## _wp15132

## Source details

**Canonical URL:** [_wp15132](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp15132.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp15132.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp15132.pdf.json)

---

### I. INTRODUCTION — scope and framing
- Traces development and implementation of the inflation-forecast targeting (IFT) approach to monetary policy.
- Observations about IFT:
  - Adopted by many central banks in advanced and emerging market economies.
  - Generally helped establish an environment of low and stable inflation in economies that previously had a poor record for monetary stability.
  - Regime often initiated under difficult circumstances, after a crisis, or amid structural change.
  - Many practitioners are highly open economies, susceptible to external shocks, especially smaller ones.
  - The regime is in its third decade.
- Recent focus since the global financial crisis:
  - Attention on monetary policy when policy interest rates are near or at the zero lower bound (ZLB), the so-called “dark corners”.1
  - Growing emphasis on forward guidance as current interest rates are at the floor and conventional monetary policies are less effective.
  - The paper argues the IFT framework can help countries avoid dark corners.
- Broader focus: issues for successful functioning of IFT, with emphasis on transparency and communications of the central bank.
- Structure (sections described): history (II); definition and mechanics of IFT (III); elimination of deviations and ZLB application (IV); model simulations and policy strategy comparisons (V); concluding remarks (VI). Accompanying supplement contains technical background.

### II. HISTORY AND LESSONS FROM PRACTICE
- Development influenced by:
  - Trends in economic theory and pragmatic learning-by-doing.
  - Experience of other IT central banks.
- Key historical/theoretical points:
  - Kydland and Prescott (1977) and Barro and Gordon (1983) motivated search for frameworks to avoid inflation bias.
  - Early rigid IT reduced inflation but could produce instrument instability; shift toward transparent, forecast-based strategies in the 1990s.
  - Announcing explicit numerical targets and increasing communications (Monetary Policy Reports, speeches) were major clarifying steps.
- Accountability and transparency:
  - Clear delegation of responsibility coupled with transparency reduces discretionary inflation bias.
  - Publishing forecasts for inflation and output became widely accepted.
- Milestones toward full-fledged IFT:
  - Announcement of targets with multi-year horizon.
  - Precision on policy interest rate setting.
  - Transparent communications on implementation.
  - Publication of complete macro forecast (including inflation).
  - Publication of a conditional forecast path, alternative scenarios and confidence bands for the short-term interest rate.

### Box 1 — country learning examples (selected findings)
- New Zealand:
  - First to embark on IT in 1989; developed full-fledged IFT.
  - Inflation quickly reduced to less than 2 percent in 1992.
  - Initial exchange-rate instrument instability; introduced FPAS in 1997 and adopted immediate full disclosure of the central bank forecast.
- Canada:
  - Moved to IT in 1991; eventual target of 2 percent unchanged since.
  - After fiscal sustainability in 1995, long-term inflation expectations stabilized at 2 percent.
- Czech Republic:
  - Adopted IT in 1998 after collapse of fixed exchange rate policy and bank failures in 1997.
  - Inflation almost 10 percent since 1993 and accelerating at adoption; FPAS with model-based forecast by 2002.
  - Began publishing detailed quarterly forecasts, including interest rate path, in 2008; strong public confidence in 2 percent target.

### Defining IFT and FPAS requirements
- Core features of IFT used in the paper:
  - Policy based on a long-run low inflation target and a medium-term forecast path to this target.
  - Central bank has structured FPAS producing model-based staff forecast and analysis on a regular schedule.
  - Projection Team (PT) presents forecast shortly before Monetary Policy Committee (MPC) decision meeting.
  - Forecast path for the short-term interest rate is endogenous within the model; rate varies to achieve long-run inflation target and eliminate any output gap.
  - Staff forecast is a key input to MPC decisions; MPC members may incorporate other information.
  - Soon after policy decision, the associated forecast path for key macro variables is disclosed, highlighting inflation path.
  - Monetary Policy Report (MPR) explains rationale at regular intervals and outlines conditional forecast path for the short-term interest rate, with confidence bands and alternative scenarios.

### Transmission mechanism, endogenous policy response, and interest-rate forecast publication
- Transmission mechanism:
  - Expectations of future policy-rate movements in the short to medium term crucially influence longer-term interest rates and the exchange rate.
  - The policy rate path (medium-term expectations) matters more than the current policy rate for affecting borrowing costs.
- Endogenous policy response:
  - Central banks choose among responses to inflation deviations, balancing short-run output-inflation trade-offs (asymptotic approaches or planned overshoots possible).
  - Full policy description entails the entire future path of the interest rate, not just the immediately effective rate.
- Publication debate:
  - Models used in IFT should incorporate an endogenous interest rate; exogenous-rate models lack a nominal anchor.
  - Many central banks produce endogenous interest-rate paths but may not publish them to preserve policy flexibility.
  - Counterarguments: publishing conditional policy-rate paths and confidence intervals clarifies how policymakers intend to influence the yield curve.
  - The chosen path reflects preferences over the inflation-output trade-off; higher weight on output stability implies smoother interest-rate adjustments and a slower return to target.

### Full disclosure of the central bank forecast — rationale and effects
- Two main reinforcing effects of full disclosure:
  - Showing a coherent view with inflation returning to the long-run target strengthens confidence in the goal and the value of money.
  - Publishing the short-term interest rate path helps move the term structure to assist the transmission mechanism.
- Payoffs:
  - Reduced cost of eliminating deviations of actual inflation from the long-run target rate; or improved short-run inflation-output trade-off.
  - Publication also improves central bank accountability.
- Examples of central banks publishing endogenous interest rate path forecasts: Reserve Bank of New Zealand, Czech National Bank, Bank of Israel, Norges Bank, Sveriges Riksbank.

