## tarea2025100-source-pdf

## Source details

**Canonical URL:** [tarea2025100-source-pdf](https://www.imf.org/-/media/files/publications/tar/2025/english/tarea2025100-source-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/tar/2025/english/tarea2025100-source-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/tar/2025/english/tarea2025100-source-pdf.pdf.json)

---

### Preface and mission
- Mission purpose and coverage:
  - At the request of the Banco de Guatemala (BANGUAT), an in-person CAPTAC-DR technical assistance (TA) mission reviewed the institutional framework of BANGUAT’s Forecasting and Policy Analysis System (FPAS).
  - TA objectives: provide recommendations to construct an FPAS better aligned with best practices and that better guides policy decisions.
  - Mission dates: March 3 through March 7, 2025.
  - Mission participant mentioned: Mr. Juan José Ospina (Short-term Monetary and Capital Markets Department External Expert, Banco de la República, Colombia).
  - Interviewed BANGUAT senior management and staff from: Vice-President (Mr. José Alfredo Blanco), Chief Economist (Mr. Johny Gramajo), Department of Macroeconomic Analysis and Forecasting (DAMP), Department of Economic Research (DIE), Macroeconomic Statistics Department (DEM), Department of Execution of Monetary, Exchange Rate and Credit Policies (DEPMCC), and Department of Communication and Institutional Relations (DCRI).
- Topics reviewed:
  - Role of FPAS in monetary-policy decision making; coordination across departments; use of models and tools; determination and use of balance of risks.
- Principal observations:
  - Identified frictions in current system and gaps relative to FPAS best practices.
  - Recommendations to address frictions and fill gaps provided in report.

### Executive summary — context, strengths, and strategic priorities
- Context and background:
  - BANGUAT has operated under explicit inflation targeting (IT) since 2005.
  - Inflation target: 4 percent with a tolerance range of ±1 percentage point since 2013.
  - Primary policy instrument: Tasa de Interés Líder de Política Monetaria (TLPM).
- Key findings:
  - New Quarterly Projection Model (QPM) exists but is not fully integrated into FPAS.
  - Forecasting Team (FT) comprises staff from DAMP, DEM, and DIE but lacks full coordination and often operates in silos.
  - Fragmentation prevents unified, macroeconomically coherent forecasts to guide Monetary Board (MB) recommendations.
  - Monetary policy is forward-looking and uses a balance of risks, but model influence on official forecasts is limited.
- Existing strengths to support reform:
  - Strong conviction and commitment from senior management.
  - Appropriate institutional arrangements and rules to support implementation.
  - Capable staff across DAMP, DEM, and DIE using appropriate models and tools.
  - Departments already generate necessary inputs for a QPM-based forecast.
  - High internal and public credibility and a high level of transparency.
- Strategic recommendations (three broad themes):
  - Improve organization and structure of the forecasting process.
  - Fully integrate the QPM into the forecasting process.
  - Streamline processes and develop necessary infrastructure.

### Key recommendations (summary with priorities and timeframes)
- Organization and Structure:
  - 1. Establish a dedicated position to coordinate the three departments and strengthen DAMP’s lead. Priority: High. Timeframe: Short-Term.
  - 2. Establish a meeting schedule around the QPM for the FT and the Technical Committee; follow the agenda strictly. Priority: High. Timeframe: Short-Term.
  - 3. Review FT organization to ensure DAMP works closely with DIE and reduce bureaucracy. Priority: Medium. Timeframe: Medium-Term.
  - 4. Center FPAS around preparation of the MPR and publish it more promptly after MB decisions. Priority: Low. Timeframe: Long-Term.
- Incorporating the QPM:
  - 5. Use DEM GDP annual forecasts and DAMP inflation forecasts, satellite models, and expert judgements to guide gradual QPM introduction. Priority: High. Timeframe: Short-Term.
  - 6. Establish a meeting schedule on QPM assumptions and inputs to give DIE sufficient discussion time. Priority: High. Timeframe: Short-Term.
  - 7. Produce balances of risks for relevant variables using the QPM and incorporate QPM risk scenarios in policy discussion. Priority: Low. Timeframe: Medium-Term.
  - 8. Engage in peer-learning to support real-time forecast construction. Priority: Medium. Timeframe: Medium-Term.
- Streamlining processes and infrastructure:
  - 9. Reduce overlap across units and construct a common database. Priority: High. Timeframe: Short-Term.
  - 10. Build a data system that supports uploading/downloading FPAS inputs/outputs, guarantees replicability, fosters data integrity, and records judgements. Priority: Medium. Timeframe: Long-Term.
- Timeframe definitions:
  - Short term: < 12 months.
  - Medium term: 12 to 24 months.
  - Long term: more than 24 months.

### Institutional framework and current FPAS governance
- Key bodies and roles:
  - Three key bodies: Monetary Board (MB), Executive Committee (EC), and Forecasting Team (FT).
  - FT (led by the Chief Economist): produces macro forecasts and policy recommendations.
  - EC: consolidates FT outputs and advises the MB.
  - MB: decision-making body; uses EC and FT input to set TLPM.
- MB specifics:
  - Composition: President of BANGUAT and seven non-permanent members.
  - Publishes annually “Resolución JM—YYYY: Revisión de Política Monetaria, Cambiaria y Crediticia.”
  - Meeting frequency: typically meets eight times a year—usually on the last Wednesday of each month—to decide the TLPM; also convenes weekly for other matters.
- EC composition and function:
  - Composition: President, Vice President, General Manager, Economic Manager (Chief Economist), Financial Manager; supported by permanent advisors including Directors of DAMP, DEM, DIE, and DCRI.
  - Functions: implements MB policies, advises MB, formulates recommendations based on FT forecasts and balance of risks.
- FT and Technical Committee formalization:
  - Operates under Acuerdo de Gerencia General Número 132–2024 defining FT composition, roles, and responsibilities.
  - FT structured in two tiers: first tier (permanent members) and second tier (Technical Committee) led by Director of DIE, expected to meet at least once a month.

