## wp1721

## Source details

**Canonical URL:** [wp1721](https://www.imf.org/-/media/files/publications/wp/wp1721.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/wp1721.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/wp1721.pdf.json)

---

### 1.1 Background — Context, transparency and rationale for IFT
- Describes the Czech National Bank (CNB)’s experience implementing an Inflation-Forecast Targeting (IFT) regime and building a system to provide policymakers the economic information needed to implement IFT.
- Main assessment: experience has been “very successful”; main achievement was to establish confidence in monetary policy amid many disturbances this century.
- Purpose: provide guidance for other countries considering similar monetary policy reforms.

Key analytical points
- IT/IFT logic:
  - The inflation forecast is an ideal conditional intermediate target (Svensson (1997)).
  - IT implies an endogenous interest rate; a model with an exogenous interest rate lacks a nominal anchor and allows inflation to drift indeterminately after disturbances.
  - Under IFT the nominal anchor is the public expectation that inflation will converge to the announced long-run objective; monetary policy must react to return inflation to the long-run target, implying an endogenous interest-rate path.
- Transparency continuum (milestones):
  - 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 inflation (and output) forecast—clarity of intermediate target—IFT;
  - publication of a conditional forecast path and confidence bands for the short-term interest rate.
- IFT operational features (as used in the paper):
  - policy based on a long-run low inflation target and a medium-term forecast path to this target;
  - the short-term interest rate forecast is endogenous within the model, varying to achieve the long-run inflation target and eliminate any output gap;
  - the forecast is a key input into the MPC decision but not binding on members;
  - soon after policy decisions the associated forecast path for key macro variables is disclosed, highlighting inflation path;
  - rationale for policy actions explained at regular intervals (usually quarterly) in a Monetary Policy Report (MPR);
  - MPR outlines the conditional forecast path for the short-term interest rate, qualitatively or with explicit numbers;
  - risks emphasized verbally and with confidence bands for variables with numerical forecasts.
- Examples of IFT central banks publishing endogenous interest-rate forecasts:
  - Reserve Bank of New Zealand; Czech National Bank; Bank of Israel; Norges Bank; Sveriges Riksbank; U.S. Federal Reserve (noted for 2 percent long-term inflation objective announced in 2012).

### 1.3 The nominal anchor
- Credibility of the long-run inflation target underpins IFT: the nominal anchor is the firm expectation that monetary policy will keep inflation stable and near the target.
- Maintaining the anchor requires policy to respond systematically to this objective.
- Forward-looking policy adjusts the future expected path of the policy interest rate when unanticipated disturbances hit, bringing inflation back gradually to the target while limiting output disruptions.
- Policy feedback operates via an endogenous short-term interest rate (feedback from expected future inflation to the policy instrument).

### 1.4 The transmission mechanism
- Two traditional channels:
  - Domestic channel: via the short-term market interest rate, controlled closely by the central bank.
  - External channel: via the exchange rate.
- Modern emphasis on expectations: the whole "Policy Rate Path" matters, not just the current setting; impact of policy actions depends on expected duration.
- At the ELB, a credible central bank can use expectations to provide an effective anti-deflation policy (see Svensson, 2003).

### 1.5 Forecasting and Policy Analysis System (FPAS)
Core FPAS elements
- A monetary policy model with a loss function or policy reaction function under which the short-term interest rate responds endogenously to return inflation to the long-run target.
- Communication systems linking policymakers, model builders, and forecasters to produce baseline and alternative scenarios with agreed assumptions.
FPAS outputs and uses
- Baseline forecasts including an endogenous forecast path for the short-term interest rate.
- Risk assessments and alternative scenarios, each with an endogenous interest-rate path.
- Each forecast contains:
  - (a) an inflation projection showing return to the long-run target within a medium-term horizon; and
  - (b) an endogenous path for the policy interest rate.
- Relevant medium-term horizon is not fixed; gaps often substantially eliminated within 4-6 quarters, but obdurate shocks may take more time.
Institutional lessons
- Absence of FPAS is not a good argument to delay IT adoption; invest in an FPAS immediately after regime adoption.
- FPAS development is iterative—“an FPAS never stands still.”

### Box 2 — Medium-term expectations at the ELB (mechanism, amplification, credibility)
- At the ELB nominal rate cannot decline, but expected inflation can stimulate the economy: commit to holding the interest rate near the ELB for an extended period so the public expects higher future inflation, lowering longer-term real rates.
- Open-economy amplification: real exchange rate depreciation raises profits and supports asset prices; lower real discount rates boost asset prices and spending.
- Credibility conditions required:
  - persuade public nominal rate remains at the floor for extended period;
  - persuade public inflation will rise over the medium term, perhaps above the target, and eventually return to target.
- Risks if policy is passive/non-credible: expected inflation would fall, real interest rates rise, real exchange rate appreciate, asset prices fall—producing a deflation trap.
- CNB’s exchange rate policy since 2013 cited as evidence expectations-based influence can be effective under transparent IFT.

### 1.6 To publish or not to publish the forecast policy rate
Practice and concerns
- All IFT central banks publish forecast for major macro variables; disclosure of endogenous policy-rate path varies.
- Table of 13 IFT central banks (publication of policy rate path and year when started):
  - New Zealand — Yes (1997)
  - Norway — Yes (2005)
  - Sweden — Yes (2007)
  - Israel — Yes (2007)
  - Czech Republic — Yes (2008)
  - United States — Yes (2012)
  - Canada — No
  - Chile — No
  - Colombia — No
  - Hungary — No
  - South Korea — No
  - Peru — No
  - Turkey — No
- Principal worry: public might misinterpret a published explicit path as a commitment, or underestimate uncertainties.
Mitigating practices
- Publish confidence bands around the endogenous interest rate path.
- Emphasize conditional nature of projections; forecast-consistent interest rate path “should in no way be interpreted as a commitment” and represents the most probable future path given initial assumptions and information.
Experience and timing
- Central banks were cautious; practice of publishing rate forecasts typically followed many years after IT adoption (shortest gap: 4 years at Norges Bank).
- Markets adapted quickly to conditional rate paths in New Zealand (since June 1997), Norway (since November 2005), Sweden (since February 2007), Israel (since July 2007), Czech Republic (since February 2008).

### 2.2 Medium-term focus — QPM-Gap: model design, evolution and transition
Model role and adoption steps
- QPM-Gap: a small-scale calibrated monetary policy model for the Czech Republic developed with IMF technical assistance.
- Steps to introduce QPM-Gap into CNB decision making included committing staff resources, step-by-step pilot development, recalibration over ~18 months, in-sample testing, shadow forecasts, and agreeing operational rules with clear deadlines.
- Switch to QPM-Gap in July 2002 marks start of IFT in Czech Republic because it had medium-term focus, forward-looking channels, internal forward-looking interest-rate term structure, external exchange-rate channel, and an endogenous short-term interest rate.
Model structure and features
- Inflation disaggregation: separate Phillips curves for core, food and fuel inflation; administered prices treated as exogenous.
- Real side: single aggregate output gap.
- NTF tools decomposed aggregate GDP forecast into household consumption, government consumption, investment, exports and imports.
Model modifications during 6 years as core forecasting model
- First modification: recalibrated long-term equilibrium paths — increased rate of appreciation of equilibrium real exchange rate; reduced domestic real equilibrium interest rate, motivated by forecast errors (strong uptrend in real exchange rate).
- Second modification (January 2007): added real wage gap to the output gap as a component of the real marginal cost gap (drawing on experience with GPM-g3).
- Third modification: reduced second-round effects of administrative measures and short-term cost-push shocks in the Phillips curve in line with weakening pass-through and lower inflation persistence.
Transition to DSGE (QPM-g3) in 2008
- QPM-g3 differences vs QPM-Gap:
  - QPM-g3: explicit derivation using “behavioural principles”; QPM-Gap: reduced form.
  - Both use model-consistent expectations.
  - QPM-g3 preserves stock-flow consistency and works with level variables; QPM-Gap works with “gaps”.
  - QPM-g3 has basic national accounts disaggregation and a balanced-growth path; QPM-Gap has equilibrium trends and no GDP structure.
  - QPM-g3 includes implicit fiscal block; QPM-Gap has simple fiscal block.
  - Both used an inflation-forecast-based interest rate reaction function.
- Practical implications:
  - QPM-g3’s structural form provides degrees of freedom to incorporate judgment consistently.
  - QPM-g3 works with level variables and thus does not require NTF disaggregation tools for spending components; satellite models using Kalman-filter methods provided inflation detail.
  - CNB experience: improved forecasting performance for inflation on the monetary policy horizon, especially since early stages of the Great Recession; large forecast errors attributed to wrong assumptions rather than wrong model.
Output gap and inflation drivers
- QPM-Gap: output gap was main domestic driver of underlying inflation (until real wage gap added).
- QPM-g3: underlying inflation driven by real marginal costs and mark-ups; no aggregate output gap as such, though domestic mark-ups correlate highly with standard output gap estimates.
- Policymakers still receive output gap estimates via potential output estimates from non-core models (HP filter, multiequation Kalman filter, production function).