### Ownership of projections (staff versus policymaking body)
- Practical judgment: staff ownership of projections is often preferable in central banks with large policymaking bodies or divergent views.
- Advantages of staff ownership:
  - Produces consistent, coherent projection reflecting economic reasoning and specific circumstances.
  - Simplifies MPR production by presenting forecast as important staff input but not sole input.
  - Avoids inefficiency and confusion from publishing multiple projections when MPC members diverge.
- MPC members’ role:
  - Focus on broad, strategic questions, provide solid economic arguments when differing from staff.
  - Clear explanations of staff vs. MPC differences in MPR help markets understand actions and increase credibility.

### Technical production and institutional requirements
- Model-based baseline forecasts, confidence intervals, and alternative scenarios require highly specialized staff resources.
- Senior management must ensure:
  - Resources adequate to the job.
  - Confidence in technical quality of forecast and analysis.
  - Forecasting team accounts for major issues senior management deems important.
- Recommendation: regard the forecast as belonging to the staff rather than the institution; publish as staff forecast with MPC views reflected via discussion and alternative scenarios.

### Implied Forward Guidance under IFT and use of FG/QE
- FG and QE used when policy rates near zero to reduce medium/long-term rates and ease credit conditions.
- Under IFT, FG is continuous: central bank provides ongoing information on current policy actions and what may be appropriate over the medium term.
- When the ZLB binds and deflation risk is main danger:
  - IFT central bank would publish a forecast with an endogenous interest rate near the floor long enough to get inflation back on track, lowering medium- and long-term yields if markets believe the forecast.
  - Publication of the forecast acts as an additional instrument; IFT applies at all times, not only during exceptional periods.
  - Markets understanding where policy is heading will move interest rates to support policy; publishing the endogenous policy rate path clarifies MPC decisions envisage a medium-term rate profile.

### Deflation risk — Japan and the euro area (experiences and implications)
- Expectations can be shock absorbers or amplifiers depending on policy credibility and activity.
- Japan:
  - Short-term interest rate near zero when Lehman event hit in September 2008; 2009 saw marked deflation.
  - Real rate rose because nominal short-term rate could not go lower.
  - Yen appreciated 25 percent between 2007 and 2009.
  - Deflationary pressures intensified; policy change in 2012 and subsequent actions led to substantial yen depreciation and large rise in Nikkei (Nikkei almost doubled between mid-2012 and end-2014), suggesting higher longer-term inflation expectations despite no nominal rate room.
- Euro area:
  - Since 2003 objective: “price increases below but close to 2 percent”; ECB mandate includes full employment and balanced growth insofar as not conflicting with price stability.
  - Possible policy adjustments: emphasize symmetry around “close to 2 percent” and consider moving to endogenous short-term interest rate in published forecasts.
  - ECB raised policy rate to 5 ¼ percent shortly before the global crisis; post-crisis inflation well below 2 percent (less than 1 percent by end-2013 and about zero by end-2014).
  - Since 2012 measures included interest rate cuts (effectively to zero in 2014), outright asset purchases, FG on future rates, and large-scale QE; early 2015 data show a pick-up in output, employment, and credit.

### Cross-region effects and expectations
- Consensus forecasts for 2015 and 2016 (surveyed Jan 2014–Apr 2015) suggest policy differences had lasting effects:
  - Expected growth of wages and real consumption remained substantially higher in the 3 IFT-like economies (United States, Canada, Czech Republic) than in euro area and Japan.
  - Reflects impact of better-anchored inflation expectations on longer-term real interest rates.

### Key statistics (selected values preserved exactly)
- Currency appreciation (Japan): yen appreciated 25 percent between 2007 and 2009.
- Table 1 (selected values; numbers in the table are in percent; bilateral USD exchange rate normalized to 100 in 2007):
  - Japan 2007: Real GDP Growth 2.2; Real Export Growth 8.7; Output Gap 0.4; CPI Inflation 0.1; Long-term Interest Rate 1.7; Short-term Interest Rate 0.6; Bilateral USD Exchange Rate 100.0
  - Japan 2008: Real GDP Growth -1.0; Real Export Growth 1.4; Output Gap -1.4; CPI Inflation 1.4; Long-term Interest Rate 1.5; Short-term Interest Rate 0.4; Bilateral USD Exchange Rate 113.9
  - Japan 2009: Real GDP Growth -5.5; Real Export Growth -24.2; Output Gap -7.1; CPI Inflation -1.3; Long-term Interest Rate 1.4; Short-term Interest Rate 0.1; Bilateral USD Exchange Rate 125.8
  - Canada 2007: Real GDP Growth 2.0; Real Export Growth 1.1; Output Gap 1.9; CPI Inflation 2.1; Long-term Interest Rate 4.3; Short-term Interest Rate 4.2; Bilateral USD Exchange Rate 100.0
  - Canada 2008: Real GDP Growth 1.2; Real Export Growth -4.5; Output Gap 0.9; CPI Inflation 2.4; Long-term Interest Rate 3.6; Short-term Interest Rate 2.4; Bilateral USD Exchange Rate 100.7
  - Canada 2009: Real GDP Growth -2.7; Real Export Growth -13.1; Output Gap -3.5; CPI Inflation 0.3; Long-term Interest Rate 3.2; Short-term Interest Rate 0.4; Bilateral USD Exchange Rate 94.0
- Table 2. Consensus CPI Inflation Expectations (Consensus Economics Quarterly Survey (April 2015)):
  - Canada (IFT CB /1 Yes (1994)): Objective 2.0; 2016 2.1 (0.1); 2017 2.1 (0.1); 2018 2.0 (0.0); Cumulative Deviations 0.2
  - Czech Republic (IFT CB /1 Yes (2002)): Objective 2.0; 2016 1.7 (-0.3); 2017 1.9 (-0.1); 2018 1.9 (-0.1); Cumulative Deviations -0.5
  - United States /2: Objective 2.3; 2016 2.2 (-0.1); 2017 2.3 (0.0); 2018 2.3 (0.0); Cumulative Deviations -0.1
  - Euro Area: Objective 2.0; 2016 1.2 (-0.8); 2017 1.5 (-0.5); 2018 1.7 (-0.3); Cumulative Deviations -1.6
  - Japan /3: Objective 2.0; 2016 1.0 (-1.0); 2017 2.0 (0.0); 2018 1.4 (-0.6); Cumulative Deviations -1.6
  - Notes:
    - 1/ IFT CBs use consistent macro forecasts to explain instrument adjustments to achieve output-inflation objectives.
    - 2/ Implicit CPI inflation objective for the U.S. estimated by the authors at about 0.3 percentage points above the Fed's official PCE inflation objective of 2.0 percent.
    - 3/ Annual CPI expectations for Japan are heavily affected by the VAT.