### Current forecasting practices, model use, and gaps
- Department responsibilities and practical outcome:
  - DAMP: prepares inflation forecast for next eight quarters, qualitative risk assessment, and fan chart around quarterly inflation forecast.
  - DEM: produces annual GDP growth forecasts; provides quarterly GDP nowcast to DIE (internal use).
  - DIE: generates forecasts using the QPM and maintains macro models, but its QPM-based forecasts are not adopted as official projections.
  - Practical outcome: FT adopts DAMP inflation and DEM growth forecasts; DIE’s QPM-based forecasts are presented but dismissed, so official forecasts and policy recommendations are not grounded in the QPM.
- Communication and MPR timing:
  - Press conference after each MB meeting announcing policy decision; Chief Economist presents balance of risks.
  - Public presentation for economic analysts the following day.
  - Monetary Policy Report (MPR) published three times a year—March, June, and September.
  - MPR timing: publication does not coincide with MB meetings; typically released in the second week of the second month of the subsequent quarter (examples: March report published in the second week of May; June report in the second week of August; September report in the second week of November).
  - December annual document: “Evaluation of Monetary, Exchange Rate, and Credit Policy as of November YYYY, and Economic Perspectives for YYYY+1.”
- Strengths:
  - Senior management commitment; QPM regarded by DIE and some senior FT members as core model; good external communication tools; strong technical staff and credibility.
- Gaps and operational challenges:
  - FT silo mentality and weak coordination; monthly meetings insufficient; participants often present department-aligned positions.
  - Fragmented forecasting process: DAMP/DEM forecasts compete with DIE’s, producing dysfunction.
  - Lack of coherent macroeconomic narrative and internal consistency: forecasts not produced using a general equilibrium model; absence of quantitative analysis of macro implications of risks.
  - Credibility of models: DIE’s forecasts historically viewed as unreliable and often dismissed without sufficient discussion.
  - FT bylaw alone does not ensure coordination; meetings are too infrequent and unstructured.
  - Chief Economist constrained to rely on DAMP-produced forecasts due to competing forecasts and limited model credibility.
  - Resource constraints: senior staff time pressures lead to delegation and reduced participation of first-tier FT in Technical Committee meetings.

### FT meeting calendar, MPR issues, and governance recommendations
- Problems identified:
  - FT bylaw mandates one meeting per month, giving discretion that leaves forecasting process overly dependent on circumstances rather than a disciplined schedule.
  - MPR issues:
    - No fixed rule for publication time.
    - Published with delay relative to MB decision, reducing relevance.
    - Too much weight on observed data and little weight on forecast.
    - About 50 percent of MPR is about other countries with no linkage to Guatemalan forecast.
    - Inclusion of variables (e.g., corn and wheat prices) that generally do not affect the forecast.
- Key governance recommendations (paragraphs 34–46):
  - 34. Appoint a dedicated senior-level FT Coordinator with delegated authority; ideally at level between Chief Economist and department directors (e.g., Deputy Manager) or assigned to senior DAMP staff. Coordination should not fall to Director of DIE.
  - 35. Coordinate forecasting process by DAMP in practice; Chief Economist guides conceptually and referees.
  - 36. Establish a DAMP team/unit to organize workflow, ensure forecast integrity, maintain coherence, set agendas, take minutes, create shared database, and conduct annual internal evaluation.
  - 37–38. Align DAMP units’ work with QPM requirements and establish a strict meeting calendar planned working backward from MB and EC meeting dates and data release timing.
  - 39. Identify necessary QPM inputs and sequence of meetings (External Assumptions; GDP Nowcast and Quarterly Projection; Discussion on Initial Conditions, Trends, and Shocks; Short-Term Inflation Forecast; Model Run; Baseline Agreement; Balance of Risks and Alternative Scenarios; Review of Risk Scenarios; Balance of Risks and Fan Chart Development).
  - 40. Review necessity of Technical Committee; consider elimination if it reduces senior involvement.
  - 41. Use QPM for counterfactual scenarios to strengthen policy recommendations.
  - 42. Systematically document key judgments and assumptions; maintain detailed logs overseen by new DAMP unit.
  - 43. Consistently follow rigorous forecasting process regardless of macro stability.
  - 44. Organize FPAS around MPR preparation; ensure all departments contribute to a unified QPM-guided forecast.
  - 45. Publish MPR promptly after MB decision with a forward-looking tone and streamlined international context.
  - 46. Begin implementing recommendations without delay; require strong institutional commitment.

### QPM conditioning, balance of risks, and model coherence
- Status and problems:
  - QPM forecasts are often “free”: not systematically anchored to external assumptions and varying conditioning horizons across forecasts.
  - Limited input from high-frequency data; expert judgement used to incorporate shocks and current developments.
  - Models are not effectively used to produce an integrated balance of risks; GDP and inflation fan charts produced independently by DEM and DAMP, causing disconnects.
  - Weak linkage between monetary policy, exchange rate, and model analysis despite exchange rate importance.
- Suggested analytical advances for QPM:
  - Analyze policy rate influence on exchange rate absent FX intervention.
  - Analyze shocks’ impacts on policy rate and exchange rate absent intervention and under FX intervention rules.
  - Analyze how FX intervention modifies policy rate response under different shocks.
- Recommendations (C):
  - Produce a fixed two-year horizon quarterly forecast (typically eight to twelve quarters) to improve comparability and policy design.
  - Draw on DAMP and DEM forecasts and expertise to guide QPM during initial integration; discuss shocks for first forecast year and allow model to “speak” for second year.
  - Use satellite models (e.g., SVAR) to align QPM with local features such as low MP transmission and high informality.
  - Apply consistent conditioning across forecasting rounds for external variables (e.g., remittances, Federal Reserve’s policy rate).
  - Produce fan charts for all key variables using the QPM and transition toward QPM-derived joint distributions.
  - Leverage macroeconomic stability as a favorable window to integrate QPM without compromising FT credibility.
  - Use QPM to strengthen linkage between external context and macro forecast and quantify transmission mechanisms.
  - Promote peer-learning experiences (forecasting exercises at other central banks or TA mission support).

