The path of disinflation in underlying domestic price and wage pressures has generally continued, albeit to different degrees (Key judgement 1). Wage growth has declined further recently, to around 5% in May, although services consumer price inflation has remained at 4.7% in recent months. Twelve-month CPI inflation averaged 3.5% in 2025 Q2, 0.1 percentage points higher than expected in the May Report and an increase from 2.8% in Q1. Around half of this rise was accounted for by an expected reduction in the extent to which energy prices have been dragging on headline inflation. The contribution of administered prices to inflation has increased, while food price inflation has also picked up by more than anticipated. While headline inflation is projected to rise slightly further in the near term, to around 3¾% over the second half of this year, the baseline projection assumes that this will not lead to additional second-round effects on domestic inflationary pressures.

Four-quarter GDP growth is projected to remain close to its recent average level, of around 1¼%, before picking up in the second half of the forecast period. This remains dependent on a sustained fall in the household saving ratio. A margin of spare capacity is estimated to have opened up in the UK economy and the labour market is continuing to loosen gradually. This margin of excess supply is expected to build a little further, before narrowing from the end of next year onwards (Key judgement 2). A range of factors are expected to reduce spare capacity further out, including the impact on demand of the gradual loosening in the stance of monetary policy embodied in the market path of interest rates. Monetary policy still weighs on the level of demand across the forecast period, however, such that it continues to contribute to the disinflationary process.

The margin of spare capacity in the economy is expected to act against some continuing persistence in domestic prices and wages in order for CPI inflation to fall back to around the 2% target in the medium term (Key judgement 3). The August CPI projection is somewhat higher than the profile in May during the first and second years of the forecast period, and broadly similar in the medium term. There remains considerable uncertainty around the calibration of the Committee’s judgement on the path of second-round effects in domestic prices and wages. Overall, the MPC judges that the upside risks around medium-term inflationary pressures have moved slightly higher since May. This is in part due to the risk of inflation expectations impacting price-setting as illustrated in the inflation persistence scenario set out in the May Report.

Box A: Assessing the restrictiveness of the monetary policy stance

The MPC increased Bank Rate significantly between December 2021 and August 2023, from 0.1% to 5.25%, before starting to reduce Bank Rate in August 2024. This box sets out a range of measures that the MPC has used to assess the restrictiveness of monetary policy. The first part of the box presents evidence on the degree to which past increases in Bank Rate continue to weigh on the level of demand. The second explores the extent to which the current monetary policy stance, taken as a whole, remains restrictive and will hence weigh on demand in the future.

Real and nominal interest rates are elevated relative to end-2021.

Both real and nominal interest rates play an important role in influencing demand and inflation (Burr and Willems (2024)). But real interest rates, defined as nominal interest rates adjusted for expected inflation, are usually judged to be more important when assessing the stance of monetary policy. That is because, when making spending and saving decisions, firms and households tend to factor in their expectations about future inflation (Box C of the February 2024 Monetary Policy Report).

Yields on UK government bonds are significantly higher than in late 2021 (Chart A). Nominal gilt yields rose notably over 2022–23, consistent with increases in Bank Rate over this period. Real yields, obtained from inflation-linked UK government bonds, also rose significantly. Both real and nominal government bond yields are lower than their peaks in 2023 but have been relatively stable over recent quarters. That partly reflects the impact of international developments, which can have a large influence on UK government bond yields (Cesa-Bianchi et al (2021) and Mann (2022)).

Chart A: Real and nominal yields remain elevated relative to end-2021

Bank Rate and select real and nominal government bond yields ^(a)

Footnotes

• Sources: Bloomberg Finance L.P., Tradeweb FTSE Gilt Closing Data and Bank calculations.
• (a) Real yields are inflation-linked UK government bond yields. The final data points shown are for 29 July 2025.

Past increases in Bank Rate are estimated to be weighing on the current level of demand, although this impact is around its peak.

The MPC raised Bank Rate between late 2021 and mid-2023 to curb the sharp rise in inflation over that period (Chart A). As described in Box C of the August 2024 Monetary Policy Report, those increases in Bank Rate have weighed on demand and contributed towards the sharp rise in the household saving ratio over recent years (Section 2.3), as well as the more recent loosening in the labour market (Section 2.4).

The impact of interest rates on demand, estimated using models that feed into the MPC’s baseline projections, is shown by the aqua bars in Chart B. Higher interest rates are estimated to have been weighing on the level of demand since late 2022. Before that, a period of lower interest rates, which had been required to stimulate the economy, had been boosting demand.

Because monetary policy affects demand and inflation with a lag (Burr and Willems (2024)), Bank staff estimate that past increases in Bank Rate and the market-implied path for future interest rates continue to weigh on the level of demand. The impact of past increases in interest rates on demand is now estimated to be around its peak, however.

Once that peak has passed, monetary policy will start to contribute positively to GDP growth, as its effect on demand starts to fade. Nevertheless, past increases in Bank Rate will still weigh on the level of demand, such that they will continue to support the disinflationary process over the coming years.

There are large uncertainties around the size and timing of the estimates in Chart B. The relationship between Bank Rate and the real economy can vary over time (Box C of the August 2024 Monetary Policy Report). In addition, these estimates capture the effects on demand of changes in Bank Rate and the market-implied path for future interest rates. This may not reflect a change in restrictiveness, which is also dependent on movements in the equilibrium interest rate, the level of interest rates that neither stimulates nor depresses the level of GDP (further details below).

Chart B: Past increases in interest rates are estimated still to be weighing on demand

Estimated contribution of the overnight indexed swap (OIS) curve to the level of GDP ^(a)

Footnotes

• Sources: Bank of England, ONS and Bank calculations.
• (a) The estimates for the contribution of the OIS curve to the level of UK GDP are the sum of the impacts on consumption, business investment, housing investment and the net imports offset in the MPC’s baseline projections. The net imports offset component reflects that part of the reduction in consumption from higher interest rates will arise from lower spending on imported goods, which mechanically pushes up GDP. These estimates are based on the standard treatment of the impact of changes in Bank Rate and the OIS curve in the Bank’s forecasting models. Bank staff estimate the impact of changes in the OIS curve between each Monetary Policy Report on the level of GDP, and these are then summed to obtain estimates of the total contributions. The estimates incorporate an impact from changes in the OIS curve on the level of GDP that persists over several years. The level of the aqua bars in the first data point in 2021 Q3 therefore reflects the estimated impact on demand of changes in the OIS curve from several years prior to that date. The effects of the estimated impact of changes in the OIS curve on sterling exchange rates have been excluded. The final data are for 2025 Q2.

Evidence from several channels of monetary policy transmission is consistent with the impact of past increases in Bank Rate on the level of GDP being around its peak.

One of the clearest indications that higher interest rates have weighed on demand has been the rise in the household saving ratio (Chart 2.10), which stood at 10.6% in 2025 Q1, up from 4.6% in 2022 Q2. Higher interest rates have incentivised additional household saving and lower borrowing over recent years. However, there are some tentative signs that the impact of higher interest rates on household saving may have peaked. Evidence from the March 2025 Bank of England/NMG survey suggests that fewer households across most income quintiles are increasing saving due to higher interest rates, relative to the peaks observed over 2023–24 (Chart C, left panel). And the saving ratio fell in 2025 Q1, although this largely reflected a reduction in real household incomes (Section 2.3).

Net secured lending to households has picked up in recent quarters, notwithstanding some recent volatility associated with changes in Stamp Duty Land Tax (SDLT). That is consistent with a rise in demand for borrowing due to the reduction in household interest rates since mid-2024 (Section 2.2), and has been associated with a pick-up in aggregate broad money growth (Box D of the May 2025 Monetary Policy Report). Lenders responding to the Credit Conditions Survey have noted an increase in demand for house purchases over the past few quarters, although that may have also reflected increased activity ahead of the changes to SDLT.

Evidence is also consistent with the impact of higher interest rates on house price inflation having peaked. House prices have fallen relative to household income over recent years, consistent with a drag from restrictive monetary policy (Box C of the August 2024 Monetary Policy Report). But house price inflation picked up in 2024, and estimates by Bank staff suggest that the drag on house price inflation from higher interest rates has fallen back a little over recent quarters.

Higher interest rates have also weighed on some components of housing investment, but there have been signs of recovery over recent quarters. Investment in new dwellings was weak over 2022–23, consistent with lower house prices having reduced the real returns on building new homes. And a lower volume of housing transactions has weighed on the transfer costs component of housing investment, which includes costs related to moving home, such as legal fees. But the recovery in house price inflation over recent quarters, and growing demand for house purchases, suggests that the monetary policy impact on the level of housing investment may be around its peak. Consistent with that, annual growth in private new housing construction output turned positive in late 2024, having been negative since early 2023. Despite this recent pick-up, intelligence from the Bank’s Agents suggests that growth in housing investment is likely to remain subdued, relative to historical norms, over the coming year.

Business investment is particularly sensitive to changes in interest rates, making this channel an important aspect of monetary policy transmission (Shah et al (2024)). Responses to the DMP Survey indicate that higher interest rates have suppressed firms’ capital expenditure growth since Bank Rate started to rise in late 2021. But this effect now appears to have peaked, with the proportion of firms stating that the cost of finance would constrain their capital expenditure over the next year falling from 46% to 38% between January 2024 and January 2025. There has been a gradual pick-up in net lending to corporates over recent quarters (Section 2.2 and Box D of the May 2025 Monetary Policy Report).