### Box 3 — medium-term inflation expectations as absorber or amplifier
- At the ZLB, higher expected future inflation offers a channel for FG to stimulate the economy even if nominal short rates cannot fall.
- Under credible, vigorous policy:
  - Promise to hold nominal rate near zero long enough raises expected future inflation, lowering longer-term real rates and cushioning contractionary shocks.
  - Open-economy amplification via real exchange rate depreciation and asset price increases reinforces stimulus.
- Under passive, non-credible policy:
  - Expected inflation falls (expected deflation rises), real rates rise at the ZLB, real exchange rate appreciates, asset prices fall, amplifying the shock (deflation trap).
- Empirical illustration: Czech National Bank experience since 2013 indicates influencing expectations under transparent IFT can be realistic.

### Box 4 — price-level-path stability and inflation overshooting; policy comparisons and ZLB implications
- Price-level-path stability:
  - Medium-term overshooting of the long-run inflation target after disinflation can help restore average medium-term inflation to target and keep the price level closer to stable growth path.
  - Non-overshooting (asymptotic return) leaves the price level permanently lower.
- Credibility and overheating risks:
  - Concern that planning to overshoot could unanchor expectations is countered by the point that policy can always raise rates if overheating occurs; no upper-bound constraint analogous to the ZLB exists for tightening.
- Simulated policy responses to shocks:
  - Demand shocks:
    - Loss-minimization (DM1) cushions the real economy better, with some inflation overshoot; federal funds rate increases more delayed under DM1 than IFB.
  - Supply shocks (illustrative "nasty" shock = 1-standard-deviation Phillips curve shock → 1.38 percent q/q inflation jump over 2015Q1–2015Q3):
    - All policy options simulate considerable rate hikes: up to 4 ½ percent for optimal control loss functions; up to 5 ½ percent for the Taylor rule.
    - Taylor rule: wide negative output gap; quick return of inflation to 2 percent without undershoot.
    - IFB reaction: policy rate ~4.8 percent for a couple of quarters; smaller cumulative output gap; inflation returns to target as quickly as Taylor.
    - Optimal control loss functions: bring inflation back faster but with undershoot; smaller and more gradual rate hikes; more negative output gaps; DM2 (low weight on output gap) implies a bigger output gap than DM1.
    - Price level benefit: optimal control returns price level to pre-shock levels; policy rules may leave price level altered.
- Summary on policy approaches and the ZLB:
  - Reaction functions can perform reasonably in normal times but struggle near the ZLB.
  - Quadratic optimal control loss function performs better near the ZLB because it commits to keeping rate at the floor long enough to raise inflation expectations, potentially overshooting the long-run target.
  - Where deflation risk is high, an aggressive counter-deflation (loss-minimizing) policy is prudent.
  - Recommendation: pay more attention to loss-minimization approaches while using Taylor rule and IFB as crosschecks.
- Confidence intervals and scenario practice:
  - Confidence intervals and alternative scenarios are critical for communicating forecast risks.
  - Findings from simulation figures:
    - Overshoots are very small relative to 30–70 percent confidence bands.
    - Loss-function approach implies more aggressive interest rate responses to negative demand shocks: wider confidence bands for the interest rate and much narrower bands for output and inflation than IFB.
    - Under IFB there is a 15 to 20 percent probability of deflation one year into the future.
    - Confidence bands for policy rate under loss-function approach are skewed because of the ZLB.
    - Volatility greater with initial output gap of -4 percent versus -2 percent.
  - DM1 vs DM2: output gap confidence band narrower for DM1 than DM2; DM1 and DM2 provide narrower bands for goal variables than IFB at cost of greater interest rate variance.
- Policy communication:
  - Central banks should present alternative scenarios for major contingencies in MPRs (e.g., large durable oil price shocks) to address key risks.

### Policy implications and recommendations (explicit)
- Publish a coherent staff forecast including an endogenous path for the short-term policy rate to:
  - Strengthen confidence in the inflation target and the value of money.
  - Aid the transmission mechanism by aligning the term structure of interest rates with policy objectives.
  - Improve accountability by clarifying intended policy path and responses to alternative developments.
- Maintain staff ownership of the baseline projection in many central banks to ensure coherence and clarity, while reflecting MPC members’ views via alternative scenarios and discussion of uncertainties in the MPR.
- Use published forecasts as an ongoing instrument under IFT (not only a crisis tool) to shape expectations and support policy, reducing reliance on ad hoc forward guidance switches.
- Consider symmetric emphasis around inflation objectives where appropriate (example: ECB “close to 2 percent”) and moving from exogenous to endogenous short-term interest rates in published forecasts to better align policy communication and expectations.
- Pay increased attention to loss-minimization (optimal control) approaches in both normal and abnormal circumstances, especially when the ZLB is a material risk.