### Operational processes, data, and infrastructure
- Current status and assessment:
  - Each department builds its own database; model codes and data are organized with version control and replicability.
  - Significant functional overlap across departments reduces efficiency and risks coherence (e.g., multiple units monitoring international developments; DIE downloads data already covered by DAMP).
  - No unified database; inputs exchanged mainly via email increasing risk of errors and inconsistencies.
  - Key variable categorization not harmonized (e.g., different core inflation measures used by DAMP and DIE).
- Operational recommendations:
  - Reorganize DAMP into four units (secciones):
    - Unit 1 (International): analyze international variables.
    - Unit 2 (Inflation): detailed local inflation analysis and short-term inflation forecasts.
    - Unit 3 (Macro analysis): collaborate with DEM to produce nowcast and short-term growth forecasts and organize macro narrative for EC, MB, and MPR.
    - Unit 4 (Forecasting process management and evaluation): oversee forecasting process, set meeting agendas, keep minutes, organize written reports, ensure MPR coherence, build shared database, and conduct annual internal evaluation.
  - FT must agree on core variables for forecast and MPR; DIE to work with DAMP and DEM inputs.
  - Agree on a core inflation measure for models, policy documents, and communications; treat alternatives as robustness checks.
  - DIE should cease constructing separate FPAS data for official forecasting; such data construction should be coordinated with DAMP.
  - Establish a centralized database with permissions, integrity, confidentiality, and capability to record DAMP, DEM, and DIE forecasts and conditioning variables to enable replication and evaluation; discontinue data exchange via email.

*Source: tarea2025100-source-pdf - Preface and excerpts from IMF Technical Assistance Report.*

### Preface ................................................................................................................

### Preface

### Mission purpose and coverage
- At the request of the Banco de Guatemala (BANGUAT), an in-person CAPTAC-DR technical assistance (TA) mission reviewed the institutional framework of BANGUAT’s Forecasting and Policy Analysis System (FPAS).
- TA objectives: provide recommendations to construct an FPAS better aligned with best practices and that better guides policy decisions.
- Mission dates: March 3 through March 7, 2025.
- Mission participant mentioned: Mr. Juan José Ospina (Short-term Monetary and Capital Markets Department External Expert, Banco de la República, Colombia).
- Interviewed BANGUAT senior management and staff from: Vice-President (Mr. José Alfredo Blanco), Chief Economist (Mr. Johny Gramajo), Department of Macroeconomic Analysis and Forecasting (DAMP), Department of Economic Research (DIE), Macroeconomic Statistics Department (DEM), Department of Execution of Monetary, Exchange Rate and Credit Policies (DEPMCC), and Department of Communication and Institutional Relations (DCRI).

### Topics reviewed
- Role of FPAS in monetary-policy decision making, benefits and frictions of current system.
- Coordination across departments involved in forecasting and policy analysis.
- Use of models and tools for forecasting and analysis.
- Determination and use of balance of risks in policy discussions.

### Principal observations from mission interactions
- Identified frictions in the current system, institutional arrangements, and coordination mechanisms across departments.
- Noted gaps relative to FPAS best practices.
- Recommendations to address frictions and fill gaps are provided in this report.
- Mission acknowledges cooperation of BANGUAT staff, logistical support, CAPTAC-DR funding and support.

---

### Executive Summary

### Context and background
- Since 2005, BANGUAT has operated under an explicit inflation targeting (IT) regime.
- Inflation target has remained at 4 percent with a tolerance range of ±1 percentage point since 2013.
- Primary policy instrument: Tasa de Interés Líder de Política Monetaria (TLPM).
- BANGUAT developed models and tools to monitor and forecast macroeconomic variables, particularly inflation, to guide the interest rate path.

### Key findings
- BANGUAT introduced a new Quarterly Projection Model (QPM) intended as the primary forecasting model but it is not yet fully integrated into the FPAS.
- The Forecasting Team (FT) comprises staff from DAMP, DEM, and DIE but lacks full coordination and often operates in silos.
- Fragmentation prevents full utilization of the QPM and other tools to produce a unified, macroeconomically coherent forecast to guide policy recommendations to the Monetary Board (MB).
- Monetary policy is forward-looking and uses a balance of risks around inflation, but model influence on official forecasts remains limited.
- Existing strengths that can support reform:
  - Strong conviction and commitment from senior management.
  - Appropriate institutional arrangements and rules to support implementation.
  - Capable staff across DAMP, DEM, and DIE using appropriate models and tools.
  - Departments already generate necessary inputs for a QPM-based forecast.
  - High internal and public credibility and a high level of transparency.

### Strategic recommendations (three broad themes)
- Improve organization and structure of the forecasting process.
- Fully integrate the QPM into the forecasting process.
- Streamline processes and develop necessary infrastructure.

### Organizational recommendations (high-level)
- Establish a dedicated FT Coordinator to align assumptions, coordinate inputs, and oversee deliverables; formally integrate the role and evaluate on measurable coordination outcomes.
- Strengthen DAMP to lead the process and create a dedicated DAMP team to organize and oversee forecasting workflow.
- Streamline FT structure to reduce bureaucracy and increase senior staff participation.
- Align the FPAS to produce a more timely, forecast-focused, and coherent Monetary Policy Report (MPR).

### QPM integration recommendations (high-level)
- Design and adopt a structured meeting schedule centered on the QPM to discuss inputs, assumptions, and outputs with subject-matter experts.
- Use annual forecasts from DEM (GDP) and DAMP (inflation), satellite models, and expert judgments to initially guide QPM forecasts and introduce the QPM gradually.
- Produce balances of risks and incorporate QPM-based risk scenarios into policy discussions.