Analysis of growth in businesses’ gross value added (GVA), the total output of goods and services less the value of inputs used in production, also suggests that the impact of higher interest rates may have peaked. Bank staff analysis based on responses to the DMP Survey suggests that, between end-2022 and mid-2024, growth in GVA for the most interest rate-sensitive industries, including the utilities and construction sectors, was notably weaker than for other industries. But since mid-2024, GVA growth has converged across sectors with different estimated interest rate sensitivities, suggesting that the impact of higher interest rates may have peaked (Chart C, right panel). That said, past increases in Bank Rate continue to depress the level of demand for goods and services in the most interest rate-sensitive sectors. The most and least interest rate-sensitive sectors are difficult to identify accurately, and so this analysis should be treated with caution.

Chart C: Indicators of demand suggest that the impact of tighter monetary policy may be around its peak

Percentage of NMG survey respondents saving more than usual due to higher interest rates; annual GVA growth for industries that are most and least sensitive to higher rates ^{(a) (b)}

Footnotes

• Sources: Bank of England/NMG survey, DMP Survey, ONS and Bank calculations.
• (a) The data are from NMG surveys between March 2023 and March 2025. The data show the percentage of households in the NMG survey that reported that they had saved more than usual over the last 12 months and that higher interest rates on savings had been one of the main reasons for this increase. Households that responded ‘prefer not to state’ have been excluded from the calculation.
• (b) The latest data are for April 2025. The most and least interest rate-sensitive industries were identified from responses to the DMP Survey, using firms’ estimates of how much higher interest rates had reduced their sales. The most and least rate-sensitive industries are those in the upper and lower quartiles respectively of estimated interest rate sensitivity.

But the impact of higher interest rates on demand via the cost of existing borrowing continues to build.

While the effects of many channels of monetary policy transmission appear to be around their peaks, the household cash-flow channel continues to build. This channel captures the effects on consumption of increases in interest rates paid by existing borrowers and those received by savers. All else equal, increases in effective mortgage rates, relative to rates paid on household deposits, tends to weigh on household consumption.

The effective interest rate on the stock of UK mortgages is expected to rise a little further over the next year. Most borrowers hold their mortgage on a fixed-rate basis over a period of between two and five years. So, while the effective rate on the stock of UK mortgages has risen from around 2% to 4% since early 2022, the long lags in mortgage refinancing mean that the effective mortgage rate is expected to continue to rise (Chart D). Meanwhile, effective deposit rates have been falling since mid-2024, as reductions in Bank Rate have passed through to household saving rates. Overall, past rises in interest rates are expected to weigh further on consumption via the cash-flow channel over coming quarters.

Chart D: The effective interest rate on the stock of mortgages is expected to rise further over coming quarters

Effective interest rates on the stock of UK mortgages ^(a)

Footnotes

• Sources: Bank of England, Financial Conduct Authority Product Sales Data (PSD) and Bank calculations.
• (a) The data from March 2025 to June 2026 are quarterly projections consistent with the MPC’s August 2025 baseline forecast. The latest historical data are from the December 2024 PSD. The projections are informed by the Bank’s quoted mortgage rate data and assume that fixed-rate mortgagors re-fix at a new rate, the average of projected two and five-year fixed mortgage rates, once their incentive period ends. Variable-rate mortgagors are assumed to remain on variable-rate contracts.

Measures of the current monetary policy stance suggest that the degree of restrictiveness has fallen to some extent over the past 18 months.

While indicators of economic activity can be used to assess the extent to which past rises in Bank Rate continue to weigh on demand, they are less informative as to whether the current stance of monetary policy remains restrictive. Forward-looking estimates of the real-rate gap measure the difference between the current and future expected real policy rate, r, and the equilibrium real interest rate, r*. The latter is the time-varying real interest rate that, if the economy started from a position of no output gap and inflation at target, would sustain output at potential and inflation at target (Box C of the February 2024 Monetary Policy Report). A positive real-rate gap indicates that the current monetary policy stance will weigh on demand, once lags in the transmission process are accounted for, and is hence restrictive.

Bank staff monitor a range of indicators of the real-rate gap (Chart E). The degree of policy restrictiveness implied by each of these measures is highly dependent on the level of r* embodied within them. The equilibrium real interest rate cannot be observed directly, and its definition varies between the approaches used. Estimates encompassing a higher r* will, all else equal, signal a lower degree of restrictiveness.

Real-rate gap indicators are also sensitive to the measure of inflation expectations used in their estimation. Nominal measures of the interest-rate gap, which do not account for inflation expectations, can therefore provide a useful cross-check on the degree of restrictiveness.

Chart E: Forward-looking measures of the real-rate gap suggest that the degree of restrictiveness has fallen

Average real-rate gaps in February 2024 and August 2025 ^(a)

Footnotes

• Sources: Bank of England, Bank of England/Ipsos Inflation Attitudes Survey, Bank of England Market Participants Survey (MaPS), Bloomberg Finance L.P., Consensus Economics, Davis et al (2024), ONS, TradeWeb FTSE Gilt Closing Data and Bank calculations.
• (a) The February 2024 estimates for the macroeconomic models are for 2023 Q4. Corresponding estimates for the macrofinancial model are for 2024 Q1, and the estimates for financial markets and the MaPS are for February 2024. The August 2025 estimates for the macroeconomic models are for 2025 Q2. Corresponding estimates for the macrofinancial model are for 2025 Q3, and the estimates for financial markets and the MaPS are for August 2025. The macroeconomic models ranges are produced using estimates from two models. The first is a dynamic semi-structural IS curve model that relates the current output gap to past and expected future output gaps and the real-rate gap. It is estimated with data from 1988 Q3 to 2025 Q2. Average real-rate gaps over a three-year horizon are attained from a model using the Kalman filter, with associated ranges inferred from a dynamic factor model. The second set of macroeconomic model estimates are generated from one of the main models underlying the baseline projections in this Report (Albuquerque et al (2025)). Estimates from this model are also average implied real-rate gaps over a three-year horizon. The real policy rates used in both macroeconomic models are produced using the same model that underlies most of the baseline projections in this Report. The macrofinancial model ranges are estimated using the state-space model from Davis et al (2024), with a 95% confidence interval applied around the central estimates of r*. The real policy rates used in the macrofinancial model ranges are ex-post real rates, as described in Taylor (2025). The financial market ranges use estimates of the nominal neutral rate derived from three term premia models: the benchmark models in Malik and Meldrum (2016), Vlieghe (2016) and Meldrum and Roberts-Sklar (2015). These estimates are adjusted by average six-year to ten-year inflation expectations from half-year Consensus surveys to obtain real neutral rate estimates. The real policy rates in the financial markets ranges are calculated as the average instantaneous forward OIS curves over three-year horizons, deflated by annual Consensus survey inflation expectations over this period. These inflation expectations are linearly interpolated over the three-year horizon. The instantaneous forward OIS curves are estimated using the average paths for interest rates in the 15 UK working days to 23 January 2024 and 29 July 2025, on which the baseline projections for the February 2024 and August 2025 Monetary Policy Reports are conditioned, respectively. The ranges for the MaPS show the 10th–90th percentile of perceptions of the average implied real-rate gaps over a three-year horizon. The implied real-rate gaps are calculated as the difference between each MaPS respondent’s weighted mean nominal Bank Rate expectation, adjusted by their expectations for CPI inflation over the same period, relative to their perception of r*. The r* component is calculated from nominal equilibrium rate perceptions from the MaPS, applying two alternative assumptions around the deflator: the 2% inflation target or three-year CPI expectations from the MaPS. All measures of the real-rate gap presented in this chart show the implied degree of monetary policy restrictiveness over a three-year horizon. But to the extent that monetary policy can impact expectations of interest rates over longer horizons, changes to the monetary policy stance beyond the three-year point are not captured within the estimates on this chart.

Bank staff use macroeconomic models to estimate the degree of monetary policy restrictiveness. These estimates are based on dynamic structural and semi-structural models, which are underpinned by an inverse relationship between the output gap and the real-rate gap described by the investment-saving (IS) curve (Box C of the February 2024 Monetary Policy Report). These models estimate the current degree of monetary policy restrictiveness based on the average real-rate gap over the next three years. There is uncertainty around the results from these models, given that the estimation of the output gap itself is uncertain.

Estimates from these macroeconomic models suggest that monetary policy remains moderately restrictive (Chart E, aqua bar in right panel), but the degree of restrictiveness has fallen relative to its peak in early 2024. Reductions in Bank Rate have contributed towards a narrower real-rate gap, and hence lower restrictiveness, over that period. Meanwhile, the estimated level of spare capacity over the forecast period is currently smaller than was projected at the time of the February 2024 Monetary Policy Report, which has pushed up the implied level of r*. The rise in inflation expectations over recent quarters (Section 2.5) has also contributed to the fall in the estimated real-rate gap.

