## Monetary Policy Issues in the UK

## Source details

**Canonical URL:** [Monetary Policy Issues in the UK](https://www.imf.org/-/media/files/publications/selected-issues-papers/2024/english/sipea2024028.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/selected-issues-papers/2024/english/sipea2024028.pdf.md)
- [Structured JSON version](/-/media/files/publications/selected-issues-papers/2024/english/sipea2024028.pdf.json)

---

### Abstract and Overview
- After hiking rates 14 consecutive times between December 2021 and August 2023, the Bank of England (BoE) has held rates at 5.25 percent since then.
- Headline inflation fell from double digits in 2023Q1 to near-target in 2024Q2; headline inflation was 2.3 percent in April 2024 and is expected to durably return to the 2 percent target in early 2025.
- Core questions addressed:
  - How have macroeconomic and financial effects of BoE tightening compared with other major AEs and previous UK cycles?
  - What is the impact of US Fed decisions on UK monetary transmission and implications for BoE communications?
  - How do model-based optimal UK policy paths compare with staff’s recommended path in the 2024 Article IV consultation?
- Main findings:
  - Monetary transmission largely mirrored previous episodes and other major AEs, except the mortgage channel (slower due to higher share of fixed-rate mortgages).
  - Fed announcements have an outsized impact on UK financial markets, raising the importance of BoE communications when diverging from the Fed.
  - Optimal model predictions are close to staff’s recommended path; if the BoE places a high weight on avoiding prolonged above-target inflation and de-anchoring, a slower pace of cuts would be warranted.

### A. Monetary Policy Transmission: Is the Current UK Cycle Different?
- Character of the current tightening cycle:
  - BoE raised rates 14 times at successive MPC meetings from December 2021, peaking at 5.25 percent in August 2023.
  - The current cycle is the longest and largest since BoE independence in 1997; the cycle length is 30 months as of May 2024 (starting from December 2021).
  - Cumulative increase in Bank Rate: 5.15 ppt (Table 1: Duration 30 months; Total Rate increases 5.15 ppt; # of Rate Hikes 14; Inflation (highest) 11.1%; Inflation (lowest) 2.3%).
- Inflation and expectations:
  - Peak headline inflation around 10–11 percent in late 2022/early 2023.
  - Market-implied 5-year ahead inflation expectations (adjusted for RPI–CPI wedge) around 2.5 percent, slightly above pre-cycle levels.
  - Long-run inflation expectations have remained well-anchored throughout the cycle.
- Macroeconomic outcomes and drivers:
  - Real GDP growth much weaker than in previous cycles; weak growth mainly reflects very weak household consumption.
  - Investment has held up better in recent years, including public investment, despite an extended period of weak investment.
  - Labor market tighter than in previous cycles (partly due to less flexible labor market since Brexit), easing gradually since early 2022 peak.
  - Real effective exchange rate depreciated initially and appreciated in recent quarters, driven by interest rate differentials with the US.
- Financial conditions and transmission channels:
  - Financial Conditions Index (FCI) impact was somewhat stronger initially than in previous cycles but has eased recently; index response adjusted for size of total rate hikes (elasticity of FCI to cumulative rate increases).
  - Interest rate channel:
    - Response of “new” bank lending rates for corporate and household loans similar to previous cycles.
    - “Effective” rates on corporate loans and household non-mortgage loans consistent with previous cycles.
    - Increase in “effective mortgage rates” significantly undershot previous cycles due to higher share of fixed-rate mortgages (slower repricing).
    - Corporate bond yields initially over-responded but settled at a passthrough around 0.7.
    - Gilt yields more responsive to policy increases than in previous cycles.
  - Asset price channel:
    - Equity prices (FTSE100) held up relatively well.
    - House prices more subdued, which may help explain weaker household consumption.
  - Credit channel:
    - Bank lending has almost remained flat in the current cycle, a clear break from the past.
    - Weak bank lending may reflect structural post-GFC break and increased COVID-era borrowing under government schemes.
    - Corporate capital issuance volatile but not significantly deviating from previous cycles.
- Comparison with other major AEs in the current cycle:
  - Other major central banks (US Fed, ECB) also raised rates significantly.
  - BoE was first to raise rates but more gradual early in the cycle than the US Fed.
  - Controlling for size of tightening, the UK’s weak GDP and consumption responses are comparable to peers; investment was relatively stronger.
  - Fiscal deficits relatively larger in the UK than peers, reflecting stronger COVID support.
  - Wage growth similar to peers but more persistent in the UK.
- Key takeaways:
  - Monetary transmission mostly worked as expected and similar to other AEs.
  - Notable exception: shallower increase in effective mortgage rates due to higher fixed-rate mortgage share.
  - Household consumption very weak despite wage growth pickup in 2023, driven by high cost of living and weak confidence.
  - Government tax incentives supported business investment, but uptake varied across firms; profitability drove manufacturing investment.