*Italic source: Excerpt from IMF working paper _wp15132 (conclusion and related sections) as provided.*

### References

### References

### I. INTRODUCTION
- The paper traces the development and implementation of the inflation-forecast targeting (IFT) approach to monetary policy.
- Observations about IFT:
  - The approach has been adopted by many central banks in advanced and emerging market economies.
  - It has generally helped to establish an environment of low and stable inflation in economies that previously had a poor record for monetary stability.
  - The regime was often initiated under difficult circumstances, after a crisis, or in the midst of structural change.
  - Many practitioners of IFT have been highly open economies, susceptible to external shocks, especially the smaller ones.
  - The regime is in its third decade.
- Recent focus since the global financial crisis:
  - Much attention on how monetary policy contributes to dealing with a weak economy when policy interest rates are near or at the zero lower bound (ZLB), the so-called “dark corners”.1
  - Growing emphasis on the importance of forward guidance as current interest rates are at the floor and conventional monetary policies are less effective.
  - The paper argues that the IFT framework can help countries avoid dark corners.
- Paper’s broader focus:
  - Several issues important for the successful functioning of an IFT strategy, many related to transparency and communications of the central bank.
  - Views based in part on analysis in this and other papers and on the experience of a number of advanced economy and developing/emerging economy central banks.
- Structure (as described):
  - Section II: history of Inflation Targeting (IT) and challenges leading to adoption of IFT.
  - Section III: background and definition of IFT; role of expectations as a nominal anchor; transmission mechanism of monetary policy; importance of an endogenous interest rate in forecasting; discussion of publishing a staff forecast in the Monetary Policy Report.
  - Section IV: how IFT would eliminate differences between current inflation and the long-run target; application to situations where the zero interest rate floor or ZLB is binding. IFT central banks provide forward guidance by publishing a macro forecast with an endogenous interest rate path and then either publishing the interest rate path or describing it verbally.
  - Section V: simulations with a simple model of the U.S. economy; examples of policy-making strategies and approaches to interest rate setting (Taylor rule, inflation-forecast-based reaction function, minimization of a loss function or optimal control policy making); comparison of projected Fed funds rate paths under the three approaches; note that optimal policy strategy could in certain circumstances result in a planned overshoot of inflation from its long-term target.
  - Concluding remarks in Section VI.
  - An accompanying supplement provides detail and technical background; all references to annexes refer to the accompanying supplement.

Notes and cross-references from the Introduction:
- 1: The term inflation targeting or IT is commonly used; IFT better describes behavior of dual-mandate central banks that focus on forecast of future inflation and on actions to bring future inflation back to target. IFT is used throughout the paper.
- 2: See Blanchard (2014).
- 3: For applications to the United States and the Czech Republic, see Alichi and others (2015a, b).

### II. SOME HISTORY OF INFLATION TARGETING
- Development of IT as a monetary policy regime was influenced by:
  - Trends in economic theory.
  - Pragmatic learning-by-doing.
  - Experience of other IT central banks.
- The paper notes:
  - Some useful examples of how IT central banks have transitioned to full-fledged IFT are illustrated in Box 1.
- Footnotes and references:
  - 4: While some central banks have reduced their policy rate to slightly below zero, the term zero lower bound or ZLB is continued to be used to refer to situations where the policy interest rate is very near to zero in either direction.
  - 5: For a brief introduction to the essential ingredients of inflation targeting, see Freedman and Laxton (2009a, 2009b, 2009c).

*Source: References section of the provided IMF paper content.*

### Box 1: Some Examples of Learning from Experience

### Box 1: Some Examples of Learning from Experience

### Country experiences with inflation targeting (IT) and learning
- New Zealand
  - First country to embark on IT in 1989 and later developed a full-fledged IFT regime.
  - Monetary policy credibility rose over time.
  - Inflation was quickly reduced to less than 2 percent in 1992.
  - Initial use of the exchange rate as the main instrument led to instrument instability.
  - Introduced the forecasting and policy analysis system (FPAS) in 1997 and adopted immediate full disclosure of the central bank forecast.
- Canada
  - Moved to IT in 1991 after examining New Zealand’s experience.
  - Elements of FPAS were already being put in place; previous policy emphasized price stability (understood to involve a long-run inflation objective below 2 percent).
  - IT program announced in early 1991 had an eventual target of 2 percent, unchanged since.
  - After fiscal policy was put on a sustainable footing in 1995, long-term inflation expectations stabilized at the 2 percent target.
- Czech Republic
  - Adopted IT in 1998 following collapse of a fixed exchange rate policy and widespread bank failures in 1997.
  - Inflation had been running at almost 10 percent since 1993 and was accelerating at the time IT was adopted.
  - With assistance from Bank of Canada staff and the IMF, CNB was using an FPAS with a model-based forecast by 2002.
  - Began publishing detailed quarterly forecasts, including the forecast path for the interest rate, in 2008.
  - Surveys over the past decade show strong public confidence in the 2 percent target; internationally noted for transparency.

### Historical and theoretical context
- Post-1970s inflation and time-consistency theory
  - Kydland and Prescott (1977) emphasized discretionary monetary policy bias toward inflation, prompting search for more robust frameworks.
  - Barro and Gordon (1983) proposed commitment to time-consistent price-stability policy to avoid short-run inflationary temptations.
- Early IT practice and evolution
  - Early adopters (e.g., New Zealand) used rigid approaches that reduced inflation but could produce instrument instability.
  - During the 1990s, emphasis shifted from rigid target adherence to transparent strategies for eliminating deviations over time.
  - Announcing an explicit numerical target was a major clarifying step.
  - Increased use of communications (Monetary Policy Reports, speeches, media briefings) accompanied IT adoption.
  - Transparent pursuit of low-inflation objectives by politically accountable central banks functioned as a constraint on discretion, reducing inflation bias.