### Process and infrastructure recommendations (high-level)
- Reduce overlapping responsibilities across departments and units.
- Create a shared, centralized database to guarantee consistent and accessible data for all FPAS contributors.
- Develop infrastructure to support replicability, data integrity, ex-post evaluation, and recording of judgements.

---

### Recommendations

### Key recommendations and implementation priorities (as presented)
- Organization and Structure:
  - 1. Establish a dedicated position responsible for coordinating the three departments involved in the forecasting process. Strengthen the role of the DAMP in leading the forecasting process and create a team within the DAMP to organize and oversee the forecasting workflow. Priority: High. Timeframe: Short-Term.
  - 2. Establish a meeting schedule around the QPM and along the lines explained in this report for the FT and the Technical Committee to build the forecast. Follow the agenda strictly. Priority: High. Timeframe: Short-Term.
  - 3. Review the organization of the FT to guarantee that the DAMP works closely with the DIE, reducing bureaucracy and guaranteeing participation from senior members and experts. Review whether the Technical Committee is needed. Priority: Medium. Timeframe: Medium-Term.
  - 4. Center the FPAS around the preparation of the MPR. The MPR should be published more promptly following the MB’s policy decision and focus more directly on the forecast and Guatemala’s domestic macroeconomic outlook, rather than primarily on known facts or international developments. Priority: Low. Timeframe: Long-Term.

- Incorporating the QPM in the Forecast:
  - 5. Use the Gross Domestic Product (GDP) annual forecasts produced by the DEM and the inflation forecasts produced by the DAMP as well as outputs from satellite models and inputs from experts (judgements) to initially guide the QPM forecast and make a gradual introduction of the QPM. Priority: High. Timeframe: Short-Term.
  - 6. Establish a meeting schedule around the assumptions and inputs required by the QPM. Guarantee sufficient time for discussions that can help the DIE produce a unified, credible forecast. Priority: High. Timeframe: Short-Term.
  - 7. Produce a balance of risks for all relevant variables using the QPM and incorporate risk scenarios produced with the QPM in the policy discussion. Priority: Low. Timeframe: Medium-Term.
  - 8. Engage in a peer-learning experience that supports the team in constructing a forecast in real time. Priority: Medium. Timeframe: Medium-Term.

- Streamlining Processes and Building Infrastructure:
  - 9. Streamlining processes: reduce overlap of activities across units(secciones) and departments and construct a common database. Priority: High. Timeframe: Short-Term.
  - 10. Build a data system that allows the FT to download and upload inputs and outputs for the FPAS, guarantees replicability, fosters data integrity, and enables ex-post evaluation. Record and store judgements made during the forecast. Priority: Medium. Timeframe: Long-Term.

- Timeframe definitions:
  - Short term: < 12 months.
  - Medium term: 12 to 24 months.
  - Long term: more than 24 months.

---

### Introduction — diagnostic overview

### Historical and institutional context
- BANGUAT has conducted monetary policy under explicit IT since 2005.
- Inflation target: 4 percent with a tolerance range of ±1 percentage point since 2013.
- Primary instrument: TLPM, influencing short-term interest rates and key macroeconomic and financial variables.

### Current FPAS status and limitations
- The new QPM exists but is not yet used to produce the official forecast; the FT does not work in a well-coordinated manner.
- Official inflation and GDP forecasts rely on satellite models developed by DAMP and DEM plus expert judgement, rather than on the QPM.
- The QPM lacks credibility among key FT members; some EC meetings occurred where the QPM-based forecast was not discussed.
- Past TA: 2021 mission recommended a new semi-structural model and internal FPAS process review; 2024 mission assisted in developing the new QPM. Some 2021 recommendations remain unimplemented.

### Purpose and structure of the report
- Purpose: identify frictions and gaps in forecasting processes and institutional arrangements; provide actionable insights to design an FPAS aligned with best practices.
- Method: evaluate FPAS role in monetary decision-making, organization and coordination of teams, and the use of models for forecasts, balances of risks, and MPR communication.
- Report organization: three themes — organization and forecasting process; use of models and QPM integration; efficiency of processes and supporting infrastructure. Each theme includes: status quo, assessment, and recommendations.

*Source: tarea2025100-source-pdf - Preface (IMF Technical Assistance Report).*

### 7.      Monetary policy in Guatemala is conducted through a well-defined institutional framework

### 7. Monetary policy in Guatemala is conducted through a well-defined institutional framework

### Institutional roles and interactions
- Three key bodies: the MB, the EC, and the FT, each with distinct but interconnected roles.
- FT (led by the Chief Economist): responsible for producing macroeconomic forecasts and formulating policy recommendations.
- EC: reviews and consolidates FT outputs and acts as a bridge between technical staff and the MB.
- MB: Guatemala’s highest authority on monetary, exchange rate, and credit policy; uses EC and FT input to set the policy rate (TLPM) and guide overall policy decisions.
- Process character: forward-looking, with the balance of risks around the inflation forecast playing a central role.
- Gap noted: forecasts and risk assessments are not produced using tools that ensure internal consistency and a coherent macroeconomic narrative (for example, general equilibrium models).

### The Monetary Board (MB)
- Role: decision-making body at BANGUAT responsible for monetary, exchange rate, and credit policies.
- Composition: President of BANGUAT and seven non-permanent members representing various sectors of Guatemalan society.
- Annual publication: “Resolución JM—YYYY: Revisión de Política Monetaria, Cambiaria y Crediticia,” outlining intervention rules and parameters for the foreign exchange market and the schedule of monetary policy meetings for the upcoming year.
- Meeting frequency:
  - Typically meets eight times a year—usually on the last Wednesday of each month—to decide on the policy interest rate TLPM.
  - Also convenes weekly to address other matters.
- Reliance: MB relies on recommendations provided by the EC for monetary policy decisions.