### B. UK MPC Decisions and Announcements: The Role of Fed Policy Spillovers
- Motivation and timing issues:
  - UK financial markets are sensitive to international news, notably FOMC decisions, which affect UK monetary transmission channels.
  - Back-to-back sequencing of FOMC and MPC announcements can lead to MPC announcements being dominated by preceding FOMC announcements.
  - MPC decisions are voted prior to the FOMC announcement in some cases; time differences mean MPC voters may not observe FOMC decisions or market reactions at the time of voting.
- Empirical example (December 13–14, 2023):
  - FOMC announcement on December 13, 2023 (including potential 75bps rate cuts in 2024) sent a dovish signal after UK markets closed.
  - On December 14, the 2-year gilt yield dropped 20 bps within 15 minutes after markets opened.
  - Later that day, BoE MPC announced a hawkish hold; three out of nine MPC members voted for a further rate hike; no press conference followed since it was a non-MPR round.
  - After the MPC decision, markets scaled back rate-cut expectations slightly but still expected more than 100bps rate cuts in 2024 (up from 75 bps the day before).
  - The 2-year gilt yield increased 3 bps within 30 minutes around the MPC announcement.
  - Market’s reaction to the FOMC was inconsistent with the MPC’s intended direction; absence of a press conference limited MPC’s ability to articulate views.
- Frequency of back-to-back meetings:
  - Identified 41 FOMC announcements occurring one day before MPC announcements from 1999 to mid-2023.
  - Many back-to-back meetings occurred after MPC moved to eight decisions a year post-2016.
  - In 2023 and 2024, five MPC meetings each year have decisions announced one day after FOMC decisions; three of them do/do not have press conferences.
- Methodology for identifying spillovers:
  - Monetary policy surprises identified using high-frequency market indicator changes within narrow windows around announcements.
  - For MPC announcements, used BoE’s UK Monetary Policy Event-Study Database (UKMPD) providing intraday data on monetary policy surprises since 1997.
  - FOMC decisions typically announced at 2PM US Eastern Time (UK markets closed); proxied UK market reaction by differences between opening prices on the next day and closing prices on the day of FOMC announcements.
- Implications:
  - FOMC announcements often have a nontrivial impact on UK financial markets; this can affect the transmission of MPC decisions into domestic demand.
  - Given the potential for FOMC-driven market moves to dominate UK market reactions, there is scope for adjusting MPC decision-timing and communications to preserve the efficacy of MPC policy signals.

### Factor identification and interpretation
- Factor analysis on seven indicators (first four quarterly short sterling futures and the 2-, 5-, and 10-year gilt yields) yields three MPC factors:
  - Target: highest correlation with short-term rates; represents news on the policy rate.
  - Path: highest correlation with the 4-quarter ahead sterling future rate; represents MPC forward guidance.
  - QE: highest correlation with the 10-year gilt yield; represents news on unconventional monetary policy (QE/QT).
- Two FOMC spillover factors identified for UK market responses to FOMC decisions:
  - Spillover 1: impact on short-term interest rates.
  - Spillover 2: impact on medium- to long-term interest rates.
- Notes:
  - Factors using the seven UK asset price changes corresponding to all FOMC announcements since 1999 show mixed correlations across the yield curves.
  - FOMC announcements can occasionally produce opposing shocks on the same day relative to MPC factors.
  - FOMC announcements can have sizable impacts on short-term rates (Spillover 1), while impacts on medium- to long-term rates are relatively smaller.