### Accountability, transparency, and institutional arrangements
- Central bank accountability
  - Central banks are typically accountable for monetary policy to government or parliament and implicitly to the public.
  - Clear delegation of responsibility associated with large increases in transparency.
  - Accountability without transparency is ineffective.
- Communications and disclosure practices
  - Debate centers on what to disclose beyond the inflation target and current policy interest rate, notably elements of quarterly macroeconomic forecasts.
  - Publishing forecasts for inflation and output is widely accepted.
  - The evolution toward inflation-forecast targeting (IFT) involved broader disclosure and use of forecasts as intermediate targets.

### Milestones on the road to full-fledged IFT
- Announcement of targets with a multi-year horizon—clarity of target.
- Precision on the policy interest rate setting—clarity of instrument.
- Transparent communications on policy implementation.
- Publication of a complete macro forecast (including inflation)—clarity of the intermediate target (IFT).
- Publication of a conditional forecast path, alternative scenarios and confidence bands for the short-term interest rate (full-fledged IFT).

### Defining Inflation-Forecast Targeting (IFT) and FPAS requirements
- Core features of IFT as used in the paper:
  - Monetary policy is based on a long-run low inflation target, and a medium-term forecast path to this target.
  - The central bank has a structured FPAS that maintains relevant data bases and produces a model-based staff forecast and associated economic analysis on a regular schedule.
  - Under the FPAS, the Projection Team (PT) presents the forecast shortly before a decision meeting of the Monetary Policy Committee (MPC).
  - The forecast path for the short-term interest rate is endogenous within the model, with the rate varying to achieve the long-run inflation target and to eliminate any output gap.
  - The staff forecast is a key input into MPC decisions, but MPC members may incorporate other information.
  - Soon after the policy decision, the associated forecast path for key macroeconomic variables is disclosed, highlighting the path for the inflation rate.
  - The rationale for policy actions is explained at regular intervals (usually quarterly) in a Monetary Policy Report (MPR).
  - The MPR outlines the conditional forecast path for the short-term interest rate, either with explicit numbers or a qualitative description.
  - Forecast presentations show confidence bands around the baseline and consider alternative scenarios.
- Majority of IT central banks would be considered inflation-forecast targeters (examples listed in source).

### Transmission mechanism, endogenous policy response, and publication of interest-rate forecasts
- Transmission mechanism
  - Expectations of future policy-rate movements in the short to medium term crucially influence longer-term interest rates and the exchange rate.
  - The policy rate path (medium-term expectations) matters more than the current policy rate for affecting borrowing costs.
- Endogenous policy response
  - Central banks choose among possible responses to inflation deviations, balancing short-run output-inflation trade-offs; options include asymptotic approaches or planned overshoots.
  - Example: Following a sudden drop in world oil prices, the PT simulates impacts and proposes an interest-rate path that returns inflation to target over the medium term while considering output costs; alternative responses can be simulated with confidence bands.
  - A full policy description entails the entire future path of the interest rate, not just the immediately effective rate.
- Publication of endogenous interest-rate forecasts
  - Models used in IFT logically should incorporate an endogenous interest rate; exogenous-rate models lack a nominal anchor.
  - Many central banks produce endogenous interest-rate paths in their models but decide not to publish them to preserve policy flexibility and avoid signaling a commitment.
  - Counterarguments for publication: because policy affects longer-term borrowing through expectations, publishing conditional policy-rate paths and confidence intervals can clarify how policymakers intend to influence the yield curve.
  - The path chosen reflects preferences over the inflation-output trade-off; higher weight on output stability implies smoother interest-rate adjustments and a slower return to the inflation target.

*Source: _wp15132 - Box 1: Some Examples of Learning from Experience*

### conclusion that the interest rate decision, at any point in time, envisages a time profile for the

### _wp15132 - conclusion that the interest rate decision, at any point in time, envisages a time profile for the

### Full disclosure of the central bank forecast: rationale and effects
- Full disclosure of the central bank forecast could reinforce the effectiveness of monetary policy in two ways:
  - By showing a coherent view of the future, with inflation returning over the medium term to the desired long-run target rate, confidence in the goal, a reliable value of money, would be strengthened.
  - The published path for the short-term interest rate would help move the term structure of interest rates in a way that would assist the transmission mechanism.
- In terms of objectives, the payoff from this reinforcement of policy effectiveness would be:
  - a reduced cost of eliminating deviations of actual inflation from the long-run target rate, or equivalently,
  - an improved short-run inflation-output trade-off.
- A connected argument: publication of the forecast interest rate path further improves the accountability of the central bank.
- Leading central banks that publish endogenous interest rate path forecasts include the Reserve Bank of New Zealand, the Czech National Bank, the Bank of Israel, the Norges Bank, and the Sveriges Riksbank.

### Ownership of the projection (Box 2): staff versus policymaking body
- Practical judgment: in many cases it would be better for the staff to have ownership of the projection, especially for central banks with relatively large policymaking bodies or divergent policymaker views.
- Advantages of staff ownership:
  - Producing a consistent and coherent projection that reflects good economic reasoning and specific economy circumstances.
  - Simplifying production of the Monetary Policy Report (MPR) by presenting the forecast as an important staff input to decision making, but not the sole input.
  - Avoiding the inefficiency and external confusion that would arise from publishing multiple projections if voting MPC members have divergent views.
- Role of MPC members:
  - MPC members should focus on broad, strategic questions and not be closely connected to the production process.
  - Members must have solid economic arguments when differing from staff views (examples: persistence of inflation pressures; timing of easing when baseline projects a slowdown but data not yet reflecting it).
- Communication benefits:
  - Clear explanations of differences between staff and MPC views in the MPR would help market participants understand actions (or inaction) and could increase central bank credibility by exposing the debate and likely reactions to future data.
- Real-world example (Czech National Bank statement): the forecast is drawn up by the CNB’s Monetary Department; the forecast is the key, but not the only, input to the Bank Board’s decision-making; the Bank Board’s final decision may not correspond to the message of the forecast due to new information or divergent views.