Bank staff also use data from financial markets and the MaPS to monitor investors’ perceptions of monetary policy restrictiveness (Box C of the February 2024 Monetary Policy Report). In line with the result from macroeconomic models, MaPS-implied estimates of the real-rate gap suggest that the restrictiveness of monetary policy has fallen since February 2024 (Chart E, gold bars). The market-implied range of the real-rate gap has widened over the same period (Chart E, purple bars). While both market and MaPS-implied estimates of the nominal interest-rate gap remain positive, estimates of the real-rate gap are now closer to zero or negative in some cases, suggesting that market participants judge that the monetary policy stance may no longer be restrictive.

There are caveats to these measures, however. Similar to other indicators, financial market and MaPS-implied measures of the real-rate gap are highly dependent on the measure of inflation expectations used. Moreover, financial market measures may provide more of a signal about deviations of the real policy rate from the long-run real equilibrium interest rate, R*, rather than the shorter-run concept of r* (Box C of the February 2024 Monetary Policy Report). That is because they incorporate expectations about future rates over long horizons. Estimates of R* derived from financial markets have risen by more than corresponding estimates from macroeconomic models over recent years (Box A of the February 2025 Monetary Policy Report).

The real-rate gap can also be estimated using macrofinancial models, which combine higher-frequency information from asset prices with lower-frequency trends in growth and inflation to derive estimates of r*. The real-rate gap is then calculated as the real policy rate less this estimate of r*. Estimates of the real-rate gap obtained using the macrofinancial model described in Davis et al (2024) suggest that the degree of monetary policy restrictiveness has fallen slightly relative to early 2024 (Chart E, orange bars). But monetary policy is estimated to remain restrictive based on this measure. And macrofinancial estimates of the nominal interest-rate gap, which do not adjust for inflation expectations, remain clearly positive.

Overall, while monetary policy continues to weigh on demand because of lags in transmission, the degree of policy restrictiveness has fallen.

Past rises in Bank Rate have contributed towards the disinflationary process by weighing materially on demand. Given lags in transmission, those past rises in Bank Rate are estimated still to be weighing on demand and will continue to do so over the forecast period, such that they contribute towards the ongoing disinflationary process. As the disinflationary process has continued and Bank Rate has been reduced, forward-looking measures of restrictiveness based on the real-rate gap indicate that the degree of restrictiveness has fallen.

Box B: Reviewing the process of quantitative tightening

Since February 2022, the MPC has been reducing the stock of assets held in the Bank of England’s Asset Purchase Facility (APF) for monetary policy purposes, a process known as quantitative tightening (QT). In line with the commitment made by the MPC in the minutes of its August 2022 meeting, this box sets out the MPC’s annual assessment of QT over the past year.

The Bank’s balance sheet grew after the financial crisis and the pandemic, as the MPC used quantitative easing (QE) to meet its 2% inflation target and to support the economy.

Following the 2008 global financial crisis, Bank Rate was reduced to close to zero to support economic activity and inflation. To further stimulate the economy when interest rates were at this level, the MPC began a programme of asset purchases known as QE (Chart A). QE works largely by reducing longer-term interest rates in the economy, ultimately boosting demand (Busetto et al (2022)).

Further QE programmes were conducted after the EU referendum and during the Covid pandemic. The size of the stock of assets held in the APF for monetary policy purposes peaked at £895 billion at the turn of 2022.

Chart A: The Bank’s APF holdings grew markedly through QE but have shrunk since 2022

Stock of gilts held for monetary policy purposes in the APF ^(a)

Footnotes

• (a) Gilt value is measured as the ‘initial purchase proceeds’ value.

Since 2022, the stock of purchased assets has been reduced and will reach £558 billion by September 2025.

In December 2021, the MPC started to increase Bank Rate in order to return CPI inflation sustainably to the 2% target. In 2022, once Bank Rate had been raised to a level that provided scope to reduce it as may be required, the Bank started to unwind its holdings of assets previously purchased as part of QE, a process known as QT.

From February 2022, the Bank stopped reinvesting maturing assets in the APF. Sales of gilts commenced in November 2022.

Reducing the size of the APF has the benefit of reducing the risk of a ratchet upwards in the size of the Bank’s balance sheet as a consequence of repeated asset purchases over time, were the economy to return to needing support if Bank Rate were to be at the effective lower bound. QT therefore increases the headroom and flexibility available to the Bank to use its balance sheet in the future if needed.

QT has comprised a mix of sales of gilts, maturing gilts and sales of corporate bonds. The size of each of these has varied from year to year (Chart B). Between October 2022 and September 2023, £20 billion of corporate bonds were sold. Gilt maturities have varied between £35 billion and £87 billion per year, reflecting the maturity profile of assets held by the APF. The balance of QT in 2024–25 was particularly skewed to maturities rather than sales, and the quantity of maturities will fall back in coming years. The size of the APF will fall to £558 billion by September 2025.

Chart B: The QT process has reflected a mix of sales and maturing assets

Initial purchase proceeds value of APF maturities and sales and maturities in future gilt stock reduction periods ^(a)

Footnotes

• (a) Each year shows maturities in the period between October and September of the following year (ie a yearly QT review cycle). For October 2022–September 2023 to October 2024–September 2025 these bars show the target gilt stock reduction voted for by the MPC, including both maturities and active sales, as well as sales of assets purchased under the Corporate Bond Purchase Scheme. In periods from October 2025–September 2026 onwards only the path of expected maturities is shown, as the Committee is yet to vote on the target gilt stock reduction for these periods.

The MPC’s QT process has been guided by three key principles…

The MPC has previously set out key principles under which QT should be conducted.

First, the MPC has used Bank Rate as its active policy tool when adjusting the stance of monetary policy.

Second, sales have been conducted so as not to disrupt the functioning of financial markets, and only in appropriate market conditions.

Third, to help achieve that, sales have been conducted in a relatively gradual and predictable manner over a period of time.

…which have helped to reduce the impact of QT on financial conditions.

The main way through which QT affects financial conditions is through the so-called portfolio rebalancing channel. This is because market demand for assets is not perfectly elastic, so the price of gilts, and other similar maturity assets, falls as the available supply increases. QT increases the supply of gilts to the market relative to a counterfactual where these gilts remain in the APF, although QT represents a relatively low share of the overall gilt market. Lower longer-duration asset prices mean long-term rates are higher than they would otherwise be. Long-term rates are affected by a wide range of factors, of which QT is only one, and movements in these rates have tended to be dominated by global factors over the past year.

The principles under which QT has been conducted help to limit financial market impacts outside this direct channel. Conceptually, QT could also affect gilt rates by changing market expectations for the future path of Bank Rate. This would be the case if market participants believed news about QT contained information about the future stance of monetary policy. As the MPC has been using Bank Rate as the active tool of monetary policy, this should not be the case. And market intelligence suggests that market participants’ Bank Rate expectations have not been affected by QT announcements. QT could, in principle, also have effects on gilt market liquidity and thus affect liquidity risk premia. However, as sales have been gradual, predictable and only conducted in appropriate market conditions, there has been no marked effect on market liquidity.

Estimates of the impact of QT on term premia suggest a modest increase in long-term interest rates of between 15–25 basis points since QT started.

Bank staff have updated their analysis of the effects of QT on gilt yields and the wider macroeconomy. While difficult to measure precisely, Bank staff continue to judge that the impact of QT on gilt yields has been modest.

Longer-term gilt rates are the sum of short-term rate expectations and a term premium, the additional interest return that investors require to hold long-term assets compared with purchasing equivalent short-term assets in sequence over the life of the long-term bond. Given that QT has been conducted in a way that does not communicate any changes to future Bank Rate, any impact of QT on gilt rates should come through higher term premia.

Between QT commencing in February 2022 and the end of June 2025, 10-year gilt rates rose by 325 basis points. Term structure model estimates indicate that around 150 basis points of this increase was driven by term premia, with the remaining rise accounted for by higher expected rates. Higher term premia have been driven by a wide range of factors including a structural reduction in future domestic demand for long-term government debt, for example due to a decline in demand from defined benefit pension schemes.

Since the 2024 QT review, government bond yields have risen across countries, particularly long-term rates. Over the past year, 10-year gilt yields have risen by around 55 basis points and 30-year gilt yields have risen by around 95 basis points in the UK. Similar, albeit somewhat smaller, moves have been seen across countries. For example, in the US 10-year yields rose a little over 20 basis points and 30-year yields rose 60 basis points. In Japan, 10-year and 30-year yields rose by around 50 basis points and 100 basis points respectively. And in the euro area, 10-year and 30-year yields have risen by around 25 basis points and 60 basis points respectively.

Market intelligence and term structure models point to higher term premia having more than accounted for higher yields in the UK and internationally, with models estimating that the total term premium for 10-year gilts over the last year has increased by around 70 basis points in the UK. Isolating the specific effect of QT is not simple because term premia are influenced by a wide range of domestic and international factors, and these factors change over time. One way to estimate the specific effect is to look at movements in term premia in short periods directly around announcements of QT policies, such as when the amount of QT planned for the forthcoming year is announced, or around the time of QT auctions, where bonds are sold to the market. This helps to identify the impact of QT because gilt market pricing tends to adjust quickly in response to market news, and the short window helps to limit the influence of other factors on the estimates.

It is challenging to identify the causal impact of QT policy announcements because there have been relatively few of them and so the sample size is limited. And, as QT is being operated in a predictable manner, QT announcements and auctions should contain little news for the market.