### Impact of FOMC spillovers on UK asset prices (regression results)
- Regression specification:
  - ∆Y_t = α + β Spillover1_t + γ Spillover2_t + ε_t
  - ∆Y_t denotes the change in the relevant asset price.
  - Spillover1_t and Spillover2_t are the previously estimated FOMC monetary spillover shocks.
- Key regression findings (Table 2: United Kingdom: Impact of the FOMC Spillovers; N = 41 for all regressions):
  - Dependent variables: (1) FTSE100, (2) FTSE250, (3) FTSE_All, (4) GBP/EUR, (5) USD/GBP
  - Spillover 1 coefficients (standard errors in parentheses):
    - FTSE100: -0.493 (0.54)
    - FTSE250: -4.036 (4.15)
    - FTSE_All: -0.105 (0.32)
    - GBP/EUR: 0.002 (0.01)
    - USD/GBP: 0.009 (0.02)
  - Spillover 2 coefficients (standard errors in parentheses):
    - FTSE100: 0.535 (0.99)
    - FTSE250: 4.065 (7.53)
    - FTSE_All: 0.142 (0.58)
    - GBP/EUR: -0.012 (0.02)
    - USD/GBP: -0.082*** (0.04)
  - Constant terms (standard errors in parentheses):
    - FTSE100: 0.0253 (0.03)
    - FTSE250: 0.258 (0.24)
    - FTSE_All: -0.0182 (0.02)
    - GBP/EUR: -0.0010 (0.00)
    - USD/GBP: 0.003*** (0.00)
  - R-squared by regression:
    - FTSE100: 0.024
    - FTSE250: 0.027
    - FTSE_All: 0.004
    - GBP/EUR: 0.015
    - USD/GBP: 0.124
  - Notes: Standard errors in parentheses. * p<0.1, ** p<0.05, *** p<0.01. GBP/EUR is the value of British pound per euro (an increase indicates sterling depreciation), and USD/GBP is the value of US dollar per British pound (an increase indicates sterling appreciation).
- Interpretation:
  - A positive Spillover 2 is associated with a slight depreciation of sterling against the dollar (USD/GBP coefficient -0.082*** indicates statistical significance at p<0.01).
  - Spillovers from FOMC announcements have little impact on the UK stock markets or bilateral exchange rates with the euro.
  - Mechanism: FOMC forward guidance can push up the long end of the US and UK yield curves, with a stronger impact on the US curve, widening interest rate differentials and leading to sterling depreciation.

### Monetary transmission to UK macro aggregates (LP-IV results)
- Estimation approach:
  - Use Stock and Watson (2018) local projection using external instruments (LP-IV).
  - Instrumental variables: three MPC shocks (Target, Path, QE) and two FOMC shocks (Spillover 1 and Spillover 2).
  - Baseline instruments: Target for changes in Bank Rate; Path for changes in 1-year gilt yield. FOMC shocks are initially excluded, then added in subsequent estimates (Spillover 1 for Bank Rate; Spillover 2 for 1-year yield). Both FOMC shocks are included in control variables when added.
  - Controls: 12 lags of dependent variables and external instruments.
  - Monthly data used for Bank Rate, 1-year and 10-year gilt yields, investment-grade corporate bond yield, FTSE All Share index, GDP volume index, and CPI over period 1997 June to 2019 December.
  - FTSE All Share index, GDP volume index, and CPI are in log levels; all other variables in percentage points.
- Key impulse response findings:
  - A 100-bps increase in Bank Rate (Target monetary policy shock) leads to about 1.2 percentage points decline in real GDP within 30 months.
  - The effect is persistent and remains at about 1-1.2 percentage points 36 months after the shock.
  - Including FOMC spillovers:
    - FOMC spillovers strengthen monetary transmission of Bank Rate by pushing forward the GDP impact.
    - The peak GDP impact reaches about six months earlier compared with baseline without FOMC shocks.
- Caveats:
  - Results are based on pre-COVID data; very few observations since 2021 limit analysis of the current cycle.
  - High volatility during COVID-19 adds noise; some abnormalities observed in the current cycle (e.g., December 14, 2023) when market reactions to FOMC and MPC decisions were inconsistent.

### Key takeaways and policy recommendations
- Findings:
  - FOMC decisions impact UK financial markets and affect transmissions of immediately following MPC decisions.
  - Pre-COVID empirical work shows FOMC announcements have generally supported MPC monetary transmission when market reactions were reinforcing.
  - When market reactions to FOMC and MPC decisions diverge (example: December 2023), transmission via FOMC spillovers can work against MPC objectives.
  - Not all MPC decisions are accompanied by a press conference; when absent the MPC lacks opportunity to elaborate or caveat its views if market reaction to Fed decisions is inconsistent with MPC policy direction.
- Policy considerations:
  - The current approach—MPC makes decision before FOMC decision but announces after market absorption of Fed decision—may be suboptimal because the MPC decision and communication do not consider the Fed’s impact on domestic financial conditions.
  - Consider streamlining information published alongside the MPC decision to enable announcing the MPC decision on the same day.
  - Consider delaying publication of MPC minutes and the MPR by a week or so (similar to FOMC practice).
  - The MPC should consider holding a press conference after each decision (as is done by other major central banks) to explain decisions in context of evolving FOMC spillovers and to avoid perceptions that BoE decisions are overshadowed by the Fed.