### The Executive Committee (EC)
- Composition: BANGUAT’s President, Vice President, General Manager, Economic Manager (Chief Economist), and Financial Manager; supported by permanent advisors including the Directors of DAMP, DEM, DIE, and DCRI.
- Functions:
  - Implements policies set by the MB.
  - Advises the MB and formulates monetary policy recommendations based on economic forecasts and the balance of risks prepared by the FT.
  - Operates under leadership of the President of BANGUAT.

### The Forecasting Team (FT) and Technical Committee
- Leadership and responsibility:
  - FT led by the Chief Economist; advises the EC on monetary, exchange rate, and credit policies.
  - FT typically meets once a month to discuss and reach consensus on economic forecasts and the balance of risks.
  - Chief Economist approves the final outputs of the FT (BANGUAT’s official macroeconomic forecasts) and presents them to the EC with a policy-rate recommendation.
  - FT evaluates and approves significant modifications to the Bank’s core forecasting models.
- New formalization:
  - Operates under Acuerdo de Gerencia General Número 132–2024, which defines FT composition, roles, and responsibilities.
  - FT includes representatives from DIE, DAMP, DEM, and Financial Stability; supervised by the Chief Economist.
  - FT structured in two tiers:
    - First tier: permanent members including the Chief Economist and directors of the four departments.
    - Second tier: Technical Committee, led by the Director of the DIE, responsible for preparing forecasts and running models; expected to meet at least once a month.

### Current forecasting practices and workflow
- Departments and outputs:
  - DAMP: prepares the inflation forecast for the next eight quarters and conducts a qualitative assessment of risks to the inflation outlook; constructs a fan chart around the quarterly inflation forecast based on historical inflation distribution adjusted for identified balance of risks.
  - DEM: produces annual GDP growth forecasts for the current and following year; provides detailed sectoral activity and national accounts data; supplies a quarterly GDP nowcast to DIE (used internally, not published).
  - DIE: generates forecasts for key macroeconomic variables using the QPM; maintains macroeconomic models including the QPM.
- Practical outcome:
  - FT adopts the inflation forecast from DAMP and the growth forecast from DEM, while dismissing DIE’s forecasts.
  - DIE’s QPM-based forecasts are presented alongside DAMP and DEM outputs but are not selected as official projections.
  - Result: BANGUAT’s official inflation and GDP forecasts, the balance of risks, and policy recommendations are not grounded in the QPM or other macroeconomic models developed by DIE.
  - Role in MPR: DIE plays no role—the DEM drafts most sections related to economic activity, while the DAMP prepares the rest of the report.

### Communication and transparency
- Communication instruments:
  - After each MB meeting on policy rates, a press conference announces the policy decision; indicates whether the decision was unanimous; Chief Economist presents the balance of risks and rationale; bank management addresses questions.
  - The presentation to the public generally mirrors that delivered to the MB.
  - The following day, a separate public presentation is made for economic analysts.
- Publications:
  - Monetary Policy Report (MPR) published three times a year—March, June, and September.
  - Timing: MPR publication does not coincide with MB meetings; typically released in the second week of the second month of the subsequent quarter (examples: March report published in the second week of May; June report in the second week of August; September report in the second week of November).
  - December annual document: “Evaluation of Monetary, Exchange Rate, and Credit Policy as of November YYYY, and Economic Perspectives for YYYY+1” — provides a comprehensive review of economic developments and forecasts for the following year; includes recommendations on parameters for exchange rate intervention and foreign reserves accumulation; outlines BANGUAT’s plans, initiatives, and accomplishments in enhancing policymaking.

### Assessment — Strengths
- Commitment and intent:
  - BANGUAT has a strong foundation to enhance its FPAS and develop it in line with international best practices.
  - Senior management (including the Vice President and Chief Economist) is convinced enhancements are necessary, particularly better incorporating general equilibrium models.
  - Goal: establish a refined version of the FPAS as a permanent institution within BANGUAT.
- Institutional rules and convictions:
  - New FT bylaw (Reglamento EP) formally defines FT structure, roles, responsibilities, and appointment of a coordinator and replacements—fosters better inter-departmental coordination (though not guaranteed).
  - QPM regarded as the core general equilibrium model for forecasting by DIE and some senior FT members (including Vice President and Chief Economist), increasing likelihood of QPM integration into FPAS.
- Communication pillars:
  - Internal: relatively good communication and coordination between DAMP and DEM supporting annual forecasts and MPR drafting.
  - External: clear and consistent communication practices with press, public, and market analysts; predictable tools (press conferences, public presentations, MPR); publication of the annual evaluation document and sharing of presentation materials that mirror those used in policy discussions.
- Technical credibility and forward-looking approach:
  - Strong technical staff track record and credibility with public and MB.
  - Policy process places inflation risks at the center, with EC-FT adopting a forward-looking perspective in recommendations to the MB.