Organizational changes and forecasting process
- Early separation into Real Economy Division and Economic Modeling Division caused tensions; restructured in 2004 into a single Macroeconomic Forecasting Division with a division director managing the whole forecast process.
- Forecasting team of about 10 people; team head chosen from core model operators and rotates usually after 2-4 forecast exercises.
- Forecasting timetable ~ seven weeks with precise meetings and deadlines (Issues meeting / 1; Forecast approval / 5; CNB Board’s monetary policy meeting / 7; Post mortem / 7/8).
- Since 2004: NTF estimates used only for nowcasts and one-quarter-ahead forecasts; NTF prepares regulated prices, indirect tax changes and government consumption as exogenous inputs.

### 2.5 The human resource input — team composition, skills, incentives and workload
Team composition and skills
- Effective FPAS team need not be large but requires complementary skills; narrow specialization is unhelpful; staff should have acquaintance with modern open-economy macro theory and some experts in the field.
- Valuable non-technical assets: strong economic intuition, close understanding of data, ability to synthesize and draft intelligible narratives for diverse audiences.
- Useful software and technical knowledge: MATLAB, IRIS, DYNARE, SIRIUS, PYTHON; univariate and multivariate filtering (HP, bandpass, Kalman); solution methods for linear and nonlinear DSGE models.
CNB staffing evolution
- By 2001 FPAS development underway with 3-4 economists in modeling and projections and 6-8 sectoral specialists; in-house training supported by IMF technical assistance.
- Modeling and projections team grew to six economists; created rotational system of two 3-member teams rotating annually between forecasting and model development.
Workload and retention
- Forecasting rounds involve daily meetings, rigid deadlines, unexpected problems and significant stress; maintain manageable staff turnover and redundancy (at least 2 individuals capable of each position).
- Automate regular processes (data management, chart/table/presentation production) to reduce time pressure and human error.
Research integration and incentives
- FT members rotate between forecasting and research; internal research by FT members more fruitful for core model improvements.
- Senior management support and recognition of FPAS impact on decisions boosted morale and retention.

Policy performance and loss function
- CNB reputation: successful IFT country despite frequent deviations of actual inflation from announced targets and output gaps often far from zero; data show bias to the downside with inflation more often below target.
- Loss-function weighting used: weight of 1, 1, and 0.5 on squared deviation of inflation from target, output gap, and change in the policy rate, respectively.
- Until 2006 inflation targeting error was more costly; since then output gaps dominated in prolonged periods (2006-08, 2009-14). Interest rate variability negligible throughout.
- During Global Financial Crisis both inflation and output gaps made a large contribution to the loss.

Transparency and interest rate path publication
- Initial NTF forecasts assumed constant interest rates and exchange rate over horizon—internally inconsistent and provided no quantified guidance.
- QPM-Gap produced an endogenous interest rate path for 3-month PRIBOR; initially described qualitatively.
- CNB published coefficients of its interest rate reaction function.
- Risk-of-confusion argument (Mishkin, 2004) weakened with capacity to produce confidence bands.
- Examples of verbal communication effects: April 2007 (market path shifted toward CNB endogenous path even though policy rate held constant) and July 2007 (market stayed well below CNB forecasted >200 basis points increase while market anticipated ~100 basis points).
- Decision to publish 3-month PRIBOR path in a fan chart with confidence bands announced on 8 March 2007; practice started early 2008.

### 4.2 Episodes of interest-rate-forecast publication — market reactions and lessons
February 2008
- Forecast narrative: after final up-cycle step in early-2008, rate-cutting expected later in the year; much of inflation increase viewed as temporary due to food/admin prices and indirect tax changes; second-round effects expected muted.
- Market reaction: pre-meeting market priced flatter outlook; immediate reaction to publication: no immediate reaction; published forecast showed a drop of 100 basis points by end-2008.
- Communication issues encouraging market discounting: over-emphasis on conditionality; Board emphasized “larger than usual” uncertainties and published an alternative scenario with stronger pass-through showing an endogenous path above market outlook.
- Ex post: Global Financial Crisis depth unforeseen; eventual rate cuts much deeper than February forecast suggested.

November 2008
- Context: divergence at short end of market outlook pre-meeting due to rise in money market risk premiums post-Lehman.
- After meeting: longer end of market outlook converged reasonably close to new published forecast.
- Interpretation: numerical publication affects expectations but not mechanically; market treats published path as conditional.

May 2009 and November 2012 — positive experiences
- May 2009: publication of even lower rates in CNB forecast pushed down market curve.
- November 2012: forecast path went close to zero; Board issued forward guidance “... rates will remain at this level over a longer horizon until inflation pressures increase significantly.” Market outlook moved down immediately to near-zero policy rate.

November 2009 and February 2011 — awkward experiences
- November 2009: forecast showed need for further cut, but Board voted 4:3 to keep rates unchanged; market outlook moved up away from the published path. Confusion partly resolved by 25 basis point cut at next meeting.
- February 2011: baseline path almost flat despite hawkish market outlook; 3 of 7 Board members voted for 25 basis point increase; market outlook moved above published forecast, placing more weight on Board voting and communications.

The ELB period
- Inflation Report III/2013: published PRIBOR path implied hypothetically negative policy rates (did not obey ELB constraint); Board communicated since September 2012 it was not considering cutting below zero and would rely on exchange rate depreciation instead. Market ignored published path; 3-month PRIBOR outlook stayed at 50 basis points, consistent with binding ELB.
- Inflation Report II/2014: CNB used exchange rate as policy instrument; communicated no interest rate increase before exit from unconventional policy. Forecast assumed renormalization in Q1 2015 with rising interest rates, but disinflationary pressures mounting. Market outlook remained extremely close to CNB forecast paths as forecasts postponed timing of exit.

Summing up
- Published interest rate path can affect market outlook and support transmission, but only when other communication elements are not confusing.
- If Board assessment or alternative scenarios point strongly elsewhere, market likely ignores the staff baseline.
- Perceived commitment problem has not materialized; market tends to downplay central bank baseline forecast and weighs Board communications heavily.
- November 2009 identified as a clear example of confusion; serves as a disciplining lesson.

### 4.3 Other aspects of transparency and communications
- Dincer and Eichengreen transparency index (2014): CNB ranked second most transparent of >100 central banks with score of 14.5 in 2014 (maximum 15); missing ½ point relates to explanation of policy decisions.
- CNB practices:
  - Press release immediately after decisions; Governor press conference same afternoon.
  - Monetary policy decision explained with new macro forecast (four times a year) or inter-forecast risk assessment (four times a year since 2008).
  - Presentations give votes cast by Board members on interest rate decisions; Minutes published eight days after meeting (with individual votes since 2008).
  - Since 2014 Governor provides a written explanation followed by Q&A.
  - Quarterly meetings with market analysts one day after policy decision; Executive Summary of Inflation Report released with detailed forecast table.
  - Full transcript of policy meetings published with a six-year delay.
- Debates on publishing individual votes: former Governor Tůma (2010) expressed reservations; some Board members in 2013 expressed discomfort with publishing individual votes or communications of individual opinions on exchange rate in extraordinary circumstances.

### 4.4 Ownership of the forecast
Definitions and CNB practice
- Forecast ownership reflects management structure, decision-making, policymakers’ involvement, and accountability.
- CNB Inflation Reports wording: forecast drawn up by CNB’s Monetary Department; forecast is key but not sole input to Bank Board’s decision; Board’s final decision may not correspond to forecast due to new information or divergent views.
- Operationally: CNB forecast is a staff forecast, not the official central bank forecast.
Practical constraints and implications
- Board has wide management responsibilities limiting time for monetary policy (example: a Vice-Governor estimated spending only 10 percent of time on monetary policy).
- Only two meetings of staff with Board during each forecasting exercise; decision-making often individualistic with split votes.
- Staff forecast typically the only fully coherent macro projection and serves as starting point for members’ outlooks and policy discussions.