### Model-based assessment of alternative policy paths (COPP toolkit)
- Toolkit and model:
  - COPP (Constrained Optimal Policy Projection) toolkit used to assess staff baseline BoE policy path versus a benchmark optimal control policy.
  - Optimal policy generated using the Svensson (2000) model (Philips curve, IS curve, and UIP condition) calibrated to match monetary transmission estimates from standard VARs for advanced economies.
- Loss function and calibration:
  - Quadratic loss function: losses depend on deviations of inflation from target, output gap, and first difference of the interest rate.
  - Weights in loss function (λπ, λy, λi) determine trade-offs.
  - Table 3: The Coefficients in the Loss Function
    - Weights / Baseline Loss Function / High Inflation Weight / High Output Weight / High Interest Rate Weight
    - λπ: 1 / 5 / 1 / 1
    - λy: 1 / 1 / 5 / 1
    - λi: 1 / 1 / 1 / 5
    - Loss Function (relative to staff projection): 0.53 / 0.66 / 0.54 / 0.56
- Simple policy rules used for comparison:
  - Taylor Rule I: coefficient of 0.9 on lagged Bank Rate (S), 1.5 on annual CPI inflation (π), 1 on output gap (y).
  - Taylor Rule II: coefficient of 0.75 on lagged Bank Rate, 2 on annual CPI inflation, 0.5 on output gap.
  - Taylor rule form: S_t = ρ S_{t-1} + (1−ρ)(S* + φ_π π_t + φ_y y_t).
- Optimal policy results:
  - Staff’s baseline policy projection is close to optimal, but with differences in timing of rate cuts.
  - Optimal interest path implies much earlier reduction in policy rate, followed by a period of rates around 4 percent in 2025-6.
  - Faster normalization in optimal path aims to bring forward closure of the output gap, accounting for transmission lags.
  - Inflation converges to the 2 percent target by 2025 under all loss-function assumptions.
- De-anchoring risk and sensitivity analyses:
  - If a prolonged period of above-target inflation risks de-anchoring expectations, optimal path moves closer to staff projection (more gradual normalization).
  - Assumption for de-anchoring scenario: fraction a = 0.5 of agents have perfect foresight with inflation expectations at the target; this fraction decays at the exponential rate each quarter of the projection horizon.
  - Constrained optimal policy under de-anchoring risk is closer to staff baseline.
  - Targeting core inflation (higher inertia relative to headline inflation) yields a similar constrained optimal policy path to the baseline headline-inflation target; the “cut, hold, cut” prescription is robust to higher inflation inertia assumptions.

### Annex I — Micro underpinnings: consumption and investment
- Household consumption (Bank of England/NMG household survey analysis):
  - From 2019 to 2022:
    - General increase in median consumption across all income groups, largely due to higher costs of living.
    - No strong income growth across all income groups (only in the top two income groups).
    - Higher consumption largely financed by withdrawal of savings or more borrowing (more likely for lower income groups).
  - From 2022 to 2023:
    - Across-the-board increases in household income, but these were largely saved, likely due to weak consumer sentiment.
  - Mortgages:
    - Less than 30 percent of households in the survey have mortgages.
    - About 50 percent of households in the highest income group have mortgages.
    - Less than 10 percent in the lowest income group have mortgages.
    - Distribution of mortgages repricing each year follows a similar pattern (more concentrated in higher-income groups).
    - Seemingly larger increases in mortgage payments in lower income groups in 2022 (Group 1) and 2023 (Group 2) likely due to other factors (such as new mortgages) rather than repricing of existing mortgages.
    - Near-term: despite continued flows of fixed-rate mortgages being subject to higher rates, this is not expected to be a major drag on consumption; continued strong wage growth and improving consumer sentiment likely support consumption recovery.
- Business investment (Worldscope firm-level data):
  - Sectoral capital expenditure changes:
    - Utilities and communications had the highest increases in capital expenditures from 2019 to 2022; manufacturing declined over that period.
    - From 2022 to 2023, utilities and communications continued strong investment and manufacturing rebounded.
  - Factors supporting manufacturing rebound:
    - Windsor Framework agreement reduced Brexit-related uncertainty.
    - Pent-up investment demand from early COVID period.
    - Full tax-expensing of business investment in plant and machinery introduced in the 2023 Spring budget and made permanent in the 2023 Autumn statement.
  - Sector differences:
    - Transport and communications appear to have utilized tax incentives (shown as other operational cash flows) to finance capital expenditures.
    - Representative manufacturing firms: recovery of profits (net income from operation) is main driver of 2023 capital expenditure recovery; sources of long-term financing (including equity) also help boost capital expenditures.
  - Outlook:
    - Permanent tax incentives will support business investment, but positive profitability outlook and access to long-term financing are crucial for sustained business investment growth.