Notwithstanding the challenges in estimating these impacts, new analysis by Bank staff points to an estimated total increase in 10-year gilt rates from cumulative QT to date of 15–25 basis points. This estimate is based on modelling the change in the 10-year gilt rate term premium on the day of QT policy announcements, combined with the effects on the day of QT auctions. Given the overall increase in term premium for 10-year gilt yields, this suggests QT has driven between a tenth and a fifth of that total.

Bank staff’s most recent estimate of the impact of QT on term premia is slightly higher than in the 2024 QT review (Box A of the August 2024 Report), when the estimated range was 10–20 basis points. This is likely to reflect a small impact on term premia from the additional QT since September 2024. The estimated rise in term premia due to this additional QT is much smaller than the overall rise in term premia over this period.

There is uncertainty over these estimates of the impact of QT. One drawback of event study estimates is that they cannot identify effects of QT that fall outside of the short window used for the estimation. For example, Bank staff analysis of past auction sales suggests that rate impacts have tended to be short lived and reverse over subsequent days and so the component of the estimates that reflects impacts on auction days could be overstated. In contrast, were the impact of QT to only build up slowly over time, or only operate with a lag after policy announcements or auctions, the total impact of QT on gilt rates could be underestimated.

Long-term gilt yields, such as those for 30-year gilts, have risen by more than 10-year gilt yields. Bank staff analysis suggests that the larger increase in longer-term rates has been driven primarily by global factors, rather than being attributable to QT. Furthermore, analysis of changes in term premia for 30-year gilts, based on the same model used for the estimated range above, suggests that QT impacts have been similar to those for 10-year gilts. Lower market liquidity for these assets means that the model-based estimates are more uncertain for this maturity.

Besides its impact on long-term interest rates through the portfolio rebalancing channel, staff analysis and external research suggests that QT could affect bank lending, especially as central bank reserves become less abundant in the system (Altavilla et al (2024)). The process of QT reduces reserves in the banking system which, among other purposes, are used by banks to insure against liquidity risk. As reserves are risk-free and the most liquid form of asset, fewer available reserves could weigh on bank lending growth. The Bank is currently transitioning to a repo-led demand-driven framework for supplying reserves, which will ensure that banks can meet their demand for reserves through use of the Bank's repo facilities (Bank of England (2024)). This framework should mitigate risks to bank lending from QT through this channel.

QT auctions have had little impact on market functioning.

There is little evidence of a material impact of QT auctions on broader market conditions. For example, the average range over which gilt rates vary on days with QT auctions is less than one basis point higher than the average range observed on days without QT auctions. Were auctions to affect market liquidity negatively, we would expect to see larger variation in rates at those times. The small impact on market liquidity will partly reflect the fact that market participants are able to adjust for the effects of QT auctions in advance, since they are announced well ahead of time. In line with the principle that QT should be conducted in appropriate market conditions, in April 2025 the Bank amended the schedule of gilt sales by delaying sales of long maturity bonds following a period of international market volatility.

External estimates of the impact of QT on rates are broadly consistent with Bank staff’s analysis, but comparing with estimates for other countries is difficult due to differences in the implementation of QT programmes.

External research on the impact of QT programmes is broadly consistent with Bank staff’s analysis. For example, Du et al (2024) find that QT programmes in various countries have had limited impact on government bond yields, market functioning and liquidity. External estimates of the impact of QT rely heavily on event study estimates. That means they have similar limitations to the Bank staff analysis outlined above. This is particularly important for the UK given the gradual and predictable approach taken for QT (Lu and Valcarcel (2024), D’Amico and Seida (2024)).

More generally, differences in the implementation of QT between the UK and other jurisdictions make it difficult to compare estimates of the impact of QT internationally. For instance, the Federal Reserve has indicated that shrinking its balance sheet through QT is an active monetary policy tool (December 2021 FOMC minutes), in contrast to the MPC’s principles outlined above.

When the MPC sets Bank Rate, it takes account of financial market conditions that reflect the effects of announced and expected APF reductions.

The small rise in market interest rates driven by QT will have a slight contractionary effect on GDP and will slightly reduce inflation. There are large uncertainties around these estimates, but Bank staff analysis based on the UK's experience of QE suggests that the impact of a 10 basis point increase in yields would be associated with a negative impact on GDP and inflation of less than 0.2% and 0.1 percentage points respectively. To the extent that asset prices at the time of the Committee's policy assessments incorporate the impacts of announced and expected APF reductions, the MPC takes account of these impacts when taking its decisions on Bank Rate. The MPC will adjust Bank Rate such that overall financial conditions are appropriate to bring inflation back to the 2% target. It retains the ability to set the monetary stance to meet the 2% inflation target at all times.

The effect of QT may vary over time and estimates of its impact are uncertain.

Global economic policy uncertainty, high issuance of government bonds across countries and structural changes within the domestic bond market which have reduced demand for long-term government debt, have all pushed up on bond term premia.

These same shifts in the gilt market could pose a risk that QT has a greater impact on market functioning than previously. In an environment with lower demand for long-term assets, for example, QT could have a larger impact on market liquidity.

Measures of gilt market functioning suggest that the market has generally continued to function in an orderly manner. However, reflecting changes in the profile of underlying demand for gilts and long-term cost and risk factors, the Debt Management Office (DMO) amended its 2025–26 financing remit by reducing sales of long-term bonds and increasing short-term financing. The DMO also increased the proportion of issuance for which an issuance method has yet to be decided, in order to provide additional flexibility to respond to evolving demand and market conditions during the financial year.

The MPC will continue to monitor the impact of QT. The analysis summarised in this box will support the MPC’s decision over the appropriate pace of gilt stock reduction over the year ahead.

At its September meeting the MPC will vote on the target for the reduction in the stock of UK government bonds held for monetary policy purposes over the 12-month period from October 2025 to September 2026. Bank Rate remains the MPC’s active tool of monetary policy and will be set to meet the 2% inflation target sustainably in the medium term.

Box C: Monetary policy since the May 2025 Report

At its meeting ending on 18 June 2025, the MPC voted by a majority of 6–3 to maintain Bank Rate at 4.25%. Three members preferred to reduce Bank Rate by 0.25 percentage points, to 4%.

There had been substantial disinflation over the past two years, as previous external shocks had receded, and as the restrictive stance of monetary policy had curbed second-round effects and stabilised longer-term inflation expectations. This had allowed the MPC to withdraw gradually some degree of policy restraint, while maintaining Bank Rate in restrictive territory so as to continue to squeeze out existing or emerging persistent inflationary pressures.

Underlying UK GDP growth appeared to have remained weak and the labour market had continued to loosen, leading to clearer signs that a margin of slack had opened up over time. Measures of pay growth had continued to moderate and, as in May, the Committee expected a significant slowing over the rest of the year. The Committee remained vigilant about the extent to which easing pay pressures would feed through to consumer price inflation.

Twelve-month CPI inflation had increased to 3.4% in May from 2.6% in March, in line with expectations in the May Monetary Policy Report. The rise had been largely due to a range of regulated prices and previous increases in energy prices. Consumer price inflation was expected to remain broadly at current rates throughout the remainder of the year before falling back towards target next year.

Furthermore, global uncertainty remained elevated. Energy prices had risen owing to an escalation of the conflict in the Middle East. The Committee would remain sensitive to heightened unpredictability in the economic and geopolitical environment, and would continue to update its assessment of risks to the economy.

CPI inflation was 3.6% in June, 0.2 percentage points higher than expected in the May Report. Mainly reflecting unexpected strength in food prices, CPI inflation is now projected to peak at 4.0% in September before falling back to 3.6% by the end of the year.

Services consumer price inflation remains elevated. While underlying disinflation in services prices seems on balance to be continuing, any progress is being masked temporarily by upward pressure from administered prices and the impact on some prices of the increase in employer NICs. Indicators of households’ inflation expectations remain elevated, while inflation expectations of firms have been more stable.

Pay growth continues to ease from elevated rates and is projected to slow to around 3¾% by the end of this year from around 5% currently.

Developments in the labour market are consistent with a continued gradual loosening. Underlying employment growth has been around zero in recent months and the degree of slack in the labour market appears to have widened. Spare capacity within firms has also increased.

Underlying momentum in UK GDP has been subdued but is expected to pick up a little in the near term. The household saving ratio remains elevated relative to its pre-Covid pandemic average.

Developments in trade policy in the US and elsewhere continue to be a key influence on global GDP growth. The future constellation of policies remains difficult to predict, but US tariffs are currently expected to be lower on average than was assumed at the time of the May Report.

Sterling had appreciated slightly in the period leading up the August Report, compared with three months earlier. Over the same period, risky asset prices had risen and the market-implied paths for policy rates in the UK, US and euro area had been little changed.

Box D: Global uncertainties and their implications for the UK economy

The MPC’s baseline projections have been produced conditional on various external assumptions, including international energy prices developing in line with market futures pricing and global tariff rates remaining at their levels as of 29 July. However, both energy prices and trade policy announcements have varied significantly since the publication of the May Report.