### 4.5 Communications — internal and external: lessons and ELB implementation
Core lessons from CNB experience
- forecasts must be communicated in digestible economic language;
- staff must incorporate judgment and relevant events into model-based forecasts;
- policymakers should see the value added of model-based forecasts via alternative scenarios and sensitivity analyses;
- models need adaptability to changes (ELB, unconventional instruments);
- improved forecasting accuracy builds confidence in FPAS.
Clear language & judgment input
- Managers edit reports to balance technical rigor and digestibility.
- Model changes (QPM-Gap to QPM-g3) required several rounds of non-technical presentations to Board members.
- Incorporating judgements: useful to apply adjustments that account for real-world events even if no material effect on policy outlook (example: “cash-for-clunkers” subsidies affecting exports and koruna selling pressure).
Reckoning for uncertainty and scenario analysis
- High-uncertainty episodes motivated commissioning of alternative scenarios and sensitivity analysis (example: February 2011 Board requested 3 alternative scenarios using global NiGEM then GPM-g3).
- February 2011 scenarios and outcomes:
  - “German engine”: more optimistic GDP but more exchange rate appreciation and little change in inflation or interest rates.
  - Higher commodity prices: currency depreciation relative to baseline, immediate interest rate increases, higher inflation for a while, weaker growth.
  - Escalation of euro area debt crisis: much weaker euro area, depreciation relative to baseline, offsetting impact on inflation, no near need to change interest rates.
ELB — systematic forward guidance and exchange rate as instrument
- Context: by autumn 2012 inflation below 2% target and economy below potential; short-term policy rates reached the ELB in November 2012.
- Policy response: weaken nominal exchange rate to support inflation target; Board signaled use of exchange rate starting September 2012 and introduced forward guidance on policy rate in November 2012 (“Interest rates will remain at this level ... over a longer horizon until inflation pressures increase significantly”).
- Modeling implementation:
  - Mid-2012 to early-2013: GPM-g3 adjusted to provide model-consistent ELB scenarios using the nominal exchange rate as additional instrument.
  - Technique: simulate desired future nominal exchange-rate path consistent with keeping inflation close to target by deriving fully-anticipated shocks to the exchange rate equation via a constrained optimization routine.
- Model simulations (real-time versions produced in 2013) show effects of a fully-anticipated 5% weakening of the exchange rate for various ELB durations:
  - The longer the ELB binding, the higher the effect of depreciation on inflation and the more positive the impact on the real economy.
  - Simultaneous weak exchange rate and zero interest rate → negative real interest rate → growing real consumption.
  - Depreciation supports real exports; combined channels produce significant positive effect on real GDP growth.
- Main policy implication: weakening the exchange rate combined with forward guidance should produce pronounced positive effects on inflation, especially when policy is credibly committed to such a policy.

Summing up and V. Conclusions (extracted points)
- Over less than a decade, Czech monetary policy progressed to the frontier by:
  - ongoing investment in human capital;
  - development of up-to-date models for forecasting and policy analysis;
  - institutional reorganizations for efficient economic intelligence provision;
  - improved dialogue between economists and policymakers;
  - bold opening of external communications with complete disclosure of the central bank’s economic forecast.
- Forecasts should not simply reflect senior management views; economists need freedom to allow productive internal discussions.
- Model-based forecasts provide policymakers with a starting point for deliberations, confidence that actions reflect thorough consideration, and a means to explain actions with a coherent narrative.
- Core model requirements for FPAS under IFT:
  - an endogenous short-term interest rate determined by monetary policy;
  - an important role for forward-looking expectations of the central bank and the public.
- Core models must be adapted over time; gap models and DSGE models each capture key principles but are incomplete; judgments and complementary models remain essential.
- Practical guidance:
  - newcomers to IFT can start with a simple gap model while researching DSGE models;
  - informed judgment is most important for near-term forecasts (sectoral experts outperform models up to 2 quarters);
  - combine near-term sector-expert forecasts with medium-term model-driven forecasts under a single manager.
- Institutional outcomes:
  - better informed media coverage and academic interest;
  - emergence of a generation of highly qualified, policy-oriented monetary theorists and model builders strengthening recruitment;
  - CNB became an internationally recognized contributor of technical assistance despite small size.

*Source: wp1721 — IMF Working Paper (selected sections: 1.1 Background; 1.3 The nominal anchor; 1.4 Transmission mechanism; 1.5 FPAS; 1.6 Publication of policy rate; 2.2 QPM-Gap; 2.5 Human resource input; 4.2 Episodes; 4.3–4.5 Communications and conclusions).*

### 1.1  Background ........................................................................................................

### 1.1 Background

### Context and purpose
- Describes the Czech National Bank (CNB)’s experience implementing an Inflation-Forecast Targeting (IFT) regime, and the building of a system for providing the economic information that policymakers need to implement IFT.
- States that this experience has been “very successful” and that its main achievement has been to establish confidence in monetary policy in the face of many disturbances this century.
- Intends to provide guidance for other countries considering monetary policy reforms of the same kind.

### Evolution in central banking transparency
- During the 1990s, central bankers recognized that the better their policies were understood, the more effective they were.
- Debate at inflation-targeting (IT) central banks centered on what to disclose beyond the inflation target and current policy interest rate—especially which elements of the quarterly macroeconomic forecast to release.
- Publishing the forecast for inflation and output has not been controversial; the central bank’s inflation forecast represents an ideal conditional intermediate target (Svensson (1997)).
- The flexible IT regime is operationally the same as IFT.

### Logical implications of IFT
- IT logically implies an endogenous interest rate; a model with an exogenous interest rate lacks a nominal anchor and allows inflation to drift indeterminately after disturbances.
- Under IT, the nominal anchor is the expectation that inflation will converge to the announced long-run objective; monetary policy must react to return inflation to the long-run target, implying an endogenous interest rate path.
- Many central banks produce an endogenous path for the interest rate in their forecasting models but most have so far decided not to publish the path, arguing the policy rate must be free to respond to contingencies and publishing could be misconstrued as a commitment.

### Rationale for full disclosure of the endogenous interest-rate path
- A leading group of central banks, including the CNB, has adopted full disclosure for practical and transparency reasons.
- Monetary policy affects spending primarily through expectations for future policy interest rates (term structure), not just the current short-term rate.
- At any policy decision, policymakers must have in mind a conditional path for the future policy rate—a band of possible paths rather than a single line—reflecting preferences over the inflation-output trade-off.
- Full disclosure of the central bank forecast can reinforce policy effectiveness by:
  - Strengthening confidence in the long-run inflation goal through a coherent view showing inflation returning to target.
  - Helping move the term structure of interest rates to assist the transmission mechanism via the published short-term interest rate path.
- The payoff from reinforcement is a reduced cost of eliminating deviations of actual inflation from the long-run target, or an improved short-run inflation-output trade-off.

### The transparency continuum (milestones)
- Milestones along the continuum of openness and accountability include:
  - 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 inflation (and output) forecast—clarity of intermediate target—IFT;
  - publication of a conditional forecast path and confidence bands for the short-term interest rate.

### Definition and operational features of IFT (as used in the paper)
- IFT means that:
  - monetary policy is based on a long-run low inflation target, and a medium-term forecast path to this target;
  - the forecast path for the short-term interest rate—the policy instrument—is endogenous within the model, with the rate varying to achieve the long-run inflation target and to eliminate any output gap;
  - the forecast is a key input into the MPC decision, but only an input—members of the MPC need not agree with the forecast, but use it as an important basis for explaining their own views;
  - 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 in greater depth 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 a qualitative description, or with explicit numbers;
  - the central bank emphasizes the risks to its assessments, verbally, and with confidence bands for those variables for which numerical forecasts are published.

### Examples of IFT central banks publishing endogenous interest-rate forecasts
- Reserve Bank of New Zealand
- Czech National Bank
- Bank of Israel
- Norges Bank
- Sveriges Riksbank
- The U.S. Federal Reserve is described as one of the best recent examples of an IFT central bank given its dual mandate and 2012 announcement of a 2 percent long-term inflation objective.

### Accountability and communication under IFT
- Under IFT, central banks communicate not just a possible path for the future policy rate and special measures, but also:
  - how the path might change with developments;
  - the rationale for policy actions.
- This enhances both accountability and policy effectiveness.

### Illustrative citations and references within the text
- Svensson (1997) on the inflation forecast as an intermediate target.
- Woodford (2005) on management of expectations; Svensson (2007) advocating publishing the central bank’s forecast interest rate path.
- Clinton and others (2015) on FPAS and developing analytical frameworks.
- Blanchard and Galí (2007) and Laxton and N’Diaye (2002) on costs and trade-offs.
- Freedman and Laxton (2009) discussed interest-rate publication issues.

### Boxed historical notes (leaders in transparency – highlights)
- New Zealand: first adopter of a simple IT in 1989; developed FPAS in 1997 and immediately disclosed the policy interest-rate path.
- Czech Republic: adopted IT in 1998 after fixed-exchange-rate collapse and financial distress; using an FPAS with a model-based forecast by 2002; began publishing detailed quarterly forecasts in 2002 and started publishing the policy-rate path in a fan chart in early 2008; public confidence in the 2 percent target has been strong.
- Sweden: switched to IT after a currency crisis in the early 1990s; from 2007 started publishing its own endogenous forecast for the policy rate in the Monetary Policy Report.
- Israel: adopted IT in 1997; began publishing its endogenous interest-rate forecast in 2007.
- Norway: adopted a 2.5 percent inflation target in March 2001; began publishing its policy-rate forecast in November 2005.
- United States: announced a long-run goal of 2 percent PCE inflation in 2012; debate on transparency continues, with some advocacy for releasing participants’ projections and policy paths.

*Source: wp1721 - 1.1 Background (IMF working paper).*

### 1.3 The nominal anchor

### 1.3 The nominal anchor

### Key findings and analytical points
- The credibility of the long-run inflation target underpins IFT: the nominal anchor is the firm expectation of the public that monetary policy will in the long run keep inflation stable and near the target rate.
- Maintaining the anchor requires that policy responds systematically to the requirements of this objective.
- With a forward-looking policy, the future expected path of the policy interest rate is adjusted when unanticipated disturbances hit the economy, to bring inflation back to the target gradually over a period that limits disruptions to output.
- Policy feedback operates via an endogenous short-term interest rate (represented as feedback from expected future inflation to the policy instrument).

