consequence of the behavior of an evolving investor base in the gilt market, a topic I discussed in my speech on the UK's external exposures (Mann, 2026).

Chart 9: Volatility of asset price factors

Rolling standard deviation of shocks

• Source: Mann (2025c), Bloomberg Finance L.P. and Bank calculations.
• Notes: The calculations are based on a structural VAR identified using sign and magnitude restrictions following Brandt et al. (2021). The model is extended to analyze monetary policy spillovers between the UK and its two most important trading partners and dominant currency blocks, the US and the euro area (as described in Mann (2024a)).

We can also directly observe the increase in financial market volatility in how UK overnight index swap (OIS) curves have evolved since the onset of the Middle East conflict. Following the outbreak of the Iran war, OIS curves have become increasingly volatile, as the widened range in their movements post-conflict (aqua) relative to pre-conflict (purple) in Chart 10 (left panel) shows. During this period, these large increases in OIS rates have been passed-through rapidly into mortgage rates faced by households: interest rates on two- and five-year fixed rate mortgages with a 75% loan-to-value ratio have risen by 87 and 73 basis points, respectively (right panel in Chart 10). The asymmetry in price and wage developments I highlighted earlier – more robust adjustments to upside shocks and relatively slow adjustment to downside shocks – can also be observed in financial markets, as increases in OIS rates pass through to mortgage rates more quickly than do decreases.

As I noted in my 200-word paragraph in the latest MPC minutes, higher financial market volatility acts as a headwind to both consumption and investment. For firms, increased uncertainty around financing conditions raises the hurdle for investment. For households, more volatile inflation can result in uncertainty about future inflation, which tends to encourage precautionary behavior and the maintenance of higher savings buffers (Fischer et al., 2025). At the same time, my preferred measure of financial conditions (see Burr, 2023), has not tightened that much. Taken together, just like other parts of the economy, financial conditions are sending mixed signals, too.

Chart 10: Volatility in OIS curves and pass-through to mortgage rates

UK instantaneous forward OIS curves (left panel) and daily quoted mortgage rates and OIS rates (right panel)

• Source: Bloomberg Finance L.P., Bank of England and Bank calculations.
• Notes: Swathes of UK instantaneous forward OIS curves in the left panel cover an equivalent period of time. Range of moves in OIS curves pre-Iran war (purple swathe) observed from 28 October 2025 until 27 February 2026. Range of moves in OIS curves since the Iran war (aqua swathe) observed from 27 February to 26 June 2026. Right panel shows 2-year and 5-year OIS rates and 2-year and 5-year fixed rate mortgages with 75% loan to value ratios. Latest observation for daily quoted mortgage and OIS rates: 26 June 2026.

Research informs my monetary policy strategy

Against the backdrop of the data and prospects discussed above, two research strands inform my strategy. The first strand emphasizes the importance of leaning against upside risks to inflation to reduce risks of de-anchoring and to ensure that monetary policy remains credible. I’ve discussed research on persistent upside risks to inflation and the consequences in numerous previous speeches.^{footnote [3]} The second research strand emphasizes the speed with which firms adjust their expected pricing strategies and the capacity for Bank Rate decisions to feed through relatively quickly into expectations, firms’ price-setting behavior, and overall inflation, a topic that I addressed last in 2023 (see Mann, 2023a).

Let’s look at what’s new in the first research strand. This research on de-anchoring examines how responsive medium- to long-term inflation expectations are to changes in short-term inflation expectations by regressing the former on the latter alongside a set of control variables.^{footnote [4]} A regression coefficient close to zero indicates firmly anchored expectations – in that case, expectations about inflation in the medium- and long-term are insensitive to short-term fluctuations. By contrast, a coefficient close to one suggests that expectations are de-anchored. Chart 11 shows the analysis for households, firms, and financial markets.

Data from household and firm surveys show an increased responsiveness of their inflation expectations to near-term inflation developments since the Covid pandemic and the energy shock following Russia’s invasion of Ukraine, suggesting that these two groups are not firmly anchored (left and center panel in Chart 11). Financial market participants’ expectations are more anchored relative to households and firms, despite the responsiveness of their long-term expectations to short-run expectations being more volatile (right panel in Chart 11). Taken together, these findings point to some erosion in credibility and potential risks of de-anchoring among households and firms.