The first part of this box investigates the sensitivity of the MPC’s baseline projections to an unanticipated rise in global energy prices. The second part investigates the sensitivity of the baseline projections to both a higher level of global tariffs and to different assumptions around the impact of higher tariffs, including via higher trade policy uncertainty and a greater impact of trade restrictions on world export prices. The higher tariff assumption amplifies the economic impact of different modelling assumptions and therefore demonstrates their relative significance.

1: Higher energy prices

Conflict in the Middle East led to rises in energy prices in June. While energy prices have fallen since then, market pricing suggests upside risks to prices remain.

Military action between Iran and Israel, later joined by the US, in June 2025, led to a peak increase in dollar Brent crude spot oil prices of around 20%. This was a large rise, though not outside of the normal range of historical oil price changes (Chart A). Spot natural gas prices rose a little further, by around 25%.

A ceasefire came into effect, and energy prices have fallen back. Brent crude spot oil prices are around $70 per barrel. However, geopolitical tensions remain. Option pricing for oil futures, which can be used to calculate a market-implied estimate of the probability of oil reaching a given price in the future, is consistent with there being a risk that renewed conflict pushes oil prices up again.

Wider conflict that affects both Iran and the Strait of Hormuz could have large effects, as this is a key route through which oil and natural gas are exported from other major energy producing countries. Supplies through the Strait of Hormuz were not disrupted during the conflict in June and the risk of disruption appears to be low, but a range of external estimates of the impact of widespread disruption or the cessation of oil exports through this route suggests global oil prices could rise to between $100–$130 per barrel in this case. Liquified natural gas prices would also be materially affected.

Chart A: The oil price increase in June 2025 was large, but even larger oil price increases have occurred before

Frequency distribution of historical four-week US dollar oil price changes ^(a)

Footnotes

• Sources: Bloomberg Finance L.P. and Bank calculations.
• (a) Data cover June 1987 to 29 July 2025. Oil prices are dollar Brent crude. Changes are calculated for each business day compared to 20 business days earlier. The bars represent oil price changes in 5 percentage point buckets, the number shown represents the upper bound of that bucket with the exception of the rightmost bar.

Higher energy prices would push up UK CPI inflation through direct effects on fuel and utility prices, as well as indirect effects through higher production costs for firms.

As part of its recent discussions, the MPC considered the sensitivity of its baseline projections to an unanticipated future rise in energy prices. In doing so, the Committee drew on experience from the rise in energy prices following Russia’s invasion of Ukraine in 2022. Higher energy prices would push up consumer price inflation and reduce GDP.

Energy prices have a direct effect on CPI inflation through petrol and diesel prices, and utility bills. Motor fuel pump prices make up 2.8% of the consumer price inflation basket of goods and services. And gas and electricity make up a further 3.4%. Changes in oil prices tend to be passed through quickly and in full to pump prices. Wholesale gas prices feed through directly, albeit with a lag, to household electricity and gas bills. This is because the Ofgem energy price cap is set quarterly with reference to previous wholesale prices. The impact on utility bills would be expected to take around three quarters to fully take effect.

Energy prices also have an indirect effect on CPI inflation through the use of oil and gas as inputs for other goods and services. Higher energy prices push up production costs, which would be expected to be passed through to consumer prices over time. For example, higher gas prices raise UK electricity costs, but since most businesses have fixed-term contracts, it will take time for these to affect businesses’ costs.

The shorter-term impacts of a hypothetical 10% rise in oil prices on CPI inflation and GDP, through both the direct and indirect channels outlined above, are depicted in Chart B. The middle panel shows the modelled response of UK CPI inflation to a global oil price shock from one of the key models that feeds into the baseline projections outlined in this Report, alongside empirical estimates from a local projections specification (Albuquerque et al (2025)).

In this estimation, a 10% peak increase in oil prices (Chart B, left panel) is associated with a peak increase in CPI inflation of 0.5%. That rise in CPI inflation occurs swiftly after the rise in oil prices. These results assume that the oil price shock evolves in line with historical patterns, which show that oil price shocks tend to be temporary, with prices falling back after peaking.

The estimates in Chart B show the impact of a rise in oil prices. Additional gas price rises would exacerbate any subsequent increase in CPI inflation. Previous Bank staff analysis shows that the importance of gas prices for CPI inflation and GDP has grown relative to oil prices over time.

Chart B: Oil price shocks are estimated to raise UK CPI inflation and reduce GDP

Impulse responses to a global oil price shock ^(a)

Footnotes

• Sources: Albuquerque et al (2025) and Bank calculations.
• (a) DSGE refers to the estimates taken from one of the main models which underlies the baseline projections in this Report. The local projections estimates are based on a sequence of regressions where the response of economic outcomes, including GDP and CPI inflation, is estimated conditional on energy price shocks. The dashed lines are 90% confidence bands around the local projection estimates. The results show impulse responses to a 10% global oil price increase. Estimates are based on data from 1987–2023.

The impact of higher energy prices on inflation could be exacerbated by second-round effects on domestic wages and prices, for example if inflation expectations were to rise.

The impact of energy price changes on CPI inflation will depend on broader economic conditions, including the behaviour of inflation expectations and the level of spare capacity in the economy.

Some evidence suggests that inflation expectations may be particularly sensitive to rises in energy costs. Bank staff analysis finds evidence of asymmetrical formation of inflation expectations, with rises in prices shifting expectations more than falls in prices. A high inflation environment can mean households pay more attention to news about inflation. And evidence shows that the prices of some products, including petrol, are particularly important for households when they form their inflation expectations (Chart C in Box E).

If inflation expectations were to rise in response to an energy cost increase by more than was typical in the past, this could lead to additional inflation persistence, prolonging the period of higher inflation. Higher inflation expectations for households could lead to upward pressure on aggregate wage growth, which would in turn raise businesses’ costs and potentially lead to further price rises. And higher inflation expectations for firms could be internalised into their own price-setting where anticipated future cost increases are reflected in current prices. Higher inflation expectations at a time of tight labour markets were a significant factor in the greater inflation persistence that followed the 2022 energy price shock.

The estimates in Chart B account for the average response of CPI inflation to higher energy prices over the past, and so implicitly capture the average impact of energy prices on inflation expectations and subsequent feedthrough to wage and price-setting. Any additional sensitivity of inflation expectations in the current economic environment would lead to more inflation persistence, to which monetary policymakers would respond appropriately.

In an environment where demand is weak relative to potential supply, however, firms may be more likely to absorb additional energy costs into their margins than if there is little spare capacity. As noted in Section 2.4, there is growing evidence of spare capacity within the economy. This could act against any additional inflation persistence from the interaction between higher inflation expectations and firms’ price-setting behaviour.

Higher energy prices also reduce GDP and can generate a trade-off for monetary policy between returning inflation to the target and supporting GDP growth and employment.

Alongside their effect on inflation, higher energy prices would also reduce UK GDP. Chart B (right panel) shows the response of UK GDP using the same baseline projection model and empirical estimates.

Following a swift hypothetical rise in energy prices, GDP falls in the subsequent quarters before recovering. There is a large degree of uncertainty over the magnitude of the GDP impact. In the DSGE estimates, the impact on GDP is relatively muted, with a temporary rise in energy prices of 10% reducing GDP by around 0.1%. That muted response reflects the fact that firms and households are assumed to substitute quickly towards non-energy inputs and goods as energy becomes more expensive. The local projections estimates indicate much larger declines in GDP, consistent with historical correlations.

As the experience of the 2022 energy price shock demonstrated, one of the challenges in understanding the impact of energy price rises is assessing how they are transmitted through the economy. The extent to which firms and households can substitute away from using energy is important and might change over time. As most UK homes rely on natural gas for heating their home, one simple example of this issue is the temperature experienced during winter. Cold weather could reduce households’ ability to reduce their gas usage.

Lower GDP from higher energy prices reflects both weaker supply and demand. Demand would be expected to fall by more than supply, however, such that the degree of spare capacity widens. The model underlying the baseline projections suggests that supply is reduced by around a fifth of the overall reduction in GDP. Combined with higher inflation, this creates a trade-off for monetary policy between returning inflation to the target and supporting GDP growth and employment.

There is uncertainty over the size of any impact of higher energy prices on supply, however, and hence spare capacity. The impact on supply could be greater than implied by this model if the energy shock is very large and results in supply chains needing to be reorganised to account for higher energy prices.

The appropriate monetary policy response to an energy price shock will depend on the size and timing of the impact on CPI inflation, as well as on the level of the output gap. To the extent that CPI inflation rises while the output gap widens, the shock will create a trade-off for monetary policy makers. If the shock is short-lived, as in the example above, monetary policy may not need to respond. But if the shock is larger and more persistent, as occurred after the 2022 Russian invasion of Ukraine, a higher level of Bank Rate could be required to meet the inflation target.

2: Greater global trade restrictions

There remains significant uncertainty over future global trade policy.

Since March 2025, the US government has announced a range of new tariffs for countries and products. The US effective tariff rate at the time the baseline projection was produced was estimated at around 14%, 12 percentage points higher than in January, but 8 percentage points lower than at the time of the May Report (Chart C).

As part of its recent discussions, the MPC considered the sensitivities in the baseline projections to both higher global tariffs and to different assumptions around the impact of higher tariffs, including via higher trade policy uncertainty and a greater impact of trade restrictions on world export prices. For the purposes of these sensitivities, the US effective tariff rate was assumed to climb to over 35% (Chart C). Other countries were assumed to respond to higher US tariffs, with significant retaliation from China and a more limited retaliation from other countries.