---

### 1.4 The transmission mechanism

### Major themes
- Two traditional channels of monetary policy influence:
  - Domestic channel: via the short-term market interest rate, controlled closely by the central bank.
  - External channel: via the exchange rate.
- Distinction between endogenous, forward-looking policy reaction functions (IFT) and approaches with exogenous interest rate paths or simple Taylor rules: the latter lack explicit feedback from expected future inflation to the policy instrument and would remove the feedback arrows in the model.
- Modern models emphasize expectations: policy affects what interest rates businesses and households face through the impact of policy rates expected in the future—the whole "Policy Rate Path" matters, not just the current setting.
- The size of the impact of a policy action depends on the expected duration of that action.

### Policy communication and the ELB
- Recognition of expectations’ role has increased central banks’ emphasis on transparent communications.
- When constrained by the effective lower bound (ELB), a credible central bank can affect expectations to provide an effective anti-deflation policy (see Svensson, 2003).

---

### 1.5 Forecasting and Policy Analysis System (FPAS)

### Core elements of the FPAS
- A monetary policy model containing a loss function or policy reaction function under which the short-term interest rate responds endogenously to return inflation to the long-run target.
- A system of communication between policymakers and model builders to ensure the model structure captures key features of the transmission mechanism as viewed by policymakers.
- A system of communication between policymakers and forecasters before and during forecast production to develop a consensus set of assumptions for a baseline and for alternative scenarios.

### FPAS outputs and uses
- The forecast necessarily includes an endogenous short-term interest rate under the control of the central bank that responds to developments affecting inflation to return it, over a medium-term horizon, to the long-run target.
- The system provides:
  - Baseline forecasts—including an endogenous forecast path for the short-term interest rate;
  - Risk assessments surrounding the baseline;
  - Alternative scenarios to the baseline—each with an endogenous interest rate path.
- Each FPAS forecast—baseline and alternative scenarios—contains:
  - (a) an inflation rate projection that shows a return to the long-run target (in the event of a deviation) within a medium-term horizon; and
  - (b) an endogenous path for the policy interest rate.
- The relevant medium-term horizon for getting inflation back on target is not fixed; it depends on the source and size of shocks. In many cases gaps might be substantially eliminated within 4-6 quarters, but more obdurate shocks may take more time.

### Institutional lessons
- Every central bank embarking on IT started with policy information systems not specifically designed for the job and adapted existing procedures.
- The absence of a suitable FPAS is not a good argument for delaying adoption of IT; rather, invest in such a system immediately after regime adoption.
- FPAS development since the mid-1990s has proceeded more rapidly by learning from other institutions; an FPAS is continuously modified—“an FPAS never stands still.”

---

### Box 2 — Medium-term inflation expectations as shock absorbers or amplifiers

### Mechanism at the ELB
- At the ELB the nominal rate cannot decline, but expected inflation can provide a channel through which forward guidance stimulates the economy.
- If monetary policy is active and credible, the central bank can persuade the public that it will eventually get inflation back to target, committing to hold the interest rate near the ELB for an extended future period, leading the public to expect increased inflation in the future and thereby reducing longer-term real rates despite the nominal rate being at the ELB.
- These expected-inflation movements reduce longer-term real rates, serve as a buffer to the shock, and may be represented in a forecast in which inflation overshoots before returning to the long-run target.

### Open-economy amplification
- The real exchange rate would depreciate and asset prices would rise, amplifying the real interest rate channel:
  - Depreciation raises profits and supports asset prices.
  - Lower real discount rates boost asset prices and stimulate spending.

### Credibility conditions required
- The central bank must persuade people that:
  - the nominal interest rate will remain at the floor for an extended period;
  - the rate of inflation will rise over the medium term, perhaps above the long-run target;
  - the rate of inflation will eventually return to target.
- The Czech National Bank’s exchange rate policy since 2013 is cited as evidence that under a transparent IFT framework influencing expectations can be effective.

### Risks under passive or non-credible policy
- If monetary policy is passive and not credible, expected inflation would fall, not rise; at the ELB, real interest rates would rise, the real exchange rate would appreciate, and asset prices would fall—producing a deflation trap.

---

### 1.6 To publish or not to publish the forecast policy rate

### Practice and concerns
- All IFT central banks publish the forecast for major macroeconomic variables that accompanies their inflation forecast; disclosure of the endogenous policy-rate path varies.
- Table of 13 IFT central banks (publication of policy rate path and year when started):
  - New Zealand — Yes (1997)
  - Norway — Yes (2005)
  - Sweden — Yes (2007)
  - Israel — Yes (2007)
  - Czech Republic — Yes (2008)
  - United States — Yes (2012)
  - Canada — No
  - Chile — No
  - Colombia — No
  - Hungary — No
  - South Korea — No
  - Peru — No
  - Turkey — No
- Principal worry: the public might misinterpret a published explicit path as a commitment, or underestimate uncertainties.

### Mitigating practices
- Publishing confidence bands around the endogenous interest rate path.
- Emphasizing the conditional nature of projections: forecasts are conditional on information available at the time and will change as new information arrives.
- Example CNB quote: “It is vital, however, that all users of the central bank’s forecasts are aware that the published forecast-consistent interest rate path should in no way be interpreted as a commitment of the central bank to set interest rates in line with the forecast... the forecast represents the most probable future path of interest rates under given initial assumptions and information. New information on the domestic and global economy that comes in after the forecast is drawn up can change the interest rate outlook.”

### Experience and timing
- Central banks have generally been cautious in moving to published rate forecasts; in most cases the practice followed many years after adoption of IT—the shortest gap was 4 years at the Norges Bank.
- To publish an endogenous interest rate path, central banks first built models with endogenous policy rates and then used them as the basic forecasting tool; policymakers had to gain confidence in FPAS outputs before publication.
- Market participants adapted quickly to published conditional rate paths; experience in New Zealand (since June 1997), Norway (since November 2005), Sweden (since February 2007), Israel (since July 2007) and the Czech Republic (since February 2008) suggests perceived communication risks were overblown.

---

*Source: IMF Working Paper — "1.3 The nominal anchor" (chapter/section extracted from wp1721 PDF).*

### 2.2 Medium-term focus—QPM-Gap

### 2.2 Medium-term focus—QPM-Gap

### Model development and institutional adoption
- QPM-Gap is a small-scale calibrated monetary policy model for the Czech Republic developed with IMF technical assistance.
- Steps taken to introduce QPM-Gap into CNB policy decision making:
  - commit staff resources: the model development and forecasting team; the team head; the model operators; and relevant sectoral specialists.
  - proceed step-by-step from a pilot version focusing on key channels of the policy transmission mechanism.
  - improve calibration by bringing the model closer to data; this process took about 18 months and involved testing dynamic properties, evaluating model-consistent estimates of unobserved variables, and identifying key demand- and supply-side shocks.
  - test forecasting properties via in-sample simulations and shadow forecasts; consult sectoral specialists on plausibility and magnitudes/timing of model responses to shocks.
  - agree on operational rules specifying timeframe of the forecast, personal responsibilities and clear deadlines for each stage.
- The switch to QPM-Gap in July 2002 is regarded as the start of IFT in the Czech Republic because QPM-Gap:
  - had a medium-term focus,
  - included forward-looking channels,
  - had an internal channel (forward-looking interest rate term structure) and an external channel (exchange rate),
  - included an endogenous, policy-determined, short-term interest rate.

### Model structure and forecasting features
- Inflation disaggregation:
  - separate Phillips curves for core, food and fuel inflation,
  - administered prices treated as exogenous (smoothing transition from the previous NTF inflation forecasting method).
- Real side representation:
  - single aggregate output gap.
- NTF team developed tools to decompose the aggregate GDP forecast from QPM-Gap into expenditure components requested by policymakers: household consumption, government consumption, investment, exports and imports.
- QPM-Gap style models are suitable for countries adopting or early in inflation targeting:
  - provide features to facilitate forward-looking monetary policy discussion,
  - adaptable to evolving country circumstances,
  - not too demanding on human and other resources.

### Material modifications during QPM-Gap’s 6 years as core forecasting model
- First modification:
  - recalibrated assumptions about long-term equilibrium paths,
  - increased rate of appreciation of the equilibrium real exchange rate,
  - reduced domestic real equilibrium interest rate,
  - motivated by forecast errors, notably a strong uptrend in the real exchange rate without visible negative effect on the real economy.
- Second modification:
  - explicitly recognized impact of domestic wage developments on CPI inflation,
  - added the real wage gap to the output gap as a component of the real marginal cost gap (extension drew on experience with the new GPM-g3 model).
- Third modification:
  - reduced second-round effects of administrative measures and other short-term cost-push shocks in the Phillips curve,
  - in line with observed weakening of pass-through effects and lower inflation persistence emerging in work on the new GPM-g3 model.
- These modifications contributed to improved forecasting performance and illustrate model development informed by assessment of past forecast errors.