*Source: IMF Selected Issues Paper (SIP/2024/028), “Monetary Policy Issues in the United Kingdom,” prepared by Agnese Carella, Ruo Chen, Katherine Dai, Gloria Li, Ruy Lama, and Roland Meeks, July 2024.*

### 4. This paper is also

### Monetary Policy Issues in the UK

### Abstract and Overview
- After hiking rates 14 consecutive times between December 2021 and August 2023, the Bank of England (BoE) has held rates at 5.25 percent since then.
- Headline inflation fell from double digits in 2023Q1 to near-target in 2024Q2; headline inflation was 2.3 percent in April 2024 and is expected to durably return to the 2 percent target in early 2025.
- The paper addresses three core questions:
  - How have macroeconomic and financial effects of BoE tightening compared with other major AEs and previous UK cycles?
  - What is the impact of US Fed decisions on UK monetary transmission and implications for BoE communications?
  - How do model-based optimal UK policy paths compare with staff’s recommended path in the 2024 Article IV consultation?
- Main findings:
  - Monetary transmission largely mirrored previous episodes and other major AEs, except the mortgage channel (slower due to higher share of fixed-rate mortgages).
  - Fed announcements have an outsized impact on UK financial markets, raising the importance of BoE communications when diverging from the Fed.
  - Optimal model predictions are close to staff’s recommended path; if the BoE places a high weight on avoiding prolonged above-target inflation and de-anchoring, a slower pace of cuts would be warranted.

### A. Monetary Policy Transmission: Is the Current UK Cycle Different?
- Character of the current tightening cycle:
  - BoE raised rates 14 times at successive MPC meetings from December 2021, peaking at 5.25 percent in August 2023.
  - The current cycle is the longest and largest since BoE independence in 1997; the cycle length is 30 months as of May 2024 (starting from December 2021).
  - Cumulative increase in Bank Rate: 5.15 ppt (Table 1 shows 2021-Dec to present: Duration 30 months; Total Rate increases 5.15 ppt; # of Rate Hikes 14; Inflation (highest) 11.1%; Inflation (lowest) 2.3%).
- Inflation and expectations:
  - Peak headline inflation around 10–11 percent in late 2022/early 2023.
  - Market-implied 5-year ahead inflation expectations (adjusted for RPI–CPI wedge) around 2.5 percent, slightly above pre-cycle levels.
  - Long-run inflation expectations have remained well-anchored throughout the cycle.
- Macroeconomic outcomes and drivers:
  - Real GDP growth much weaker than in previous cycles; weak growth mainly reflects very weak household consumption.
  - Investment has held up better in recent years, including public investment, despite an extended period of weak investment.
  - Labor market tighter than in previous cycles (partly due to less flexible labor market since Brexit), easing gradually since early 2022 peak.
  - Real effective exchange rate depreciated initially and appreciated in recent quarters, driven by interest rate differentials with the US.
- Financial conditions and transmission channels:
  - Financial Conditions Index (FCI) impact was somewhat stronger initially than in previous cycles but has eased recently; index response adjusted for size of total rate hikes (elasticity of FCI to cumulative rate increases).
  - Interest rate channel:
    - Response of “new” bank lending rates for corporate and household loans similar to previous cycles.
    - “Effective” rates on corporate loans and household non-mortgage loans consistent with previous cycles.
    - Increase in “effective mortgage rates” significantly undershot previous cycles due to higher share of fixed-rate mortgages (slower repricing).
    - Corporate bond yields initially over-responded but settled at a passthrough around 0.7.
    - Gilt yields more responsive to policy increases than in previous cycles.
  - Asset price channel:
    - Equity prices (FTSE100) held up relatively well.
    - House prices more subdued, which may help explain weaker household consumption.
  - Credit channel:
    - Bank lending has almost remained flat in the current cycle, a clear break from the past.
    - Weak bank lending may reflect structural post-GFC break and increased COVID-era borrowing under government schemes.
    - Corporate capital issuance volatile but not significantly deviating from previous cycles.
- Comparison with other major AEs in the current cycle:
  - Other major central banks (US Fed, ECB) also raised rates significantly.
  - BoE was first to raise rates but more gradual early in the cycle than the US Fed.
  - Controlling for size of tightening, the UK’s weak GDP and consumption responses are comparable to peers; investment was relatively stronger.
  - Fiscal deficits relatively larger in the UK than peers, reflecting stronger COVID support.
  - Wage growth similar to peers but more persistent in the UK.
- Key takeaways (Section A):
  - Monetary transmission mostly worked as expected and similar to other AEs.
  - Notable exception: shallower increase in effective mortgage rates due to higher fixed-rate mortgage share.
  - Household consumption very weak despite wage growth pickup in 2023, driven by high cost of living and weak confidence.
  - Government tax incentives supported business investment, but uptake varied across firms; profitability drove manufacturing investment.