This exercise was not intended to model a particular likely outcome for global tariff rates. Indeed, some of the assumptions outlined in the footnote to Chart C are different to those that have been announced, including the US-EU deal on 27 July. That deal would lead to US tariff rates on goods from the EU at 15% rather than 30%, although its impact on the sensitivities below would be small. Nevertheless, the analysis remains valuable for exploring both the impact of a higher set of tariffs and the accompanying model sensitivities.

Chart C: US tariff rates have increased since the start of this year

Estimated US effective tariff rates ^(a)

Footnotes

• Sources: US Bureau of Economic Analysis, US International Trade Commission, White House and Bank calculations.
• (a) The alternative assumptions diamond was calculated as the combination of the US tariff letters sent in the first half of July, which would have raised the US effective tariff rate by around 4 percentage points, as well as US product-specific tariffs, such as pharmaceuticals, and a re-escalation of trade tensions with China. The May 2025 Report diamond is the tariff level assumed for the May Report baseline projections. And the August 2025 Report diamond is the same for this Report and is based on tariff rates implemented as of 29 July.

There are uncertainties in how any given set of global trade policies will impact UK GDP…

As set out in Box C of the May 2025 Monetary Policy Report, higher bilateral global tariffs between the US and other countries would be expected to reduce global demand for UK exports and UK GDP. This reflects: a direct reduction in US imports from the UK due to higher tariffs; a decline in real incomes in affected countries that reduces demand for imports more broadly; and estimates for shifts in global trade patterns such as countries selling their goods into new markets.

Estimates of the relationship between trade restrictions and GDP from different economic models vary. Using the same estimated relationships that underlie the baseline projections, the higher global tariffs set out above would reduce UK GDP by a little over an additional 0.2% over the next three years. Around two thirds of that impact would be expected to materialise over the course of the next year, with the remaining impact appearing evenly over the following two years. There are also likely to be longer-run impacts on global and domestic productivity growth.

There are important uncertainties around these estimates, including over the relative importance of different mechanisms through which higher tariffs impact the economy as well as around model estimates of their impact.

One area that continues to make economic impacts difficult to estimate is the extent to which uncertainty over future trade policy reduces GDP. In an environment where trade policies are uncertain, businesses may be less willing to invest in new production capacity and there could be a wider impact on business and household sentiment. Measures of policy uncertainty have fallen back from their peaks in April but could rise or fall in the future. If measured global trade policy uncertainty were to rise back to the peak earlier in the year and gradually decline over the next year and a half, estimates suggest that would reduce UK GDP by around 0.1%.

…as well as prices.

Within the models used for this analysis, the quantitatively largest channel through which trade restrictions affect UK prices is their effect on world export prices. Weaker global demand for exports puts downward pressure on world prices which then passes through into UK import prices. Based on the higher tariff levels discussed above, world export prices could fall by an additional 2% over the next three years, which would cause a peak reduction in UK CPI inflation of around 0.1 percentage points.

Another channel through which global trade restrictions affect domestic prices is the impact of weaker global demand on spare capacity in the UK. By raising prices of imports, higher tariffs reduce real incomes in those countries where they have been increased, which will weigh on demand for UK exports. All else equal this will lead to additional spare capacity in the UK. Higher tariff levels discussed above could reduce annual inflation by a little under 0.1 percentage points through this channel.

One uncertainty regarding the future path of world export prices relates to a possible change in the pattern of trade flows due to new trade restrictions. Specifically, the impact of lower global export prices on UK inflation could be compounded if countries formerly exporting goods to a third country attempt to shift some of this trade to the UK. This may necessitate a fall in the prices of these goods. This effect could be amplified if other countries respond to this additional competition by also reducing their export prices to the UK.

The extent to which export flows shift to the UK in response to US and retaliatory tariffs is difficult to project. Bank staff analysis of the impact of tariff increases on trade flows during previous US administrations suggests limited evidence for significant shifts in export flows to the UK in previous episodes of higher tariffs. This suggests that there may be limited impacts on UK import prices beyond the effects on global export prices mentioned above. However, as announced tariff increases have been larger than those previous examples, there could be larger effects in the current period. Monitoring high-frequency trade flows data is important for understanding how these effects may materialise in the current economic environment.

One way to assess the potential importance of this channel is to adjust the assumed import price elasticities underlying the baseline projections. It is possible the lost US demand could lead to bigger reductions in exporters’ prices if demand from the rest of the world proves less responsive to price cuts. Assuming a non-US import price elasticity near the bottom of the range of empirical estimates, combined with spillovers from shifting trade patterns leading to a greater increase in global price competition, would be enough to further reduce UK CPI inflation from the assumed higher tariff level above by around 0.1 percentage points in 2027.

The extent to which financial market responses amplify or offset higher global trade restrictions is also uncertain.

The sensitivities above are also dependent on the financial market reaction to changes in trade restrictions. In particular, the assumed trade policy retaliation of other countries to higher US tariffs is important for the estimated financial market effects of higher tariffs.

One important financial market channel for the impact of global trade restrictions on UK GDP and inflation is how much exchange rates move in response to those restrictions. It is difficult to predict movements in the sterling exchange rate in response to future tariff developments, however, and any movements will depend on the extent of retaliation as well as broader market risk sentiment.

Broader financial market moves could plausibly amplify the negative impact of higher global trade restrictions on UK GDP and inflation. A combination of higher uncertainty and weaker global growth would be expected to reduce risky asset prices, for example equity prices. These changes in risky asset prices would also drive a reduction in GDP. Plausible estimates of this channel, based on the higher tariffs in Chart C, suggest a further reduction in GDP of around 0.2% after three years and a small reduction in CPI inflation. These estimates were derived from the ECB-Global semi-structural model and the Bank’s main forecasting model.

The MPC continues to monitor the effects of changing global trade restrictions and the associated impact on the UK.

Significant uncertainty remains over future global trade policies, as well as the extent to which changing trade policies affect the UK inflation outlook. The MPC continues to monitor the effects of trade restrictions as they materialise in the UK. This includes the extent to which the economic impacts appear to be in line with model-based estimates and are materialising through the expected channels outlined in this box. The MPC will learn more about the true impacts over time, as some effects are likely to emerge relatively slowly.

Box E: The outlook for food price inflation

Food consumer price inflation rose to 4.5% in June, materially higher than expected at the time of the May Report. While food price inflation remains much lower than the double-digit rates seen in 2022 and 2023, it is well above its pre-Covid average of around 1½%. Food accounts for about 11% of the overall consumption basket of UK households, and there is evidence that food prices can have an outsized effect on households’ inflation perceptions and expectations (Anesti et al (2025) and Bonciani et al (2024)).

This box discusses the current drivers of food price inflation, including global agricultural commodity prices as well as domestic factors such as rising labour costs and regulatory changes. The box then sets out an outlook for food price inflation, informed by conversations with the Bank’s Agents’ contacts in the food industry, before discussing the link between food price inflation and household inflation expectations.

Some agricultural commodity prices have increased sharply this year, pushing up food price inflation.

UK food price inflation is materially affected by movements in global agricultural commodity prices. That is because many food items are globally traded goods, with their prices largely determined in international markets. For instance, the UK imports large amounts of agricultural products that cannot be produced domestically, such as cocoa and coffee beans. But there is also significant trade in agricultural goods that are domestically produced. For example, the UK imported around 240 thousand tonnes of beef in 2024, mainly from Ireland, while exporting 130 of the 900 thousand tonnes of beef produced domestically, mainly to continental Europe.

The S&P Global GSCI global agricultural price index has increased by around 8% over the past year. That is materially above the average annual increase of 2.8% since 1995, but well below the increase of over 60% observed in 2021.

Within this, cattle prices have risen by almost 20%. Several contacts noted that dry weather conditions are pushing up the production costs of beef and some dairy products in the UK and elsewhere, as cows must be fed silage earlier in the year due to less grass growth.

There have also been large increases in the wholesale prices of coffee beans and cocoa over the past two years. Droughts in major coffee-producing regions like Brazil have led to reduced crop yields, while heavier than usual rainfall and the outbreak of plant diseases have contributed to smaller cocoa harvests in West Africa. At the same time, there is evidence that growing demand for coffee in emerging markets such as China has contributed to the rise in global coffee bean prices.

In contrast, the wholesale prices for sugar and olive oil have fallen materially. The latter is driven mainly by more favourable weather conditions in Spain, which produces almost half of the world’s olive oil, with production recovering after droughts in recent years.

Changes in wholesale prices are passed through to retail prices with a lag, in part reflecting delivery contracts that are made a year in advance. That said, pass-through for perishable goods such as food has historically tended to be faster than for durable consumption goods (Box D of the November 2024 Monetary Policy Report). The pace of wholesale price increases has slowed in recent months, which is expected to reduce pressures on retail food price inflation in 2026 if sustained.

Movements in global agricultural prices have been important drivers of recent food consumer price inflation in the UK. In June, UK CPI subcomponent inflation rates for beef, butter, chocolate and coffee were elevated, while the inflation rates of olive oil and sugar were negative (Chart A), broadly consistent with recent movements in global wholesale prices.