### Transition to DSGE (QPM-g3) and comparative points
- In 2008 the CNB switched to a DSGE model, QPM-g3, as its core quarterly projection model, following debate, testing and shadow forecasting.
- Key differences between QPM-g3 and QPM-Gap:
  - QPM-g3: explicit derivation using “behavioural principles”; QPM-Gap: reduced form.
  - Both use model-consistent expectations.
  - QPM-g3 preserves stock-flow consistency and works with level variables; QPM-Gap works with “gaps” from pre-filtered trends.
  - QPM-g3 has basic national accounts disaggregation and a balanced-growth path with technology trends; QPM-Gap has equilibrium trends and no GDP structure.
  - QPM-g3 includes an implicit fiscal block; QPM-Gap has a simple fiscal block.
  - Both used an inflation-forecast-based interest rate reaction function.
  - QPM-g3 uses carefully chosen “structural shocks”; QPM-Gap used residuals for each equation.
- Practical forecasting implications:
  - QPM-g3’s disaggregated structural form provides degrees of freedom for incorporating judgment consistently.
  - QPM-g3 works with level variables and thus does not require the NTF disaggregation tools for major spending components; however, because it does not itself contain inflation components, satellite models using Kalman-filter methods were developed to provide inflation detail.
  - Despite questions about DSGE models for forecasting, the CNB’s experience was positive, with improved forecasting performance for inflation on the monetary policy horizon, particularly since the early stages of the Great Recession; large forecast errors during that time were attributed to wrong assumptions rather than a wrong model.
- Output gap and inflation drivers:
  - In QPM-Gap: the output gap was the main domestic driver of underlying inflation (until the real wage gap addition in January 2007).
  - In QPM-g3: underlying inflation is driven by real marginal costs and mark-ups; there is no aggregate output gap as such, though mark-ups in the domestic intermediate sector correlate highly with standard output gap estimates.
  - Policymakers still receive output gap estimates via potential output estimates from a suite of non-core models (e.g., univariate HP filter, multi-equation Kalman filter inherited from the QPM-Gap period, and a production function).

### Organizational and procedural changes to integrate near-term forecasters and macro modelers
- Forecasting organization issues:
  - Early separation into two divisions (“Real Economy Division” and “Economic Modeling Division”) created tensions during integration of near-term and medium-term forecasts; forecast revisions could reflect changes in “bargaining power” between groups.
  - A restructuring in 2004 merged the two teams into a single Macroeconomic Forecasting Division; a division director now manages the whole forecast process to ensure cooperative input.
  - A department-wide forecasting team of about 10 people ensures active involvement of other divisions; the team head is chosen from core model operators and rotates usually after 2-4 forecast exercises.
  - The forecasting team presents work at departmental meetings and the team head gives forecast presentations at CNB Board meetings.
- Forecasting process timetable and practice:
  - The forecasting process lasts about seven weeks and follows a detailed timetable with precise deadlines and responsibilities.
  - Outline timetable (Meeting / Week):
    - Issues meeting / 1
    - Meeting on forecasting techniques and past forecast evaluation / 2
    - NTF presentation to the department / 3
    - Meeting on initial conditions and forecast assumptions / 3
    - Meeting with the Board on initial conditions / 4
    - First version of the forecast / 4
    - Meeting with the Board on alternative scenarios / 5
    - Approval of the forecast / 5
    - Drafting of the Situation Report / Inflation Report / 6
    - Meeting on the monetary policy recommendation / 7
    - CNB Board’s monetary policy meeting / 7
    - Post mortem meeting / 7/8
  - Original practice had taken 3-4-quarter-ahead NTF forecasts as given, which led to problems because persistence in QPM-Gap meant NTF input dominated policy-horizon forecasts.
  - Since 2004 practice changed: CNB forecasts use NTF estimates only for the most recent quarter (nowcasts), and/or for one-quarter-ahead forecasts.
  - NTF team prepares forecasts for regulated prices, indirect tax changes and government consumption, which are exogenous inputs into the model over the whole forecast horizon; similar treatment applies to the foreign economic outlook and fiscal impulse estimates.
  - Expert judgments influence forecasts when NFT experts make convincing arguments; the general-equilibrium core model framework implements such judgments consistently and transparently.
- Transition challenges and lessons:
  - Initial application of the new system caused Board discomfort and produced an inaccurate forecast, but over time the system reduced forecasting errors.
  - Experiences emphasize the importance of:
    - clear communications between staff and policymakers from the outset of a quarterly projection round so modifications to models or assumptions do not take policymakers by surprise,
    - open communication and close cooperation across divisions so that all members work from an agreed set of assumptions.
  - Appropriate modalities for good communication and cooperation can be refined over time; occasional hiccups can reveal system weaknesses that lead to improvements.

*Source: wp1721 - 2.2 Medium-term focus—QPM-Gap (IMF).*

### 2.5 The human resource input

### 2.5 The human resource input

### Team composition and required skills
- An effective FPAS team does not have to be large, but needs a balance of complementary skills.
- Specialized modeling skills are not acquired quickly, either through training or recruitment; match between job requirements and individual skills is more important than filling positions without delay.
- Narrow specialization is unhelpful; diversification of skill profiles is achieved by encouraging staff to take interest in broader aspects of team work and through timely job rotation.
- Purely technical knowledge is necessary but not sufficient: all members should have at least some acquaintance with modern, open-economy, macroeconomic theory, and some should be experts in the field.
- Valuable non-technical assets:
  - Strong economic intuition.
  - Close understanding of the data.
  - Ability to synthesize various strands and draft intelligible narratives tailored to audiences (policymakers or academics vs. journalists or the general public).
- Useful software and model-related technical knowledge (as listed):
  - MATLAB, IRIS, DYNARE, SIRIUS, and PYTHON.
  - Univariate and multivariate filtering (HP, bandpass, and Kalman).
  - Solution methods for solving linear and nonlinear dynamic stochastic general equilibrium models.

### CNB historical experience on staffing and skills
- Pre-1998 setup:
  - Most MD economists were sectoral specialists applying NTF methods; level of short-term interest rate was widely believed to have negligible effect on aggregate demand.
  - Very limited monetary policy autonomy under a pegged exchange rate to a 2-currency basket (weighted 65 percent German mark, and 35 percent U.S. dollar).
  - Economics departments of Czech universities modernized courses only gradually after 1990, so supply of graduates with modern macroeconomics knowledge was very limited.
  - Result: MD output in late 1990s was largely descriptive and statistically oriented.
- Strengths in Czech education:
  - High-quality education in mathematics and technical subjects produced technically strong economists.
  - By the turn of the century, graduates well-schooled in modern macroeconomics were emerging and CNB became a successful recruiter.

### Staffing evolution, training, and internal organization
- Senior management priority and IMF technical assistance:
  - By 2001 the FPAS development was well underway with 3-4 economists in the modeling and projections team, and 6-8 sectoral specialists.
  - In-house training, supported by IMF technical assistance, created a critical mass of knowledge and staff; assistance covered state-of-the-art modeling and FPAS modus operandi (roles, timing, incorporating off-model information).
- Team growth and rotation:
  - The modeling and projections team grew over time to six economists.
  - Enlargement allowed creation of a rotational system composed of two 3-member teams, which rotate on an annual basis between forecasting and model development.
  - During each calendar year one of the 3-member teams is responsible for the forecast and the other team does model development and economic research to support the FPAS.
- Workload, deadlines, and stress:
  - Forecasting rounds involve daily meetings, rigid deadlines, unexpected economic or technical problems, compromises among staff views, preparation of presentations for policymakers, and quarterly forecast error evaluations.
  - The nature of the work puts enormous pressure on the forecasting team and can result in significant stress given the tight deadlines.
- Retention and redundancy:
  - Important to keep staff turnover at manageable levels.
  - At least 2 individuals should be capable of performing in each position to provide backup—an argument for job rotation.
  - Models and data sets should be fully documented and stored, accessible to the whole team.

### Research integration and knowledge transfer
- FT members regularly switch between forecasting and research to focus on model development (e.g., extensions with new blocks) and other projects with longer planning horizons.
- Experience suggests internal research by FT members is more fruitful for core model improvements due to lower costs of knowledge transfer; open-ended research projects can produce unexpectedly valuable ideas.

### Incentives and career development
- Senior management support and acknowledgment that FPAS input affected decisions boosted morale and provided incentives to remain with the team.
- Opportunity to acquire valuable know-how and to add to individual human capital—economists who assisted the launch of the FPAS became leading experts internationally—served as a motivating factor.

### Operational improvements
- Time pressure on staff can be reduced by automating regular processes (data management, production of charts, tables and presentations), requiring significant initial investment by technically skilled staff but yielding long-run benefits and reduced human error risk.
- Quarterly forecast evaluations based on detailed model-consistent analysis of factors contributing to forecast errors are presented to policymakers and help identify priority areas for model improvement.