### Assessment — Gaps
- Silo mentality and weak coordination:
  - FT operates with a silo mentality; the three departments (DAMP, DIE, DEM) function as two or three distinct, uncoordinated teams despite being one FT on paper.
  - Monthly meetings are insufficient; discussions on shocks and economic narratives are limited and too infrequent.
  - Participants often feel pressure to present department-aligned positions rather than integrated views.
- Fragmented forecasting process:
  - DAMP’s and DEM’s forecasts often compete with DIE’s, leading to dysfunction.
  - DIE’s contributions—including the QPM, joint forecasts, and risk scenarios—have not been meaningfully integrated into FPAS or MPR.
- Lack of coherent macroeconomic narrative and internal consistency:
  - Forecast, balance of risks, and macro analysis often lack a coherent underlying narrative.
  - Without forecasts generated using a general equilibrium model, there is no assurance that projections for GDP, inflation, and policy implications are internally consistent in direction or magnitude.
  - Quantitative analysis of potential macroeconomic and policy implications of risks in the balance of risks is absent.
- Credibility and usage of models:
  - Historically, model-based forecasts sometimes “did not make sense” or were not deemed reliable; DIE’s forecasts lacked credibility in policy terms and were frequently dismissed without sufficient discussion.
  - Time constraints and leadership schedules reduced attention to models, diminishing their perceived usefulness and integration.
  - Despite assertions that the QPM is the official model, under current conditions it is unlikely to be adopted in practice: models have lost credibility with the Chief Economist and other senior FT members.
- Limitations of regulation alone:
  - FT bylaw (Acuerdo General Número 101–2020 previously, and now Acuerdo de Gerencia General Número 132–2024) does not guarantee necessary coordination and engagement across teams.
  - Technical Committee and FT meetings are too infrequent, lack structure, and do not ensure participation of key personnel from DAMP and DEM.
- Challenges for the Chief Economist:
  - Competing forecasts place the Chief Economist in a difficult position when expected to explain and defend forecasts that are not well-understood or easy to justify.
  - Consequently, the Chief Economist typically relies on DAMP-produced forecast and balance of risks when communicating with decision-makers.
- Resource and participation constraints:
  - Senior management within FT faces numerous obligations and stiff time constraints (notably the Chief Economist and Director of DAMP), limiting effective coordination.
  - Delegation has led to junior staff attending some FT meetings; first-tier FT often does not participate in Technical Committee meetings, reducing the quality and acceptance of QPM-based forecasts.
  - FT’s hierarchical structure excludes some experts and senior staff from discussions, limiting input from sectoral experts and senior economists.
- Insufficient interaction and meeting cadence:
  - FT and Technical Committee meet only once a month—typically the day before the forecast is due—resulting in an “output” meeting with too few “input” meetings to develop a robust projection.
  - Limited preparation time between Technical Committee output meetings and the Chief Economist meeting reduces the likelihood of a convincing macroeconomic narrative.
- Weak discipline in routine review:
  - EC or FT sometimes do not engage in thorough discussion of the forecast when there are no significant changes in observed variables or projected outcomes, weakening routine use of the QPM and risking oversight of emerging risks.
  - Quarterly projections (such as those generated by the QPM) require more frequent and in-depth discussions.

*IMF Technical Assistance Report | excerpts from chapter on monetary policy framework*

### 32.        The FT’s meeting calendar is too open and lacks sufficient structure. The FT bylaw

### FT meeting calendar and forecasting process (excerpts)

### Problems with the FT meeting calendar and MPR timing
- 32. The FT’s meeting calendar is too open and lacks sufficient structure. The FT bylaw (Reglamento EP) mandates only one meeting per month. While this may suffice as a general guideline, in practice it leaves too much discretion to the DIE Director to determine when meetings take place. As a result, the forecasting process becomes overly dependent on the current context or circumstances, rather than following a consistent and disciplined schedule.
- 33.i. The MPR does not have a fixed rule for the time of its publication.
- 33.ii. The MPR is published with a delay relative to the MB’s decision. This reduces the relevance of the MPR as a communication tool that explains in detail how the central bank is understanding the economic outlook and the implications for MP. In the presence of shocks or new data, the MPR may end up outdated at the time of publication.
- 33.iii. The MPR has too much weigh on observed data and little weight on the forecast. This, and the fact that the forecast is not produced with a macro model such as the QPM, makes it hard to see and understand a comprehensive, coherent macro story.
- 33.iv. International developments have too much weight in the MPR. About 50 percent of the MPR is about other countries, and there is no linkage between those developments and the forecast for the Guatemalan economy. There is, however, a conceptual link between international developments and risks and the inflation balance of risks.
- 33.v. In the MPR there are variables shown and discussed such as corn and wheat prices that in general do not affect the forecast. They are there due to historical reasons. This excess of potentially irrelevant information is in part the result of not having a unified framework, such as the QPM, for producing the macro-projections.