Chart A: Inflation rates for some common food items have reflected movements in global prices

Annual inflation rates for selected food subcomponents ^(a)

Footnotes

• Sources: ONS and Bank calculations.
• (a) The chart shows annual inflation rates for selected subclasses (COICOP 5) of the CPI basket in June 2025. The labels include the subclass weights within food and non-alcoholic beverages.

To the extent that food prices are driven by global agricultural prices, they should tend to move in the same direction across countries. As Chart B illustrates, UK and euro-area CPI food price inflation rates have moved closely together in recent years. Staff analysis also finds that in recent months, the retail prices of individual food items in the UK and euro area have moved in the same direction, pointing to underlying commodity prices being a common driver.

Chart B: Food price inflation has recently picked up by more in the UK than in the euro area

Annual food price inflation ^(a)

Footnotes

• Sources: Eurostat, ONS and Bank calculations.
• (a) The food price inflation rates shown here include non-alcoholic beverages. The final data points are for June 2025.

Nevertheless, food price inflation is currently over 1½ percentage points higher in the UK than in the euro area. Compositional differences between the UK and euro-area CPI food baskets could in principle explain some of that difference. But reweighting euro-area food inflation with UK item weights suggests that this effect is currently small. There are instead several UK-specific factors that could account for the difference, namely higher increases in labour costs and the forthcoming introduction of the framework for the Extended Producer Responsibility (EPR) for packaging discussed below.

Higher labour costs are contributing to food price inflation in the UK.

In addition to global agricultural commodity prices, domestic labour costs are currently an important driver of food price inflation. UK wage growth has moderated but remains significantly higher than in the euro area (Section 2.5). And annual regular pay growth has been particularly strong in the food manufacturing sector (7.0%) and in the retail sector (7.6%), compared with 4.9% for aggregate private sector regular AWE growth in the three months to May. That is in part because a relatively high proportion of staff in these sectors are paid at or close to the National Living Wage (NLW), which increased by 6.7% in April. Furthermore, overall labour costs of supermarkets are likely to have been disproportionately affected by the lower threshold at which employers start paying NICs, in part because a relatively high proportion of supermarket staff is employed part-time.

These material increases in labour costs are likely to have pushed up food prices. According to the Bank’s Agents’ contacts, most of the labour cost increases associated with the NLW, but only some of the costs associated with the employer NICs increase, have so far been passed on to consumers, which has added around 1%–2% to food prices.

Labour costs are expected to continue to push up food prices in the second half of the year, although firms may not achieve full pass-through. Contacts noted that consumers were already responding to high food price inflation by trading down, limiting retailers’ ability to increase prices. For example, sales of own-label items had increased at the expense of branded products. Customers were also increasingly purchasing cheaper cuts of meat and buying food staples in larger value packs. To reduce the need for higher prices, many firms along the supply chain were trying to mitigate cost increases, including through reductions in headcount. Many contacts also noted that cost pressures were putting further pressure on profit margins, which were already compressed.

A new framework to reduce packaging waste is expected to push up food price inflation further in the second half of the year.

The Environment Act 2021 introduced the EPR framework for packaging to help the UK achieve its environmental goals. This framework transfers the costs of packaging waste management from local government to the businesses that produce the packaging. By placing these costs on producers, the policy is intended to incentivise reductions in the use of packaging materials and encourage recycling. The UK government announced the final details of the scheme in June, and affected firms are due to receive their first EPR invoices in October. A similar packaging waste regulation has been introduced in the EU and will come into force in August 2026, which may push up euro-area food price inflation somewhat next year.

Because packaging costs represent a higher share of food and drinks costs compared with many other goods, the packaging levy is expected to affect CPI inflation mostly via food prices. Bank staff estimate that the scheme could add a little over ½% to the level of food prices if these costs were fully passed through to consumers.

The Bank’s Agents’ contacts report that some suppliers started passing on the expected costs of EPR in the first half of the year, ahead of the scheme’s implementation. This may be because some UK grocers are increasingly operating on longer-term contracts with major UK food suppliers, which is leading suppliers to be more forward-looking in their pricing strategies. That said, most of the pass-through is still likely to occur towards the end of the year, which will push up somewhat on food price inflation.

Food price inflation is expected to rise to around 5.5% by the end of the year, before falling back gradually in 2026.

Bank staff expect food price inflation to be around 5% in 2025 Q3, 1½ percentage points higher than expected in the May Report, before rising further to 5½% by the end of this year (Chart 2.19). Food price inflation is projected to account for about 0.5 percentage points of the expected deviation in CPI inflation from the 2% target in H2. Bank staff expect food price inflation to fall back gradually next year as pressures from labour cost increases fade and global wholesale food price inflation returns to historical averages. This is broadly in line with intelligence from the Banks’ Agents, who report that most contacts expect food inflation to approach 5%–5.5% in the second half of the year before falling back to around 2%–3% in 2026.

There are material uncertainties around this projection. Consumers could continue switching to lower cost items to a greater degree than expected, which would put additional pressure on retailers to limit price increases. Acting in the opposite direction, global agricultural commodity prices could increase further, for instance if higher energy prices were materially to raise key input costs, or extreme weather events due to climate change were to negatively impact food production (Parker (2025)). Several contacts also noted that consumers had become more conscious of the environmental and animal welfare implications of their food consumption. This change in consumer preferences was already contributing to higher price pressures and was likely to continue to do so in the future. Higher-welfare animal products are more expensive to produce, and the necessary changes to supply chains can take several years and require substantial investment.

Households’ inflation expectations can be strongly influenced by developments in food prices.

The MPC monitors food prices closely, in part because they can have an outsized influence on inflation expectations. Household expectations for future inflation comove closely with perceptions of current inflation, which in turn are sensitive to price changes observed in people’s daily lives, and particularly to changes in salient items such as food (D’Acunto et al (2021) and Anesti et al (2025)).

The currently elevated level of households’ near-term inflation expectations (Section 2.5) is likely to have been driven in part by high food price inflation. Recent Bank staff analysis extending Anesti et al (2025) shows that a 1 percentage point increase in the contribution of food prices to CPI inflation was historically associated with a roughly 1 percentage point increase in the Citi/YouGov measure of one year ahead household inflation expectations. However, since 2022 this effect has increased to 1.8 percentage points. That is consistent with previous Bank staff work, which suggested that households’ inflation expectations had become more sensitive to aggregate price changes since the most recent period of very high inflation (Section 2.5 of the May 2025 Monetary Policy Report). Household inflation expectations have also become more sensitive to petrol prices but remain relatively less sensitive to broader core inflation (Chart C).

Chart C: Households’ near-term inflation expectations have been particularly sensitive to developments in food price inflation

Estimated average response of household inflation expectations to changes in selected CPI components ^(a)

Footnotes

• Sources: Citigroup, ONS, YouGov and Bank calculations.
• (a) The chart shows the estimated coefficients from regressions of changes in Citi/YouGov one year ahead household inflation expectations on changes in the contributions of food price inflation, energy price inflation and core CPI inflation to headline CPI inflation. The pre-2022 estimation period runs from 2005 Q1 to 2021 Q4, while the post-2022 estimation period runs from 2022 Q1 to 2025 Q2.

The MPC will continue to monitor food price inflation, including potential effects on inflation expectations and any resulting risks to the disinflation process.

CPI inflation has risen in recent months, driven in part by higher food price inflation. While households’ near-term inflation expectations have fallen back slightly in the latest data, they remain elevated. That is likely to be due, in part, to the rise in food price inflation combined with households’ attentiveness to food prices.

The expected further increases in food price inflation, against a backdrop of elevated headline inflation, means that inflation expectations could remain elevated or even increase further in the second half of the year. This could raise the risk that economic mechanisms such as those outlined in the higher inflation persistence scenario in Box A of the May 2025 Monetary Policy Report could begin to operate. The MPC will monitor closely developments in inflation expectations measures, including any risks they could pose to the ongoing disinflation process. The MPC will respond appropriately to ensure that inflation expectations remain anchored, and inflation returns to the 2% target in the medium term.

Box F: Agents’ update on business conditions

This box presents a summary of the Agents’ intelligence gathered in the six weeks to early July that was considered by the MPC at its August meeting.

Sentiment remains subdued across retail, manufacturing, construction and business services, unchanged from the update given in the Agents’ summary of business conditions – 2025 Q2. Weak demand is expected to continue for the rest of 2025. Uncertainty continues to weigh on export expectations, even though the impact of US tariffs has so far been milder than feared.

Agents report a slightly looser labour market, both in terms of employment intentions and easing recruitment difficulties. Intelligence on pay settlements for 2025 continues to remain broadly in line with the Agents’ January pay survey at 3.7%. Contacts note that lower inflation and a looser labour market than over the past few years are driving lower settlements.

Food price inflation is now expected to peak at 5.0%–5.5% in the autumn. And consumer goods inflation is rising as retailers pass on their supply chain and own cost increases. Consumer services inflation remains elevated and stable.

Demand for consumer goods and services remains weak. Contacts expect this to continue for the rest of 2025, citing low consumer confidence.