### Policy performance and loss function
- Czech Republic developed reputation as a successful IFT country despite frequent deviations of actual inflation from announced targets and output gaps often far from zero; data show a bias to the downside with inflation more often below the target than above it.
- According to the loss function until 2006 the inflation targeting error was the more costly facet; since then output gaps have been more dominant for prolonged periods (e.g., 2006-08, and 2009-14). Interest rate variability was a negligible factor throughout.
- During the years of the Global Financial Crisis both inflation and output gaps made a large contribution to the loss.
- Loss function weighting used:
  - Assigns a weight of 1, 1, and 0.5 on the squared deviation of inflation from its target, output gap, and change in the policy rate, respectively.

### Credibility and communication
- Early undershooting of the inflation target likely helped dispel memories of double-digit inflation before 1998 and supported endogenous credibility-building.
- Anchoring of inflation expectations aided by:
  - Consistent, forward-looking approach to monetary policy.
  - Systematic adjustments of monetary policy instruments to bring inflation back to target over time.
  - Transparent communications and the use of defendable, up-to-date analytical tools.
- The capacity to provide a coherent narrative of how interest rates and the economy might behave over the medium term was crucial for building credibility of the Czech IFT regime.

### Experience with full transparency and the interest rate path
- Initial NTF forecasts assumed interest rates and the exchange rate constant over the forecast horizon—this ignored the nominal anchor and was internally inconsistent, and provided no quantified guidance on direction and speed of actions required to achieve objectives.
- Following the introduction of the initial QPM-Gap model in 2002, the forecast produced an endogenous interest rate path for the 3-month PRIBOR; initially described qualitatively with limited information on slope.
- CNB published coefficients of its interest rate reaction function to allow analysts to understand systematic behavior of monetary policy.
- Risk-of-confusion argument (Mishkin, 2004) against publishing numerical policy paths has weakened given IFT central banks’ capacity to produce confidence bands.
- Partial disclosure results:
  - Verbal descriptions and publication of reaction-function coefficients often affected market expectations similar to numerical publication but without explicit commitment.
  - There were instances where verbal communication did not fully align market outlook with the forecast path (example: July 2007).

### April and July 2007 examples (verbal communication effects)
- April 2007 monetary policy meeting example:
  - Market path shifted toward the CNB endogenous interest rate path between dates reflecting information from the prior meeting and the immediate policy decision—suggesting the forecast was credible and effective in shaping market expectations.
  - The day-after shift indicates the announcement and press conference contained new information not priced-in by the market; the actual policy rate was held constant (although with two members voting for an increase).
- July 2007 example:
  - Between meetings modest increase in market interest rate outlook.
  - July decision raised the actual policy rate by 25 basis points (with one Board member voting for 50 basis points).
  - Press release said: “Consistent with the macroeconomic forecast and its assumptions is growth in nominal interest rates.”
  - After the meeting market rate outlook did not change at all and stayed well below the new unpublished interest rate path—CNB forecasted an increase of more than 200 basis points over a 1-year horizon, while the market anticipated little more than 100 basis points.
  - This illustrates potential strengthening of monetary policy transmission through full publication of the interest rate path.

### Move to full publication and fan charts
- Starting in early 2008, CNB decided to publish the path for the 3-month PRIBOR in a fan chart with confidence bands based on past forecast errors to:
  - Enhance transparency of the CNB forecast and associated monetary-policy decisions.
  - Increase effectiveness of monetary policy transmission.
- The relatively wide confidence bands illustrated the degree of uncertainty and the conditional nature of the published path.
- Institutional meeting frequency:
  - Until the end of 2007, the CNB’s Bank Board met to discuss monetary issues once a month.
  - Since the beginning of 2008 there are only eight such meetings a year.
- Decision announcement date:
  - The decision to publish the path in a fan chart was announced on 8 March 2007.

*Source: wp1721 - 2.5 The human resource input (IMF PDF chapter).*

### 4.2 Some interesting episodes for interest-rate-forecast publication

### 4.2 Some interesting episodes for interest-rate-forecast publication

### 4.2.1 February 2008
- Forecast narrative: after a final step in the up-cycle in early-2008, rate-cutting would start later during the year; much of the increase in inflation was viewed as temporary due to increased food and administered prices and indirect tax changes, and second-round effects expected to be muted given the softening output forecast.
- Market reaction:
  - Pre-policy meeting: market had priced in a flatter interest rate outlook.
  - Immediate reaction to publication: no immediate reaction; published forecast showed a drop in the interest rate of 100 basis points by the end of 2008.
- Communication issues that likely encouraged market discounting of the published rate projection:
  - Prior effort to explain conditionality may have been overdone—gave a reason to ignore the published rate projection.
  - The Board emphasized uncertainties—“larger than usual” with “risks on both sides”.
  - CNB published an alternative scenario, using the previous version of QPM-Gap, with much stronger inflation pass-through effects, which showed an endogenous interest rate path above the market outlook.
- Ex post note: the Global Financial Crisis depth was unforeseen and eventual rate cuts were much deeper than the February forecast had suggested.

### 4.2.2 November 2008
- Context: short end of market outlook had diverged from the forecast before the meeting due to a sharp rise in money market risk premiums, post-Lehman collapse.
- After the policy meeting: longer end of market outlook converged reasonably close to the new published forecast.
- Interpretation:
  - Numerical publication of the interest rate path tends to affect market expectations, but not one-to-one and not mechanically.
  - Market treats published path as conditional, not as a commitment; CNB had stated that actual path can deviate due to new information and differing Board views.
  - Absence of mechanical market reaction reduces risk of falsely perceived commitment.
  - Market expectations are affected by the whole “package” of communication: forecast and interest rate path, alternative scenarios, Board assessment of risks, voting ratio, Minutes (published with an eight-day lag), etc.

### 4.2.3 May 2009 and November 2012 — positive experiences
- May 2009:
  - Post-crisis economic weakness and disinflation: market had been pricing further rate cuts pre-meeting.
  - Publication of even lower rates in CNB forecast pushed down the market curve.
- November 2012:
  - Forecast path went to close to zero.
  - Board issued strengthened forward guidance: “... rates will remain at this level over a longer horizon until inflation pressures increase significantly.”
  - Market outlook moved down immediately to a level consistent with the forecast, almost zero policy rate (given the market risk premium) and the forward guidance.
- Additional note: numerous other positive examples cited (Inflation Reports I/2009, III/2009, II/2010, IV/2010, II/2011, and to a lesser extent II/2012 and III/2012).

### 4.2.4 November 2009 and February 2011 — awkward experiences
- November 2009:
  - Forecast showed need for a further rate cut, but the Board disagreed (vote was 4:3 for unchanged rates).
  - Market interest rate outlook moved up, farther from the published interest rate path.
  - Confusion partly resolved by a 25 basis point cut at the next meeting.
- February 2011:
  - Rise in current inflation (mainly increases in energy prices and indirect tax rates) and a swift recovery led to a hawkish market outlook.
  - CNB baseline interest rate path was almost flat for the one-year-ahead horizon (reflecting “the effects of fiscal consolidation ... and still low level of foreign interest rates in the short run”).
  - Staff prepared three alternatives; only one implied rate increases.
  - Three out of 7 Board members voted for a 25 basis point increase at the meeting.
  - Market reaction: moved interest rate outlook farther above the published CNB forecast, placing more weight on Board voting and communications than on the forecast.
  - Retrospective assessment: staff forecast provided a better guide.

### 4.2.5 The ELB period
- Inflation Report III/2013 (August meeting):
  - Published interest rate path for 3-month PRIBOR would imply hypothetically negative policy rates given the money market premium—forecast did not obey the ELB constraint.
  - Board had been communicating since September 2012 that it was not considering cutting rates below zero and would instead rely on exchange rate depreciation if further easing were needed.
  - Market response: ignored published interest rate path; 3-month PRIBOR outlook stayed at 50 basis points, consistent with a binding ELB.
- Inflation Report II/2014 (May meeting):
  - CNB using exchange rate as policy instrument; communicated that no interest rate increase would take place before exit from this unconventional policy.
  - Uncovered interest parity implied strong and credible forward guidance on interest rates; market one-year outlook consistently flat.
  - Forecast assumed renormalization in first quarter of 2015 with rising interest rates, but disinflationary pressures were mounting.
  - Board assessed risks as “being slightly anti-inflationary,” noting “the probability of a later exit from the exchange rate commitment was increasing”.
  - Market reaction: slightly decreased its interest rate outlook.
  - Subsequent CNB forecasts further postponed assumed timing of exit; market outlook remained extremely close to CNB forecast paths.

### 4.2.6 A summing up
- Published interest rate path can affect market outlook and support monetary policy transmission, but only when other communication elements are not confusing.
- If Board assessment of risks or an alternative scenario points strongly in another direction, market is likely to ignore the central bank baseline forecast and may move away from it.
- Perceived commitment problem has not materialized; market tends to downplay the central bank baseline forecast.
- Market places high weight on Board communications, reflecting understanding that policymakers (not staff forecasts) make final decisions.
- November 2009 is identified as the only clear example of confusion—serving as a disciplining lesson for future communications.