### Technical production and institutional requirements
- Deriving model-based baseline forecasts, confidence intervals, and alternative scenarios requires highly specialized staff resources.
- Senior management must ensure:
  - resources are adequate to the job,
  - confidence in the technical quality of the forecast and associated analysis,
  - the forecasting team accounts for major issues for the outlook as senior management sees them.
- Recommendation: regard the forecast as belonging to the staff rather than the institution; advantages of publishing the forecast as a staff forecast are discussed in Box 2.

### Implied Forward Guidance (IFT) and forward guidance (FG)
- FG and QE have been used to reduce medium-term and long-term interest rates and ease credit conditions after policy rates were cut almost to zero.
- Under IFT, FG is part of a continuous process where the central bank provides a continuous flow of information on current policy actions and on what actions may be appropriate over the medium term.
- During periods when the ZLB is binding and the main danger is deflation, an IFT central bank would publish a forecast with an endogenous interest rate near the floor long enough to get inflation back on track; if this forecast affects market expectations, medium- and long-term rates fall below long-run equilibrium values.
- Publication of the forecast becomes an additional instrument, akin to the Fed’s FG since 2008, but the IFT strategy applies at all times, not only during exceptional periods like ZLB.
- Principle: markets understanding where policy is heading will move interest rates in a way that supports policy; publishing the endogenous policy rate path clarifies that MPC decisions envisage a medium-term rate profile, not merely a rate until the next meeting.
- Under IFT, the central bank communicates not only a possible future policy-rate path but also how this path might change in response to developments and the rationale for policy actions—improving accountability and policy effectiveness.
- IFT avoids the need to give special guidance about when policy approaches switch on/off or trigger values for inflation/unemployment; contrasts with the Fed’s changing FG policy since 2008 which might give an impression of improvisation.

### Deflation risk: Japan and the euro area — experiences and implications
- Expectations can act as shock absorbers or amplifiers depending on whether the policy regime is credible and active.
- Japan:
  - Short-term interest rate near the zero floor when the Lehman event hit in September 2008.
  - 2009 saw marked deflation.
  - The real rate rose because the nominal short-term rate could not go lower.
  - The yen appreciated 25 percent between 2007 and 2009.
  - Deflationary pressures intensified, in part from a sharp drop in exports, possibly amplified by public beliefs that policy would not offset the contractionary shock.
  - 2012 policy change and subsequent actions led to substantial depreciation of the yen and a strong rise in the Nikkei Index (Nikkei almost doubled between mid-2012 and end-2014), suggesting raised longer-term inflation expectations even though the Bank of Japan had no room to reduce the policy interest rate.
- Euro area:
  - Since 2003 monetary policy objective: “price increases below but close to 2 percent.” The ECB mandate also calls for full employment and balanced growth insofar as they do not conflict with price stability.
  - A possible change: emphasize symmetry around the “close to 2 percent” inflation objective and consider moving to an endogenous short-term interest rate in published forecasts (from current exogenous rate taken from market forward rates).
  - ECB raised policy rate to 5 ¼ percent shortly before the global crisis as inflation rose; post-crisis saw severe weakness and inflation well below 2 percent (less than 1 percent by end-2013 and about zero by end-2014).
  - Since 2012, measures included interest rate cuts (effectively to zero in 2014), outright asset purchases, FG on future rates, and large-scale QE; early 2015 data on output, employment, and credit show a pick-up.
- Cross-region effects:
  - Consensus forecasts for 2015 and 2016 (surveyed Jan 2014–Apr 2015) suggest cross-region policy differences had a lasting effect: expected growth of wages and real consumption remained substantially higher in the 3 IFT-like economies (United States, Canada, Czech Republic) than in the euro area and Japan.
  - This reflects, among other things, the impact of better-anchored inflation expectations on longer-term real interest rates.

### Key statistics and comparative indicators (selected from text)
- Currency appreciation (Japan): yen appreciated 25 percent between 2007 and 2009.
- Table 1 (selected values, note: numbers in the table are in percent; bilateral USD exchange rate normalized to 100 in 2007):
  - Japan 2007: Real GDP Growth 2.2; Real Export Growth 8.7; Output Gap 0.4; CPI Inflation 0.1; Long-term Interest Rate 1.7; Short-term Interest Rate 0.6; Bilateral USD Exchange Rate 100.0
  - Japan 2008: Real GDP Growth -1.0; Real Export Growth 1.4; Output Gap -1.4; CPI Inflation 1.4; Long-term Interest Rate 1.5; Short-term Interest Rate 0.4; Bilateral USD Exchange Rate 113.9
  - Japan 2009: Real GDP Growth -5.5; Real Export Growth -24.2; Output Gap -7.1; CPI Inflation -1.3; Long-term Interest Rate 1.4; Short-term Interest Rate 0.1; Bilateral USD Exchange Rate 125.8
  - Canada 2007: Real GDP Growth 2.0; Real Export Growth 1.1; Output Gap 1.9; CPI Inflation 2.1; Long-term Interest Rate 4.3; Short-term Interest Rate 4.2; Bilateral USD Exchange Rate 100.0
  - Canada 2008: Real GDP Growth 1.2; Real Export Growth -4.5; Output Gap 0.9; CPI Inflation 2.4; Long-term Interest Rate 3.6; Short-term Interest Rate 2.4; Bilateral USD Exchange Rate 100.7
  - Canada 2009: Real GDP Growth -2.7; Real Export Growth -13.1; Output Gap -3.5; CPI Inflation 0.3; Long-term Interest Rate 3.2; Short-term Interest Rate 0.4; Bilateral USD Exchange Rate 94.0
- Table 2. Consensus CPI Inflation Expectations (Objective and forecasts with deviations in parenthesis; source: Consensus Economics Quarterly Survey (April 2015)):
  - Canada (IFT CB /1 Yes (1994)): Objective 2.0; 2016 2.1 (0.1); 2017 2.1 (0.1); 2018 2.0 (0.0); Cumulative Deviations from Inflation Objectives 0.2
  - Czech Republic (IFT CB /1 Yes (2002)): Objective 2.0; 2016 1.7 (-0.3); 2017 1.9 (-0.1); 2018 1.9 (-0.1); Cumulative Deviations -0.5
  - United States /2: Objective 2.3; 2016 2.2 (-0.1); 2017 2.3 (0.0); 2018 2.3 (0.0); Cumulative Deviations -0.1
  - Euro Area: Objective 2.0; 2016 1.2 (-0.8); 2017 1.5 (-0.5); 2018 1.7 (-0.3); Cumulative Deviations -1.6
  - Japan /3: Objective 2.0; 2016 1.0 (-1.0); 2017 2.0 (0.0); 2018 1.4 (-0.6); Cumulative Deviations -1.6
  - Notes in Table 2:
    - 1/ IFT CBs use consistent macro forecasts to explain how they are adjusting their instruments to achieve their output-inflation objectives.
    - 2/ The implicit CPI inflation objective for the U.S. is estimated by the authors at about 0.3 percentage points above the Fed's official PCE inflation objective of 2.0 percent.
    - 3/ Annual CPI expectations for Japan are heavily affected by the VAT.