### Key recommendations to strengthen coordination and governance (paragraphs 34–46)
- 34. Appoint a dedicated senior-level FT Coordinator with delegated authority to align forecast assumptions, ensure cross-departmental consistency, and oversee timely delivery of outputs. Desired attributes and scope:
  - Primarily focused on the forecasting process; strong knowledge of the Guatemalan economy.
  - Understanding of models used by the DIE and data/empirical tools used by the DAMP and the DEM.
  - Seniority to command credibility with team and the Chief Economist; strong listening and consensus-building skills; authority to make decisions when consensus cannot be reached.
  - Ideally a new position at a level between the Chief Economist and department directors (e.g., Deputy Manager (Subgerente)) or formally assigned to a senior DAMP staff member (ideally the Director).
  - If assigned to the Director of DAMP, the mandate should be explicit to ensure authority and respect across the FT.
  - Coordination of the FT and Technical Committee should not fall to the Director of the DIE.
- 35. Coordinate the entire forecasting process (preparing the forecast, presenting it with the balance of risks, and publishing the MPR) by the DAMP rather than the Chief Economist in practice. Role of Chief Economist or Deputy Manager:
  - Guide the forecast conceptually, serve as referee and interlocutor, ensure analytical soundness, challenge assumptions, and foster dialogue.
- 36. Establish a team (new unit/Sección within DAMP) to organize workflow, ensure forecast integrity, and maintain coherence across forecast, balance of risks, policy advice, and the MPR. Responsibilities:
  - Systematically set meeting agendas, take minutes, organize written reports, ensure MPR coherence, create a shared database for all FT members, and conduct an annual internal evaluation of both forecast and process.
- 37. Align DAMP units’ work with the structure and requirements of the QPM and the MPR: units must provide essential QPM inputs and contribute to the macroeconomic narrative and MP formulation, including discussion of underlying shocks and data interpretation.
- 38. Agree on and establish a meeting calendar for collaboratively building the forecast, to be followed systematically and strictly, guided by QPM input requirements. Coordination of the agenda is the responsibility of the new DAMP unit. Meetings should be planned working backward from MB and EC meetings and data release timing.
- 39. Identify necessary QPM inputs and a sequence of meetings as an initial framework:
  - i. External Assumptions (Unit 1 of the DAMP);
  - ii. GDP Nowcast and Quarterly Projection (DEM and Unit 3 of the DAMP);
  - iii. Discussion on Initial Conditions, Trends, and Shocks (DAMP measures such as GDP gap and neutral rate; DIE using QPM to analyze shocks);
  - iv. Short-Term Inflation Forecast (Unit 2 of the DAMP);
  - v. Model Run (Corrimiento) — deep discussion of model forecast, shocks for conditioning assumptions, macro story, and key judgments;
  - vi. Baseline Agreement — finalize and agree baseline scenario;
  - vii. Balance of Risks and Alternative Scenarios — plan how to build risks and alternatives in QPM and which risks to illustrate/quantify;
  - viii. Review of Risk Scenarios — review QPM-constructed scenarios for alignment with outlook and policy;
  - ix. Balance of Risks and Fan Chart Development — construct the fan chart around the QPM forecast and discuss balance of risks.
- 40. Review necessity of the Technical Committee and consider elimination; the FT requires active involvement of the senior coordinator (Deputy Manager), senior DAMP officials, and relevant experts to ensure correct shocks and a sound credible forecast.
- 41. Use the QPM to strengthen policy recommendations by incorporating counterfactual scenarios reflecting identified risks in the balance of risks.
- 42. Systematically document key judgments and assumptions made in constructing each forecast; maintain detailed logs of expert judgments at each stage; documentation to be coordinated and overseen by the new DAMP unit.
- 43. Consistently follow a rigorous forecasting process: all meetings in the forecast agenda should always take place, even in periods of macroeconomic stability; this discipline supports risk identification, continuous improvement, and reduces complacency.
- 44. Organize the FPAS around preparation of the MPR, which should convey the core macroeconomic narrative of the forecast. All departments (DAMP, DEM, DIE) should contribute to writing the MPR, ensuring a unified QPM-guided forecast.
- 45. Publish the MPR promptly after the MB decision meeting with a more forward-looking and prospective tone; emphasize macroeconomic story and forecast rather than past developments; streamline international context to what is relevant; domestic section should focus on the forecast and its implications for MP; a separate section could describe known facts.
- 46. Begin implementing recommendations without delay; reforms require strong institutional commitment and alignment at senior leadership and staff levels. Prompt action is essential to sustain momentum and institutional commitment.

### Use of models in the forecasting process and QPM incorporation (paragraphs 47–53)
- 47. Status quo: DAMP, DIE, and DEM each use distinct forecasting models and tools; outputs are combined with expert judgment to produce forecasts. No single tool is relied upon; because macroeconomic models are not utilized to produce the official forecast, there is no guarantee of quantitative or theoretical consistency and coherence of the final forecast.
- 48. DAMP responsibilities and tools:
  - Produces inflation forecasts and publishes indicators to assess the stance of monetary policy and financial conditions (calendar year, monthly, and quarterly).
  - Employs 'artisan models,' VARs, and ARIMA models; uses high-frequency data (precios mayoristas) and expert judgment.
  - Tools are particularly useful for short-term projections and can provide valuable inputs for the QPM. The index of monetary conditions and the monetary policy stance measure can be useful QPM inputs.
- 49. DEM responsibilities and tools:
  - Produces GDP forecasts across horizons and frequencies; official GDP forecast is annual for a calendar year and is primarily based on national accounts data and expert judgment.
  - Forecast developed by sector at the CIIU level and aggregated up.
  - Generates quarterly nowcast and forecast using statistical models (e.g., ARIMA) combined with expert judgment; quarterly projections are sent to the DIE as inputs for QPM runs, but there is limited FT discussion of these quarterly forecasts.
- 50. DIE responsibilities and tools:
  - Produces macroeconomic forecasts and occasionally alternative or risk scenarios using the QPM; significant progress made with 2024 TA mission.
  - Developed alternative models such as SVAR to complement QPM and identify weaknesses.
  - DIE constructs its own database from external and internal sources and receives inputs from DAMP and DEM; discussions of inputs, underlying shocks, and macro implications remain limited.
  - Alternative scenarios have occasionally been produced using the QPM but have not significantly influenced balance of risks or policy recommendations.
  - Technical software used: IRIS, Matlab, Dynare; models are meticulously maintained with version tracking for replicability.
- 51. Strengths supporting QPM integration:
  - The QPM (developed with IMF TA) is regarded by DIE as a significant improvement.
  - Teams already produce necessary inputs for a complete QPM-based forecast.
  - Staff systematically estimate and monitor unobservable variables (output gap, neutral rate, monetary financial conditions) key to integrating the model into the forecasting process.
- 52. Staff capacity:
  - Teams have deep technical expertise and extensive knowledge of the Guatemalan economy; appropriate models and tools are used with expert judgment; BANGUAT’s role in producing GDP data strengthens FT insights; DIE has technical capacity to apply QPM.
- 53. Gaps:
  - QPM forecasts lack a fixed forecast horizon. Currently the FT's forecast is annual, with horizon varying by time of year. QPM forecasts project either six or eight quarters ahead at different times. This inconsistency introduces variability in outputs and communication, complicates MP design, analysis of shock impact/persistence, and formulation of policy response.

*https://www.imf.org/-/media/files/publications/tar/2025/english/tarea2025100-source-pdf.pdf*

### 54.        QPM forecasts are not consistently conditioned on external assumptions over the entire

### tarea2025100-source-pdf - 54.        QPM forecasts are not consistently conditioned on external assumptions over the entire

### QPM conditioning and use of high-frequency information
- QPM forecasts are often “free”: not systematically anchored to key pieces of information and the horizon over which external variables are conditioned varies from one forecast to another.
- The model often receives limited input from high-frequency data that could reflect important developments outside the model’s structure.
- Expert judgement is crucial to incorporate high-frequency data and to introduce shocks into the forecast that reflect the FT’s understanding of the current economic environment and likely future developments.