### 4.3 Other aspects of transparency
- Dincer and Eichengreen transparency index (updated 2015) places the CNB as the second most transparent central bank in a sample of more than 100 central banks, with a score of 14.5 in 2014 (maximum 15).
  - The ½ point missing relates to explanation of policy decisions.
- Evolution and features of CNB communications:
  - Board issues a press release immediately after decisions; Governor gives a press conference the same afternoon.
  - Monetary policy decision explained in context of a new macroeconomic forecast (four times a year) or an inter-forecast risk assessment (also four times a year since 2008).
  - Presentations give votes cast by Board members on interest rate decisions.
  - Since 2014, Governor provides a written explanation of the decision, followed by a Q&A session.
  - Minutes published eight days after policy meeting (with individual votes since 2008).
  - Inflation Report contains full forecast detail; structure evolved to emphasize forward-looking content and deliver a more concise message.
- Other communications:
  - Quarterly meetings with financial market analysts (local and foreign) one day after policy decision, with release of the Executive Summary of the Inflation Report including a detailed forecast table.
  - CNB publishes a full transcript of policy meetings with a six-year delay.
- Debates on transparency:
  - Former Governor Tůma (2010) expressed reservations about publishing Board votes by name, arguing it may lead to opportunistic behavior and less flexible voting.
  - In 2013 some Board members felt discomfort when the Board decided not to publish individual votes and to restrict communication of individual opinions on the exchange rate—intended to avoid public confusion in extraordinary exchange-rate-as-policy circumstances.
  - Publishing individual votes is not a must for an IFT central bank.

### 4.4 Ownership of the forecast
- Definitions and distinctions:
  - Forecast ownership reflects management structure, decision-making process, policymakers’ involvement in assumptions/judgments, and accountability requirements.
  - Forecast may be labelled Board/MPC forecast, staff forecast, or central bank forecast; labels overlap and the institution is ultimately responsible for production and publication.
- CNB practice and wording:
  - CNB Inflation Reports: “The forecast for the Czech economy 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. At its meetings during the quarter, the Bank Board discusses the current forecast and the balance of risks and uncertainties surrounding it. The Bank Board’s final decision may not correspond to the message of the forecast due to the arrival of new information since the forecast was drawn up and to the possibility of asymmetric assessment of the risks of the forecast and divergent views of some board members...”
  - Operationally: CNB forecast is a staff forecast, and not the official forecast of the central bank.
- Practical constraints at the CNB:
  - Board has wide management responsibilities (including financial system supervision) limiting time for monetary policy; example: a Vice-Governor estimated spending only 10 percent of time on monetary policy.
  - Only two meetings of staff with Board during each forecasting exercise—one on assumptions and initial conditions, another on alternative and sensitivity scenarios.
  - Decision-making and Board communication can be individualistic, often reflected in split votes.
  - Board sometimes differs from the staff forecast (instances noted in Section 4.2).
- Implications:
  - Staff forecast is likely the only fully coherent macroeconomic projection and serves as the point of departure for members’ outlooks and policy discussions.
  - A staff forecast applies when the Board/MPC has other management responsibilities, limited involvement in technical forecast aspects, wants to retain discretion to deviate, and can publicly distance itself from staff analyses—the CNB record suggests these conditions apply.

*Source: wp1721 - 4.2 Some interesting episodes for interest-rate-forecast publication (IMF working paper content provided).*

### 4.5 Communications—internal and external

### 4.5 Communications—internal and external

### Overview and core lessons
- From the outset in 2002, CNB policymakers provided steadfast support to the IFT FPAS, with its model-based forecasts; this support survived complete turnover of members of the Board level.  
- Key lessons from the Czech experience:
  - forecasts should be communicated to policymakers in digestible economic language that is not overwhelmed by technicalities;
  - staff must incorporate judgment and relevant actual events into model-based forecasts;
  - policymakers should be able to see the value added of model-based forecasts for making decisions under uncertainty (e.g., through alternative scenarios; alternative policy options; sensitivity analyses, etc.);
  - the models need to be adaptable to changes in the economic environment (e.g., the ELB, and unconventional policy instruments);
  - improved forecasting accuracy builds confidence in the FPAS.

### 4.5.1 Clear language and explanation
- Modelers do not necessarily write well for a non-technical audience; at the CNB, managers of the Monetary Department (economists with good communications skills) edit reports to strike a balance between technical rigor and digestibility.
- Model changes require careful explanation:
  - QPM-Gap (pre-mid-2008) made inflation a function of the output gap, import prices and cost-push shocks.
  - The QPM-g3 model (introduced mid-2008) replaced the output gap with endogenous real marginal costs of producing consumer goods and with firms’ mark-ups.
  - The change was difficult for Board members to digest and required several rounds of non-technical presentations; presentations are now routinely given to new Board members.

### 4.5.2 Judgment input
- Brůha and others (2013): supplement model structure with judgements that often affect the optimal future course of monetary policy.
- It can be useful to apply adjustments that have no material effect on the policy outlook if they account explicitly for real-world events likely to capture attention.
  - Example: post-Global Financial Crisis “cash-for-clunkers” subsidies in Western Europe moderated the decline in Czech exports and reduced selling pressure on the koruna; the overall forecast assessment for Czech inflation and interest rates was neutral, but explicitly incorporating the subsidies mattered for credibility and completeness of the forecast narrative.

### 4.5.3 Reckoning for uncertainty
- High-uncertainty episodes (early Global Financial Crisis, start of the European sovereign debt crisis, first contact with the ELB) lead the Board to commission alternative scenarios and sensitivity analyses to see how the monetary policy outlook changes with different assumptions.
- A model-based forecasting system facilitates the rapid production of alternative scenarios; under an NTF-based system, producing one scenario would require time-consuming iterations among sectoral specialists.
- Forecasting tools that help policymakers promptly while weighing risks and communications aspects have special value.

- Example: February 2011 (high uncertainties)
  - The economy was rebounding from the Great Recession, supported by strong German performance, but foreign demand outlook was not encouraging and domestic fiscal policy was restrictive.
  - The baseline forecast showed a slowdown.
  - The Board requested 3 alternative scenarios to the baseline (for all three scenarios, an alternative outlook for foreign variables was first simulated using the global NiGEM model; in the second step, it was used in the GPM-g3 model to produce alternative forecasts).
  - Scenario outcomes:
    - “German engine” scenario: more optimistic GDP forecast, but associated with more pronounced exchange rate appreciation and hence little change in the inflation or interest rate outlook.
    - Higher commodity prices scenario (adverse supply shock): implied currency depreciation relative to baseline, immediate interest rate increases, higher inflation for a while, and somewhat weaker economic growth.
    - “Escalation of euro area debt crisis” scenario: much weaker euro area and renewed downturn, depreciation in the exchange rate relative to baseline; offsetting impact on inflation and hence no need soon to change interest rates.
  - Outcome at the February 2011 Board meeting: split vote, with the majority deciding to keep rates unchanged and a minority voting for a 25 basis point increase. All Board members appreciated the quantitative insights from model-based alternative scenarios.

### 4.5.4 ELB — systematic forward guidance under IFT
- Timeline and context:
  - By 2010 the Czech economy seemed to be recovering from the Global Financial Crisis; another slowdown began in late 2011 due to euro zone demand slowdown and domestic fiscal consolidation.
  - In autumn 2012 observed inflation was below the CNB’s 2% target, the economy was operating below its potential, and the CNB’s inflation forecast signaled undershooting of the target.
  - Short-term policy rates reached the ELB in November 2012.
- Policy response:
  - With conventional policy constrained by the ELB, the CNB concluded weakening the nominal exchange rate would be an effective tool to support the inflation target as a credible nominal anchor.
  - Starting in September 2012, the Board signaled consideration of using the exchange rate as an additional tool.
  - Forward guidance on the policy rate was introduced in November 2012; Governor Singer noted on November 1 that “Interest rates will remain at this level (i.e. technical zero) over a longer horizon until inflation pressures increase significantly”.
  - Verbal interventions initially prevented nominal exchange rate appreciation, but their effectiveness diminished by autumn 2013.
- Modeling and implementation:
  - Mid-2012 to beginning-of-2013: CNB modeling and projections team adjusted the GPM-g3 model to provide model-consistent scenarios in the ELB environment using the nominal exchange rate as an additional instrument.
  - By the beginning of 2013 the forecasting team could operate GPM-g3 using the exchange rate as instrument instead of the short-term nominal policy rate.
  - Technique: simulate the desired future path of the nominal exchange rate consistent with keeping inflation close to target by deriving fully-anticipated shocks to the exchange rate equation (a hybrid version of the uncovered interest rate parity) via a constrained optimization routine; the algorithm was developed iteratively until results were considered sufficiently plausible for policy debate.
- Effectiveness and model simulations:
  - Public expectations are critical to ELB policy effectiveness because the impact on the real economy derives from a drop in the real interest rate.
  - Model simulations (real-time versions produced in 2013; Figure 15 presents an improved version) show effects of a fully-anticipated 5% weakening of the exchange rate for various lengths of stay of the policy rate at the ELB:
    - The longer the ELB binding, the higher the effect of a depreciation of the exchange rate on inflation and the more positive the impact on the real economy.
    - Simultaneous weak exchange rate and zero interest rate (resulting in a negative real interest rate) is the main factor behind growing real consumption.
    - Depreciation of the real exchange rate supports real exports.
    - Combined real exchange rate and interest rate channels produce a significant positive effect on real GDP growth.
  - Main policy implication: weakening the exchange rate combined with forward guidance should produce pronounced positive effects on inflation, especially when monetary policy is credibly committed to such policy.
- The modified forecasting framework allowed staff to assess ELB implications for the baseline forecast and to explore effects of unconventional policy of exchange rate depreciation; the Board requested model-based ex post assessments of the exchange rate instrument’s impact.