### Policy implications and recommendations
- Publish a coherent staff forecast including an endogenous path for the short-term policy rate to:
  - Strengthen confidence in the inflation target and the value of money.
  - Aid the transmission mechanism by aligning the term structure of interest rates with policy objectives.
  - Improve accountability by clarifying the central bank’s intended policy path and its reaction to alternative developments.
- Maintain staff ownership of the baseline projection in many central banks to ensure coherence and clarity, while allowing MPC members’ views to be reflected via alternative scenarios and discussion of uncertainties in the MPR.
- Under IFT, use published forecasts as an ongoing instrument (not just a crisis tool) to shape expectations and support policy, reducing the need for ad hoc forward guidance switches.
- Consider symmetric emphasis around inflation objectives where appropriate (example: ECB “close to 2 percent”) and moving from exogenous to endogenous short-term interest rates in published forecasts to better align policy communication and expectations.

*Italic source: Excerpt from IMF working paper _wp15132 (conclusion and related sections) as provided.*

### Box 3: Medium-Term Inflation Expectations as Shock Absorber or Amplifier

### Box 3: Medium-Term Inflation Expectations as Shock Absorber or Amplifier

### ZLB transmission via medium-term inflation expectations
- At the zero lower bound (ZLB), the nominal interest rate cannot decline, but expected inflation offers a channel through which forward guidance (FG) can stimulate the economy.
- If monetary policy is active and credible:
  - The central bank can persuade the public it will eventually return inflation to the long-run target.
  - A promise to hold the nominal interest rate near zero for an extended period raises expected future inflation.
  - Higher expected future inflation reduces longer-term real interest rates even though the nominal rate is at the ZLB.
  - These movements act as a buffer to a contractionary shock.
- Under credible, vigorous policy, the central bank might present a stimulative forecast in which inflation overshoots over the medium term before returning to the long-run target.

### Open-economy amplification: real exchange rate and asset prices
- The real exchange rate depreciates under higher expected medium-term inflation, reinforcing the stimulative effect.
- Asset prices rise because:
  - A reduction in real interest rates lowers the real discount rate, boosting asset valuations.
  - Exchange rate depreciation positively affects profits, further supporting asset prices.
- Higher asset prices stimulate spending, amplifying the real interest rate channel in an open economy.

### Credibility versus passivity: absorber versus amplifier
- If monetary policy is credible and active → expectations act as an absorber of shocks:
  - Expected inflation rises, real long-term rates fall, real exchange rate depreciates, asset prices rise, and spending is stimulated.
- If monetary policy is passive and not credible → expectations act as an amplifier of shocks:
  - Expected inflation falls (or expected deflation rises), real interest rates rise at the ZLB, the real exchange rate appreciates, asset prices fall—leading to a classic deflation trap.
- The difference between the two regimes is illustrated by a flowchart (Box Figure 3.1).

### Practical example and credibility evidence
- The exchange rate policy used by the Czech National Bank since 2013, which relied heavily on influencing expectations under a transparent inflation-forecast-based (IFT) framework, suggests that influencing expectations in this way can be a realistic prospect.

*Source: IMF staff calculations*

### Box 4: Price-Level-Path Stability and Inflation Overshooting

### Box 4: Price-Level-Path Stability and Inflation Overshooting

### Price-level-path stability and overshooting
- If the economy has been mired in an undesired disinflation, an overshooting of the long-run inflation target in the medium term is harmless because it helps restore the average medium-term rate of inflation to target and holds the price level closer to a stable long-run growth path (illustrated by the red line vs. the blue line for non-overshooting IT).
- A strongly stabilizing IFT policy would have an outcome similar to price-level targeting (example: Canada per Kamenik and others, 2013).
- If inflation returns asymptotically to target after a deflationary period (non-overshooting), the price level path remains permanently lower (blue line).
- A contractionary shock with a non-credible passive policy at the ZLB can amplify the shock via expectations (yellow line).

### Central-bank credibility, expectations, and risks of overheating
- Some central bankers fear that planning to inflate beyond target could unanchor expectations.
- Counterpoint: if the economy overheats, the central bank would raise the policy interest rate to counter inflationary effects; there is no equivalent to the ZLB to constrain interest rate increases.
- Credibility grants policymakers scope to be flexible in achieving objectives.
- The inflation objective is symmetric; there is no reason people would be more concerned by overshoots than undershoots, especially when overshoots follow a series of undershoots.