Supermarkets say turnover growth is being driven by price rises, with sales volumes at best up only marginally on last year. Contacts continue to report strong sales of premium ready meals, as customers switch from eating out. Retailers report that higher footfall is generally being offset by lower spend per customer. Heightened consumer caution continues to inhibit sales of more expensive durable goods.

Restaurants report flat volumes, although some pub chains report relatively strong growth supported by good weather. Fast food retailers see continued evidence of trading down as they gain higher-income customers but lose those on lower incomes with volumes falling overall. Home delivery volumes continue to grow strongly. The experience of hotels and visitor attractions is mixed but suggests a relatively flat market. Airports, ports, rail, and tour operators now report low growth in volumes and expect a further slowing.

Investment intentions remain subdued though regular replacement and update investment continues.

Uncertainty related to weak demand, tax, regulation and wider government policy (for example, inheritance tax, business assets relief, Employment Rights Bill, Extended Producer Responsibility (EPR) regulation), as well as geopolitical instability continues to cause contacts to hold back or delay investment. Concerns about soft demand and growth prospects mean contacts generally do not need to expand capacity and are tending to stretch out the replacement cycle for machinery and vehicles. Contacts are reluctant to take on new borrowing even though they also tend not to have much financial headroom to fund investment.

The cost of labour and desire to grow without increasing headcount means that investment in software solutions, AI and low-cost automation is quite widespread and, because it can generally be financed incrementally, it is treated more like operational expenditure than capital spending. Although mention of AI is prevalent its use tends to be more experimental rather than a wholesale implementation.

Manufactured goods exports volumes weakened slightly further on last year owing to US tariffs. Services export values growth has held up.

Contacts report that the pace of decline in exports to the EU continues to ease, although recovery remains constrained by post-Brexit trade frictions and weak demand. Growth in exports to the US is weaker, with softer demand linked to higher tariffs. Demand from China and the Middle East is slightly softer. Defence and aerospace exports continue to grow strongly as do exports of advisory services. Tourism from Asia remains subdued, and the number of US visitors has declined further.

The direct impact of US tariffs is milder than feared but more general tariff-related uncertainty still weighs on sentiment, though less so recently. Some contacts paused exports in April but have since resumed them. Some report increased Chinese supply into Europe, mainly in finished goods, but few observe this as putting downward pressure on price inflation. Goods exporters to the US are increasingly prioritising supply chain resilience and face some pressure to cut prices. Contacts welcomed new UK trade deals with the US and India, and Common Understanding with the EU.

Business services volumes growth remains weak. Contacts are still cautious about the outlook, expecting only a modest pickup in volume growth into 2026.

Demand for audit, advisory and tax services remains strong. There are some tentative signs of merger and acquisition activity recovering, which is expected to continue should domestic uncertainty subside and the cost of finance fall. Restructuring and insolvency activity continue to edge up on the same period last year. Demand for IT services is still mixed. Although clients are cautious about spending on larger projects, there is strong growth in AI and cybersecurity. Growth in corporate hospitality and marketing remains weak, as businesses limit discretionary spend. Logistics and wholesaling contacts report broadly stable volumes on last year. Recruitment activity remains down on last year, but there are some signs of stabilisation as businesses address pockets of skill shortages or seek advice on reducing their workforce.

The annual rate of decline in manufacturing volumes eased owing to the slight lessening in trade uncertainty. Despite this, the weak domestic market and subdued new orders mean confidence remains weak. Some expected improvement in weaker subsectors should lead to modest overall manufacturing output growth in 2026.

While supply chains have normalised from the high stock levels built up after the pandemic, reducing stock and delaying orders are common themes as companies seek to manage costs.

Contacts report that US tariffs and electric vehicle competition from overseas producers is dampening automotive sector output. A few construction suppliers to house builders are seeing modest order growth, but other output is flat or falling as projects, for example in the renewables sector, continue to be delayed. Output of consumer goods and food and drink remains subdued as consumers limit discretionary spend. Manufacturers supplying energy infrastructure projects are more positive. The aviation sector has good order books, but lack of availability of components and related phasing of delivery has constrained the pace of growth. Strong demand persists in the defence sector.

The annual pace of decline in construction output is slightly worse, mainly due to a further softening in commercial activity. While the medium-term outlook is positive, some remain cautious about the extent of recovery in 2026.

Commercial construction projects continue to be delayed or shelved, alongside a more general slowing in commercial activity. Contacts reported delays to public sector projects ahead of the Spending Review. While the review has increased clarity around government and infrastructure projects, some remain concerned that these delays will persist. House builders now mostly report growth in new build rates, despite frequent mention of planning delays. High-rise development is hampered by new planning regulations and higher costs. Housing association spending remains focused on repair and maintenance. Outside of the High Speed 2 rail network and the Hinkley Point power station, infrastructure spending remains subdued for the most part.

Employment intentions are slightly weaker than in previous rounds owing largely to higher labour costs, related to higher National Insurance contributions (NICs). However, indications are that considerable adjustment has taken place already with many firms having reduced headcount over the last year.

Larger firms and those in consumer-facing sectors have been the most active in reducing headcount. While recruitment freezes are the most common means, Agents are starting to hear more mentions of redundancies, though mainly voluntary. Contacts also report that new employment rights legislation may dampen labour demand further.

Many contacts view automation, AI, or the offshoring of some roles as a solution to increased labour costs, or a way to grow output without increasing headcount. Some manufacturers see production overseas as better value.

Agents see further evidence of recruitment difficulties easing. Many contacts observe that the balance of power in the labour market has shifted more towards employers, and that recruitment conditions are now normal for less skilled occupations. Even in specialist roles where skills shortages remain, contacts report some improvement in their ability to recruit.

Intelligence on pay settlements for 2025 continues to remain broadly in line with the Agents’ January pay survey at 3.7%.

Contacts note that lower inflation and a looser labour market than over the past few years are driving lower settlements, with some freezing pay or delaying pay rises, although total labour costs are growing owing to NICs. Most contacts are not yet able to give a figure for 2026, but some report that they see pressures easing further and are budgeting for settlements in the 2%–4% range with a rough average of 3%–3.5%. Potential upside risks to this include the impact of the current period of above-target CPIU inflation, unions bargaining for higher increases, and next year’s increase in the National Living Wage (NLW).

Higher prices, lower margins, lower pay settlements and reduced labour input are the most commonly cited channels to mitigate NICs and NLW cost pressures. Some firms facing lower margins this year will potentially take further action on prices, pay, and headcount into 2026.

Contacts are reporting some increase in spare capacity, particularly in manufacturing industries where demand is weak. For some, automation and investment are providing additional capacity.

The combination of higher labour costs and EPR has put pressure on firms to recover their margins via price increases, although many are not able to recover as much as they would like. Consequently, pressures to restore margins will persist.

Weak demand is making it difficult for firms to increase prices to the extent they would like in the face of these higher costs. This is leading many contacts to try to reduce other costs. This strategy may rebuild their margins, but it could take several quarters.

Materials’ cost inflation is modestly positive as is inflation of imported finished goods, reflecting a pickup in the cost of energy, labour, freight and some commodities. The 90-day pause on tariffs has made it difficult to gauge the impact tariffs will have on supply chains but the effects to date have been modest. Most contacts find themselves in a holding pattern waiting for a response. Until tariffs are fully enacted, there remains significant uncertainty around their impact on costs and how firms will respond.

Manufacturers’ domestic prices are increasing as they pass on some of their higher input and labour costs, where demand allows. Business-to-business price inflation is easing gradually, but the pace of decline is mixed across sectors.

Consumer goods inflation is rising as retailers pass on their supply chain and own cost increases. This is most evident in food price inflation which is now expected to peak between 5.0%–5.5% in the autumn. Although higher NLW and NIC costs are already being reflected in higher food prices, the majority of EPR costs are still to be felt. There are limited signs yet of competition in the grocery sector taking the edge off this expected peak in food price inflation. Non-food goods price inflation remains modest, reflecting soft input costs and weak demand.

Consumer services inflation remains elevated and stable. The moderation in services inflation has paused as companies have passed on higher labour and other input costs to the extent demand allows.

Corporate credit availability remains close to normal in aggregate, with increased competition for the more creditworthy. Credit demand remains weak, but there are signs of some recovery.

Bank lending appetite is positive but targeted. Competition for safer lending is intensifying, with profitable firms able to refinance or increase borrowing at better spreads. However, credit remains tight for those sectors struggling financially. Alternative finance is accessible, but at higher rates. Early-stage finance is tighter than usual.

Credit demand remains weak. Contacts are cautious around the timing of market refinancing, due to high market volatility and lower liquidity. Higher capital gains tax has dampened transactional demand since April, but rising enquiries suggest some recovery in 2025 H2. Demand for working capital and asset finance is holding up.

Demand is weaker in the housing market now that the temporary increase in activity ahead of stamp duty changes in England has fallen away.

Secondary market demand is subdued, while supply is picking up. Demand is strongest at the cheaper end, while expensive homes are selling more slowly. Estate agents have mixed views, but most expect prices to remain flat over 2025.

Demand for rental properties continues to outstrip supply as more buy-to-let landlords exit the sector adding to the supply of homes for sale. This supply squeeze is supporting rent inflation. Some landlords are concerned arrears might rise once the Renters Reform Bill becomes enacted in law.