### 4.5.5 Summing up
- Establishing good two-way communication between staff and the Board over the model-based FPAS was not easy, and there was awkwardness when staff moved to a new model, but the process ultimately worked reasonably smoothly.
- Open, two-way dialogue helped staff provide useful information, increasing Board confidence in the model-based FPAS; mutual support is important for successful implementation of IFT.

### Key methodological and institutional conclusions (V. CONCLUSIONS — extracted points)
- Over less than a decade, Czech monetary policy progressed from limited experience to the frontier of the state of the art through:
  - ongoing investment in human capital;
  - development of up-to-date models for forecasting and policy analysis;
  - institutional reorganizations to allow efficient provision of relevant economic intelligence;
  - improved dialogue between economists and policymakers;
  - bold opening of external communications, with complete disclosure of the central bank’s economic forecast.
- Forecasts in a central bank should not simply reflect senior management views; economists must have a degree of freedom to allow productive internal discussions.
- Model-based forecasts provided policymakers with:
  - a starting point for deliberations;
  - confidence that actions reflected thorough consideration;
  - a means to explain actions with a coherent economic narrative.
- Core model requirements for an FPAS under IFT:
  - an endogenous short-term interest rate determined by monetary policy;
  - an important role for forward-looking expectations of the central bank and the public.
- Core models must be adapted over time; no one model provides all answers:
  - gap models and DSGE models each capture key macroeconomic principles but are incomplete; judgments and complementary models remain essential.
- Practical guidance:
  - newcomers to IFT can start with a simple gap model if appropriate adjustments are made while research on DSGE models continues;
  - informed judgment is most important for near-term forecasts (group of sectoral experts outperforms models up to 2 quarters);
  - combine near-term (sector-expert) forecasts with medium-term (model-driven) forecasts, ideally under a single manager overseeing the entire horizon.
- Institutional outcomes:
  - better informed media coverage and growth of academic interest in monetary policy;
  - emergence of a generation of highly qualified, policy-oriented monetary theorists and model builders strengthening recruitment;
  - CNB became an internationally recognized contributor of technical assistance despite small size.

*Source: wp1721 - 4.5 Communications—internal and external*

### References

### wp1721 - References

### Key topics covered
- Inflation targeting and inflation-forecast targeting.
- Monetary policy frameworks and rules for small, open, and emerging economies.
- Central bank transparency, communication, and publication of interest rate projections.
- Forecasting systems, DSGE and structural models, and policy analysis systems.
- Exchange rate as an instrument at zero interest rates and unconventional policy options.
- Evaluation of forecast quality, forecast decomposition, and dealing with data uncertainty.

### Notable working papers and studies (selected citations preserved verbatim)
- Alichi, A., H. Chen, K. Clinton, C. Freedman, M. Johnson, O. Kamenik, T. Kışınbay, and D. Laxton, 2009, “Inflation Targeting Under Imperfect Policy Credibility,” IMF Working Paper No. 09/94.
- Alichi, A., J. Benes, J. Felman, I. Feng, C. Freedman, D. Laxton, E. Tanner, D. Vavra, and H. Wang, 2015a, “Frontiers of Monetary Policymaking: Adding the Exchange Rate as a Tool to Combat Deflationary Risks in the Czech Republic,” IMF Working Paper No. 15/74.
- Alichi, A., K. Clinton, C. Freedman, M. Juillard, O. Kamenik, D. Laxton, J. Turunen, and H. Wang, 2015b, “Avoiding Dark Corners: A Robust Monetary Policy Framework for the United States,” IMF Working Paper No. 15/134.
- Arbatli, E., D. Botman, K. Clinton, P. Cova, V. Gaspar, Z. Jakab, D. Laxton, C. Lonkeng Ngouana, J. Mongardini, and H. Wang, 2016, “Reflating Japan: Time to Get Unconventional?” IMF Working Paper No. 16/157.
- Clinton, K., C. Freedman, M. Juillard, O. Kamenik, D. Laxton, and H. Wang, 2015, “Inflation-Forecast Targeting: Applying the Principle of Transparency,” IMF Working Paper No. 15/132.
- Laxton, D., D. Rose, and A. Scott, 2009, “Developing a Structured Forecasting and Policy Analysis System,” IMF Working Paper No. 09/65.
- Freedman, C. and D. Laxton, 2009, “Inflation Targeting Pillars: Transparency and Accountability,” IMF Working Paper No. 09/262.
- Obstfeld, M., K. Clinton, O. Kamenik, D. Laxton, Y. Ustyugova, and H. Wang, 2016, “How to Improve Inflation Targeting in Canada,” IMF Working Paper No. 16/192.

### Country- and institution-specific studies (selected)
- Andrle, M., T. Hlédik, O. Kameník, and J. Vlček, 2009, “Implementing the New Structural Model of the Czech National Bank,” Prague, CNB Working Paper, No. 2/2009.
- Brázdik, F., Z. Humplová, and F. Kopřiva, 2014, “Evaluating a Structural Model Forecast: Decomposition Approach,” Prague, CNB Research and Policy Note, No. 2/2014.
- Brůha, J., T. Hlédik, T. Holub, J. Polanský, and J. Tonner, 2013, “Incorporating Judgments and Dealing with Data Uncertainty in Forecasting at the Czech National Bank,” Prague, CNB Research and Policy Note, No. 3/2013.
- Coats, W., D. Laxton, and D. Rose, 2003, “The Czech National Bank’s Forecasting and Policy Analysis System,” Prague, Czech National Bank.
- Franta, M., T. Holub, P. Král, I. Kubicová, K. Šmídková, and B. Vašíček, 2014, “The Exchange Rate as an Instrument at Zero Interest Rates: The Case of the Czech Republic,” Research and Policy Notes 3, Czech National Bank.
- Šmídková, ed., 2008, “Evaluation of the fulfilment of the CNB’s inflation targets 1998–2007,” Prague, Czech National Bank.

### Methodology, forecasting, and model adoption (selected)
- Isard, P. and D. Laxton, 2000, “Inflation-Forecast Targeting and the Role of Macroeconomic Models,” in Inflation Targeting in Transition Economies: The Case of the Czech Republic, ed. by Warren Coates (Prague: Czech National Bank), pp. 105-137.
- Laxton, D. and P. Pesenti, 2003, “Monetary Rules for Small, Open, Emerging Economies,” Journal of Monetary Economics, vol. 50(5), pp. 1109–1146.
- Fukač, M. and A. Pagan, 2006, “Issues in Adopting DSGE Models for Use in the Policy Process,” CNB Working Paper, No. 6/2006.
- Skořepa, M. and V. Kotlán, 2006, “Inflation Targeting: To Forecast or To Simulate?” Prague Economic Papers, University of Economics, Prague, vol. 2006(4), pages 300–314.
- Brázdik, F. et al., 2014, “Evaluating a Structural Model Forecast: Decomposition Approach,” Prague, CNB Research and Policy Note, No. 2/2014.

### Communication, transparency, and policy-path discussion (selected)
- Archer, D., 2005, “Central Bank Communications and the Publication of Interest Rate Projections,” A paper for a Sveriges Riksbank conference on inflation targeting, Stockholm, June 2005.
- Woodford, M., 2005, “Central-Bank Communication and Policy Effectiveness,” presented at the Federal Reserve Bank of Kansas City Symposium, Jackson Hole, Wyoming, August 25–27.
- Dincer, N. N. and B. Eichengreen, 2014, “Central Bank Transparency and Independence: Updates and New Measures,” International Journal of Central Banking, Vol. 10, No. 1: 189-259, March.
- Mishkin, F. S., 2004, “Can Central Bank Transparency Go Too Far?” NBER Working Paper No. 10829.
- Svensson, L. E. O., 1997, “Inflation Forecast Targeting: Implementing and Monitoring Inflation Targets,” European Economic Review, 41(6), pp. 1111–46.
- Svensson, L. E. O., 2003, “Escaping from a Liquidity Trap and Deflation: The Foolproof Way and Others,” NBER Working Paper No. 10195.

*References list as provided in the source content.*

---


_Source: https://www.imf.org/-/media/files/publications/wp/wp1721.pdf_