### Demand shocks (V.3.2) — IFB reaction function vs. DM1 optimal control
- Simulated shocks: mild contractionary (green), large contractionary (red), expansionary (black); baseline repeated from Figure 7 (blue).
- Loss-minimization (DM1) cushions the real economy better against contractionary demand shocks, with some overshoot in inflation under DM1.
- The increase in the federal funds rate is more delayed under DM1 than under the IFB reaction function.
- Rationale: loss-function approach places high marginal cost on widening the negative output gap of initial conditions; optimal control focuses on disinflationary pressures and deflationary risks rather than a potential future inflation overshoot.
- Reactions to a positive demand shock are similar across IFB and DM1:
  - Policy rate rises to 3 percent.
  - Output gap goes through a mild cycle toward zero.
  - Inflation has a modest overshoot.

### Supply shocks (V.3.3) — illustrative "nasty" supply shock
- Definition: "Nasty" = 1-standard-deviation shock to the stochastic term in the Phillips curve, equivalent on impact to a 1.38 percent jump in quarter-on-quarter inflation; duration 2015Q1-2015Q3.
- Starting point: zero output gap; 2 percent inflation; stable equilibrium real interest rate.
- All four policy options simulate considerable policy rate hikes:
  - Up to 4 ½ percent for the optimal control loss functions.
  - Up to 5 ½ percent for the Taylor rule.
- Comparative outcomes:
  - Taylor rule: wide negative output gap; returns inflation to 2 percent quickly without undershoot.
  - IFB reaction function: outperforms Taylor rule; policy rate ~4.8 percent for a couple of quarters; smaller cumulative output gap; inflation returns to target as quickly as Taylor rule because IFB treats inflation impact as one-off.
  - Optimal control loss functions:
    - Bring inflation back to target faster than reaction functions, but with an undershoot.
    - Interest rate hikes are smaller and more gradual.
    - Output gaps are more negative because an overshoot implies lower inflation later, producing higher real rates compared to reaction functions.
    - DM2 (low weight on output gap) implies a bigger output gap compared to DM1.
    - Benefit: price level is brought back to pre-shock levels under optimal control; under policy rules the price level can permanently change and behave like a random walk over long sequences of shocks.
- Loss-minimizing strategies produce slightly more negative output gaps but smaller inflation gaps; because disutility from higher inflation increases disproportionately with the level of inflation, larger peak inflation errors under IFB are heavily penalized.
- Overall superiority of loss-minimizing approach reflected in much narrower output and inflation confidence bands than those for IFB, though comparative advantage is not large for this particular shock.

### Summary on policy approaches and the ZLB (V.3.4)
- Reaction functions can perform reasonably in normal times but struggle in abnormal times, especially near the ZLB.
- Quadratic optimal control loss function performs better as ZLB approaches because it commits to keeping the rate at the floor long enough that inflation will rise, possibly above the long-run target, thereby boosting inflation expectations and reducing the real medium-term interest rate even when nominal short rates cannot fall further.
- Where there is little risk of sustained inflationary pressure but a high risk of getting stuck in deflation, such an aggressive counter-deflation policy is prudent.
- Recommendation: pay more attention to loss-minimization approaches in both normal and abnormal circumstances, while using Taylor rule and IFB as crosschecks.

### Confidence intervals and alternative simulations (V.4)
- Confidence intervals and alternative scenario simulations are critical for communicating forecast risks.
- Figures 14–16: left panels show confidence intervals for fed funds rate, output gap, and year-on-year inflation under IFB and two calibrations of optimal control; right panels show results if initial output gap is believed to be -4 percent rather than -2 percent.
- Shaded bands: confidence intervals from 10 to 90 percent in 10 percentage-point increments.
- Findings:
  - Overshoots are very small relative to, for example, the 30 to 70 percent confidence bands.
  - Loss-function approach implies more aggressive interest rate responses to negative demand shocks, resulting in:
    - Wider confidence bands for the interest rate.
    - Much narrower bands for output and inflation than IFB.
  - Under IFB there is a 15 to 20 percent probability of deflation one year into the future.
  - Confidence bands for policy rate under loss-function approach are skewed because of the ZLB.
  - Volatility is greater with an initial output gap of -4 percent versus -2 percent because the likelihood that the floor impedes effective stabilization is higher.
- Figures 17–19 compare loss-function responses to small and large supply shocks versus IFB; the large supply shock doubles the shocks in the small supply shock scenario.
- Additional comparative notes:
  - Output gap confidence band is narrower for DM1 than DM2 (reflecting lower DM2 weight on output gap), though the difference is not large.
  - DM1 and DM2 provide narrower bands for goal variables than IFB at the cost of greater interest rate variance.

### Policy communication and scenario practice
- Central banks can communicate risks by presenting alternative scenarios for contingencies (e.g., large durable oil price shocks) in Monetary Policy Reports to address the two or three most important risks.

### Conclusion (VI) — implications for inflation targeting and transparency
- Inflation targeting has been used for almost 25 years and is widespread, indicating robustness.
- Key principles underpinning robustness include transparency:
  - Central banks should publish an explicit path for the inflation rate forecast, showing how quickly inflation will be returned to the long-run target and what this implies for the interest rate policy instrument.
  - Transparency influences medium-term expectations of nominal interest rates and inflation, aiding real interest rate movements and making transmission more effective.
  - A coherent macroeconomic explanation of policy intentions strengthens credibility and the nominal anchor.
- A credible IFT framework allows flexibility: QE, exchange-rate intervention, and other extraordinary measures can be placed in a longer-term context; FG is continuous via regular disclosure of expected future interest-rate paths.
- When constrained by the ZLB, influencing medium-term inflation expectations is crucial; an aggressive counter-deflation strategy may involve deliberate medium-term overshooting of the long-run target and, if properly explained, need not weaken confidence in the long-run objective.

*Source: IMF staff calculations and analysis as presented in Box 4: Price-Level-Path Stability and Inflation Overshooting.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp15132.pdf_