Chart 11: Responsiveness of medium/long-term inflation expectations to short-term inflation expectations

Regression coefficients (percentage points)

• Source: Bank of England/Ipsos Inflation Attitudes Survey, Decision Maker Panel, Market Participants Survey and Bank calculations.
• Notes: Household measure shows the slope of 5-year-ahead inflation expectations on 1-year-ahead inflation expectations from the Bank of England/Ipsos Inflation Attitudes Survey. Firm measure shows the slope of CPI 3-year-ahead expectations on CPI 1-year-ahead expectations from the Decision Maker Panel. Financial Markets measure shows the slope of 5-year-ahead inflation expectations on 1-year-ahead inflation expectations from the Market Participants Survey. Shaded areas denote 95% confidence bands.

Staff research has also used the Boosted Inflation Model – a machine learning model based on boosted decision trees that is designed to capture non-linear relationships between a range of indicators and CPI inflation (Buckmann et al., 2025). It can decompose inflation into its underlying drivers, reflecting the price Phillips curve as well as potential drift of that curve. Chart 12 shows the model’s decomposition of predicted inflation relative to the two percent target. The contribution of the trend inflation component (aqua), which comes from elevated household inflation expectations has started to increase since the Iran war, but remains below the levels last seen in 2023 and 2024. Since household expectations can feed into both wage and price developments, this can be a harbinger of second-round effects.

These developments increase upside risks to the inflation process and raise the risk of
de-anchoring, particularly given the extended period of above-target inflation and the increase in inflation volatility. As discussed in Mann (2022), once expectations drift from the two percent target, monetary policy needs to tighten more, to both control expectations and to moderate inflation itself. So, considering this research strand alone would typically point towards a tighter policy stance.

Chart 12: Decomposition of CPI inflation from Boosted Inflation Model

Annualized and smoothed (left) and annualized month-on-month (right)

• Source: Buckmann et al. (2025).
• Notes: Colored bars show the block-wise predictive contributions to 1-month-ahead annualized month-on-month CPI inflation (white line), around 2% target. Blockwise boosted inflation model estimated via cross-validation over 1989m1-2026m5 sample. Dashed line: CPI inflation, annualized. 

What’s new in the second strand of research – the speed of transmission of monetary policy. Evidence points to a relatively more rapid transmission of monetary policy to borrowing costs, but also directly to inflation expectations as compared to what is embodied in most traditional slack-driven models of inflation process. Although even the more familiar approach finds that monetary transmission to GDP has been quicker.^{footnote [5]}

With regard to the first stage of transmission of monetary policy,^{footnote [6]} the relatively quick increase in mortgage rates in response to the rise in inflation and market expectations of tighter monetary policy that I showed in Chart 10 illustrates a key channel through which the financial environment will affect households considering a new mortgage or seeking to remortgage by making their monthly repayments more expensive. Note as well that the moderation from the peak has been slower than the advance, pointing to asymmetry in the institutional response to the policy environment.

Besides the fast pass-through to financial variables, Brandt et al. (2026) use
high-frequency credit and debit card spending data to show that monetary policy shocks can affect real economic activity very quickly, too. They show that consumer spending declines within a few days in response to a contractionary monetary policy shock, while web searches for terms related to unemployment rise. This result suggests that a fast adjustment in households’ expectations about their economic prospects can drive a change in consumption behavior.

Firm-level research by Yotzov et al. (2026) also shows that firms update both their CPI inflation perceptions and expectations as well as their own-price growth expectations upon CPI releases. Specifically, the authors find that between 2022 and 2024, a one percentage point increase in headline CPI inflation raises firms’ expected year-ahead own-price growth by 0.3 percentage points. On CPI perceptions, they respond quickly within a few hours of data releases. Firms’ expected own-price growth also increases a few hours after a positive change in CPI inflation, although to a smaller extent.

Research by Di Pace et al. (2025) analyses how firms update their price expectations in response to monetary policy announcements by the Bank of England. They find that within days following the Bank Rate decision, firms revise down their year-ahead price growth expectations if the increase in Bank Rate is sufficiently large. As Chart 13 shows, Bank Rate increases of 50 basis points or more result in a statistically significant and economically sizeable downward revision in firms’ expected price growth for the year ahead. While this pattern is evident across the distribution of firms’ expected own-price changes, it is more pronounced in the left tail. This means that firms revise the lower-price outcomes in their pricing plans more than the higher-price outcomes, suggesting an asymmetric updating of expectations. Chart 13 also shows that firms do not seem to significantly alter their price growth expectations in response to announcements of Bank Rate changes smaller than 50 basis points.