### B. UK MPC Decisions and Announcements: The Role of Fed Policy Spillovers
- Motivation:
  - UK financial markets, as a global financial center, are sensitive to international news, notably FOMC decisions, which affect UK monetary transmission channels.
  - Back-to-back sequencing of FOMC and MPC announcements can lead to MPC announcements being dominated by preceding FOMC announcements.
- Empirical observations and examples:
  - Example: FOMC announcement on December 13, 2023 (including potential 75bps rate cuts in 2024) sent a dovish signal after UK markets closed.
    - On December 14, the 2-year gilt yield dropped 20 bps within 15 minutes after markets opened.
    - Later that day, BoE MPC announced a hawkish hold; three out of nine MPC members voted for a further rate hike; no press conference followed since it was a non-MPR round.
    - After the MPC decision, markets scaled back rate-cut expectations slightly but still expected more than 100bps rate cuts in 2024 (up from 75 bps the day before).
    - The 2-year gilt yield increased 3 bps within 30 minutes around the MPC announcement.
  - Timing issue: MPC decisions are voted prior to the FOMC announcement in some cases; due to time differences, MPC voters may not observe FOMC decisions or market reactions at the time of voting.
  - In the December 13–14 case, the market’s reaction to the FOMC was inconsistent with the MPC’s intended direction, and absence of a press conference limited MPC’s ability to articulate views.
- Frequency and scope of back-to-back meetings:
  - Identified 41 FOMC announcements occurring one day before MPC announcements from 1999 to mid-2023.
  - Many back-to-back meetings occurred after MPC moved to eight decisions a year post-2016.
  - In 2023 and 2024, five MPC meetings each year have decisions announced one day after FOMC decisions; three of them do/do not have press conferences.
- Methodology for identifying spillovers:
  - Monetary policy surprises identified using high-frequency market indicator changes within narrow windows around announcements (Gurkaynak et al. 2005; Swanson 2021; Braun et al. 2023).
  - For MPC announcements, used BoE’s UK Monetary Policy Event-Study Database (UKMPD) providing intraday data on monetary policy surprises since 1997.
  - FOMC decisions typically announced at 2PM US Eastern Time (UK markets closed); proxied UK market reaction by differences between opening prices on the next day and closing prices on the day of FOMC announcements.
- Implications:
  - FOMC announcements often have a nontrivial impact on UK financial markets; this can affect the transmission of MPC decisions into domestic demand.
  - Given the potential for FOMC-driven market moves to dominate UK market reactions, there is scope for adjusting MPC decision-timing and communications to preserve the efficacy of MPC policy signals.

*Source: IMF Selected Issues Paper (SIP/2024/028), “Monetary Policy Issues in the United Kingdom,” prepared by Agnese Carella, Ruo Chen, Katherine Dai, Gloria Li, Ruy Lama, and Roland Meeks, July 2024.*

### 12.      Then, we use factor analysis to identify underlying factors capturing policy decision

### 12.      Then, we use factor analysis to identify underlying factors capturing policy decision shocks

### Factor identification and interpretation
- Factor analysis on price movements of seven indicators (the first four quarterly short sterling futures and the 2-, 5-, and 10-year gilt yields) yields three MPC factors:
  - Target: highest correlation with short-term rates; represents news on the policy rate.
  - Path: highest correlation with the 4-quarter ahead sterling future rate; represents MPC forward guidance.
  - QE: highest correlation with the 10-year gilt yield; represents news on unconventional monetary policy (QE/QT).
- Two FOMC spillover factors (Spillover 1 and Spillover 2) are identified for UK market responses to FOMC decisions:
  - Spillover 1: impact on short-term interest rates.
  - Spillover 2: impact on medium- to long-term interest rates.
- Notes:
  - Factors using the seven UK asset price changes corresponding to all FOMC announcements since 1999 show mixed correlations across the yield curves.
  - FOMC announcements can occasionally produce opposing shocks on the same day relative to MPC factors.
  - FOMC announcements can have sizable impacts on short-term rates (Spillover 1), while impacts on medium- to long-term rates are relatively smaller.