### Balance of risks and coherence across models
- Models are not effectively used to produce or inform a consistent and integrated balance of risks.
- GDP and inflation fan charts are produced independently by the DEM and the DAMP, respectively, causing a disconnect:
  - Risks reflected in the inflation forecast are not incorporated into the GDP forecast intervals, undermining coherence of the macroeconomic narrative.
  - There is little quantitative exploration of how risks to inflation might affect other key variables, particularly GDP growth and the policy rate.
- This limits the usefulness of risk analysis for policymaking and for communicating the central bank’s outlook in the MPR.

### Monetary policy, exchange rate, and model linkage
- Linkage between the forecast/analysis of monetary policy and the exchange rate is weak despite the exchange rate’s crucial role in the Guatemalan economy and in the conduct of MP.
- Suggested analytical advances for the QPM:
  - Analyze how changes in the monetary policy rate influence the exchange rate in the absence of FX intervention.
  - Analyze how various shocks impact both the policy rate and the exchange rate without intervention and under what conditions they would trigger intervention based on BANGUAT’s pre-established parameters.
  - Analyze how FX intervention modifies the policy rate’s response to different shocks.
- Integration is increasingly necessary as the Guatemalan economy transitions to an IT regime with a flexible exchange rate.

### Recommendations (C)
- Produce a fixed two-year horizon quarterly forecast:
  - Aim: a consistent forecast horizon—typically eight to twelve quarters—to enhance clarity, comparability over time, and credibility.
  - Benefits: improve design of monetary policy and analysis of shocks and medium-term impacts.
- Draw on DAMP and DEM forecasts/expertise to guide the QPM during initial integration:
  - Increases ownership among FT members and credibility.
  - During the first year of the forecast horizon, thoroughly discuss shocks used to incorporate guidance; may allow the model to "speak" for the second year.
- Use satellite models to support QPM:
  - Example: structural VAR models developed by the DIE can provide insights aligning QPM forecasts with local features (e.g., low transmission of MP influenced by underdeveloped credit and financial markets and high informality).
  - Over time, such exercises can adapt QPM to ensure greater relevance and realism.
- Apply a consistent approach to conditioning across forecasting rounds:
  - External variables—such as remittances or the Federal Reserve’s policy rate—should be incorporated as conditioning assumptions over the entire forecast horizon.
  - Current practice is inconsistent (e.g., short-term inflation forecasts sometimes used, assumptions about the Federal Reserve’s interest rate path inconsistently applied).
- Produce fan charts for all key forecast variables using the QPM:
  - Initially, generate charts using current techniques applied to the QPM forecast.
  - Transition toward producing fan charts directly from the QPM to leverage its general equilibrium structure and jointly determined distributions.
- Leverage current macroeconomic stability to incorporate the QPM into the forecasting framework:
  - This context offers a favorable window for timely implementation without compromising FT’s credibility with the EC or the MB.
- Use the QPM to strengthen the link between the external context and the macroeconomic forecast:
  - External assumptions in the QPM should align with those reported in the balance of risks, policy discussions, and the MPR.
  - QPM should explain and quantify transmission mechanisms of changes in external variables and implications for the forecast, policy recommendations, and decisions.
- Promote peer-learning experiences:
  - Options: participate in a forecasting exercise at another central bank or receive a TA mission supporting real-time forecast construction.

### Operational processes, efficiency, and infrastructure — status and assessment
- Status quo:
  - Each department builds its own database tailored to its models.
  - Strong organization of model codes and data across departments; runs are replicable and version control is maintained.
- Assessment:
  - Significant overlap in functions within and across departments reduces efficiency and risks coherence.
    - Example: all three units of the DAMP monitor international developments; the DIE independently downloads data on inflation and external variables already covered by DAMP.
  - No unified database across departments; fragmented approach can lead to discrepancies.
  - Inputs are exchanged primarily via email, increasing risk of errors and inconsistencies.
  - Categorization and analysis of key variables are not harmonized (example: different measures of core inflation used by DAMP and DIE, producing conflicting signals).

### Operational recommendations
- Reorganize DAMP to reduce overlap and shift emphasis away from excessive focus on international developments:
  - Proposed structure into four units (secciones):
    - Unit 1 (International): Responsible for analyzing international variables.
    - Unit 2 (Inflation): Focused on detailed local inflation analysis and short-term inflation forecasts.
    - Unit 3 (Macro analysis): In collaboration with the DEM, produce a nowcast and short-term growth forecasts and organize the macroeconomic narrative for the EC, the MB, and the MPR.
    - Unit 4 (Forecasting process management and evaluation): Oversee the forecasting process, set meeting agendas, keep minutes, organize written reports, ensure MPR coherence, build a shared database for FT members, and conduct an annual internal evaluation of the forecast and process.
- FT must reach consensus on variables to be analyzed for the forecast and the MPR:
  - DIE must work with inputs from DAMP and DEM.
  - Agree on a core inflation measure for use in models, policy documents, and communications; treat alternative measures as robustness checks or complementary inputs.
- DIE should cease constructing its own data for FPAS purposes intended for the FPAS:
  - Constructing separate data consumes time, duplicates DAMP functions, and risks inconsistencies.
  - DIE may explore alternative data for research, but adoption for FPAS should be coordinated with DAMP.
- Establish a centralized database and discontinue sending data via email:
  - Database must include appropriate permissions, ensure integrity and confidentiality, and facilitate timely data submission.
  - Database should record DAMP, DEM, and DIE forecasts and conditioning variables to enable future replication and evaluation of the forecasting process.

*IMF Technical Assistance Report — excerpted content*

---


_Source: https://www.imf.org/-/media/files/publications/tar/2025/english/tarea2025100-source-pdf.pdf_