Chart 13: Responsiveness of firm price expectations to bank rate changes

Regression coefficients

• Source: Di Pace et al. (2025).
• Notes: The chart plots regression results for different moments of the distribution of firms’ price expectations on monetary policy shocks within a 5-day window around policy announcements. Monetary policy shocks are measured using actual Bank Rate changes. The series is normalized so that coefficients correspond to the response to a 25 basis point shock. BR change < 50 bps indicates a Bank Rate change below 50 basis points in absolute value. BR change >= 50 bps indicates a Bank Rate change equal or above 50 basis points in absolute value. Analysis uses data from November 2016 until December 2023. Error bars denote 95% confidence intervals.

Implications for monetary policy strategy

How do I use the underlying data and these two strands of research to inform my policy strategy? Consider the following three dimensions generally, but with specific reference to my June decision.

First, the importance of a disaggregated assessment. Disaggregated data is always key to how I evaluate the economic conjuncture. In June, these data exhibited mixed signals, which provided a more nuanced view of the inflation-activity trade-off than the macro data. Wage growth has differed between private and public sectors. Pricing behavior of firms has varied in speed of responsiveness. The administered one-off changes in prices masked underlying trend inflation. Disaggregated labor signals in some sectors were less weak than the overall unemployment rate. Tracking fiscal outturns has been important, especially given the gap between GDP and market-sector output, and the evidence from the Forecast Evaluation Report.

Second, on the topic of volatility and noise versus signals. Whereas both higher inflation and financial market volatility do make it harder to distinguish signals from transitory movements, because inflation volatility goes hand-in-hand with inflation rates, and financial volatility is reflected in risk premia, the noise contributes to the signal. It puts upside pressure on inflation and tightens financial conditions, at least in nominal terms. The source of inflation pressures – from costs versus from demand matters. The source of financial risk premia – from domestic drivers versus foreign spillovers – matters. In June, the cost-oriented inflation pressures were being offset by domestic-oriented financial restrictiveness. Bearing these factors in mind has been key to many of my policy decisions.

Third, on the topic of timing. Energy price developments, profit margin behavior, and the underpinnings of the 2027 wage negotiations will be critical in determining whether current cost pressures become more embedded. All these conditions are still unfolding. Given the seasonality of wage negotiations and their dependence on previous inflation and inflation expectations, data outturns (including the expectations for one-year-ahead) in the second half of this year are particularly important for my future decisions.

Considering my June decision, by the disaggregated data assessment, there was more upside risks to inflation compared to downside risks for activity. On that basis and considering the research about risks of de-anchoring alone, tightening Bank Rate could have made sense. But, at the time of the June meeting, financial markets had tightened considerably at least in nominal terms.

Research on the responsiveness of pricing and consumer behaviors to CPI announcements and to monetary policy shocks is clear evidence that monetary policy can have quick effects on the key variables of inflation expectations and CPI inflation, which is the Bank of England’s remit. In June, observing these forward-looking and high-frequency data in the context of both strands of the research meant that an activist hold was the appropriate decision.

Going forward, as I evaluate the disaggregated data, the sources of signal-to-noise in financial conditions, and the inputs to firms’ pricing strategies, I am confident that if outturns – especially in expectations – are unfavorable to the underlying inflation process, an activist move can bring inflation expectations and outcomes toward the two percent target.

Acknowledgements

I would like to thank Hassana Babangida and Christoph Herler for their help in the preparation of this speech.

I would also like to thank Andrew Bailey, Nick Bate, Lennart Brandt, Jonathan Bridges, Marcus Buckmann, Sam Christie, Rohan Churm, Ruslana Datsenko, Giulia Giardin, Michael McLeay, Galina Potjagailo, Harry Rigg, Martin Seneca, Krishan Shah, Daniel Steel, Michal Stelmach, Elyse Sullivan, James Tasker, Gellert Turkevi-Nagy, Kamran Vaishnav, James Walkington, Boromeus Wanengkirtyo, Carleton Webb, Teresa Wukovits-Votzi, and Ivan Yotzov for their comments and help with data and analysis.

References

Bank of England (2024). ‘Monetary Policy Report – May 2024’.

Bank of England (2026a). ‘Bank Rate maintained at 3.75% - June 2026 Monetary Policy Summary and Minutes’.

Bank of England (2026b). ‘Forecast Evaluation Report – January 2026’.

Bank of England (2026c). ‘Monetary Policy Report – February 2026’.

Bank of England (2026d). ‘Monetary Policy Report – April 2026’.