### Impact of FOMC spillovers on UK asset prices (regression results)
- Regression specification:
  - ∆Y_t = α + β Spillover1_t + γ Spillover2_t + ε_t
  - ∆Y_t denotes the change in the relevant asset price.
  - Spillover1_t and Spillover2_t are the previously estimated FOMC monetary spillover shocks.
- Key regression findings (Table 2: United Kingdom: Impact of the FOMC Spillovers; N = 41 for all regressions):
  - Dependent variables: (1) FTSE100, (2) FTSE250, (3) FTSE_All, (4) GBP/EUR, (5) USD/GBP
  - Spillover 1 coefficients (standard errors in parentheses):
    - FTSE100: -0.493 (0.54)
    - FTSE250: -4.036 (4.15)
    - FTSE_All: -0.105 (0.32)
    - GBP/EUR: 0.002 (0.01)
    - USD/GBP: 0.009 (0.02)
  - Spillover 2 coefficients (standard errors in parentheses):
    - FTSE100: 0.535 (0.99)
    - FTSE250: 4.065 (7.53)
    - FTSE_All: 0.142 (0.58)
    - GBP/EUR: -0.012 (0.02)
    - USD/GBP: -0.082*** (0.04)
  - Constant terms (standard errors in parentheses):
    - FTSE100: 0.0253 (0.03)
    - FTSE250: 0.258 (0.24)
    - FTSE_All: -0.0182 (0.02)
    - GBP/EUR: -0.0010 (0.00)
    - USD/GBP: 0.003*** (0.00)
  - R-squared by regression:
    - FTSE100: 0.024
    - FTSE250: 0.027
    - FTSE_All: 0.004
    - GBP/EUR: 0.015
    - USD/GBP: 0.124
  - Notes: Standard errors in parentheses. * p<0.1, ** p<0.05, *** p<0.01. GBP/EUR is the value of British pound per euro (an increase indicates sterling depreciation), and USD/GBP is the value of US dollar per British pound (an increase indicates sterling appreciation).
- Interpretation:
  - A positive Spillover 2 is associated with a slight depreciation of sterling against the dollar (USD/GBP coefficient -0.082*** indicates statistical significance at p<0.01).
  - Spillovers from FOMC announcements have little impact on the UK stock markets or bilateral exchange rates with the euro.
  - Mechanism: FOMC forward guidance can push up the long end of the US and UK yield curves, with a stronger impact on the US curve, widening interest rate differentials and leading to sterling depreciation.

### Monetary transmission to UK macro aggregates (LP-IV results)
- Estimation approach:
  - Use Stock and Watson (2018) local projection using external instruments (LP-IV).
  - Instrumental variables: three MPC shocks (Target, Path, QE) and two FOMC shocks (Spillover 1 and Spillover 2).
  - Baseline instruments: Target for changes in Bank Rate; Path for changes in 1-year gilt yield. FOMC shocks are initially excluded, then added in subsequent estimates (Spillover 1 for Bank Rate; Spillover 2 for 1-year yield). Both FOMC shocks are included in control variables when added.
  - Controls: 12 lags of dependent variables and external instruments.
  - Monthly data used for Bank Rate, 1-year and 10-year gilt yields, investment-grade corporate bond yield, FTSE All Share index, GDP volume index, and CPI over period 1997 June to 2019 December.
  - FTSE All Share index, GDP volume index, and CPI are in log levels; all other variables in percentage points.
- Key impulse response findings:
  - A 100-bps increase in Bank Rate (Target monetary policy shock) leads to about 1.2 percentage points decline in real GDP within 30 months.
  - The effect is persistent and remains at about 1-1.2 percentage points 36 months after the shock.
  - Including FOMC spillovers:
    - FOMC spillovers strengthen monetary transmission of Bank Rate by pushing forward the GDP impact.
    - The peak GDP impact reaches about six months earlier compared with baseline without FOMC shocks.
  - Caveats:
    - Results are based on pre-COVID data; very few observations since 2021 limit analysis of the current cycle.
    - High volatility during COVID-19 adds noise; some abnormalities observed in the current cycle (e.g., December 14, 2023) when market reactions to FOMC and MPC decisions were inconsistent.

### Key takeaways and policy recommendations
- Findings:
  - FOMC decisions impact UK financial markets and affect transmissions of immediately following MPC decisions.
  - Pre-COVID empirical work shows FOMC announcements have generally supported MPC monetary transmission when market reactions were reinforcing.
  - When market reactions to FOMC and MPC decisions diverge (example: December 2023), transmission via FOMC spillovers can work against MPC objectives.
  - Not all MPC decisions are accompanied by a press conference; when absent the MPC lacks opportunity to elaborate or caveat its views if market reaction to Fed decisions is inconsistent with MPC policy direction.
- Policy considerations:
  - The current approach—MPC makes decision before FOMC decision but announces after market absorption of Fed decision—may be suboptimal because the MPC decision and communication do not consider the Fed’s impact on domestic financial conditions.
  - Consider streamlining information published alongside the MPC decision to enable announcing the MPC decision on the same day.
  - Consider delaying publication of MPC minutes and the MPR by a week or so (similar to FOMC practice).
  - The MPC should consider holding a press conference after each decision (as is done by other major central banks) to explain decisions in context of evolving FOMC spillovers and to avoid perceptions that BoE decisions are overshadowed by the Fed.

### Model-based assessment of alternative policy paths (COPP toolkit)
- Toolkit and model:
  - COPP (Constrained Optimal Policy Projection) toolkit used to assess staff baseline BoE policy path versus a benchmark optimal control policy.
  - Optimal policy generated using the Svensson (2000) model (Philips curve, IS curve, and UIP condition) calibrated to match monetary transmission estimates from standard VARs for advanced economies.
- Loss function and calibration:
  - Quadratic loss function: losses depend on deviations of inflation from target, output gap, and first difference of the interest rate.
  - Weights in loss function (λπ, λy, λi) determine trade-offs.
  - Table 3: The Coefficients in the Loss Function
    - Weights / Baseline Loss Function / High Inflation Weight / High Output Weight / High Interest Rate Weight
    - λπ: 1 / 5 / 1 / 1
    - λy: 1 / 1 / 5 / 1
    - λi: 1 / 1 / 1 / 5
    - Loss Function (relative to staff projection): 0.53 / 0.66 / 0.54 / 0.56
- Simple policy rules used for comparison:
  - Taylor Rule I: coefficient of 0.9 on lagged Bank Rate (S), 1.5 on annual CPI inflation (π), 1 on output gap (y).
  - Taylor Rule II: coefficient of 0.75 on lagged Bank Rate, 2 on annual CPI inflation, 0.5 on output gap.
  - Taylor rule form: S_t = ρ S_{t-1} + (1−ρ)(S* + φ_π π_t + φ_y y_t).
- Optimal policy results:
  - Staff’s baseline policy projection is close to optimal, but with differences in timing of rate cuts.
  - Optimal interest path implies much earlier reduction in policy rate, followed by a period of rates around 4 percent in 2025-6.
  - Faster normalization in optimal path aims to bring forward closure of the output gap, accounting for transmission lags.
  - Inflation converges to the 2 percent target by 2025 under all loss-function assumptions.
- De-anchoring risk and sensitivity analyses:
  - If a prolonged period of above-target inflation risks de-anchoring expectations, optimal path moves closer to staff projection (more gradual normalization).
  - Assumption for de-anchoring scenario: fraction a = 0.5 of agents have perfect foresight with inflation expectations at the target; this fraction decays at the exponential rate each quarter of the projection horizon.
  - Constrained optimal policy under de-anchoring risk is closer to staff baseline.
  - Targeting core inflation (higher inertia relative to headline inflation) yields a similar constrained optimal policy path to the baseline headline-inflation target; the “cut, hold, cut” prescription is robust to higher inflation inertia assumptions.

### Annex I — Micro underpinnings: consumption and investment
- Household consumption:
  - Analysis of changes in household consumption from 2019 to 2022 and from 2022 to 2023 and financing sources (income vs saving/borrowing), accounting for changes in debt service, using Bank of England/NMG household survey data.
  - From 2019 to 2022:
    - General increase in median consumption across all income groups, largely due to higher costs of living.
    - No strong income growth across all income groups (only in the top two income groups).
    - Higher consumption largely financed by withdrawal of savings or more borrowing (more likely for lower income groups).
  - From 2022 to 2023:
    - Across-the-board increases in household income, but these were largely saved, likely due to weak consumer sentiment.
  - Mortgages:
    - Less than 30 percent of households in the survey have mortgages.
    - About 50 percent of households in the highest income group have mortgages.
    - Less than 10 percent in the lowest income group have mortgages.
    - Distribution of mortgages repricing each year follows a similar pattern (more concentrated in higher-income groups).
    - Seemingly larger increases in mortgage payments in lower income groups in 2022 (Group 1) and 2023 (Group 2) likely due to other factors (such as new mortgages) rather than repricing of existing mortgages.
    - Near-term: despite continued flows of fixed-rate mortgages being subject to higher rates, this is not expected to be a major drag on consumption; continued strong wage growth and improving consumer sentiment likely support consumption recovery.
- Business investment:
  - Firm-level data (Worldscope) indicates average (asset-weighted) sectoral capital expenditure changes:
    - Utilities and communications had the highest increases in capital expenditures from 2019 to 2022; manufacturing declined over that period.
    - From 2022 to 2023, utilities and communications continued strong investment and manufacturing rebounded.
  - Factors supporting manufacturing rebound:
    - Windsor Framework agreement reduced Brexit-related uncertainty.
    - Pent-up investment demand from early COVID period.
    - Full tax-expensing of business investment in plant and machinery introduced in the 2023 Spring budget and made permanent in the 2023 Autumn statement.
  - Sector differences:
    - Transport and communications appear to have utilized tax incentives (shown as other operational cash flows) to finance capital expenditures.
    - Representative manufacturing firms: recovery of profits (net income from operation) is main driver of 2023 capital expenditure recovery; sources of long-term financing (including equity) also help boost capital expenditures.
  - Outlook: Permanent tax incentives will support business investment, but positive profitability outlook and access to long-term financing are crucial for sustained business investment growth.

*Source: Excerpt from sipea2024028 (IMF Selected Issues Paper).*

---


_Source: https://www.imf.org/-/media/files/publications/selected-issues-papers/2024/english/sipea2024028.pdf_