Brandt, L., Saint Guilhem, A., Schröder, M., and Van Robays, I. (2021). ‘What drives euro area financial markets? The role of US spillovers and global risk’, ECB Working Paper Series, No. 2560.

Brandt, L., Fischer, J. J., Horn, C.-W., Miranda-Agrippino, S. and Pallotti, F. (2026). ‘The Short-Term Effects of Monetary Policy’, mimeo.

Brignone, D., and Piffer, M. (2026). ‘Structural forecast analysis’, Bank of England Staff Working Paper, No. 1,165.

Buckmann, M., Potjagailo, G. and Schnattinger, P. (2025). ‘Blockwise Boosted Inflation: Non-linear determinants of inflation using machine learning’, Bank of England Staff Working Paper, No. 1,143.

Bunn, P., Bloom, N., Menzies, C., Mizen, P., Thwaites, G., and Yotzov, I. (2026a). ‘State- and time-dependent pricing’, Bank of England Staff Working Paper, No. 1,166.

Bunn, P., Bloom, N., Mizen, P., Thwaites, G., and Yotzov, I. (2026b). ‘How UK firms are responding to the war in Iran: Early evidence from the Decision Maker Panel’, VoxEU.org.

Burr, N. (2023). ‘The challenge of measuring financial conditions’, Bank Underground.

Burr, N., and Willems, T. (2024). ‘About a rate of (general) interest: how monetary policy transmits’, Bank of England Quarterly Bulletin.

Datsenko, R. and Wanengkirtyo, B. (2026). Inflation Shock Pass-Through under State-Dependent Pricing with Asymmetric Rigidities, mimeo.

Di Pace, F., Mangiante, G., and Masolo, R. M. (2025). ‘Do firm expectations respond to monetary policy announcements?’, Journal of Monetary Economics, 149, pp. 1-17.

Fischer, J. J., Herler, C. and Schnattinger, P. (2025). ‘When the fog clears: the effect of reduced inflation uncertainty on households’ financial behaviour’. Bank of England Staff Working Paper, No. 1,133.

Mann, C. L. (2022). ‘Inflation expectations, inflation persistence, and monetary policy strategy’, speech given at the 53^rd Annual Conference of the Money Macro and Finance Society, University of Kent, 5^th September.

Mann, C. L. (2023a). ‘Expectations, lags, and the transmission of monetary policy’, speech given at the Resolution Foundation, 23^rd February.

Mann, C. L. (2023b). ‘Inflation models and research: distilling dynamics for monetary policy decision-making’, speech given at the Canadian Association for Business Economics, 11^th September.

Mann, C. L. (2024a). ‘Policy spillovers when external shocks persist and domestic activity diverges’, speech given at the Central Bank Research Association, National Bank of Poland and Bank of Lithuania 5th biennial conference ‘Macroeconomic adjustments after large global shocks’ in Vilnius, Lithuania, 20^th September.

Mann, C. L. (2024b). ‘The Great Moderation 20 years on’, speech given at the Annual Conference of the Society of Professional Economists, London, 14^th November.

Mann, C. L. (2025a). ‘Explaining the consumption gap’, speech given at the Resolution Foundation, 9^th October.

Mann, C. L. (2025b). ‘Five ‘C’s for Central Bank Research’, speech given at The Future of Central Banking conference on the occasion of the 100th Anniversary, Banco de México, 26^th August.

Mann, C. L. (2025c). ‘Holding the anchor in turbulent waters’, speech given at the conference on ‘35 years of flexible inflation targeting: Opportunities and challenges’, Wellington, New Zealand, 6^th March.

Mann, C. L. (2026). ‘Old exposures, new actors: implications for monetary policy of the UK’s external imbalances’, speech given at the London School of Economics and Political Science, 13^th May.

Office for Budget Responsibility (2026). ‘Forecast evaluation report – June 2026’.

Weber, M., Candia, B., Afrouzi, H., Ropele, T., Lluberas, R., Frache, S., Meyer, B., Kumar, S., Gorodnichenko, Y., Georgarakos, D., Coibion, O., Kenny, G., and Ponce, J. (2025). ‘Tell Me Something I Don’t Already Know: Learning in Low- and High-Inflation Settings’, Econometrica, 93 (1), pp. 229-264.

Yotzov, I., Bloom, N., Bunn, P., Mizen, P., and Thwaits, G. (2026). ‘The Speed of Firm Response to Inflation’, Journal of the European Economic Association, 24 (2), pp. 736-768.